Monitoring Technologies for HVDC Transmission Lines

Abstract

HVDC transmission systems are becoming more prevalent because of the advantages they offer. They are more efficient and environmentally friendly and are becoming preferred in distributed power generation. The reliable operation of HVDC transmission lines requires distributed, online monitoring, which is not as well-developed as those in an HVAC system. The advancement of HVDC systems will require online monitoring that provides information on the operating and environmental conditions, mechanical stress on the conductors and the structure, vegetation clearance, and security of the system. This perspective paper provides a thorough overview of the state-of-the-art technologies that are applicable to the monitoring of HVDC transmission lines. The challenges and future direction in the development of sensors for HVDC applications are highlighted. One of the key challenges unique to HVDC transmission lines is energy harvesting from the transmission line conductors to provide power for the monitoring equipment. This paper reviews the potential technologies for energy harvesting from HVAC transmission lines and their suitability for employment in HVDC transmission lines.

1. Introduction

Traditional benefits and drawbacks of HVDC transmission compared to HVAC are well known. HVDC transmission lines (see Figure 1) can operate more efficiently than HVAC by having reduced line losses and increased current capacity using the same transmission line conductors. Limitations to the large expansion of HVDC transmission are historically due to the high cost of the HVDC converter station equipment, and some technological limitations [1]. These limitations have been mainly attributed to the lack of HVDC circuit breaker technology and the converter technology needed for multi-terminal HVDC networks. With modern developments in HVDC circuit breakers [2], and higher-capacity IGBT-based VSC converter technology [3], many of the obstacles to the development of an HVDC grid are being alleviated.

Decarbonization in the electrical power grid is a key driver for the expansion of HVDC transmission [4]. Societal and environmental pressures to reduce carbon emissions will generate considerable incentives to move toward HVDC transmission because it serves to reduce transmission losses and maximize the integration of renewable energy sources. Globally, a growing number of large HVDC and UHVDC transmission lines are currently in development or planned for future development. HVDC transmission will occupy a greater portion of the global bulk power grids. Additional HVDC transmission may be obtained from converting existing HVAC lines to HVDC [5]. Such conversion provides economic benefits from increased current capacity in HVDC over HVAC. It is estimated that such conversion achieves 3–5 times greater capacity.

Implementation of an HVDC grid would benefit renewable energy integration in several ways. For offshore wind generation applications, where it is necessary to transmit HVDC power via underwater cables, an HVDC grid would be more efficient by removing the need to convert the power to HVAC immediately at the onshore terminal. Another example where an HVDC grid is beneficial for renewable energy generation is with remote locations for hydro, wind, and photovoltaic power generation. Because these sites are commonly in remote geographical regions, and far from major urban centers, HVDC transmission is desirable because it imposes less disruption to nature as HVDC transmission lines require smaller transmission towers and corridors than an HVAC system of the same capacity.

To assure reliability and fully optimize renewable energy in the bulk transmission system, digitalization, and online monitoring is expected to play a key role in HVDC transmission. Online monitoring has become more common for HVAC transmission in recent years. Online monitoring improves reliability by providing early warning of fault risks for transmission lines. Environmental factors such as fire and smoke, or ice and snow buildup can be monitored enabling the risk to be managed by remediation or arranging contingency in the network. Monitoring also provides optimization for renewable energy on the power grid by enabling controlled overloading. HVAC line loading limits, traditionally set based on seasonal maximum temperatures, are changing to focus on real-time thermal ratings collected from online monitoring data. This concept of dynamic line rating (DLR) has been integrated into a US Federal Regulatory Order 881 [6]. One of the goals of this is to maximize the utilization of renewable energy sources on the power grid. If transmission circuits may be overloaded above traditional seasonal limits in a controlled manner through monitoring, then the system may ingest more renewable power when required.

With the anticipated expansion of HVDC transmission and the development of an HVDC grid, digitalization will necessitate new online monitoring technologies for HVDC transmission lines (see Figure 2). Similar to the HVAC application, HVDC transmission line monitoring will be used to ensure the reliability of critical HVDC transmission infrastructure. In a multi-terminal HVDC grid, controlled overloading of transmission lines will also be desired, and online monitoring for the conductor temperature and sag will be needed to estimate capacity for DLR.

In this perspectives paper, potential technologies for online monitoring of HVDC overhead transmission lines are considered. The motivation for this work is to examine the future need for online monitoring in the future bulk power HVDC transmission system and to consider the possible range of technology solutions. To date, limited research has been conducted for online monitoring of HVDC overhead transmission lines. The industry trends toward decarbonization and digitalization will have a significant impact on HVDC transmission since HVDC is a critical element in reducing carbon emissions. The operation of an HVDC grid will require online monitoring to assure reliability and control to optimize the utilization of clean energy resources. For the reliability of HVDC transmission, operating conditions, environmental conditions, and mechanical stresses are analyzed based on the risk they pose to HVDC transmission lines and the potential technologies available for online monitoring of these parameters.

In some instances, HVDC transmission online monitoring will be very similar to HVAC implementations. For example, monitoring of environmental conditions such as conductor sag, and temperature may be implemented very similarly to those for HVAC lines. Conversely, monitoring parameters, such as line monitoring for voltage and current, electromagnetic fields, or UV and air corona scanning may differ from HVAC implementations. For example, industry limits for visual air corona and radio interference voltage limits (RIV) for transmission line hardware are based on HVAC transmission [7,8]. However, HVDC air corona may have much lower repetition rates, thus reducing intensity and RIV. Furthermore, accumulation of space charges around energized conductors and hardware is unlikely to occur in HVAC transmission but does occur for HVDC. Other key differences in requirements for HVDC transmission online monitoring versus HVAC are related to the breakdown mechanisms in HVDC. Due to HVDC transmission lines air corona producing accumulated concentrations of space charge in the vicinity of the high voltage conductors, some HVDC transmission lines incur higher than expected flashovers by phenomena not well understood. These flashover events are referred to as unexplained flashovers. There is some evidence to suggest that the rate of unexplained flashovers is increasing over time. Figure 3 shows the growing number of unexplained flashovers recorded by Manitoba Hydro.

In this paper, online monitoring of HVDC parameters: electrical corona and environmental factors is proposed to better understand unexplained flashovers in the HVDC transmission system. Another key difference in HVDC transmission line monitoring relates to power harvesting. Unlike HVAC transmission where energy for monitoring devices may be harvested by inductive coupling to the transmission line, solutions for HVDC will require other multiple sources to power online monitoring. A combination of solutions from solar, wind, vibration, and wireless power transformers are considered in this paper. Powerline harvesting with wireless power transfer is an area in need of further research to enable the deployment of HVDC monitoring. This paper also provides a detailed explanation of the advantages and disadvantages of each technology when adopted for an HVDC system.

2. Operating Conditions

The transmission line operating conditions refer to a range of parameters and external influences, which affect the performance and reliability of the transmission line in transmitting electrical power. The task of a group of sensing devices installed on HVDC transmission line systems is to record and monitor the operating conditions of the transmission line. The range of functions for these sensors includes line voltage and current, conductor temperature, physical security and investigation, ultraviolet (UV) imaging, infrared (IR) scanning, and monitoring of leakage currents through the line insulators. HVDC transmission lines traverse very long distances, varying terrains and environments, which can be difficult to patrol and monitor. Therefore remote sensing techniques are of great importance to the understanding of the operation of the system. In the following section, the existing sensing technologies for the operating conditions are reviewed and discussed.

2.1. Electric Voltage and Current

Voltage and current measurement in HVDC cannot be implemented using conventional passive transformer devices used ubiquitously in HVAC transmission because HVDC power lacks a time-varying electromagnetic field on which basic instrument transformers are useful. For this reason, HVDC voltage and current sensors are more complex and for practical reasons have been mainly limited to the terminal ends of the HVDC line. Voltage and current measurement in HVDC are based on technologies such as shunt resistors, magnetic field sensors, optical CTs, or piezoelectric sensors for the measurement of current and resistive-capacitive dividers for the measurement of voltage. They all require installation in close proximity to the conductors and that could be a limiting factor [9]. However, with the future development of HVDC multi-terminal networks/grids, there will be a demand for distributed sensor nodes. It is expected in the future that hybrid optical voltage and current sensors will be utilized [10].

2.2. Conductor Temperature

For safe, reliable operation of the HVDC transmission line, the temperature of the conductor is a key parameter for the calculation of conductor sag and the resulting clearances [11]. Furthermore, dynamic line rating (DLR), which is a method of controlled overloading of the transmission line, has widely been used in HVAC transmission and employs the conductor temperature as input data. Extension of DLR to HVDC multi-terminal networks/grids can maximize the transmission line power transfer and operate the conductors above their available ampacity in the cold season.

