Analysis of Model Predictive Control-Based Energy Management System Performance to Enhance Energy Transmission

Abstract

A supervisory control system using Model Predictive Control (MPC) has been designed to evaluate the efficiency of wind and solar power and is consistent with the cost function in the supervisory MPC optimization problem. A two-layer Economic Model Predictive Control (EMPC) framework has been developed and has improved results such as cost reductions compared to recent advanced methods. A speed Generalized Predictive Control (GPC) scheme intended for wind energy conversion systems was developed last year, with simulation results indicating superior performance over previous models. A Hierarchical Distributed Model Predictive Control (HDMPC) can work under different weather conditions with improved economic performance and keep a good balance between power delivery and load demand. An energy management system (EMS), built on the basis of MPC, can be quite lucrative for the sphere in the present climate scenario, with the selection and testing of suitable algorithms, controlled processes, cost functions, and a set of constraints as well as with proper optimizations carried out. Previous research indicates that an MPC-based EMS has the potential to be a good solution to manage energy well and also introduced it to the world experimentally. The key intention of this research study is to explore the existing advances that have been introduced and to analyze their performance in terms of cost function, different sets of constraints, variant conversion processes, and scalability to achieve more optimized operation of MPC-based EMS.

1. Introduction

Climate change has become an alarming issue for the present world, characterized by ongoing shifts in temperatures and weather patterns. These alterations can happen due to natural phenomena such as variations in solar activity or volcanic eruptions. However, since the 1800s, the actions of humans have been the principal reason for climate change, primarily through the combustion of fossil fuels like oil, coal, and gas. These fuels produce the emission of greenhouse gases which work as a layer all around the earth to raise temperatures. Here, the main greenhouse gases are carbon dioxide and methane. These are produced by the gasoline used for driving cars, gas emissions of industries, or coal to heat buildings, etc. Cutting down a huge number of trees is also a big reason for an increased amount of carbon dioxide. Farming, oil, and gas operations are the main sources of methane releases. Industry, energy, agriculture, transport, buildings, and the use of land are among the key matters causing greenhouse gases [1].

Renewable energy sources (RES) can be a massive relief for getting rid of these emissions and taking a U-turn from the current global warming era. These energies are derived from our natural resources, which can be refilled at a greater rate than they are spent. Sunlight and wind are such kinds of sources that are continually being restocked. So, sources of renewable energy are plentiful, and they are everywhere all the time. We need an energy storage system so that we can make good use of energy sources when we actually need them. Energy can be found in different forms such as radioactive, chemical, gravitational potential, electrical potential, electricity, elevated temperature, heat, and kinetic. Different techniques and technologies are available to collect and store different varieties of energy. The selection of technology is usually decided upon via application, finances, incorporation within the system, and most importantly, the obtainability of resources. To balance our energy usage concerning stored energy or energy production, we need an EMS—Energy Management System. An EMS assists in understanding the usage configuration and patterns; production and distribution decisions can be made depending on that.

While considering EMS as being a great boon for advantageous energy usage, integrating it with MPC brings about the most strategic performance from RES. However this integration has a lot of challenges, and not much research has been carried out to solve them. This paper proposes to point out the functionality of this integration. The key intention of this research study is to explore recent advances that have been introduced and to analyze the performance of MPC-based EMS in terms of the cost function, different sets of constraints, variant conversion processes, and scalability to achieve a more optimized operation of MPC-based EMS.

2. Energy Management System

Energy is mainly about having the capability to do work like moving an object by using some source of force. Energy can be of many different types, like kinetic energy, electrical energy, mechanical energy, nuclear energy, etc., but all energies fall into two basic forms: kinetic and potential [2]. Another matter of fact is all the many kinds of energy resources we have fall into another two criteria, which are renewable or nonrenewable energy sources [3]. Here, renewable energy sources will be discussed, as these are the ones that are needed for the energy management system that is being analyzed.

2.1. Renewable Energy Sources

The sources of renewable energy are sustainable energy, which is not going to finish, or is limitless, like what we get from the sun—solar energy. There is another term called ‘alternative energy’ which also commonly refers to renewable energy sources, which means energy sources that are substitutes for most usually consumed non-sustainable sources such as coal. Presently, the most common sources of renewable energy are as follows:

• Solar energy;
• Wind energy;
• Hydro energy;
• Tidal energy;
• Geothermal energy;
• Biomass energy [4].

The energy is derived from natural sources that can be refilled at a better rate than they are spent. Sunlight and wind are such kinds of sources that are continually being restocked. So, renewable energy sources are abundant, and they are everywhere all the time. On the contrary, fossil fuels such as oil, coal, and gas are non-renewable resources that take millions of years for their formation. Also, when they are burned to produce energy, they originate detrimental greenhouse gas releases like carbon dioxide. However, the generation of renewable energy forms far fewer emissions than burning fossil fuels. The switch from fossil fuels, which are presently responsible for a huge portion of emissions, to renewable energy is significant for taking into account the environmental contingency. An important fact is that renewables have become cheaper in many nations and can perform three times more work than fossil fuels [5].

