Numerical Simulation and Experimental Study of the Pneumo-Electric Hybrid-Driven Pipeline Inspection Robot in Low-Pressure Gas Pipeline

Abstract

Intelligent pipeline inspection is necessary to operate submarine pipelines safely. At present, speed excursion and blockage are the challenges in the inspection of low-pressure gas pipelines. Accordingly, this study proposes a novel pneumo-electric hybrid-driven scheme to improve the traveling stability of inspection robots. To adapt to different working conditions, building blocks and CFD numerical simulation methods are used to study the throttling pressure control flow field of the robot. The results proved that the flow clearance had the most evident effect. The flow clearance was reduced from 30 to 5 mm, and the differential pressure of the prototype increased from 0.3 to 17 kPa. The skeleton diameter has a small effect on the differential pressure. The differential pressure increases as the gas velocity increases. By analyzing the prototype in different positions, it was found that the differential pressure of the prototype while passing the elbow decreased by 45% at 45°, which quantified the fluid-driven force gap of the prototype while passing through the elbow. Finally, by comparing the speed of prototype with that of fluid-driven pig, it is demonstrated that a pneumo-electric hybrid-driven scheme is an effective solution to the problem of unstable inspection operation of low-pressure gas pipelines.

1. Introduction

With the rapid development of the subsea oil and gas exploitation field, the scale of subsea transmission pipeline systems continues to increase [1]. Pipeline inspection gauges (PIGs) and pipeline inspection robots (PIRs) are important tools for regular and comprehensive pipeline inspection. New baseline pipelines, as well as some natural gas pipelines, are characterized by low pressure, and this pipeline condition brings great challenges to in-pipe inspection engineering [2]. Speed excursion, oscillation, blockage, and other phenomena are often observed in pipeline inspection operations [3,4], seriously affecting the validity of pipeline inspection data [5,6]. Further, speed excursion may lead to pipeline rupture, seriously threatening the safe operation of the pipeline [7].

In low-pressure gas pipelines, the instability of a PIG’s travel speed is mainly caused by the gas compressibility and frequently changing travel friction in the pipeline [8,9]. The PIG can travel at a steady speed in normal frictional positions; however, in positions where friction increases instantaneously, such as elbows, welds, positions where wall thickness changes, etc., the PIG’s travel speed undergoes a complex cycle of oscillations [10]. In the other pipe sections where abnormal increases in frictional force are present, the PIG must travel through these abnormal frictional force sections by building up a higher differential pressure [11]. When traveling from the abnormal frictional force section to the uniform frictional force section with a significant driving force, the pressure difference of the PIG is much more significant than that of the frictional force, causing the PIG to accelerate [12]. Low-pressure conditions can also prevent the PIG from building up sufficient drive-differential pressure in a timely and efficient manner. Unstable velocity oscillations can easily cause the PIG to be blocked in certain positions [13].

Currently, the drive mode of PIGs and PIRs mainly adopts two types: passive, which is fluid-driven, and active, which is motor-driven [14]. The differential pressure of the gas medium at the nose and tail of the PIG drives the fluid-driven type. A fluid-driven PIG without bypass travels at a speed that is extremely dependent on the gas flow rate and is considered an unrestricted system with extremely difficult speed control [15]. A fluid-driven detector with a bypass allows it to travel below the gas velocity, achieving the speed control goal to a certain extent [16]. The speed control method is primarily achieved by controlling the bypass leakage area to control the pressure difference between the tail and nose of the PIG, changing the relationship between the driving force and frictional force, and controlling the traveling speed of the PIG [17]. The main difficulty in controlling the speed of bypass valve leakage is in determining the pressure loss coefficient of the bypass valve leakage. Numerous scholars have conducted studies on different bypass structures through experiments and simulations [18,19,20]. The leakage pressure drop equation and computational fluid dynamics calculations have proven effective methods. The transient dynamic characteristics of a PIG can be characterized by studying the bypass leakage valve. Some control strategies may suppress velocity oscillations to a certain extent. However, fluid-driven detectors must build up a certain pressure difference in low-pressure gas pipelines whenever the instantaneous friction changes are large. Inevitably, high differential pressure builds up, and the transient reduction in friction causes an unbalanced PIG driving force [21,22]. Avoiding speed excursions in elbows, weld seams, and where wall thickness and friction changes are problems that are currently difficult to solve in fluid-driven internal detectors [23].

In recent years, actively driven pipeline inspection robots have been used to inspect new baseline pipelines [24]. The multiple drive wheels of a PIR are evenly distributed around the center axis, and the motors provide kinetic energy to the drive wheels, allowing the PIR to travel autonomously through the pipe. Drive wheels are passively operated using springs or connected using a linkage actively operated by an actuator. Supporting by springs has the advantage of simplifying construction and allowing smooth adaptation to changes in pipe diameter [25]. However, this does not allow the pressure against the pipe wall to be controlled. Therefore, movement through inclined or vertical pipes is problematic. In addition, self-driven robots are limited by the energy density of the battery and one-time discharge power [26]. The distance the battery can support the inspection operation is very limited, and it cannot accomplish inspection tasks in long-distance pipelines.

Therefore, it is important to design a method that can perform inspection operations in low-pressure gas pipelines, realize stable operation of the entire pipe, and avoid blockage and speed excursion in abnormal frictional force sections, such as elbows.