Methods for sensing the conductor temperature include the surface mount thermistor (SMT) direct temperature sensors, power donuts, distributed temperature sensors with optical fibers, surface acoustic wave sensors, and conductor temperature modeling [11].

2.3. Security and Investigation

Video cameras can be used for in-field monitoring of transmission lines for security and investigation. Video and still images can be utilized for obtaining information in remote locations, of HVDC transmission lines, which may be difficult to access. Examples of issues that can be investigated with video cameras are conductor icing, vegetation, smoke and fire, unknown system operating issues such as faults, bird, and wild life issues, criminal activities, conductor galloping, tower hardware issues, and high water levels or icing (Figure 4) around infrastructure [12].

HVDC transmission lines are normally longer than HVAC transmission lines, spanning over thousands of kilometers, and therefore more difficult to inspect and patrol, and some regions of the line may not have cellular signals available. As an example, cameras were utilized in 2021 in very remote areas of Manitoba Hydro’s 500 kV HVDC bipolar lines (BPI and BPII) to investigate unexplained flashovers [13,14] which occur every summer, mainly in the warmest part of the day and only on the negative poles. Eighty to 100% of the flashovers on BPI and BPII are of the unexplained type. As many as 46 unexplained flashovers occurred in one summer period. The unexplained flashovers only occur from May to September, peaking in July and August, and mainly occur after 11 a.m., predominately in the afternoon, but ending by midnight.

The camera systems were powered by solar panels with battery storage. Data can be retrieved by satellite communications as there is no cellphone signal in this area. An event that was captured on 21 July 2021 in the midspan on a negative pole at 4:08 p.m. on a sunny afternoon is shown in Figure 5. LiDAR analysis indicated the flashover occurred over a distance greater than understood by present standards and requirements. Environmental sensors demonstrated that the flashover occurred at a temperature maximum, and wind and humidity minimum. The HVDC fault locators, based on the time difference of the incoming wave fronts to the stations at the ends of the line, have been proven to be able to locate faults on an HVDC transmission line to within a span. Figure 6 displays the equipment installed on the towers with cameras and LiDAR, solar panels, a weather station, and a control box containing the batteries and computer hardware.

The unexplained flashovers result in the inability to obtain safety hold-offs (block of restarts) to perform live work for up to 6 months of the year. The cost of a planned outage on a typical HVDC transmission pole can be as costly as $400 k per day depending on the season and generation conditions. A forced and unplanned outage resulting from an event can be expected to be significantly more costly when purchasing emergency power in the short-term market. HVDC transmission lines tend to carry large amounts of power and are critical to the provider, and as such the cost of additional monitoring systems is economically justifiable. For example, the cost of adding one unit of monitoring system (shown in Figure 6) for the unexplained flashover camera/sensor project was $45 k. A complete understanding and mitigation of the unexplained flashovers remains a high priority for future improvements.

2.4. Leakage Current

Pollution on HVDC insulators can be more critical than on HVAC insulators [15]. Although the insulator length for HVAC is dictated by switching and lightning transient overvoltages, for HVDC, the length of the line insulators is determined by the pollution severity. Monitoring of the leakage current can be utilized to determine the condition of energized transmission insulators.

Although leakage current monitoring on HVAC transmission lines has been well documented in the literature, HVDC leakage monitoring is a more recent topic. Not all AC methods can be utilized for HVDC as there is no alternating current. One method that works for both HVDC and HVAC systems is the use of a shunt resistor which can be installed between the insulator end and the tower. The voltage can be measured across the resistor by a measurement device and/or data logger and translated to a current using Ohm’s law. The shunt resistor method is not ideal for the field as it requires the installation of an additional insulator for the setup [16].

Other methods use a fluxgate magnetometer and a DC leakage current sensor. Fluxgate magnetometers can be utilized to measure static and low-frequency magnetic fields. They output a signal which is proportional to the external magnetic field. A very good correlation was found when comparing the shunt resistor method with the fluxgate magnetometer. A DC current sensor was also found to give good results, but its utilization is more intrusive than the fluxgate magnetometer method [17]. A drawback of the fluxgate magnetometer is that it can be impacted by external magnetic fields. This was resolved in [18] as a differential magnetometer method is utilized to eliminate the background magnetic fields.

2.5. Electric Field and Corona Discharge

High electric field stress on the HVDC transmission lines or their associated line hardware and insulators can lead to corona discharges. Corona, an electrical discharge, is caused by an electric field that is strong enough to ionize the gaseous medium surrounding the conductor at a lower voltage than the breakdown voltage of the gas insulation [19,20]. Excessive corona discharges present several issues for the transmission line operation. The first issue is that corona discharges induce power losses on the line. The second is that it causes a nuisance to the public due to audible noise as well as producing electromagnetic interference with radio communications; called Radio Interference Voltage (RIV) [21]. Corona discharges are also known to cause degradation of transmission line insulators by eroding the cement and pin regions of glass and porcelain insulators, or the seals and silicone rubber sheds of polymeric insulators [21]. This damage can pose serious risks of failure on the HVDC transmission line. Although the HVDC transmission line conductors, insulators, and line hardware designs are usually tested in high voltage laboratories to verify their electrical performance in terms of corona discharge [7], operating and environmental conditions after installation in service, e.g., limited clearances to the ground due to conductor sag, and/or overgrown vegetation, can lead to corona discharges. The common method for online monitoring of corona discharges on transmission lines is UV monitoring (see Section 2.6). For monitoring of RIV, instruments designed for ground-level measurements of the radio interference are common [8].

For HVDC transmission specifically, corona discharges can lead to unexplained flashovers due to a buildup of space charges in and around the transmission line. This mechanism is not observed for HVAC lines since the alternating field in HVAC does not produce the same density of space charges. Unexplained flashovers are a rare occurrence on HVDC transmission lines, and their mechanism is not fully understood. Literature [22] has suggested a negative space charge phenomenon, termed a fast flashover, where a positive streamer propagates from the ground due to negative space charge from an HVDC electrode at lower humidity than for pollution-type flashovers, which involved fiberglass-reinforced plastic poles in the air gap. This was mitigated by the use of polymer sheds or inhibitor electrodes [14,22]. It has been shown [23] that lower humidity results in more space charge accumulation on insulator surfaces while increasing humidity decreased the flashover voltage.

With improved online monitoring tools to measure corona, relative humidity, and temperature, it may be possible to correlate these factors and predict where and when unexplained flashovers are more likely to occur.

2.6. Infrared and Ultraviolet Scanning

Infrared (IR) and ultraviolet (UV) scanning is an important tool for the inspection and detection of anomalies on HVDC transmission lines. When air corona occurs on transmission lines or their associated hardware and insulators, the discharges emit ultraviolet light. This emission can be detected by ultraviolet imaging devices [12]. Ultraviolet scanning can detect corona issues at insulators and line hardware by sensing the emitted light in the UV range which is overlayed on an image. Other methods for detecting corona and partial discharges include radio disturbance meters, UHF sensors, HF/VHF sensors, capacitive/inductive methods, resistive methods for leakage current measurements, and acoustic and optical sensors [24]. Figure 7a shows an example of corona detection acquired by a UV camera and a hot spot detected by an IR camera is shown in Figure 7b. Both images have been acquired from a 500 kV HVDC transmission line.

Infrared scanning can be utilized for determining defects on insulator and line hardware of transmission lines or hot spots on conductors, commonly at splices or other conductor connectors. An infrared camera detects the heat emitted and converts it into an electronic signal, which is processed to develop an image.

Traditionally, these devices are used for periodic surveillance measurements rather than permanently installed online monitoring; however, some tower-mounted monitoring devices include UV corona imaging devices for permanent dedicated monitoring.

2.7. Challenges, Future Direction, and Perspective

HVDC transmission lines traverse long distances, difficult terrains, and varying climates and will require smart towers, contact, and contactless sensors, and metering to ensure grid stability, reliability, and power flow. Voltage and current measurement techniques will be important as protection and the location of faults will be required for HVDC grids [10]. Fault detection methods based on traveling wave fronts will not work for HVDC grids. DLR will be further utilized as HVDC grids develop and other sources of energy such as wind power or photovoltaics penetrate the grid. HVDC transmission lines in polluted environments will require leakage monitoring or consideration in the insulator design for improved performance. Advanced image recognition and the use of camera technologies with communication capabilities will be required to investigate issues in remote locations on HVDC lines. Standards, such as those related to live working, compact HVDC tower design [25], and vegetation/ground clearances, will need to further evolve to consider the differences between HVAC and HVDC that have yet to be investigated such as the impacts of space charge [23,26,27,28]. Development of power harvesting techniques from HVDC lines is required to power the required sensors and communications for operating condition monitoring. Power harvesting will allow the sensors to be installed at the location of interest on the conductor and provide more location-specific data on the operating conditions.