2.2. Energy Storage System

But if we wish to use renewable energy, we need to store it first. That is how it will become beneficial for us. And to do that we need an ESS—Energy Storage System. An energy storage system is a system for transforming electrical energy from power systems to a form that can be stored, changing it back to electrical energy depending on when it is needed [6].

An energy storage system is basically a set of procedures and tools that are used for storing energy [7]. Later on, the energy that is stored can be taken from it to do suitable jobs. Lots of renewable energy resources (such as wind, solar, or tides) are occasional. There are situations when the consumption of renewable energy is not directly when the energy is obtainable. Those are the times we need energy storage systems so that we can make good use of energy sources when we actually need them. Energy can be found in different forms such as chemical, radioactive, gravitational potential, electricity, elevated temperature, heat, and kinetic. Different applications, techniques, and technologies are available for storing different types of energy as we can see in Figure 1. The selection of technology is usually decided upon via application, finances, incorporation in the system, and most importantly the obtainability of resources. The working mechanism is shown in Figure 2. Systems of energy storage are moreover used in transforming energy from forms that are challenging to store into forms that are easier or more economical. Five vital storage technologies of energy are as follows:

• Electrochemical Energy Storage;
• Chemical Energy Storage;
• Thermal Energy Storage;
• Mechanical Energy Storage;
• Electrical Energy Storage [8].

Figure 1. Energy storage applications and technologies.

Figure 1. Energy storage applications and technologies.

Figure 2. Working mechanism of energy storage system.

Figure 2. Working mechanism of energy storage system.

To balance our energy usage with respect to stored energy or energy production, we need an EMS—Energy Management System [9]. An EMS assists in understanding usage patterns, and production and distribution decisions can be made depending on that.

2.3. Energy Management System

An energy management system (EMS) is basically a system of tools supported by the computer that is being used by the machinists of power utility grids for observing, controlling, and optimizing the production performance or broadcasting system like Figure 3. Moreover, it can also be utilized in small-scale systems such as microgrids. Due to the charging of Electric Vehicles (EVs) becoming more common, smaller domestic devices that are widespread nowadays can decide when an EV can charge depending on the overall load versus the overall capacity of an electric service [10].

An energy management system (EMS) is a structure for energy consumers, involving industrial, trade, and municipal organizations, for managing their use of energy [11]. If we think about healthcare facilities, there are three principal components for having an efficient strategy of energy management for these, which are hazard management, efficacy, and ecological sustainability.

Top Five Ideas for Effective Energy Management

• Recognize resources of energy consumption;
• Collect utility fee information;
• Investigate meter data;
• Detect prospects to conserve costs;
• Trace progress.

2.4. Battery Energy Storage System

An integral part of EMS is battery energy storage systems (BESS); these are devices that facilitate energy from renewables, such as solar and wind, to be stored and then freed depending on consumers’ needs [12].

A battery energy storage system is a sub-set of energy storage systems that makes use of an electrochemical solution [13]. Figure 4 shows BESS with its primary power components. A BESS provides an easy way to gather energy and store it to be used later, or provide power to an off-grid application, or complement a peak in demand. It is not being used to replace grid power completely in general but offers temporary responses in applications where access to grid power is intermittent or the use of a generator is inappropriate because of noise or pollution problems. Moreover, an ESS can also manage the energy that is generated by intermittent resources, like solar panels. Different kinds of battery technologies are available to be used in BESS and Figure 5 shows its architecture. However, over recent years, solutions using Lithium-ion batteries have been popular, because of the advantages of their energy storage system, such as their extended working life, greater operational range, lightweight structure, great energy efficiency, and importantly, the declining price of the technology. These, together with the low total cost of ownership and sustainability, make them attractive for numerous applications [14,15].

2.5. Model Predictive Control

While we talk about improving the battery life, range capability, and energy efficiency of the energy management system and also making it more economical, we need to consider the assistance of a model predictive controller. Model predictive control (MPC) is an innovative technique of controlling processes that are used to control processes while also satisfying a set of constraints [16]. MPC is mainly used to predict the forthcoming conduct of the controlled system through a restricted duration and calculate the input of an optimal control that reduces a previously determined cost function, while also confirming the satisfaction of determined system constraints.

To be more accurate, MPC is a very favorable control strategy that is based on numerical optimization [17]. Upcoming control inputs, as well as forthcoming responses from the plant, will be estimated using a system model as we can observe in Figure 6. It will be optimized on a well-ordered time interim basis depending on a performance index. MPC can improve controlled performances within a process; as a result, its predictive control method has become very popular in industries. MPC is also able to provide stability, optimality, and robustness to a system [18]. We can see the various useful components of MPC as shown in Figure 7.

2.5.1. Supervisory Control System MPC

The research in [19] focuses on developing a supervisory model predictive control strategy for the optimized management, as well as the operation, of hybrid standalone wind–solar energy generation systems. They have designed a supervisory control system using MPC that computes the references of power for the solar and wind subsystems at every time sample, while also reducing an appropriate cost function. The references of power are sent to a couple of local controllers that drive the pair of solar subsystems to the desired references of power. They have included some discussion of the integration of empirical contemplation, like minimizing the values of inrush or surge currents which are at their peak, into the construction of the optimization problem of MPC [19].