To realize stable driving of a PIR in low-pressure gas pipelines, this study proposes a novel pneumo-electric hybrid-driven pipeline inspection robot (PEPIR) based on a previous study conducted on electric self-driven PIRs, combined with the operating principle of a fluid-driven PIG’s pressure drop of the leakage flow.

Designing the structure of the PEPIR, establishing a reasonable driving scheme, and optimizing the PEPIR speed control strategy have become the main challenges to be resolved using this method. Therefore, this study focused on these issues. Finally, this study provides a new method for detecting low-pressure gas pipelines.

The major novelty and originality of this study are highlighted as follows. A novel PEPIR is proposed for the first time, and the throttling pressure control flow field was studied based on the new robot structure to explore the effects of different flow clearances, skeleton diameters, and pipeline flow velocities on the differential pressure of the PEPIR. The building block and CFD numerical simulation methods were used to model the pressure drop. The differential pressure of the robot at different locations of the elbow is also analyzed. An experimental pipeline system was established and dynamic tests of the PEPIR inside the pipeline were carried out to verify the validity of the research method. Finally, a complete design process is proposed for the PEPIR, which provides a foundation and guidance for future PEPIRs that can be adapted to various operating conditions.

2. Modeling of Pneumo-Electric Hybrid-Driven Pipeline Inspection Robot

The PEPIR model includes a transient fluid-driven pipe model and a transient dynamics analysis model. The transient fluid-driven pipeline model describes the effect of the PEPIR traveling in the fluid and the driving force of the fluid on the PEPIR. The PEPIR transient dynamics analysis model combines the fluid-driven model with the motor-drive model to analyze the detector motion behavior in the pipe.

The main cause of the speed excursion was a sudden change in the frictional force difference. Electrodynamic compensation compensates for these sudden changes in frictional force at specific sections such that the fluid-driven pressure differential provided by the fluid medium to the PEPIR remains uniform. This fundamentally avoids sudden acceleration caused by differential friction and allows the PEPIR to travel at a smoother overall speed.

The PEPIR proposed in this study is an innovative design that combines fluid-driven principles on the foundation of a motor-active-drive PIR. The calculation of the pressure drop and design of the self-driven motor control mechanism were used to realize operation in the new baseline and the low-pressure gas pipeline.

2.1. Working Principle and Advantages of the PEPIR

The structure of the PEPIR used in this study is shown in Figure 1. It mainly contains a double cup with a clearance leakage structure, a motor drive system, a sensor sensing system, and a power control system. The PEPIR is based on a self-driven PIR with the addition of a double-jointed power cup and a support wheel that is self-adaptive to the pipe diameter with a degree of adjustability of 40%.

The diameter of the traditional spherical sealing cup was larger than the inner diameter of the pipeline. The compression of the cup by the pipe wall isolates the gas before and after the PIG to form a pressure difference but also produces a large frictional force [10]. A schematic diagram of the fluid leakage structure of the PEPIR in this study is shown in Figure 2. The cup is different from the traditional design with an overfilling capacity, in which the cup has a certain clearance between the cup and the pipe, and the radius of the power cup is smaller than the inner diameter of the pipe δ mm. This structural design enables the fluid to discharge before and after the PEPIR to form a pressure difference for the PEPIR to provide power while avoiding the challenges of a fluid-driven PIG faced with sudden changes in the thinning of the pipe wall, cup deformation, or elbow with the increase in compressive stress, greatly reducing the traveling frictional force. The difficulty of this structure is controlling the steady-state traveling under different operating conditions, which requires a rigorous throttle pressure control flow field study.

2.2. Fluid-Driven Throttling Pressure Control Flow Field Analysis

The primary driving force of a fluid-driven PIR is the differential pressure generated by the fluid at the nose and tail of the PIR. A different cross-section along the flow path characterizes the leakage structure. As the fluid passes through the restriction, the velocity accelerates, and the pressure decreases according to Bernoulli’s principle. The PIR travels at a speed lower than the fluid velocity, and the pressure difference generated by the leakage flow becomes the driving force of the PIR. Studies on the fluid-driven PIG gas leakage pressure drop generated by the flow characteristics of the operation have been conducted, and it was found that the pressure loss coefficient and medium flow velocity determine the pressure difference between the tail and the nose of the PIR [27].

The pressure loss of the flowing gas through the leakage structure is described by Bernoulli’s equation, which evolves into the pressure loss coefficient multiplied by the dynamic pressure generated by the driving gas through the leakage structure between different cross-sections, expressed as:

$$∆P=P_{tail}-P_{nose}={\displaystyle \frac{\xi \rho }{2}}\left[{\left({\displaystyle \frac{A_0}{A_d}}\right)}^2{\left(V_{gas}-V_{pir}\right)}^2\right]$$

(1)

$$∆P=P_{tail}-P_{nose}={\displaystyle \frac{\xi \rho }{2}}{V_h}^2$$

(2)

A pressure loss investigation was conducted to investigate the pressure drop in the annular clearance leakage mode of the PEPIR. In this study, the clearance leakage was converted into an equivalent cross-sectional area coefficient to calculate the pressure loss coefficient.

$$A_{\delta }=A_0-A_d={\displaystyle \frac{\pi \delta \left(2D-\delta \right)}{4}}$$