3. Mechanical Stress

Monitoring the mechanical stress and tension on transmission lines infrastructure due to thermal expansion/contraction, vibration, or galloping of the conductors plays an important role in improving the reliability and optimum operation of a power transmission system. In the following, the established and emerging methodologies for monitoring the causes of mechanical stress are reviewed and their advantages and disadvantages are summarized in

Table 1. A comparison of proposed mechanical stress monitoring techniques.

Monitoring Parameter	Monitoring Technique	Advantages	Disadvantages
Sag	EPRI video sagometer	Low power consumption of 5W. Energized installation.	Requirement of external power supply such as solar power. Requirement of a system of diode lasers / LED illuminators to illuminate the target at night.
	NIR optical sag monitoring system	Capability to operate even without a target. Reduction of errors due to tilting of the mounting structure. Utilization of power-saving techniques.	Requirement of external power supply such as solar power. High power consumption of 50 W.
Vibration	Tribo-electric nanogenerator-based monitoring system	Self-powered. Low cost. High sensitivity of 23.5 V/mm.	Less durability due to the risk of mechanical failures during the operation. De-energized installation due to distributed monitoring.
	Fiber-optic acceleration sensor-based monitoring system	Anti-electromagnetic interference. High precision. Low power requirement. Easy installation.	High cost. Requirement of external power supply.
Galloping	Fiber-Bragg grating tension sensor-based monitoring system	Anti-electromagnetic interference. Improved accuracy. Distributed monitoring.	Less suitable for harsh weather conditions.
	Acceleration and displacement sensor-based monitoring system	Wireless communication. Simple working principle.	Requirement of many sensor nodes. High cost.
Tension	Nexan’s CAT-1 tension monitoring system	High reliability. Less vulnerable to overvoltages.	Requirement of external power supply such as solar power. Less accurate measurements at line loading below 20%.
	Fiber-Bragg grating fitting sensor	High measurement range of 60 kN. Anti-electromagnetic interference.	Requirement of external power supply. Not suitable for harsh weather conditions.

.

3.1. Sag

Transmission line conductors sag during their operation due to the increase in the conductor temperature. The loading of HVDC transmission lines is based on static line ratings, assuming critical environmental conditions, such that the safe clearance limits are not exceeded. There are a few main sag monitoring techniques that include measuring vibration, conductor inclination, target capturing, and the stopwatch method. Although there are ample real-time sag monitoring devices for HVAC transmission lines, their deployment in HVDC transmission lines is challenging because power harvesting from local constant magnetic and electric fields is not currently practical/feasible on HVDC transmission lines. For example, the application of techniques discussed in [29,30] are limited to HVAC transmission lines due to their power supply requirements that is typically harvested from the HVAC transmission lines.

To overcome the power harvesting challenge, equipment is installed on the tower and powered by solar energy. An EPRI video sagometer [31] is an example of such equipment. It employs imaging technology for line sag monitoring that captures a target on the line whose position changes as a result of line sagging. In an improvement of this technology, a new optical system has been introduced which takes advantage of a near-infrared (NIR) camera and image processing techniques. The main improvement of this device is its ability to operate in the absence of reference targets [32]. As another HVDC sag monitoring technique, the stopwatch method can be mentioned. It is based on the return wave method [33] which measures the reflection time of a mechanical wave in determining the sagging.

3.2. Vibration

Aeolian vibrations are low-magnitude, high-frequency oscillations on the power lines and optical ground wires (OPGWs) induced by wind. Typically, the frequencies of these vibrations are in the approximate range of 3–100 Hz, and the amplitudes can be reached up to 2–3 times of conductor diameter for low frequencies [34]. They cause structural failures in conductor strands or clamps, which are substantial economic losses to the utilities [34,35]. Figure 8 shows an example of damage to a conductor at the clamp due to vibration. Real-time monitoring is useful in recognizing the need for preventive maintenance actions such as the installation of vibration dampers, etc. Tribo-electric nanogenerator-based vibration monitoring device is a self-powered sensor that generates a real-time voltage signal that is proportional to the vibrations [36]. A fiber-optic acceleration sensor-based monitoring system that measures the alternating bending amplitude [37] of overhead transmission lines has been proposed in [38]. This technique can be employed on both HVAC and HVDC transmission lines. Hydro Québec, a Canadian utility, employs a vibration monitoring device called “PAVICA” which is based on strain gauges that produce signals proportional to the vibrations on the conductors [39].

3.3. Galloping

Overhead conductor galloping is a type of vibration, but it differs from aeolian vibrations by having low frequency and high amplitude movements. Generally, these kinds of vibrations are in the approximate frequency range of (0.1–3) Hz and can reach up to 300 times greater than the conductor diameter [40]. This phenomenon causes flashovers, outages as well as infrastructure damage, hence its monitoring is of high need to prevent such issues. An experimentally proven galloping monitoring system based on a fiber-Bragg grating tension sensor that transmits the tension changes of a phase conductor to a remote interrogator was proposed in [41]. Employing an interrogator that can be up to 10 km away, this device can partially overcome the challenge of supplying on-site power from HVDC transmission lines. A real-time monitoring network of acceleration and displacement sensors to monitor galloping amplitude and frequency was proposed in [40]. The sensors are mounted on the conductors and connected to a monitoring station on the tower at which the processing of the data and information is communicated through wireless communication [40].

3.4. Tension

Mechanical stress in a conductor in an overhead transmission line is referred to as conductor tension. A change in the overhead transmission line sag, due to thermal expansion/contraction, changes the conductor tension. The loading effects of wind, galloping, and icing also affect the line tension. Tension can be used to calculate the conductor sag and the average temperature of conductors of a transmission line based on the relationship between sag and tension [42]. Real-time monitoring of conductor tension is important to estimate the sag and the average conductor temperature to operate transmission lines within their maximum capacity without violating the safety clearances.

Most of the utilities use Nexans’ CAT-1 tension monitoring system that deploys conventional load cells with strain gauges installed between dead-end insulators and the tower structures to monitor conductor tension [43]. This device can continuously monitor the tension of the conductor and produce a proportional electrical signal. The monitoring system is powered by solar power [44]. As a novel monitoring technique, a fiber-Bragg grating (FBG) fitting sensor that has experimentally been tested in a laboratory and the field, has been proposed in [45]. This has some advantages such as higher accuracy, less prone to electromagnetic interference, and an improved dynamic range.

3.5. Challenges, Future Direction, and Perspective

Dynamic line rating (DLR), which is determined by the environmental and operational conditions of HVDC transmission lines is necessary to identify the optimum loading capabilities of HVDC transmission lines without violating clearance limits. Real-time conductor sag is one of the most important inputs in the calculations of the DLR of a transmission line. Currently established sag monitoring methods are highly applicable to HVAC overhead lines and cannot be practically implemented due to the limitations of energy harvesting from HVDC transmission lines. Out of a few technically capable methods, image (either visible or NIR) monitoring techniques are more suitable for the real-time sag monitoring of HVDC overhead lines. The sagging of conductors in an overhead HVDC transmission line can be determined using conductor tension. Continuous monitoring of conductor tension also has a significant value in the determination of DLR. There is a limited number of monitoring techniques currently in use that are dedicated to conductor tension monitoring.

The supply of power to the available mechanical sensors and methods is an important issue regarding HVDC transmission lines due to the lack of existing power harvesting techniques. Active power transfer techniques will be useful for transmitting harvested HVDC energy to mechanical sensors on the structures.

4. Environmental Conditions

In consideration of the developments of HVDC grids, sensors are required on HVDC transmission lines to determine how its loading capacity is impacted by ambient temperature, wind, precipitation, humidity, and solar radiation for the implementation of DLR technology. Additionally, sensors for environmental conditions such as icing, smoke, and fire are required to protect the infrastructure, its operation and ensure its reliability. In the following, the existing technologies for environmental conditions are reviewed and discussed.

4.1. Ambient Temperature

Federal Energy Regulatory Commission (FERC) Order 881 will require ambient-adjusted ratings (AARs) on the transmission lines over which public utility transmission providers supply transmission service [6]. All requirements of FERC 881 must be implemented by no later than 12 July 2025. AAR applies to a time period of not greater than one hour and reflects an up-to-date forecast of ambient air temperature across the time period to which the rating applies. Ambient-adjusted ratings are transmission line ratings that are more dynamic than static or seasonal ratings that are traditionally more common [46]. High ambient temperatures combined with high conductor loading cause the sagging of conductors and loss of clearance and could ultimately lead to damage and aging of the conductor. Ambient temperature is one of the factors impacting the thermal stress of conductors, the others being wind speed/direction and solar radiation [11]. The reduced clearance to the ground resulting from ambient temperature can increase the field stress on HVDC transmission lines and result in increased space charge generation and the possibility of flashovers.