MPC is a famous control method for its capacity to justify the input constraints and state and optimality matters in a clear and detailed manner in the assessment of control actions. A system model is used by MPC for predicting the future evolution of the system at each sampling time from the present state within a given prediction horizon. Using those predictions, the input/setpoint trajectory that minimizes a given performance index over a finite time horizon is computed in their work, to bring about an optimization problem that is subjected to constraints and more suited. They have presented simulation results that have shown how effective and applicable the proposed approach is. They planned to introduce an investigation of the long-period behavior of a hybrid wind-solar generation system, and at the same time consider information on upcoming weather forecasts and the performance of the investigated system under the conditions of the forthcoming power demand, in their further studies [19].

Later in their further research publication [20], they showed that to be compatible with the cost function in the supervisory MPC optimization problem, the mean performance cost of system operations every hour can be determined with the given equation:

$$\begin{array}{ll}J=& \frac{1}{Mt-Mi+1}{\displaystyle \sum _{i=Mi}^{Mt}}(\alpha {|P}_d^{for}(t|ti)–P_w^{ref}(t|ti)\\ & -P_s^{ref} (t|ti)+i_{b_1}^{ref}(t|ti) E_b+i_{b_2}^{ref}(t|ti)E_b|\\ & +{\beta }_1 \overline{i}{b}_1{\left(i\right)}^2+{\beta }_2\overline{ i}{b}_2{\left(i\right)}^2\\ & +{\gamma }_1 \overline{d}{b}_1{\left(i\right)}^2 + {\gamma }_2\overline{d}{b}_2{\left(i\right)}^2+\delta {\overline{P}}_s(i))\end{array}$$

Here, every term following the summation symbol correlates to a term in the cost function (of the equation below): sections may be divided by subheadings. It should provide a concise and precise description of the experimental results and their interpretation, as well as the experimental conclusions that can be drawn.

$$\begin{array}{ll}J\left(tk\right)=& \mathsf{\alpha } {\int }_{tk}^{tk+N}|P_d^{for}(t|t_k) - P_w^{ref}(t|t_k)-P_s^{ref}(t|t_k)\\ & +i_{b_1}^{ref}(t|t_k) E_b+i_{b_2}^{ref}(t|t_k)E_b|dt\\ & +{\beta }_1 {\int }_{tk}^{tk+N}{i}_{b_1}^{ref} {(t|t_k)}^2 \mathrm{dt}+{\beta }_2 {\int }_{tk}^{tk+N}{i}_{b_2}^{ref} {(t|t_k)}^{2}dt\\ & +{\gamma }_1{\int }_{tk}^{tk+N}{d}_{{\mathrm{b}}_1}{\left(t\right)}^2\mathit{dt}+{\gamma }_2{\int }_{tk}^{tk+N}{d}_{{\mathrm{b}}_2}{\left(t\right)}^2\mathit{dt}\\ & +\delta {\int }_{tk}^{tk+N}{P}_s^{ref}(t|t_k)dt\end{array}$$

with the weight factors remaining the same. Another fact is that high-frequency fluctuations are not considered in the cost evaluation. We can see in their works that the lowest performance cost is provided by the centralized MPC. With the increase in iteration number, the iterative MPC’s performance cost reduces considerably and coincides with a value that is marginally more than the value that corresponds to the sequential MPC. This can be seen from the reversed sequential MPC, in which case they adopted a solar–wind sequence; it does not, however, imply that the sequential MPC is generally bigger than the iterative MPC. It was not possible to guarantee that the iterative MPC would produce a performance cost that would converge with the centralized MPC, because of the significant nonlinearity and non-convexity of the MPC problem [20].

2.5.2. Wind Turbine MPC

A model predictive controller is greatly involved in the research of advanced wind turbine control algorithms. Its predictive properties and the explicit incorporation of restrictions in the control issue are the primary advantages over other control systems. The control legislation that is produced is also ideal for the selected goal. According to [20], a PWA model was able to represent the wind turbine dynamics’ strong nonlinearity. The MPC was designed using the generated model. The obtained simulation results have presented better performance compared to the baseline controller. As a matter of fact, a longer prediction horizon and the inclusion of wind speed prediction in the optimization problem led to significantly improved MPC outcomes. However, it also gave rise to another challenging optimization issue [21].

However, if we talk about wind turbines in general, the designing of wind turbine controllers is quite tough. The controller needs to have a good optimum power-tracking ability and tower stability and also needs to perform good component load mitigation and fault detection and compensation. To overcome the control issues, the implementation of model-based predictive control design approaches in wind energy applications was designed in [22]. The capacity to manage restrictions such as pitch rate and rotor speed, as well as the methodical design for managing the Multiple-Input–Multiple-Output (MIMO) wind turbine control problem, are the primary advantages of MPC in wind turbines. They have summarized the benefits of MPC in wind turbines as a performance in optimal power tracking and load mitigation on the tower that has been shown to be improved meaningfully compared with the PI controller, a switching solution provided for a shift and evasion of over-speed of the rotor, and the motion of the floating turbine tower dampened negatively. Upcoming research is likely to be needed on the use of the MPC design approach in wind turbine control [22].