(3)

$$\phi ={\displaystyle \frac{A_{\delta }}{A_0}}={\displaystyle \frac{2D\delta -{\delta }^2}{D^2}}$$

(4)

According to the continuity equation, the gas velocity with respect to that of the PEPIR is:

$$V_h={\displaystyle \frac{V_{gas}-V_{pir}}{\phi }}$$

(5)

The flow characteristics of the fluid differed for the different leakage structures. Therefore, many scholars have explored pressure loss coefficients for different discharge structures [28]. The “building block approach” describes the pressure loss in special variable cross-sectional pipes for complex leakage structures. Because the geometry of the PIR varies depending on its application, a building block approach was used to provide a generic framework for determining the corresponding pressure loss coefficients. The building block approach relies on the geometric decomposition of the PIR and considers the contributions of the individual components of the PIR geometry to the overall pressure loss. A building block approach was also used in this study. We exploited the symmetry of PEPIR to use a two-dimensional axisymmetric framework to determine standard flow patterns that are uncorrelated between standard flow elements, combining the characteristics of contraction expansion and expansion-contraction flow channels. The total pressure loss was initially established as a linear combination of the standard flow elements.

For the annular leakage pressure drop structure in this study, the main influences on the fluid pressure drop behavior can be simplified as the skeleton and two power cups, which are subdivided into three sets of geometric building blocks: the pressure loss coefficient of the tail cup clearance flow discharge, K_ct; the pressure loss coefficient of the flow through the chamber annulus, K_apl; and the pressure drop loss coefficient of the nose cup clearance flow discharge, K_cn. The leakage clearance of each cup was divided into three parts: sudden contraction, friction loss in the flow through the chamber, and sudden expansion.

According to the existing theoretical framework for the pressure drop of the sudden expansion and contraction of the fluid through the leakage structure, the pressure drop loss coefficient and leakage clearance can be established using the Idelchik correlation [29].

$$K_{ct}={\displaystyle \frac{∆p}{\frac{\rho w_1^2}{2}}}=\left[0.5{\left(1-{\displaystyle \frac{A_{\delta }}{A_0}}\right)}^{0.75}+\tau {\left(1-{\displaystyle \frac{A_{\delta }}{A_0}}\right)}^{1.375}+{\left(1-{\displaystyle \frac{A_{\delta }}{A_0-A_s}}\right)}^2+\lambda {\displaystyle \frac{l_c}{D_h}}\right]$$

(6)

$$K_{cn}={\displaystyle \frac{∆p}{\frac{\rho w_1^2}{2}}}=\left[0.5{\left(1-{\displaystyle \frac{A_{\delta }}{A_0-A_s}}\right)}^{0.75}+\tau {\left(1-{\displaystyle \frac{A_{\delta }}{A_0-A_s}}\right)}^{1.375}+{\left(1-{\displaystyle \frac{A_{\delta }}{A_0}}\right)}^2+\lambda {\displaystyle \frac{l_c}{D_h}}\right]$$

(7)

The annular flow pressure loss coefficient in the PEPIR internal chamber was established from the pressure loss coefficients of the annular flow discharge structure as follows:

$$\lambda ={\displaystyle \frac{1}{{(1.8logRe-1.64)}^2}}$$

(8)

$${\lambda }_{\mathrm{apl}}=\left({\displaystyle \frac{0.02d}{D_0}}+0.98\right)\left({\displaystyle \frac{1}{\lambda }}-0.27{\displaystyle \frac{d}{D_0}}+0.1\right)$$

(9)

$$K_{apl}\equiv {\displaystyle \frac{∆p}{\frac{\rho w_0^2}{2}}}={\lambda }_{\mathrm{apl}}{\displaystyle \frac{l_s}{D_h}}$$

(10)

In the circular pipe variable cross-section structure, the contraction or expansion structure is the most common. This study, for the first time, explored the in-pipe detector annular clearance leakage structure, established the friction loss of the fluid in the annular channel, and supplemented the pressure loss term of the annulus sudden contraction and expansion. The generalized framework for complex flow patterns in contracted and expanded tubes characterized by standard flow elements is established, and a generalized method for calculating the pressure loss in contracted and expanded tubes is provided.

2.3. Analytical Modeling of the PEPIR Transient Dynamics

For the movement of the PEPIR in the pipe, the forces on the PEPIR are shown in Figure 3. They include the passive driving force generated by the pressure drop of the gas leakage through the clearance formed by the cup and the pipe, the active driving force F_M generated by the operation of the motor, the driving friction F_f generated by the contact friction between the PEPIR and the pipe, the gravitational component mgsinθ of the PEPIR, and the shear force F_τ of the gas flow on the PEPIR.

$$m{\displaystyle \frac{dv}{dt}}=∆P\left(1-\phi \right)A-F_f+F_M-mgsin\theta +F_{\tau }$$

(11)

Considering the quasi-steady state motion process, Equation (11) can be appropriately simplified to obtain the PEPIR traveling speed model:

$$V_{pir}=V_{gas}-\phi \sqrt{{\displaystyle \frac{2(F_f-F_M-mgsin\theta )}{\xi \rho }}}$$

(12)

For the scheme design of the PEPIR, the most important step is first to identify the operating conditions of the pipeline to design a suitable range of the driving force of the gas pressure drop and to design the motor parameters that can satisfy the operating requirements in conjunction with the overall dynamics model.