4.2. Wind Speed and Direction

Common methods to measure wind speed and direction are rotating anemometers with wind vanes of a cup or propeller-based design. Other methods of measuring wind speed utilize ultrasonic anemometers, pressure-based sensors such as the Pitot tube, or those based on the cooling rate of a hot wire. More recent methods to measure wind speed utilize Doppler radar or LiDAR [47].

The wind speed is an important factor in determining and forecasting the loading capability of transmission lines as the wind has a cooling effect on the conductor [11]. The actual wind speed is considered in DLR methods that utilize ambient conditions. However, in AAR methods a low constant value of wind is typically assumed resulting in less than optimum loading capacities [46]. Wind speed and direction can also have a direct impact on the effects of ice and snow on the HVDC transmission from mechanical stresses, cause fires and smoke around the infrastructure, and may aid in the dispersal of space charge.

4.3. Precipitation

Precipitation can also influence cooling conductors and increase their transfer capability [11]. This effect is useful to be considered by the utilization of DLRs with precipitation sensors. The sensors available for sensing precipitation include the mechanical tipping bucket method to quantify moisture. Another sensor employs the impact precipitation sensor which can detect the impact of individual rain drops. Acoustic sensors have also been demonstrated to detect rainfall. Other sensors can determine more details of the water droplets such as their dimensions, speed, volume, and number, which include optical sensors, piezoelectric sensors, and Doppler radar sensors [48]. Precipitation sensors can also provide the utilities with an understanding of the fire or flooding risk near HVDC infrastructure. Precipitation can have an overall impact on the washing of polluted insulators or increases in leakage currents leading to dry band arcing. Precipitation also may lead to more conductive vegetation and increases the risk of space charge-related flashover from HVDC transmission lines.

4.4. Humidity

Humidity is another factor that can impact the DLR of a transmission line. The heat balance is impacted by humidity and affects the power transfer capability [49]. The humidity data are also important for acquiring knowledge of for proper simulation of outdoor environments in a laboratory setting. There are various hygrometer sensors to measure humidity. The psychrometer technique of humidity measurement utilizes a dry and wet thermometer bulb. They are very accurate and useful for calibrating other humidity sensors. Other sensors available for humidity measurement include fiber-optic sensors that sense a refractive index change from humidity, resistive sensors which sense changes in conductivity caused by the adsorption of moisture, capacitive sensors which sense changes in the dielectric constant, surface acoustic wave sensors to sense the relative change in the phase velocity, piezoresistive humidity sensors, and the magnetoelastic humidity sensor based on changes in the resonant frequency [50]. Humidity can also have a direct impact on space charge and HVDC transmission-related flashovers [14,23].

4.5. Solar Radiation

Solar radiation can increase the temperature of a conductor as more heat is gained, and this worsens when there is limited wind to cool the surface [11]. Higher solar radiation without wind will result in decreased transfer capability. DLR systems can be based on the environmental condition measurements such as ambient temperature, wind speed, precipitation, and solar radiation to calculate the conductor temperature. Thermally based methods utilize the actual conductor temperature, while mechanically based methods utilize sag, clearance, and tension to estimate the conductor temperature [51,52]. The typical method to measure the solar radiation at a horizontal surface is by way of the pyranometer sensor which measures the total global radiation. Global radiation is the total of direct and diffuse radiation. A pyrheliometer sensor measures only the direct radiation by way of using a Sun-following tracker [53].

4.6. Smoke and Fire

Sensors for detecting smoke and fire can be a useful asset for HVDC transmission lines as early detection can avoid damage to structures and conductors, avoid outages from flashovers and avoid risk to the public and workers from step and touch potentials. Fire and smoke can decrease the electrical strength of air due to high temperatures decreasing the air density, production of electrical charges, and presence of conductive particles in the smoke [54,55]. Fires can also cause increases in conductor temperature which leads to an increase in sag and reduced clearances [11]. There is also the possibility that HVDC transmission lines themselves could ignite fires beneath them due to the buildup of space charge beneath the pole resulting in a breakdown of the air gap. Fires under HVDC lines have been shown to increase radio interference levels and reduce the corona inception voltage [56].

Large particles from a fire can cause conductive paths between the fire and conductor and greatly decrease the air gap electrical strength. There was a noted reduction in the breakdown of DC voltage under negative polarity of up to 29% when considering particles [57]. The DC breakdown voltage for negative polarity has also been observed to be higher than the positive polarity for cases considering fire only [58]. It was demonstrated that under smoke conditions and low DC voltage, there is a significant increase in the conductance of the air and resulting leakage current between two parallel plates [59]. The flashover of an air gap due to fire and smoke is most likely due to the conductive nature of the flame or the particles in the smoke, and not just smoke alone [58,60,61].

There are various sensors available and emerging for the detection of fire and smoke. Heat sensing is one method where the thermal energy is detected by a heating element or an IR camera. Heat sensors can operate based on a fixed temperature, rate of rise, or rate of compensation. The fixed temperature sensors can be of a fusible-element, distributed, or bi-metal type. Distributed heat sensors can be of the electrical, optical, or sheathed thermocouple type. Gas sensors are another method to detect a fire and detect the presence of gases by measuring a sensor output change. Gas sensors include semiconductors, catalytic beads, photoionization, infrared, electrochemical, optical, acoustic, gas chromatograph, and calorimetric technologies. Flames can be sensed by nonvisual or visual techniques. Sensors for nonvisual methods include UV and IR. Visual techniques are IR cameras and visual cameras. Image processing and deep learning can be applied to detect fires. Smoke sensing also includes nonvisual methods such as photoelectric and ionization-based sensors, or a combination of both. Visual methods to detect smoke by way of cameras can also be utilized. Other alternative sensors for sensing fires include ultrasonic and microwave techniques [62].

4.7. Ice and Snow Buildup

Ice and snow buildup on transmission lines can impose damaging mechanical stress on the transmission line conductors, hardware, insulators, and transmission towers.

The presence of ice and snow on transmission line conductors can enhance aeolian vibrations, galloping, bundle rolling, and ice shedding. Figure 9 shows icing on all aspects of the HVDC transmission line including the conductors, electrodes, overhead ground wire, insulators, and structures following an icing event. The above issues will be further enhanced when wind is present and can lead to the failure of the conductors or structure.

Ice and snow buildup can result in flashovers with outages [63], dynamic loads and conductor damage, high tension, sag, impacts on summer power flow transfer capabilities (due to conductor deformity), structure fatigue and damage, and hardware wear and tear and fatigue [64]. Ice and snow buildup on transmission line insulators can be of particular concern, leading to a less reliable transmission system from flashovers [63].

System outages and damage due to ice and snow buildup could be prevented by active monitoring of assets and timely response. Monitoring for icing can be completed by utilizing environmental monitoring and knowing the present wind conditions. This can also include the use of test span monitoring. Sensors for ice detection include capacitive techniques, ultrasound methods, resonance sensing, microwave energy, impedance measurements, and light sensing [65]. Other methods include optical fiber sensing methods utilizing the OPGW wire but not all overhead wires have such fiber optics [66,67,68]. Monitoring ice buildup can also be achieved by monitoring conductor sag, tension, and temperature [69].

The most direct method of verifying the accumulation of snow and ice buildup on transmission line conductors is using tower-mounted cameras and computer vision to automatically detect the buildup. Icing on insulators can also be undertaken with camera image recognition techniques [70,71]. Other techniques include thermal or UV monitoring or using leakage current measurements [63].

Sensors such as those utilized for dynamic line rating can be utilized for icing detection as parameters such as tension, sag and clearance at low temperatures can determine the risk to the transmission line. Manitoba Hydro presently has a project installed where AC power harvesting is utilized to power direct contact equipment that includes cameras, environmental sensors, and real-time information on tilt/pitch, roll, galloping, and vibration of the conductor. Figure 10 displays an AC power harvesting system for detecting icing and wind-related events. Figure 11 demonstrates the camera of such devices capturing icing events where Figure 11a illustrates corona with icing in low light conditions and Figure 11b shows the icing in day light conditions. No such power harvesting solution exists for HVDC at this time.

Shield wires of overhead transmission lines are often the most susceptible to icing as they are the highest conductors on a transmission line. Shield wires are de-energized conductors for lightning protection, and are normally grounded at every pole or structure. The shield wires tend to be smaller and not as robust as phase conductors and can experience increased sag leading to the tripping of the transmission line after contact. Energy-harvesting techniques to power monitoring equipment are normally focused on the phase conductor as there is a source of energy. For AC transmission lines there is the possibility of utilizing the inductive current in the shield wires or the capacitive coupling to the wire for providing a power supply to monitoring equipment [72]. Since there is no inductive or capacitive coupling for steady-state HVDC, this furthers the need for active power transfer methods and HVDC power harvesting techniques.