In [22], for the wind energy conversion systems, a speed GPC scheme is provided. Their main objective was controlling the wind turbine following the Maximum Power Point-Tracking (MPPT) algorithm. To achieve that, they maintained the power coefficient at its ideal level and used a GPC regulator to modify the turbine’s speed in response to the wind speed. As a result, throughout a range of wind speeds, the wind turbine was able to operate at its full power efficiency thanks to the designed control system. The constraints on the rotor windings were integrated into the GPC through the Truncated Newton (TNC) optimizer in the conceptual control design process. The simulation results have validated that the performance of the proposed regulator in controlling the variable speed wind turbine is superior to that of the current techniques. For example, the Doubly-Fed Induction Generator (DFIG) velocity’s settling time when utilizing the suggested GPC controller stayed between 30 and 110 ms faster than when the Proportional Integral (PI) controller was used, throughout a range of wind speeds [23].

2.5.3. Hierarchical Distributed MPC

Furthermore, a Hierarchical Distributed Model Predictive Control (HDMPC) is a great way to manage large-scale, geographically dispersed, networked systems because it uses a distributed hierarchical structure to allocate various control objects/functions to different levels. An independent wind, solar, and battery hybrid power system is an example of a standalone microgrid, with numerous control task difficulties. Establishing a merged dynamic and economic performance model of a standalone wind–solar–battery hybrid power system, and then constituting a highly efficient HDMPC and using coordinated subsystem optimization, the suggested HDMPC in [24] recognized distributed energy as plug-and-play. Their simulation works under different weather conditions and has demonstrated a clear improvement in economic performance, while maintaining a stable load demand and power delivery ratio. The experiment further demonstrated its applicability. This solution can become a successful solution for controlling a complex microgrid that has clearly shown random fluctuation and intermittency [24].

They aimed to do some future research on the robustness of the overall closed-loop hybrid standalone wind–solar power generation system, which was motivated by the intermittency and uncertainty of wind and solar resources. HDMPC research can expand the resilience of the well-known linear MPC to the nonlinear hierarchical and distributed control frame situation in wind–solar power generating systems. Their current study takes into account the erratic and unpredictable nature of solar and wind energy to ensure the resilience of the suggested Hierarchical Distributed Model Predictive Control (HDMPC) [24].

2.5.4. Economic MPC

Moreover, one technique that works well for controlling isolated microgrids is economic model predictive control (EMPC). EMPC-based microgrid energy management systems (EMS) have provided performance improvements compared to regular methods. Even though centralized EMPC-based primary control algorithms have not been suggested previously for isolated microgrids, this kind of strategy can lower operating costs more than current approaches by adapting more economically to transitory events like solar array shadowing. A novel two-layer EMPC framework has been developed that is well suited for the microgrid control problem, with an experimental demonstration of the developed control framework as a controller for an isolated microgrid; EMPC was utilized by the energy management system and the principal control layer, and a comparison of the produced controller’s performance with that of previous techniques was conducted [25].

2.5.5. Residential/Building Management System MPC

After a brief hopeful discussion on improving the operating cost problem, let us focus on the conversion loss, which is a principal concern when we talk about EMS. To lower the conversion losses in the home distribution system, an innovative microgrid energy management plan has been devised. Its purpose was to determine the power strength on the DC side before transmission using a new algorithm. Only when the power was sufficient was the conversion process activated; if it was determined to be insufficient, it was stopped and placed in an auxiliary battery. Weak power conversion would result in significant loss across the transformers and converters. In this arrangement, solar PV and an auxiliary battery bank powered the DC loads, while the utility grid supplied the AC loads. The power conversion was only carried out under unavoidable situations. The solar PV-equipped hybrid AC/DC microgrid model has been designed and validated [26].

Furthermore, a hardware prototype configuration has been created and put through testing in a lab to show how crucial the intended approach is. Through the use of an automatic centralized microgrid controller, a conversion loss reduction strategy and efficient energy management has been achieved. The effectiveness of the suggested scheme in comparison to the current technologies has been shown through a comparative examination. Following the use of the suggested distribution strategy, the following victories were attained:

• Feeble power was coped with efficiently and conversion losses have declined meaningfully.
• Multiple conversion processes were reduced as a result of efficient battery-supported energy management, and battery energy efficiency increased by 8–10%.
• An enhanced energy economy is the outcome of the clever management of the demand response, BESS, and renewable resources.
• The energy usage in the microgrid context was optimized by employing a genetic algorithm, and the cost of electricity was reduced meaningfully [26].

Forecasting building parameters is a critical and challenging part of MPC since a building’s thermal model is nonlinear and linked with some issues. A learning-based MPC strategy was introduced to manage the thermal property of a four-zone office building. An Artificial Neural Network (ANN) is incorporated with the model-based control approach to address the issues. The occupancy profile predictions are generated through the ANN, then this data is supplied to the model-based controller. Energy Plus 9.4 software was used in this work to simulate a building with real materials and components and to test the proposed approach on it. The proposed learning-based approach showed a notably better performance in maintaining the residents’ comfort and reducing energy usage (40.56% energy savings), compared to the conventional MPC [27].

An intelligent control strategy for households based on MPC was proposed, with an ESS with a high improbability of electricity price, load demand, and PV generation, while considering economic dispatch and the charge and discharge of ESS interactions with the external grid and the curtailment of loads. The strategy shows a reduced operational cost of the microgrid in comparison with the day-ahead strategy. This intelligent control strategy shows much stability, with a rise in the forecast error level [28].