In Equation (12), the parameters influencing the PEPIR traveling speed are clearly shown, and the variations in the traveling frictional force and gas flow velocity are the main factors affecting the motion of the PEPIR. This study aims to realize closed-loop automatic control of the travel velocity in a pipeline by modeling the PEPIR transient dynamics. Limited by the inability of the existing technology to monitor the gas flow velocity where the detector is located in real-time, a uniform value is assumed for the gas flow velocity.

3. Numerical Setup

The building block approach has certain limitations in confronting continuous elements, where the gas interacts with multiple elements, causing some error in the sum of the individual pressure loss coefficients. The pressure drop loss of the flow through the chamber includes friction and energy losses, and the contraction pressure drop of the tail cup cannot be regarded as an ideal and uniform inlet distribution after passing through the jet of the nose cup and the vortex cycling motion of the chamber. The pressure loss coefficients established based on the Idelchik correlation may contain certain errors at low flow rates [30]. The total pressure loss is a nonlinear combination of standard flow elements.

Therefore, the characterization of the relationship of energy loss under the interaction of these elements, the complex flow pattern of interconnections between the standard flow elements, and the determination of how the total pressure drop is distributed geometrically become the research challenges of the present study of PEPIR.

In previous studies, computational fluid dynamics (CFD) finite element simulation analysis of the pressure loss coefficient was shown to be an effective analytical method. Therefore, this study incorporated the simulation results to modify the established fluid-driven model further.

3.1. Analysis of Influencing Factors of Pressure Drop

This study selected the three most important parameters for the PIR design and operating conditions for analysis. The cup and skeleton diameters influence the flow domain formed by the PIG and pipe wall, which directly affects the flow behavior of the flow relief. Thus, fixing these two values plays an important role in the design and manufacture of PEPIR. According to the structural characteristics of the existing PIR, the cup interferes as lightly as possible when traveling through the elbow to obtain larger pressure-drop characteristics. In addition, the flow velocity of the pipeline fluid directly determined the magnitude of the passively driven differential pressure of the fluid obtained by PEPIR. In this study, the clearance between the cup and the pipe wall was set to be 5–30 mm, the size of the skeleton was set to be 50–200 mm, and the flow velocity was set to be 3–7 m/s. This study performed a flow-field simulation analysis for the parameters mentioned above to determine the most suitable design scheme under different working conditions.

Under low-pressure operating conditions, the speed excursion phenomenon after traveling through the elbow is the most important problem to be solved in the inspection operation, and the differential pressure change caused by a sudden change in the frictional force during the process of traveling through the elbow is unavoidable. In this study, the cup was set as a clearance type to reduce the degree of sudden changes in frictional force while traveling through the elbow. Therefore, this study established an elbow motion model to investigate the change law of the gas differential pressure force during the passage of PEPIR through the elbow.

3.2. Simulation Model

The PEPIR structure mainly consists of a skeleton and two power cups. The skeleton diameter, skeleton length, and cup diameter (clearance distance between the cup and pipe wall) directly affect the fluid pressure drop characteristics of PEPIR. A simplified fluid domain extraction was performed to study the flow characteristics of PEPIR. Figure 4 illustrates the fluid domain model of the PEPIR established in this study. To ensure that the fluid flowed fully, the upstream and downstream lengths were set to 5 D and 15 D, respectively. Mesh generation for the numerical model is illustrated in Figure 4. Local mesh encryption was performed on the watersheds around the PEPIR.

The gas differential pressure force of the PEPIR is affected by the velocity difference between the operating speed of the PEPIR and the flow velocity in the pipe (V_gas–V_pir). To investigate the effect of the gas flow velocity in the tube on the pressure drop, boundary conditions were set with inlet velocities ranging from 1 to 7 m/s. The background of this study was low-pressure conditions and the outlet pressure boundary was set to 0.1 MPa.

Simultaneously, the flow field area changes during the passage of the PEPIR through the elbow. The modeling process was based on the attitude data acquisition of the PEPIR in the passability experiment, whereas the attitude of the PEPIR at different positions of the elbow was determined by kinetic analysis and sensor monitoring. The fluid region in the PEPIR and elbow regions was simplified and extracted to explore the change law of the gas differential pressure force of the PEPIR in the overall process of passing through the elbow.

In the elbow section, the orthogonal quality of the mesh was crucial for the accuracy of the simulation. By refining and adjusting the overall mesh, the number of meshes with orthogonal quality greater than 0.6 accounts for 82.1%, and the number of meshes between 0.2–0.6 accounts for 17.9%. We performed the mesh independence study of PEPIR in straight pipe, comparing the pressure distribution curves along the pipe axis with mesh numbers 2,365,701, 3,101,156, and 3,915,125. The comparison reveals that the average uncertainty of the numerical method in this study is 2.62%, which proves that the influence of the mesh on the numerical results is within acceptable limits.

3.3. Analysis of Simulation Results

3.3.1. Influence of the Flow Clearances

The flow clearance is the most important factor influencing the pressure drop in the flow field of the PEPIR. A flow clearance that is too large will not be able to establish sufficient differential pressure to provide the driving force of the PEPIR, and a flow clearance that is too small will greatly increase the value of the frictional force when the PEPIR passes through elbows, which can easily cause traveling stagnation or blockage. The effect of different flow clearances on the pressure drop is explored by means of given inlet velocities. Previous investigations have demonstrated that the size of the clearance has a considerable effect on the mass flow rate [31].