4.8. Challenges, Future Direction, and Perspective

Weather data monitoring services are not always available along long and often remote HVDC transmission lines in diverse climatic conditions. Distributed ambient sensors will be required, especially considering future HVDC grids. Environmental sensors will play a critical role in the DLR of future HVDC transmission grids. With climate change, more fires and smoke, as well as icing events can be expected for HVDC transmission lines and sensors will be required to protect the infrastructure and the reliability of the system. At this time, limited power harvesting methods exist for HVDC to utilize for DLR or other required environmental sensors. Since an overhead ground wire can be the most susceptible to icing events, power harvesting combined with active power transfer is required for HVDC transmission. Further research is required to fully understand the impact of smoke and particulate density near HVDC transmission lines, as well as to ensure that HVDC transmission lines are not the cause of wildfires. Environmental sensors will provide critical data as these variables can have a direct impact on the space charge effects, unique to HVDC transmission, and the lines’ overall performance. Power harvesting, combined with active power transfer will allow the sensors to be installed at the location of interest on the conductor and provide more location-specific data on the environmental conditions.

5. Vegetation and Ground Clearance

Growing vegetation is a natural hazard to the safety and reliability of overhead transmission lines. It is essential to regulate and manage vegetation growth effectively to reduce the possibility of power outages. The detection of vegetation encroachment has been made possible by a variety of technologies, including airborne photogrammetry (AP), synthetic aperture radar (SAR), and light detection and ranging (LiDAR) [73]. Monitoring methods based on visual inspections are labor-intensive and have limited detection rates because of the constrained capacity of staff to examine all types of clearance anomaly problems. Using remote sensing technologies to collect data from cameras or scanners has improved the detection rate but has not resulted in a significant saving in maintenance costs as the data analysis has remained a manual process. The objective of this section is to present the state-of-the-art of vegetation and ground clearance monitoring techniques in the research literature.

5.1. Synthetic Aperture Radar

Synthetic aperture radar (SAR) is a type of active sensor, installed on satellites, helicopters, etc., that collects the reflection of electromagnetic waves off the surfaces in the microwave range spectrum and creates 2D or 3D images of the illuminated area [74]. Microwaves can pass through clouds, making the SAR method useful for monitoring applications since it allows for image acquisition in all climatic conditions [12,75]. However, it is not always possible to obtain SAR images from the intended area because of the limited operational time of satellites by space agencies, and capturing very high-resolution images is expensive for utilities to monitor vegetation, making the spatial resolution to be limited. In addition, various factors such as speckle noise, foreshortening, or shadows can affect the quality and clarity of SAR images.

5.2. Airborne Laser Scanning

Airborne Laser Scanning (ALS) is a remote sensing technique that uses LiDAR technology mounted on an aircraft to acquire 3D point clouds of objects. A heliborne position and orientation system (POS) that uses a combination of global positioning system (GPS), and inertial measurement unit technology can be used to create a large area three-dimensional data acquisition system based on LiDAR technology. This sensing method is currently used in many countries to manage vegetation near power lines. A separation distance measuring system based on LiDAR is used in Japan to monitor vegetation located beneath the transmission lines [12]. This system uses an ALS unit to accurately determine the coordinates of the ground, vegetation, and power conductors, and then the distance between lines and trees is calculated based on the collected information. Most of the ALS-related research has focused on developing automated techniques for classifying and reconstructing powerline elements, specifically conductors.

In [76], a method for reconstructing power lines has been proposed that uses the distribution properties of powerline groups between two adjacent pylons as well as contextual information from related pylon objects rather than relying solely on the local shape of a single span. To detect power lines in a forest environment using the ALS point clouds, various image features of the 3D image dataset were utilized to classify and map the 3D images into power lines and non-power lines binary 2D images. The proposed statistical analysis involves considering a set of features, such as height, density, and histogram thresholds and a classification accuracy of 93% has been reported [77].

Ground-based laser scanning is also an available technology that can be mounted on camera stands as terrestrial laser scanning (TLS) units or vehicles to collect data closer to targets. The main advantage of stationary scanning systems is that more detailed and high-density images can be obtained because the sensor is perfectly located with respect to the target and record images from various angles. A method based on a perceptual grouping framework has been proposed to reconstruct power lines from ALS data [78]. Powerline model reconstruction is achieved by detecting powerline candidate points from 3D data through applying Hough transform and voxel-based classification, showing a 94% success rate [78]. Figure 12 displays the output of the LiDAR imagery captured from tower-mounted stationary cameras of the setup in Figure 6 to determine the conductor to vegetation distance.

5.3. Challenges, Future Direction, and Perspective:

Vegetation and ground clearance monitoring is required to be routinely performed, for HVDC transmission lines, via remote sensing techniques to prevent vegetation-induced faults. Figure 13 displays a right of way (RoW) in a remote location where the HVDC transmission lines require vegetation clearing to avoid faults and outages.

There are various technologies currently available that can be used to monitor vegetation effectively. With the advent of new technologies such as multispectral imaging, it can be expected that utilities upgrade to new detection techniques to improve the reliability of the system. As the vegetation growing is a gradual process, the main challenge is to analyze and process large datasets generated by the advanced sensors. As mentioned before, manually processing the dataset is a tedious task, and automatic methods are required to map the power lines and the vegetation around the conductors. Hybrid approaches that integrate various technologies, such as satellite image processing followed by detailed drone or mobile inspection of problematic powerline corridors, are the future trend. Processing methods based on artificial intelligence and machine learning techniques are showing great capabilities in analyzing large datasets and can be of interest in offering high-quality analyses for utilities.

6. Supply Power Requirements for Monitoring Devices

Due to the scalability, flexibility, and reliability of wireless power networks (WSN), an attractive solution is to utilize distributed wireless sensor nodes in a power network to acquire real-time information for critical power components such as transmission lines. In this approach, power grid operators can take more proactive action to changing conditions and ensure that the grid operating parameters are maintained within their limit values. However, traditional methods of providing power supplies for monitoring devices in field conditions are often ineffective, making it a limiting factor in the widespread use of WSNs. Overhead HVDC transmission lines are often located in remote areas or difficult terrains, meaning that providing a low-voltage power source may be challenging. Although large amounts of energy flow through the transmission line system, it is not easy to power up monitoring sensors directly from the grid. Energy harvesting (EH) is rapidly gaining prominence due to the remarkable progress made in low-power WSNs. Energy harvesting refers to converting ambient energy to electrical energy for use in electric circuits and devices. Typical energy sources for harvesting methods are solar, wind, microwave, magnetic field, and electric field. This section reviews energy-harvesting methods that can be efficiently implemented in transmission lines.

6.1. Energy Harvesting from Solar

Solar energy harvesting is a well-established technology that is utilized in various fields. A photovoltaic (PV) system is employed to capture the power of the Sun and convert it into usable electricity. A key challenge of relying on solar panels for energy is that weather conditions can affect the efficiency of solar cells. This can result in unstable energy output, as solar power may only be generated consistently for a limited period of time each day in different locations. For optimal utilization of incoming energy, it is crucial to enhance the performance of cells through techniques such as impedance matching and maximum power point tracking. In addition, environmental factors, such as dust, industrial contamination, ice, or snow, can also significantly degrade the performance of solar panels, leading to increased maintenance requirements. To supply a WSN using a solar energy harvester, it is necessary to utilize energy-storage solutions, such as batteries or capacitors, that require replacement [79]. In [80], a line-current transformer with energy-harvesting windings and a solar PV system was used to monitor the power transmission lines. The proposed hybrid system could offer a stable power supply of 3.3 V and 120 $\mathsf{\mu }$A on average. A hybrid-solar energy-harvesting system consisted of a solar panel to collect energy from the Sun and four tree-shaped antennas to harvest energy from RF signals. The proposed system was tested at an operating frequency of 2.45 GHz and a combined enhancement of 186% was achieved. For a generic PV module, maximum power can be expressed as [81]

$$P_{\max }(t,m)=F_{\mathrm{F}}{V}_{\mathrm{oc}}(t,m)I_{\mathrm{sc}}(t,m)N$$

(1)

where $P_{\max }(t,m)$ is the maximum power output of N PV modules connected in series, and

$$V_{\mathrm{oc}}(t,m)=V_{\mathrm{oc},\mathrm{stc}}(1+B_{\mathrm{volt}}(T_{\mathrm{cell}}(t,m)-T_{\mathrm{stc}}))$$

(2)

$$I_{\mathrm{sc}}(t,m)=I_{\mathrm{sc},\mathrm{stc}}(1+{\alpha }_{\mathrm{current}}(T_{\mathrm{cell}}(t,m)-T_{\mathrm{stc}}))\frac{\lambda (t,m)}{{\lambda }_{\mathrm{stc}}}$$

(3)

$$F_{\mathrm{F}}=\frac{V_{\mathrm{mp}}{I}_{\mathrm{mp}}}{V_{\mathrm{oc}}{I}_{\mathrm{sc}}}$$

(4)

The parameters of these equations are defined in

Table 2. Definition of the parameters used in (1) for calculating maximum power of a PV module [81].