When it comes to a home energy management system, a policy is necessary that brings comfort to the residents and also keeps the cost at a good balance. But house thermodynamics, incorrect predictions in the forecasting of PV generation, outdoor temperature, and the load demand of users make this challenging. An MPC-based RL approach is an impressive tool for handling the home energy management system problem with model errors and system uncertainties. By parameterizing the model, cost function, and constraints of MPC and training the parameters by RL, an optimal policy that satisfies both economy and comfort was obtained. This approach has the advantages of a higher sampling efficiency, the easier incorporation of predictive information, easier realization of constraints, higher robustness, and better cost-effectiveness contrasted to the typical TD3 algorithm [29].

2.5.6. Railway EMS-MPC

If we think of the real-world scenario for MPC to be able to actively work or not, then we need to appreciate the paper [30], which presents a railway energy management system based on the hierarchical coordination of energy flows from electric traction substations and on-route train energy usage. According to their proposed method, the railway system was divided into two levels: energy-efficient specific train energy intake control at the bottom, and cost-effective electric traction substation energy flow management at the top. The levels were synchronized using parametric hierarchical MPC to enhance energy efficiency and lower the entire system’s operating expenses. Through higher-level interactions with the electricity grid, the system was able to provide ancillary services and respond to a variety of grid requests. Lower-level train drivers were accustomed to operating the system at the lowest possible cost while adhering to timetables and on-route constraints. The created algorithm was validated in a realistic real-world case study scenario involving a railway operator and a train manufacturer. The final results have shown significant cost and energy consumption reductions that were achieved by the coordination of many trains supplied by the same traction substation at the same time [30].

2.5.7. Microgrid EMS-MPC

MPC-based EMS architecture is a promising method for controlling and optimizing the power flow in microgrids. The framework maximizes the use of renewable energy sources, including solar photovoltaic, wind turbines, and energy storage systems (ESS), to enable a more sustainable and efficient energy generation and distribution. In one study, microgrid management operational expenses were decreased and prediction accuracy increased with the use of an MPC-based EMS [31]. Another study gives hope that, when the load needs to exceed the capacity of the battery converter or when the stored battery energy is insufficient to meet demand during peak hours, the MPC strategy can be applied and the grid can be adjusted to act as storage to absorb any extra PV energy [32].

Renewable energy sources bring a lot of possibilities and advantages to the energy management system, but they also bring along some drawbacks. They bring some risks or unpredictability, such as cost, reliability, etc., when associated with MPC. A robust MPC framework can be a solution to that. A mixed-integer programming model for the energy management of isolated microgrids was first proposed to reduce the cost of integrated economic operation, and then a robust equivalent of this model was derived. Next, an MPC framework with an online energy-scheduling strategy was proposed for the uncertain risks of energy management, and it was found that it could enhance the reliability of the operation management of an isolated microgrid and decrease the operating cost of the system, compared to the traditional MPC framework [33].

Computational time and scalability are also a matter of concern, besides cost, reliability, and energy generation, when it comes to an MPC energy management system with microgrids. By decoupling the unit commitment (UC) and economic dispatch (ED) complications and resolving them independently, and also taking the control scheme into account for the current and predicted prices of electricity, the forecasts of loads, and the availability of renewable energy, the concerns can be diminished. The proposed method performed better and was able to handle the constraints in terms of computation time and scalability, with a slight trade-off in cost minimization, while respecting the constraints of the microgrid in comparison with conventional ones within the MATLAB environment [34].

Along with the discussed problem, forecasting error is also a major issue for an MPC-based EMS for a microgrid. As this method is gaining a lot of focus nowadays and being considered as an optimal solution, solving forecasting errors has become crucial. By generating forecasting errors in a simulation environment, it was revealed that the system vaguely differs from the optimal operation when forecasting errors exist. However, researchers emphasized developing better formulations to measure the deviation from the optimal operation [35].

Moreover, predicting the disturbances that impact the operation of a microgrid is crucial. Good predictions play a vital role in the operation of a microgrid using an MPC-based EMS. A state–space model designed using an MPC-based EMS, while including renewable generation and user demand prediction, can affect prediction performance. After proper analysis and testing of the methodology on an experimental renewable energy-based microgrid platform which is based on renewable energy sources and hydrogen storage, simulations under unlike environments demonstrate that the integration of good predictions can enhance the operation cost of running the microgrid. So, the development of good forecasting methods can reduce operational costs, and the prediction capabilities of MPC make a good way to improve microgrid operation [36].

Furthermore, because of its many useful policies, Photovoltaic (PV) has become very popular among households. A grid-connected residential microgrid model consisting of PV, battery, water tank, cleaning pump, HVAC, lighting, and other electricity usage with an advanced energy management strategy based on MPC was proposed to reorganize power consumption and attain a reduction in energy costs, as well as enhancing self-consumption ability through load shifting. A microgrid platform was deployed that included an Open Platform Communication technology-based communication system to ensure the real-time operation of advanced energy management strategies. The experimental outcome confirmed the validation and possibility of the proposed MPC-based energy management strategy [37].