Figure 5b shows a streamline of the flow field region enclosed by the inner wall of the pipe, skeleton, and power cup as control domain. Upstream of the first cup, the fluid flow rate is accelerated owing to the clearance restriction between the cup and the pipe wall. For thick orifices, the streamlines converged at the constricted vein and were redistributed to the annular flow domain. This corresponded to the initial pressure drop in the pressure contours of Figure 5a. Significant vortices were observed in the control volume, which consisted of the tail and nose cups, pipe wall, and skeleton. Comparing 5 to 20 mm flow clearance, the aerodynamic cross-section of the jet increased throughout the vortex chamber, and the jet area was much wider than that for a flow clearance of 5 mm. The vortex formed within the vortex chamber was affected by the increase in jet area, and the vortex cross-section became smaller as the jet area increased. The fluid expanded into the pipe wall after moving away from the second clearance. The flow restriction and subsequent expansion result in a loss of mechanical energy, reflected as a pressure drop after traveling through the clearance.

From Figure 6 we can clearly see that the pressure drop arises before and after the PEPIR, which is the pressure difference of the different flow clearances. Combined with the pressure distribution curve, the contribution of each type of flow restriction to the frictional force of the pressure drop was observed. As the flow clearance decreased from 30 to 5 mm, the pressure difference between the tail and nose of the PEPIR rose from 0.3 to 17 kPa. This agrees with the law of the pressure loss coefficient equation. Because a smaller flow clearance has a smaller flow area, the differential pressure is inversely proportional to the leakage area. This indicates that a smaller flow clearance must be designed to achieve a larger gas differential pressure driving force. Therefore, it is necessary to determine the operating frictional force and operating conditions of the PEPIR to determine an appropriate flow clearance value.

In the elbow section, the flow clearance was reduced as much as possible to minimize the increase in the frictional force caused by the interference of the cup. Therefore, the design guidelines for the flow clearance cannot be based solely on the criteria for obtaining the maximum gas differential pressure driving force. This ensures that drastic and sudden changes in the frictional force at the elbows are avoided as much as possible, assuming the regular pipe section can overcome the regular operational frictional force.

3.3.2. Influence of the Skeleton Diameter

The effect of the skeleton diameter on the pressure drop was analyzed. Figure 7 shows the velocity and pressure contours for a fixed inlet flow velocity. An increase in skeleton diameter from 50 to 200 mm decreased the pressure difference between the tail and nose of the PIPER from 42 to 39 kPa. This is because an increase in the skeleton diameter reduced the basin of the annular flow, increased the gas flow velocity, and increased the pressure drop. The pattern of the effect of the change in the size of the skeleton on the pressure drop corresponds well with the pressure loss coefficient Equations (6), (7) and (10).

From the pressure curve in Figure 8, it was found that the larger the skeleton diameter, the higher the pressure drop frictional force. An increased skeleton diameter results in a larger installation space inside the device. However, it is also necessary to consider the passability problem caused by an increase in the skeleton diameter. One risk associated with a larger skeleton is that the equipment may achieve a greater operational frictional force during travel through the elbows.

It can be seen from the data in Figure 9 that although both the skeleton and flow clearance affected the magnitude of the gas pressure drop, the effect of the skeleton size on the pressure drop was significantly weaker than the effect of the pressure drop caused by the flow clearance. Therefore, flow clearance should be the main factor in designing the PEPIR, considering different working conditions, whereas the design guidelines for the skeleton size should be based mainly on passability.

3.3.3. Influence of Gas Flow Velocity

The gas flow velocity primarily depends on the operating conditions of the pipeline. To enable the PEPIR to travel at a preset velocity, the change in differential pressure for different velocity differences (V_gas–V_pir) was investigated by fixing the PEPIR model at different inlet velocities. From the pressure contours in Figure 10, it is apparent that the distribution characteristics of the flow field are similar at different flow velocities. The larger the inlet flow velocity, the larger the corresponding dynamic pressure value. From the curves of the differential pressure at the tail and head of the PEPIR for different flow velocity in Figure 11, it can be clearly found that the larger the inlet flow velocity is, the larger the differential pressure of the PEPIR is. A higher flow velocity indicates greater flow loss when experiencing flow restriction. The law of flow velocity versus differential pressure is almost identical to the square relationship in the pressure loss coefficient equation.

Under actual pipeline conditions, when the flow velocity in the pipe is fixed, the PEPIR traveling speed decreases when the frictional force increases, the velocity difference between the gas flow velocities increases, and the PEPIR obtains a larger driving pressure difference to overcome the increased traveling frictional force. However, for small-diameter low-pressure pipelines, the ideal traveling speed of the PEPIR is 1–2 m/s, and the margin of the velocity difference increase is only 1–2 m/s. For a gas flow velocity of 10 m/s in a small-diameter low-pressure pipe, the pressure difference between the tail and nose of the PEPIR increased to a limited extent when the traveling speed of the PEPIR decreased, only increasing from 4 to 5 kPa. This is generally less than the extent of the sudden change in frictional force, and it is impossible to overcome the increased frictional force to travel through the sudden change in the frictional force section. Therefore, the most important task of a fluid-driven modeling study is to understand the working conditions before the internal inspection operation and to design a scheme for satisfying the regular frictional force section travel based on gas flow velocity. In engineering practice, the degree of variation in the gas flow velocity under low-pressure conditions with small pipe diameters may be even greater, thus significantly affecting the stability of the gas differential pressure.