Parameter	Description	Parameter	Description
t	Time (hour)	m	Month of the year
$P_{\max }$	PV system maximum power output	$T_{\mathrm{cell}}$	PV cell operating temperature (°C)
$F_{\mathrm{F}}$	Fill factor	$T_{\mathrm{stc}}$	PV cell temperature under standard condition
$V_{\mathrm{oc}}$	Open-circuit voltage (V) of PV cell	$I_{\mathrm{sc}}$	Short-circuit current of PV cell
$V_{\mathrm{oc},\mathrm{stc}}$	Open-circuit voltage (V) of PV cell under standard condition	$I_{\mathrm{sc},\mathrm{stc}}$	Short-circuit current of PV cell under standard condition
$B_{\mathrm{volt}}$	Voltage temperature coefficient as per the manufacturer	${\alpha }_{\mathrm{current}}$	Current temperature coefficient as per the manufacturer
$\lambda (t,m)$	Global solar irradiance ($\frac{\mathrm{W}}{{\mathrm{m}}^2}$)	${\lambda }_{\mathrm{stc}}$	Global solar irradiance ($\frac{\mathrm{W}}{{\mathrm{m}}^2}$) under standard test condition
$V_{\mathrm{mp}}$	Voltage of PV module at maximum power point	$I_{\mathrm{mp}}$	PV module current at maximum power point
N	Number of PV modules	$I_{\mathrm{oc}}$	Open-circuit current (A) of PV module

. It can be seen that a solar PV module is not a constant current or voltage source and its power output depends on ambient variables such as the intensity of sunlight, temperature, and load characteristics. Hence, scheduling plans and forecasting models should be adopted for sensor nodes to minimize the required battery capacity and maximize the utilization of available solar energy.

6.2. Energy Harvesting from Magnetic Field

The magnetic field generated by the current of transmission lines is inversely proportional to the distance from the conductor and its strength increases by the current magnitude in power lines. Hence, the magnetic field is the strongest in the vicinity of the conductors for energy harvesting, requiring the EH apparatus to be as close to the conductors as possible. Energy harvesting from the magnetic field induced by an AC is governed by Faraday’s law of induction. A typical magnetic field energy harvester is a transformer that is installed on power lines. The main disadvantage is that the output power of the harvester depends on the current magnitude of the transmission line, meaning that a minimum current flow is required to have a reasonable power output. In [82], a miniaturized permanent magnetic synchronous generator (PMSG) was used to capture power from AC power lines. The proposed harvester system uses a vibrating beam around the conductor that is connected to a linear PMSG with an array of permanent magnets with opposite polarity and experimental results demonstrated that the prototype has a power density of 3.2 mW/kA.

Under normal operating conditions with each pole carrying the rated current but in opposite directions, the generated static (DC) magnetic field magnitude is at a minimum at the ground level of the tower and a maximum at the conductor height. In Figure 14, the magnetic field generated by a bipole HVDC transmission line where each pole consists of two conductors carrying a total current of 2000 A is shown. In general, the magnetic field will be strongest at the surface of the conductors, with minimums between the bundled conductors. It is not expected that the static magnetic field will impact electronic components unless they involve magnetic materials. Relays, inductors, transformers, hall sensors, linear variable differential transformers, and motors are examples of components that may be impacted by the static magnetic fields from the saturation of magnetic cores. Disturbances caused by transients on HVDC lines are not fundamentally different from those in a power frequency HVAC system and immunity to short-duration transients will need to be determined according to standards, such as IEC 61000-4-8.

6.3. Energy Harvesting from Electric Field

The electric field generated by transmission lines only depends on the rated voltage and is not affected by the amount of current flowing through the power lines. EH, based on the generated electric field, is a feasible and well-suited method to power up sensors in HV transmission lines as it can be utilized when the power conductors are energized regardless of the current level, providing a reliable and available source of energy. Wrapping a copper sheet around the insulator is a common method for harvesting energy from power lines. The capacitance created by the copper sheet and the energized wire is then used to harvest energy. A diode is used to rectify the generated voltage and to prevent scavenged energy from being back fed. The EH unit consists of a battery/super-capacitor to store energy when the output voltage is not sufficient to supply sensors and a switch to connect the load to the EH unit when the voltage is regulated [79].

The main limitation of the electric field EH method is that the displacement current through the capacitance is limited to several hundred µA, resulting in an EH unit with a continuously available energy of less than 1 mW. To increase the ac input current to the harvester, a two-winding transformer was used to step down the voltage before rectification. Using this strategy, the maximum harvested power level increased to 370 mW [83]. Although the EH unit with transformer design improves the power output, the outcome is bulky, and this method has its constraints. In [84], a self-powered flyback converter was used to convert the energy from the capacitor harvester to a low DC voltage.

6.4. Energy Harvesting from Radio Frequency

Radio frequency (RF) technology facilitates the harvesting of energy from various RF sources, such as TV towers, Wi-Fi, and radio broadcast. The ambient RF waves have a frequency spectrum in the range of 0.2–24 GHz. By utilizing receiving antennas, RF signals can be converted to DC power to power sensor nodes in power transmission lines. Harvesting energy from RF waves can be classified as intended or ambient. The intended RF transmitter generates signals with the aim of targeting a specific location. EH based on this technology offers great potential in areas with strong RF signals to provide a consistent energy source for the application of continuous online monitoring. To maintain and track the optimal performance of the RF-EH method, the frequency and power level of the RF signal are required to be monitored [85]. Although RF-EH provides a free energy source, this method has a low power density as the emitted power is proportional to the inverse square of the distance from the radiation source. The low RF density can only be used in wireless sensors with low power consumption, such as duty-cycled operations with limited delay. This is because the system is first required to collect enough RF energy before supplying power to sensor nodes. In addition, RF signals may not always be available due to RF source outages such as the duty cycle of Wi-Fi routers. An RF-based harvester system consists of an antenna, a rectifier, and a voltage booster. Loss of signal strength along the transmission path, energy dissipation of the EH circuit, shadowing effect of the environment, interference, and multipath propagation of RF waves are the main factors that affect the effectiveness of the RF-EH approach. Battery-free operation of sensor nodes can be accomplished using an RF identification (RFID) system such that the sensor tag can harvest energy from the reader for measurement application [86]. Additionally, supercapacitors can be used in battery-less sensor platforms to store energy whenever an RF power source is available and to supply power during the communication and sensing cycles [87].

6.5. Energy Harvesting from Thermoelectric

Generating electrical power from a temperature gradient is referred to as thermoelectric energy harvesting (T-EH). Thermoelectric generators (TEGs) are commonly used to convert heat into usable electrical energy. A typical commercial TEG is made up of a series of semiconductor thermocouples that are sandwiched between two ceramic plates. The efficiency of thermal EH is dependent on environmental conditions such as weather, temperature, and the current level of power conductors, meaning that this method may not be able to provide a consistent power source in the transmission line monitoring system. To harvest the thermal energy of power lines, a TEG can be wrapped around the conductor to harvest heat flux [88].

6.6. Energy Harvesting from Vibration

The mechanical vibrations present in power lines, such as that caused by wind, can be a source for harvesting electricity [89]. Energy can be extracted from vibrations using a variety of ways, including piezoelectric, inductive, and electrostatic approaches. Electric charges are generated when piezoelectric materials such as quartz crystals are subjected to mechanical stress and the energy density can be defined as [90]

$$u=\frac{1}{2}(d_{33}\times g_{33}){\left(\frac{F}{A}\right)}^2$$

(5)

where $d_{33}$, $g_{33}$, F, and A are the piezoelectric charge factor, the piezoelectric voltage factor, force, and surface area, respectively. Another common approach is to attach a permanent magnet or Halbach array to the tip of a cantilever to generate vibrations due to the magnetic field, and convert the mechanical energy to electric energy [91,92]. However, it is required to improve the efficiency of energy conversion via optimizing the mechanical and material properties of the piezoelectric EHs, i.e., parameters $d_{33}$ and $g_{33}$ in (5). For this reason, various compositions of materials such as nanostructures, ceramics, and polymers are utilized to provide high performance in energy harvesting. In inductive-based EH, relative motion between a conductive coil and a permanent magnet induces voltage because of the change in flux density in the coil [93,94]. In electrostatic EH, energy is generated by the charge transfer due to the movement of two charged capacitor surfaces, change in the dielectric constant or the area of capacitor plates [86]. The vibration EH method offers several advantages, such as high power density, compact size, and high conversion performance over other EH techniques thanks to micro-electromechanical systems (MEMs) technology. However, the main disadvantage is that the frequency spectrum of ambient vibration sources varies, which poses a challenge to the optimal performance of the sensors [94].