A management and control strategy was applied in a housing microgrid, while satisfying the required demand and amplifying the economic benefits, lessening the degradation and rigorous use of energy storage systems by extending their valuable life, minimizing the energy exchange with the main grid, and respecting the operational constraints of the entire system. The demand response had a significant impact on electricity consumption decline, around 20.4% through peak hours, and showed that it can be a beneficial substitute for both prosumers and utilities. It can provide support in upgrading the energy management of microgrids, including several components and electrical loads. However, the further integration of new strategies, other energy sources like hydrogen, and other storage systems should be conducted, as microgrids operate with improbabilities in demand and renewable energy sources. [38]

The authors of [38] also employed an MPC-based EMS integrated with a fault detection and isolation system to resolve the fault occurrence in the system control input and to keep the system security unharmed. Solutions were brought by minimizing consumption using demand response concepts like load curtailment and reconfiguration of parameters. The proposed strategy using a real microgrid consisting of a RES, local electrical grid, and energy storage systems illustrates better initial results. A fault was identified but did not bring drawbacks to the operational performance of the microgrid, confirming that the microgrid can operate at reduced electricity demand for an extensive duration until regularity is recognized [39].

2.5.8. Power Grid MPC

The limits of conventional controllers and the difficulties in integrating intermittent renewable energy sources into the electrical power grid have been discussed in this [40] paper. MPC has been shown to be a workable way to get beyond these obstacles. After a thorough analysis of MPC, an algorithm was created to maximize the performance of renewable energy sources in the electrical grid. After that, several simulations were conducted to assess and examine the suggested controller and system. The results have shown that MPC has a lot to offer when it comes to flexibility, optimization, and integrating renewable energy sources into the electrical grid [40].

2.5.9. Smart Grid MPC

The restrictions or drawbacks of conventional methods are many but also tried and tested. We need to include digital technology in our power systems to optimize efficiency and improve decentralized energy production and consumption in a way that is called the smart grid. The authors of [41] proposed an MPC-based optimization strategy for a user-side energy management system and applied their strategy to the user side in a smart grid in Beijing. Simulation results verified the feasibility and effectiveness of the optimization strategy, as it can completely synchronize real-time electricity charge, user demand, and renewable energy forecast information, and can understand energy storage charging and discharging management and unit optimal output control, besides achieving the lowermost overall cost of the system [41].

We need to consider an MPC-based smart grid for a home energy management system, with all its uncertainties, as well as predictable input, such as the real-time price, weather conditions, and activity model. With the effort to be more realistic, studies depict that a 12% energy cost savings can be achieved by the proposed model without compromising customers’ ease. The MPC-based HEMS model is computationally competent and demands less than one minute for a 24-h MPC run, which makes it appropriate for applying practically [42].

2.6. Geopolitics–EMS

Some good advancements have happened in energy storage applications in the power supply systems of Ukraine. It is necessary to recognize significant areas for energy storage systems’ further development, including the current scenarios in Ukraine. The trend of increasing the demand for integration of ESS in Ukraine’s power systems was examined. The timely verification for mode interaction in the interfaces between the transmission system operator and distribution system operator has a problem. An approach to line capacity management based on power control of electrical energy storage (EES) for distribution system operators was proposed. Regulating the active power flow and ensuring dynamic stability under the emergency of the power system became possible with the choice of location. With the selection of EES parameters and locations in the electrical distribution network, a simulation of the Ukraine distribution system operator was conducted; it shows good practical results in increasing the technical and economic performance of the power supply system. However, some technical solutions are essential for the improvement of the technical and economic efficiency of the power grid [43].

The economy of Ukraine is quite based on energy. Updated equipment and technologies are crucial to improve productivity, as it reflects on notable economic losses while purchasing energy resources; and using conventional energy resources provides an ecological burden on the country’s ecosystem which can be solved by simulation with RES. An independent approach should be used for developing alternate energy while including the current decentralization reform and climatical irregularities of the Ukrainian regions to enhance the investment of potential investors. These approaches to the state policy on stimulating the use of RES will ease the load on the ecosystem, shrink the state and local expenses, and improve energy efficiency. Nonetheless, effective implementation and additional research are highly necessary to confirm effective cooperation between the regions, find reserves for growing electricity exports from RES, and strengthen energy efficiency. Also, state policy on renewable energy needs to be considered as a complex task, and improvement in this area is significant [44].

Energy efficiency and energy savings have become a principal concern in Ukraine because of the current geopolitical situation. Implementing an energy management system has become more urgent. Some advancements in the EMS concerning public services in EU countries like designing low-energy buildings, improving buildings, the installation of solar power plants, substituting heating systems, and encouraging energy consumption behavior amongst inhabitants were studied and the trend of energy management implementation was examined. Following the signed EU–Ukraine Energy Efficiency Directive, priority goals were recognized. With Ukraine’s integration into the EU’s political and economic structures, the implementation of energy management became a requirement under the signed European Parliament Directive 2012/27/EU [45]. The priority objectives of the signed EU–Ukraine Energy Efficiency Directives are as follows

i.

The state takes economic, practical, and sensible actions designed to quicken energy conservation.

ii.