3.3.4. Analysis of Changes in Differential Pressure Traveling through Elbow

This study mainly aims to solve the problem of the passage of PEPIR in the elbow under low-pressure working conditions. Therefore, the movement process of PEPIR in the elbow should be analyzed. Through the PEPIR in the elbow before 0.2 m, elbow 0°, 15°, 30°, 45°, 60°, 75°, and 90°, eight different positions of the analysis were used to explore the prototype through the elbow process of fluid flow behavior change law. As shown in Figure 12, the pressure contours of the PEPIR were at different positions on the elbow. It can be observed that the PEPIR traveling through the elbow during the process of differential pressure first shows a decreasing trend and then an increasing trend. While traveling through the elbow, the leakage area changed, the center of the device was no longer located along the pipeline axis, and the leakage channel no longer had a regular symmetrical distribution. This affects the overall gas flow pressure-drop behavior. The value of the differential pressure driving force acquired by the PEPIR can be obtained by integrating and solving for the pressure on the stressed surface.

Figure 13 shows the pressure distribution curves of the PEPIR at different elbow positions. It can be observed that pressure loss still mainly occurred at the tail and nose of the PEPIR, and the pressure dropped rapidly in the first cup. It can be observed from the pressure drop curve that the change in the value of the pressure difference between the tail and nose of the PEPIR in the straight section before entering the elbow is 3565 Pa, and as the PEPIR enters the elbow, is increased to 3851 Pa. PEPIR enters the front side of the elbow, and the tail skin cup begins to be compressed. As can be seen from Figure 14, the flow channel appears to be narrowed and the flow velocity in the channel is higher, which produces a larger flow loss; therefore, the differential pressure increases. After the PEPIR enters the elbow, the center of mass of the device moves inward under the mechanical action of the overall structure, the clearance between the inner PEPIR cup and pipe wall becomes smaller, or contact interference occurs, and the external flow channel increases. Under the influence of such a change in the flow region, the pressure drop of the fluid flowing through the PEPIR decreased from 3851 to 2698 Pa.

By observing the streamlines at different angles, it was found that the external flow region gradually increased as it entered the elbow, and the fluid differential pressure also showed a gradually decreasing trend, which was attributed to the larger flow region caused by the attitude of the PEPIR. After the PEPIR travels at 45°, the external flow domain gradually decreases again, gradually increasing the fluid differential pressure. In the overall process of traveling through the elbow, the differential pressure first showed a decreasing trend, minimizing at 45°, and then exhibited an increasing trend. Simultaneously, it was found that the pressure difference between the tail and nose of the PEPIR was not the same at symmetrical positions of the bend (e.g., 15° vs. 75°). Considering that the fluid after PEPIR must flow through the curved pipe, the wall curvature also limits the jet behavior compared to a straight pipe, and this effect is more complex. Through an analysis of the streamlines, the changes in the fluid differential pressure appeared to be the result of a combination of changes in the flow domain caused by the attitude of the PEPIR and changes in the flow behavior caused by the bending of the pipe wall.

4. Experimental Study

4.1. Design of Experimental Scheme

A pipeline test system was established to sufficiently evaluate the gas differential pressure variation characteristics of the pneumo-electric hybrid-driven scheme.

First, a static pressure-drop experiment of the PEPIR was conducted. Setting the clearance of the cup to 5–20 mm, the PEPIR has a travel frictional force so that it can remain blocked when the gas differential pressure is less than the start-up differential pressure. This part of the test drive motor was not included in the study. The inlet flow velocity was adjusted to 2–8 m/s by adjusting different gas flow velocities to explore the relationship between the gas flow velocity and the change in the passive differential pressure of the fluid.

Subsequently, a dynamic pressure drop test of the PEPIR was performed. The piping system was equipped with an elbow section to investigate the changes in PEPIR during travel through the elbow. The specific test system arrangement is displayed in Figure 15 and the developed PEPIR prototype was used in this experiment. The prototype traveled at a speed of 1 m/s through a speed control system. After the prototype motor started traveling, the air compressor at the head end of the pipeline started generating gas for the pipeline system. The PEPIR was launched from the launcher and traveled through a 7 m long straight section of the pipe to the elbow. The PEPIR was made to travel through the elbow process with the speed feedback control to ensure the travel speed is 1 m/s. The real-time speed was solved by the mileage wheels, and the real-time speed signal of the output was fed back to the input in the form of negative feedback and compared with the desired speed value of 1 m/s. The control law used a proportional-integral (PI) controller with PWM technique to control the speed of the BLDC motor. Inlet flow velocity was maintained constant to maintain the speed of the prototype, and the relative speed of the gas flow velocity was constant to explore the prototype in the process of traveling through the elbow in the fluid differential pressure change law. Finally, the PEPIR traveled through a section of the straight pipe to reach the receiver.