6.7. Energy Harvesting from Corona

Corona discharge in power lines dissipates energy and can cause electromagnetic interference and damage to transmission line components. In [95], a corona EH method was proposed to add an artificial corona electrode to an HVDC line tower, to generate corona pulse currents. The energy-harvesting circuit was able to provide a stable energy supply of 10.15 mW from a 22 kV DC line. The current of the needle plane electrode can be expressed as [95]

$$i_{\mathrm{c}}=i_{\mathrm{cd}}+ϵ\left(st\right)i_{\mathrm{cp}}$$

(6)

in which $i_{\mathrm{cd}}$ and $i_{\mathrm{cp}}$ are the steady-state DC current and pulse current components. $ϵ\left(st\right)$ function represents the corona state condition, whose value is 1 if the corona glow discharge is not present. The maximum available power for the proposed corona EH can be represented as:

$$P_{\max }=I_{\mathrm{C}}(V_{\mathrm{HVDC}}-V_{\mathrm{Corona}})$$

(7)

where $V_{\mathrm{HVDC}}$ and $V_{\mathrm{Corona}}$ are the rated DC voltage and the corona onset voltage of the corona electrode.

6.8. Energy Harvesting from Wind

Wind energy can be harnessed for monitoring applications and is commonly used for large-scale energy generation. This particular method of energy harvesting involves the conversion of kinetic energy into electrical energy through the utilization of turbines based on the electromagnetic induction principle [96]. In [97], a wind energy-harvesting method was proposed that was able to provide 7.86 mW at an average wind speed of 3.62 m/s. A six-bladed wind turbine was designed with a blade radius of 7 cm, generating 439 mW under 7 m/s wind speed [98]. It was observed that increasing the number of blades enhances the power output coefficient because angular velocity reduction is compensated for by the increase in torque. A wind flutter generator was proposed that operates based on the aeroelastic flutter effect, generating energy from the flow of air by the motion of the structure. This device is rotor-less and does not include gears and was tested in a wind tunnel by changing the wind speed from 3 to 20 m/s with a maximum output power of 171 mW [99]. Weather forecast information was used and transmitted to the sensor location to improve the power management by the WSN using a wind turbine with a 3 cm blade radius and a maximum power of 7.86 mW [100]. The available kinetic energy from wind can be expressed as [101]

$$P_{\mathrm{in}}=\frac{1}{2}m{v}^2=\frac{1}{2}{\rho }_{\mathrm{air}}A{v}^3$$

(8)

where $P_{\mathrm{in}}$, m, ${\rho }_{\mathrm{air}}$, A, and v are the kinetic energy of incoming wind, the mass of air over the wind turbine, the density of air, the rotor area of the turbine, and the wind speed. The airflow cannot be completely blocked by the wind turbine and the potential power can be expressed in terms of the inlet ($v_1$) and outlet ($v_2$) speeds:

$$P_{\mathrm{turbine}}=\frac{1}{2}{\rho }_{\mathrm{air}}A(v_1^3-v_2^3)$$

(9)

$$P_{\mathrm{turbine}}=\frac{1}{2}{\rho }_{\mathrm{air}}A{v}_1^3(1-\frac{v_2^2}{v_1^2})(1+\frac{v_2}{v_1})$$

(10)

giving maximum power, which is known as Betz’s limit.

$$P_{\max }=\frac{16}{27}\times \frac{1}{2}{\rho }_{\mathrm{air}}A{v}_1^3$$

(11)

The power coefficient $C_{\mathrm{p}}$ of the rotor has been introduced in

$$P_{\mathrm{turbine}}=C_{\mathrm{p}}\frac{1}{2}{\rho }_{\mathrm{air}}A{v}_1^3$$

(12)

to allow for comparing the efficiency of various wind turbines and is a function of blade pitch angle ($\theta$) and tip speed ratio ($\lambda$) [102] and $C_{\mathrm{p}}$ can be approximated as

$$C_{\mathrm{p}}=c_1((\frac{c_2}{{\lambda }_i})-c_3\theta -c_4{\theta }^{c_5}-c_6)e^{\frac{-c_7}{{\lambda }_i}}$$

(13)

$${\lambda }_i=\frac{1}{\frac{1}{(\lambda +c_8\theta )}-\frac{c_9}{{\theta }^3+1}}$$

(14)

$$\lambda =\frac{\omega r}{v}$$

(15)

where r, $\omega$, and v are blade radius, angular velocity of rotor, and incoming wind speed, respectively. The coefficients $c_1$ to $c_9$ for a typical turbine are shown in

Table 3. Approximated power coefficients for a typical turbine [102].

Coefficient	Value
$c_1$	0.6
$c_2$	160
$c_3$	0.93
$c_4$	0
$c_5$	0
$c_6$	9.3
$c_7$	9.8
$c_8$	0.037
$c_9$	0

.

A summary of energy-harvesting techniques is summarized in

Table 4. A summary of available energy-harvesting techniques, and their advantages and disadvantages.

Source	Power Density	Advantages	Disadvantages
Solar	15–100 $\frac{\mathrm{mW}}{{\mathrm{cm}}^2}$ [79,103]	High output voltage. High power density. Independent of power system operating conditions. For AC and DC grids.	The weather affects the availability. Need for battery storage. Costly due to regular maintenance.
Thermoelectric	50 $\frac{\mathsf{\mu }\mathrm{W}}{{\mathrm{cm}}^3}$ [104]	For both AC and DC systems. Highly scalable.	Depends on ambient variables. Need for efficient heat sinking.
Vibration	Inductive: 2 $\frac{\mathsf{\mu }\mathrm{W}}{{\mathrm{cm}}^3}$ Piezoelectric: 107 $\frac{\mathrm{mW}}{{\mathrm{cm}}^3}$ [79,103] Electrostatic: 45 $\frac{\mathsf{\mu }\mathrm{W}}{{\mathrm{mm}}^3}$	No External source is required for inductive and piezoelectric. High output voltage (15 V could be achieved).	May not be applicable to DC grids. Requires an external voltage source for operation (electrostatic)
Electric Field	170 $\frac{\mathsf{\mu }\mathrm{W}}{{\mathrm{cm}}^2}$ [105]	EH is possible as soon as the power conductor is energized. No need for power lines to carry current. A non-intrusive power supply for the sensors.	Implementation is difficult in DC grids. EH from low-voltage AC lines may be challenging.
Magnetic Field	280 $\frac{\mathsf{\mu }\mathrm{W}}{{\mathrm{cm}}^2}$ [103]	Easy to install on transmission lines. Simple EH structure.	Difficult to implement in DC transmission lines. Require electric current in the power conductors.
RF Waves	1 $\frac{\mathsf{\mu }\mathrm{W}}{{\mathrm{cm}}^2}$ [79,103]	Available in urban areas. For both AC and DC systems. Can operate in dark atmospheric conditions unlike solar EH method.	Low power density. RF waves may not be available in remote areas.
Corona	10.15 kW in a DC system of 22 kV [95]	Applicable to AC and DC transmission lines. Stable and high output power.	Increasing loss by adding an artificial electrode. May cause an overcurrent in the system.
Wind	439 mW with 7 cm blade radius [98]	High output power. For both DC and AC lines. Does not depend on the operating condition of the transmission line.	Intermittent and variable. Limited suitable location. Requires battery storage and maintenance.

.

6.9. Challenges, Future Direction, and Perspective

Unlike HVAC systems, harvesting energy from the electromagnetic fields of an HVDC transmission line is not possible as the electromagnetic fields are not time-varying. In the future, utilities may have sensors such as LiDar cameras in remote locations on all towers that necessitate a reliable source of power for their operation. This will require power harvesting schemes along with wireless power transfer technology to transfer the harvested power from the conductors to the tower (where the sensors are installed). Novel, reliable EH techniques are required for powering equipment installed on HVDC systems. The thermoelectric EH approach offers an attractive solution for DC systems. In addition, it is expected to have hybrid EH approaches to provide power supplies in HVDC systems such as solar cells and RF-EH, increasing the required transmission line capital investment. However, power management algorithms are of high importance in the design of monitoring systems as sensors measure a short duty cycle.