Set interim energy-saving goals for the third year of the Directive and give a review of its strategy for achieving the interim targets.

iii.

Designation of one or more new or current bodies or agencies that have overall control and responsibility for oversight and establishing the framework related to the indicators.

iv.

After reviewing and reporting on the first three years of application of this Directive, the Commission will examine whether it will be necessary to put forward a proposal for a Directive for further development.

Consequently, the Law of Ukraine No. 1818-IX “On Energy Efficiency” of 21 October 2021 [46] was signed in Ukraine based on the objectives of the signed Directive, according to which the country will implement a modern European instrument for implementing energy efficiency policy and an energy management system. At the local level, municipalities are developing energy-saving programs (energy efficiency improvement). EU grant programs for Ukraine were also examined to facilitate the implementation of an energy management system. We need to take EMS in private enterprises in the EU and Ukraine into account to analyze further the best practices of the EU countries to implement positive best practices in the field of energy management [47].

The globally second-largest oil producer and natural gas exporter, Russia’s war with Ukraine which began on 24 February 2022 has sternly affected the energy market to such an extent that crude oil prices have gone high. Though it negatively affected the global economy through some other sectors too, the energy market was hit the hardest. With Russia being a significant player in the global economy, the change in oil prices has been a matter of investigation for both industry and academic sectors [48].

Using an event analysis based on variational mode decomposition (VMD), from 1 October 2021 to 25 August 2022 as the event window, it was found that the war and its chain of events caused the West Texas Intermediate (WTI) crude oil prices to increase by USD 37.14, a 52.33% surge, and the Brent crude oil price to rise by USD 41.49, a 56.33% increase. On 7 March 2022, the WTI crude oil futures price reached USD 133.46/barrel, and the Brent crude oil futures price reached USD 139.13/barrel, the highest price since July 2008. Since then, crude oil prices have remained consistently high, only experiencing short-term fluctuations during some negotiations and proceedings [49].

To quantitatively analyze the effect of the Russia–Ukraine war on crude oil prices, an EMC analysis framework was made. The analysis window was set from 24 June 2021 to 27 October 2022, with the event window from 24 February 2022 to 27 October 2022, and the estimated window from 24 June 2021 to 23 February 2022. This framework was viewed as appropriate for determining the net impact of unexpected and temporary intense events, such as wars and geopolitical conflicts, on commodity prices, yet much more sensitive considerations needed to be given to use this methodology. When unstable situations happen for a long duration and prompt multiple chain reactions, such as the COVID-19 pandemic, it is necessary to set out multiple event windows which will be gauged independently to guarantee accurate measurements and analysis and define the research objects. The results have shown an ability to obtain only the lower limit of the impact of the war on crude oil prices. Though the causal testing showed trivial impacts of other factors, the actual impact still went beyond the value calculated in this paper [49]. Moreover, building a technique to put these factors together and conducting a comprehensive assessment of the long-term impact of the Russia–Ukraine war on oil prices is also vital [49].

Modern Industry 4.0 offers a substantial number of tools and prospects for the competent and secure use of energy; tools like blockchain, automation, coding, and cloud storage are among the most promising. Industry 4.0 means the fourth industrial revolution, stated as a new level of organization and control over the entire value chain of the life cycle of products. Though the concept is still speculative, it is a realistic concept consisting of the Internet of Things, Industrial Internet, Smart Manufacturing, and Cloud-Based Manufacturing. For the strict integration of humans in the manufacturing process to show continuous improvement, focusing on value-adding activities and avoiding waste are the main concerns of Industry 4.0 [50]. Paper [51] examines theoretical, methodological, and analytical approaches for the effective implementation of energy storage systems in the energy sector of Ukraine as a feature of Industry 4.0. It was shown that integrating ESS into the modern technologies of Ukraine would bring territorial circulation of energy resources; the deployment of energy legislation; the dynamic use of digital tools based on the Industry 4.0 system; and the lessening of energy use, minimizing costs and maximizing savings. ESS as an element of Industry 4.0 for Ukraine’s energy sector will meaningfully enhance people’s lives and improve the country’s energy sector, establish new companies, and bring light to new technologies in the energy sector [51].

3. Challenges

Though it is very beneficial for the environment and also economic for us to use RES, there are some challenges to making it happen wisely. Some key challenges in making worthy use of renewable energy sources are their storing processes, the selection of parameters for suitable management patterns, and achieving a good balance with respect to usage and storage. Different techniques and technologies are available to store different forms of energy. The technology choice is usually decided by their applications, finances, incorporation in the system, and most importantly, which resources are available or not.

An EMS based on MPC can become more lucrative in the present climate scenario, if suitable algorithms, controlled processes, cost functions, and a set of constraints can be nominated and tested, as well as if proper optimizations are carried out. Some research has been conducted in the past years to make an MPC-based EMS a good solution to manage energy well and to introduce it to the real world experimentally. But not one of these provides an actual solution or analyzes the performance by investigating it as a whole.