Pressure sensors were present in both the launcher and receiver, and a flow velocity gauge was placed at the receiver end to record the changes in the gas flow velocity in the pipe. The PEPIR had differential pressure sensors to record the pressure difference between its tail and nose. The change in the gas differential pressure driving force of the PEPIR as it traveled through the piping system was obtained by analyzing the data from the differential pressure sensors. The mileage wheel recorded the speed and position of the PEPIR during its overall movement. Dynamic parameters, such as the speed and position of the PEPIR traveling in the piping system, pressure difference between the tail and nose of the PEPIR, and flow velocity in the piping, were collected using a multi-sensor approach. A kinetic analysis equation resolved the regular relationship between each parameter in the motion process.

4.2. Analysis of Experimental Results

In the static pressure drop experiment, no additional frictional force was generated owing to the flow clearance between the cup and pipe wall in the straight section. Therefore, the frictional forces of the PEPIR with different flow clearances were the same. When air with an inlet velocity of 7 m/s entered the pipe, the pressure differences between the tail and nose of the prototype with flow clearances of 10 and 20 mm were 4318 and 819 Pa, respectively. The prototype with the flow clearance of 5 mm overcame the static frictional force and started to travel at the inlet flow velocity of 3.5 m/s, and the pressure difference formed at the inlet flow velocity of 3.3 m/s was 4380 Pa.

By calculating the gas flow velocity and pressure difference between the tail and nose of the PEPIR, the pressure loss coefficient after calculation can be solved to obtain the pressure loss coefficients after conversion when the flow clearances of 5, 10, and 20 mm were 2.83, 2.39, and 1.69, respectively.

$$\xi ={\displaystyle \frac{2∆P}{\rho V_{gas}-{V_{pir}}^2}}$$

(13)

For the same frictional force of travel, a smaller flow clearance caused the PEPIR to travel faster; a smaller flow clearance results in a larger gas differential pressure driving force on the PEPIR. Under the premise that the gas flow velocity is certain, the difference between the gas flow velocity and the travel speed of the PEPIR will be sufficiently small to make the gas differential pressure force and traveling frictional force reach an equilibrium state. This is consistent with the PEPIR kinetic analysis law. This implies that a smaller flow clearance causes PEPIR to travel faster.

Table 1. The pressure difference of different flow clearance (Formula, Simulation, Experiment).

Parameter	1	2	3
Clearance (mm)	5	10	20
Cup diameter (mm)	390	380	360
Pressure coefficient (Formula)	2.74	2.45	1.94
Pressure coefficient (Simulation)	2.45	2.2	1.74
Pressure coefficient (Experiment)	2.83	2.39	1.69

lists the calculated, simulated, and experimental pressure loss coefficient equation values for different flow clearances. It can be observed that the overall theoretical values are high relative to the simulated and experimental values. This is because the theoretical calculations do not consider the interaction relationship and must be corrected. The overall simulation results are consistent with the resulting pattern demonstrated by the experimental results. Therefore, the simulation analysis can guide the design of a pneumo-electric hybrid-driven scheme.

From Figure 16 we can see the change in the pressure difference between the tail and nose of the PEPIR with different inlet flow velocities when the flow clearance was 10 mm. With the increase of the inlet flow velocity, the pressure difference between the tail and nose of the PEPIR also increases, and the differential pressure is directly related to the speed. When the inlet flow velocity increased to more than 8 m/s, the PEPIR began to travel because the differential pressure of the PEPIR was greater than the static frictional force. After the prototype was driven, because the dynamic friction was less than the static friction, the relative velocity between the gas and prototype decreased, and the pressure difference of the prototype was maintained in the straight pipe section at a value less than the starting pressure difference. During the travel of the PEPIR in a straight pipe, the travel speed and differential pressure values are always in the process of feedback regulation.

Figure 17 shows the prototype’s pressure difference and speed curves while traveling through the elbow. The PEPIR traveled through the elbow at a steady speed of 1 m/s with very slight fluctuations in the travel speed, which are analyzed by the speed and differential pressure values collected by the sensors.

At 8 s, the PEPIR started traveling into the elbow, and the differential pressure increased temporarily. This is because the tail cup of the PEPIR entered the elbow part, and the flow domain formed by the lower part of the cup and the wall of the elbow pipe changed, which affected the flow behavior and made the differential pressure rise. Subsequently, after the PEPIR fully entered the elbow, a small interference between the skin cup and the pipe wall occurred because of the support arm of the PEPIR, and the leakage structure formed was no longer in the standard symmetrical annular shape. The pressure difference of the PEPIR decreased during its passage through the elbow. This also indicates that the differential pressure of the PEPIR decreased and then increased during the passage of the elbow when the velocity of the prototype relative to the gas was constant. When the centroid of the prototype was at 45° to the elbow, the minimum pressure difference reached 2470 Pa, which was 45% lower than the normal value. This is consistent with the results of the law of differential pressure obtained when traveling through different positions in the numerical simulation. In the engineering case, the differential pressure force of the PEPIR decreased during the passage of the elbow; however, simultaneously, owing to the increase in interference between the power cup and pipe wall, the frictional force increased, and the speed of travel decreased owing to the imbalance between the driving force and frictional force. If a stable travel speed of the PEPIR is to be maintained, an additional driving force from the motor is required to maintain the overall dynamic balance. The results of the experimental and simulation investigations enabled the acquisition of gaps in the elbow drive differential pressure values to design a motor control scheme for traveling through the elbow.