7. Wireless Power Transfer

The increasing deployment of monitoring sensors in transmission lines has made the use of wires and batteries for charging these sensor nodes inefficient. Hence, contactless solutions are required to power up sensors from energy sources to increase the effectiveness of the monitoring system. Inductive power transfer (IPT), capacitive power transfer (CPT), magnetic resonance coupling (MRC), RF power transfer (RFPT), and laser power transfer (LPT) are the most common types of wireless power transfer methods.

7.1. Near-Field WPT

IPT and CPT wireless power transfer systems are based on near-field coupling, meaning that the transmit and receive coils must be located close to each other for efficient energy transfer. This is because magnetic and electric fields attenuate rapidly with the charging distance. Additionally, the coupling factor between the primary and secondary coils of the inductive link and the resistance of mutually coupled windings, mainly determine the efficiency of the IPT method, limiting the charging distance to within 20 cm [106]. The magnetic field can induce eddy currents in nearby metallic objects, increasing the loss in the system. However, metallic objects can be utilized in CPT to transfer power, resulting in reduced costs and preventing eddy current losses in the system [107].

MRC power transfer operates based on the same principle as IPT. Two magnetically coupled coils resonate at the same frequency and can transfer energy with high efficiency and small leakage to extraneous non-resonant components. It was shown that this WPT method can charge multiple devices simultaneously [108,109]. The operating frequency is in megahertz frequency and the energy can be transferred at a longer distance compared to IPT as the quality factor is higher. In [110], this technology was implemented to deliver 20 W of power to online monitoring sensors with the application in smart grids and with a charging distance of 1.1 m in a 110-kV HV line and efficiency of 20.2%. An HV insulator string with WPT capability was designed to harvest energy from the magnetic field of AC power lines and transmit power to online monitoring sensors using the domino-resonant concept, in which coil-resonators were embedded in insulator discs to transfer power between ends of the insulator structure [111,112].

7.2. Far-Field WPT

Although the near-field WPT technologies have filled a niche in the industry market, far-field technologies such as microwave and laser power transfer methods show promising technologies in the development of WSNs [113]. The frequency of RFPT ranges from MHz to GHz. A transmit antenna, a receive antenna, an impedance matching circuit, a voltage multiplier, and a DC capacitor comprise an RF radiation power charging system. The conversion efficiency at the RF receiver is highly dependent on the RF signal density as well as the impedance matching of the voltage booster and the antenna [114,115]. This power transfer method is suitable for charging low-power monitoring sensors in a WSN or RFID system as the transmission efficiency is very low and it is not a feasible approach for high-power-sensing activities due to the potential impacts on human health [116]. In addition, the major drawback of power transfer using the RF method is that sensor nodes may have different energy-harvesting rates, which are significantly dependent on the distance between the sending antenna and the receive sensor nodes. This unfairness in the harvested power of sensors should be properly addressed using control protocols and power management algorithms in a WSN [117]. In [118], laser technology was used to power up multiple sensor nodes with a charging distance of 1.5 m and a maximum energy density of 85 $\mathsf{\mu }\mathrm{W}/{\mathrm{cm}}^2$ using solar cells as an energy harvester. Distributed laser charging (DLC) was used to provide sensors with WPT capabilities with an overall efficiency of 10–37% and a maximum charging distance of 10 m [119,120].

7.3. Challenges, Future Direction, and Perspective

With the development of reliable EH schemes, it is required to transfer power to sensor nodes efficiently. As many sensors exist in a future tower structure, transferring power to multiple sensors is of high importance to be performed simultaneously while keeping a suitable power transfer rate to have enough power for the proper operation of WSN. The transfer distance is required to be increased as higher power density is transmitted to sensors such as monitoring cameras. Concurrent power and information transfer approaches are the future trend for WSNs in power grids. With new and growing technologies such as the Internet of Things (IoT), various platforms can be designed to regulate HVDC grids that are decentralized and intelligent. Providing power for these monitoring devices in the future instrumented HVDC tower is still a challenge as no reliable EH approach exists for DC lines and far-field power transfer methods are still not efficient over large distances. The advantages and disadvantages of wireless power transfer methods are summarized in

Table 5. A summary of available power transfer techniques, and their advantages and disadvantages [113].

Wireless Charging Method	Advantages	Disadvantages
Inductive Power Transfer	Can achieve high efficiency. Has been standardized in some applications (Smartphones). Well-established technology and safe for humans.	Limited Range (0.5–40cm). Proper alignment is required. Heat generation during charging. Eddy current losses.
Capacitive Power Transfer	Low eddy current losses. Lower cost using metal objects. Several kilowatts of power output can be achieved. Flexible and scalable to be used in small-size applications.	Short charging distance (∼100mm). Limited efficiency. Affected by parasitic components.
Magnetic Resonance Coupling	Power in the kilowatts range can be transferred. Long power transfer distance (km). Less sensitive to misalignment of transmitter and receiver. Able to charge multiple devices simultaneously.	Implementation can be challenging. Efficiency is low (10%). Can be affected by physical obstacles.
RF Radiation	Long charging distance (in km range). Does not require direct line-of-sight. Highly scalable and can charge multiple devices.	Low efficiency. Safety consideration due to potential health risks. Low power-transmission rate.
Laser Power Transfer	Long-range charging distance. High power transfer. Has the potential for fast and non-contact charging applications.	Complex implementation. Requires precise beam alignment mechanism. Low efficiency (20%). Line-of-sight requirement.

.

8. Conclusions

Future enhancement and improving the reliability of HVDC transmission lines is tied to the development of online monitoring technologies adapted to operate in remote areas. The development of HVDC grids and penetration of renewables will need new sensor technologies and energy-harvesting schemes. They also enable an understanding of unresolved phenomena already occurring in the existing HVDC systems. An overall perspective of the needs of HVDC transmission lines is presented in

Table 6. An overview of the parameters already monitored in HVAC transmission lines and the value of their monitoring in HVDC transmission lines.

Category	Monitoring/Power Supply Parameter	Already in HVAC and Adaptable to HVDC	Importance			Comments
			Low	Medium	High
Operating Conditions	Electric Voltage and Current	No			x	Voltage and current measurement techniques required for HVDC grids for fault location and protection. Existing HVAC technologies are not adaptable to HVDC.
	Conductor Temperature	Yes		x		Required to support HVDC DLR and avoid conductor damage from fires.
	Security and Investigation	Yes			x	Required as HVDC transmission lines traverse long distances and remote areas. Difficult to patrol and determine issues in real time.
	Leakage Current	No		x		Shunt Resistor Method for AC and DC. HVDC insulators more prone to pollution issues and methods required to provide support to proactive mitigation.
	Electric Field & Corona Discharge	Yes		x		When correlated with environmental sensors, may allow prediction of unexplained flashover events.
	Infrared & Ultraviolet Scanning	Yes		x		Required for inspection and maintenance assessments of HVDC transmission lines.
Mechanical Stress	Sag	Yes			x	Avoidance of clearance violations on HVDC transmission lines. Sag determines maximum loading capability.
	Vibration	Yes		x		Investigation of issues and avoidance of equipment damage on HVDC transmission lines. Data can be correlated with wind and icing.
	Galloping	Yes		x		Investigation of issues and avoidance of equipment damage on HVDC transmission lines. Data can be correlated with wind and icing.
	Tension	Yes		x		Tension can be correlated with wind, icing, and sag data.
Environmental Conditions	Ambient Temperature	Yes		x		Required to support HVDC DLR and environmental parameters for investigation of unexplained flashovers.
	Wind Speed and Direction	Yes		x		Required to support HVDC DLR and environmental parameters for investigation of unexplained flashovers.
	Precipitation	Yes		x		Required to support HVDC DLR and environmental parameters for investigation of unexplained flashovers.
	Humidity	Yes		x		Important for correlation with other parameters such as leakage current, corona, and unexplained flashovers.
	Solar Radiation	Yes		x		Required to support HVDC DLR and environmental parameters for investigation of unexplained flashovers.
	Smoke and Fire	Yes		x		Required to ensure protection and reliability of the HVDC transmission line.
	Ice and Snow Buildup	Yes			x	Allows mitigating techniques to be applied to HVDC transmission lines to avoid outages and equipment damage. Critical to analyze corresponding wind data.
Vegetation and Ground Clearance	Synthetic aperture radar	Yes		x		Can be applied to HVDC lines to determine vegetation clearing programs and ensuring system reliability and avoidance of clearance violations.
	Airborne or stationary laser scanner	Yes		x		Can be applied to HVDC lines to determine vegetation clearing programs and ensuring system reliability and avoidance of clearance violations.

.