A supervisory control system based on MPC has been designed to compute the references of power for the solar and wind subsystems at every sampling time while reducing a cost function that is suited. Although it has been consistent with the cost function in the supervisory MPC optimization problem, the iterative MPC was not guaranteed to give a performance cost converging with the centralized MPC, because of the strong nonlinearity and non-convexity of the MPC problem. Predictive properties and the explicit incorporation of restrictions in the control issue are the primary advantages over other control systems. The high nonlinearity of wind turbine dynamics was approximated with a PWA model, and the simulation results have also presented better performance compared to the baseline controller. However, better results with the MPC are achievable using longer prediction horizons and integrating wind speed prediction in the optimization problem. But this will bring up an additional difficult optimization problem. Further work on it needs to be done. A speed GPC scheme for wind energy conversion systems was worked on last year, and valid simulation results were found to provide better performance than the existing ones, but the experiment was limited to simulation only.

Hierarchical Distributed Model Predictive Control can be understood as the plug-and-play of distributed energy and can also work under varied conditions of weather; it has shown obvious improvement in making it more economical, while also keeping a proper balance between power delivery and load demand. This solution can become a promising solution if the robustness of the intermittency and uncertainty of wind and solar resources is investigated. A hardware setup prototype has been made and weak power was coped with, multiple conversion processes were reduced, and battery energy efficiency was improved, but the battery energy efficiency was improved only by a little amount.

Some outcomes have shown good reductions in cost and energy consumption, through the simultaneous coordination of several trains which were from the same traction substation, and quite a low cost of system operation with timetables and on-route constraints were found. However, it cannot reach the actual cost reduction matter. A two-layer EMPC framework has been developed, and the proposed algorithm has demonstrated a considerable improvement in performance in every instance. The proposed EMPC algorithm only achieved quite low reductions in contrast to the current state-of-the-art methods. Future research is likely to be needed on the use of an MPC design approach for EMS to make it cost-effective for human beings, concerning less energy consumption, a significant improvement in battery energy efficiency, a single conversion process to offer all facilities, like understanding the production and consumption balance and having more storage even after transforming different types of energy, as well as being prepared even in uncertain conditions.

Some other research was also done on different types of grids such as microgrids, power grids, and smart grids, to investigate the current scenario of uncertainties of EMS-based MPC, which has shown the importance of selecting better optimization strategies, fault detection, and isolation systems, and also integrating different energy storage systems, as well as variant energy resources, to make a more stable and user- or customer-friendly energy management system.

The challenges of this research can be coped with by using multiple dynamic cost functions, energy resources, optimization strategies, a choice of algorithms, fault detection methods, and variant sets of constraints to find out the finest conversion process and improved battery performance, for a scalable MPC-based EMS which would be able to balance between load demand and power delivery, with the ability to have a noteworthy fault avoidance.

4. Future Research Directions

Energy is mainly about having the capability to do work, like moving an object by using some source of force. Energy can be of many different types, like kinetic energy and electrical energy. If we can become able to use the energies that surround us as efficiently as possible, we will be able to bring a lot of solutions to the foregoing climate crisis.

Different types of cost functions, fault detection strategies, and variant optimization algorithms on multiple operating parameters have been proposed and worked on by different researchers, to find out the actual solution for a much-improved EMS with an efficient BESS, while also incorporating MPC. Different types of grids were also discussed, along with global energy problems and the geopolitical situation of the energy sector.

Since the issues are still not resolved, they need further investigation, and this is why the objectives of analyzing the performance of MPC-based EMS have been taken as a principal concern. Presently, an analysis based on different types of algorithms is available, and not one is there to be trusted completely for an overall improved performance. So, here the cost function of MPC-based EMS technology needs to be examined by selecting different and dynamic parameters and variant ranges that need to be used to check how much-improved efficiency can be found with the high range capacity of different systems like wind turbine systems, solar systems, battery power system, and battery energy storage system. First, different systems with present scenarios should be checked, and then an analysis should be conducted.

An MPC-based EMS strategy should be developed that can effectively predict power usage and predictable inputs and fault disturbances, effective algorithms to make energy transmission effective and efficiently check the performance of the model, and investigations should be performed on how the model works compared to the present findings. The performance of the model needs to be checked under real circumstances to see how it works, using different control processes and under different sets of constraints.

5. Conclusions

This paper discusses how and why MPC-based EMS technology should be investigated and provides detailed steps on how to achieve that. Having discussed all the strengths and limitations of MPC-based EMS, we can reach a point where the main expectations of MPC-based EMS can be met if the choice of the cost function, set of constraints, conversion processes, prediction of the disturbances that disrupt the performance of EMS, and fault detection methods can be ascertained in an appropriate and significantly improved way, as these are the main necessitates of an MPC-based EMS. It is expected that an MPC-based EMS shows less sensitivity to constant changes in delivery and load demand; nevertheless, different sets of constraints and multiple conversion processes deployed in the same system or method do not work in support of an MPC-based EMS, as its efficiency declines. Including a better predictable input selection strategy, isolation system, energy resources like hydrogen, and Industry 4.0 technologies to our power systems to optimize efficiency and improve decentralized energy production and consumption has become incredibly necessary, however, to confirm improved performance, or else collision and noise can take place because of battery efficiency and power transformation hazards, while the number of demand increases. Moreover, some extensive research and real-time experiments on the impact of the Russia–Ukraine war on the global energy sector are also significant. MPC-based EMS can be scalable, with appropriate acknowledgment needing to put on its actual concerns.