5. Optimal Design of Pneumo-Electric Hybrid-Driven Scheme

Combining theory, numerical simulations, and experimental data makes it feasible to analyze the relationship between the effects of different skeleton diameters, flow clearances, and gas flow velocities on the gas differential pressure driving force of the PEPIR. This provides a direct reference for the design guidelines of pneumo-electric hybrid-driven schemes.

The design process first determines the operating conditions of the actual pipeline, including parameters such as pressure, flow velocity, length, and pipeline wall thickness. The normal travel frictional force value of the PEPIR at the wall thickness was obtained through a drag test. The gas pressure difference required at the preset speed was solved using the kinetic analysis model. In conjunction with the simulation analysis model, we defined the flow clearance suitable for satisfying the power requirements.

Simultaneously, for the inspection operation of new base pipelines, the pipeline operating conditions can be changed by adjusting the inlet flow velocity to find an operating solution with higher economy and travel stability.

The working mechanism of the pneumo-electric hybrid-driven scheme under different working conditions is shown in Figure 18.

The normal frictional force was measured using a dragging test in horizontal pipe sections with uniform frictional force. Hydrodynamic calculations determined the inlet gas flow rate. The driving force for the PEPIR was the fluid pressure difference, and the self-driven motor was not used in this state. In engineering applications, the effects of gas flow rate and pressure variations can be calculated by combining Equation (1) in this study.

In sections with abnormal frictional force, such as the upslope, elbow, and tee, the attitude sensor and speed measurement system determine the traveling status. At this time, the gas differential pressure force cannot satisfy the driving force requirements for the stable travel of the PEPIR. The control system supplies power to the self-crawling drive motor, and the self-driving motor starts working to provide power compensation to stabilize the PEPIR through the abnormal frictional force section.

The downslope/excessive-speed state is sensed by the attitude sensor/velocity-measuring wheel. The control system switches the motor into regenerative braking state and accesses the brake resistor to maintain PEPIR within a set travel speed range. Simultaneously, the kinetic energy is converted into electrical energy for storage. When the speed decreases, the motor starts providing kinetic energy again, transitioning from the recuperation state to the normal load state.

The passive drive of the fluid alone is always facing the problem of blockage and speed excursion in low-pressure gas pipelines. On the one hand, the annular leakage structure of this study can minimize the extent of sudden friction changes. For conventional gas pipelines, the gas velocity and pressure are more stable. The calculated design of the fluid drive enables the PEPIR to travel stably in the regular operating conditions of the pipeline. On the other hand, the annular discharge structure will still face the problem of insufficient power in the elbow and other sudden changes of friction, and then depends on the active drive of the motor to provide power compensation. By shortening the build-up time of the fluid pressure difference through motor power compensation, the PEPIR no longer needs the build-up of the gas pressure difference, and thus avoids speed excursion.

A comparison of the travel velocity curves of the prototype of this study and the real fluid-driven PIG is shown in Figure 19. As can be seen from the velocity profile, the traditional spherical cup fluid-driven PIG has obvious speed excursion after blockage during the process of traveling through the elbow in low-pressure pipeline. By comparison, the PEPIR does not require the accumulation of gas differential pressure by means of an active motor drive, which avoids the phenomenon of speed excursion after traveling through the section with a sudden change in frictional force.

This allows the PEPIR to travel constantly and ensures ideal passability. The pneumo-electric hybrid-driven mode exhibits travel stability and high passability under low-pressure working conditions. This solves the problems of speed excursion and blockage of traditional PIGs, and the PEPIR can ensure the successful implementation of intelligent pipeline internal inspection operations.

6. Conclusions

In this research, an intelligent internal inspection robot for pipelines with the application of a new hybrid pneumo-electric driven method is presented. First, a fluid pressure drop model of the PEPIR in the pipeline was established, and a numerical simulation analysis of the different factors influencing the pressure drop was conducted. The effectiveness of the proposed numerical model in characterizing the flow field properties was verified by comparing the numerical simulation results with the experimental results. The main conclusions of this study are as follows:

(1)

A fluid pressure drop model conforming to the structural features of the PEPIR was constructed using the building block approach, which can effectively characterize the influence law of different structural features on the fluid pressure drop.

(2)

A comparative analysis showed that the smaller the skeleton diameter of the PEPIR, the larger the fluid pressure drop, and the smaller the flow clearance, the larger the fluid pressure drop. It is also found that the gas flow velocity is linearly correlated with the fluid pressure drop. The results of the study fully demonstrate the design of appropriate structural parameters and operating conditions to realize stable travel of the PEPIR in the uniform section of the low-pressure pipeline travel frictional force.

(3)

Through the analysis of the flow behavior in the process of PEPIR traveling through the elbow, the fluid differential pressure tended to decrease and then increase. The differential pressure of the PEPIR is the smallest at the 45° of elbow, which was less than 45% of the value of the straight pipe section. This law clarifies the compensation scheme for the driving force gap when PEPIR passes through an elbow.

(4)

Through the combination of a fluid passive drive and a motor active drive, the stability of a PEPIR traveling in a low-pressure gas pipeline can be significantly improved, and the speed excursion phenomenon can be avoided. This improves the safety and quality of internal inspection operations. In the future, this technology can be combined with an air compressor to be applied to pipelines with no gas medium that have not yet been put into production or new pipelines, solving the difficult problem of internal inspection operations in the absence of a gas source.