Distributed energy storage and photovoltaic multi-target coordination control method

Abstract

The invention discloses a distributed energy storage and photovoltaic multi-target coordination control method, and relates to the technical field of power system operation and control, and the method comprises the steps: generating a distributed energy storage coordination control strategy according to an output probability model and a charging and discharging dynamic model, and carrying out the control; then, a power distribution network multi-target optimization model for maximizing the daily net profit of the power distribution network and minimizing the load curve variance is adopted for optimization solution, the obtained final solution is used for controlling the charging and discharging behaviors of the energy storage system, and the defect that multiple operation indexes are difficult to consider under single-target optimization is overcome; according to the method, the load curve can be effectively smoothed, the peak-valley difference can be effectively reduced, meanwhile, the operation income of the system is remarkably improved by optimizing the energy storage charging and discharging behaviors under the time-of-use electricity price mechanism, the operation reliability and economical efficiency of the power distribution area under high-proportion photovoltaic access are enhanced, and obvious superiority is achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power system operation and control, in particular to a multi-objective coordinated control method of distributed energy storage and photovoltaic. BACKGROUND

[0002] With the rapid development of renewable energy, the penetration rate of distributed photovoltaic in distribution areas is continuously increasing. However, photovoltaic power generation has intermittency and volatility, and its output and load demand often do not match in time, which directly connected to the grid can easily cause voltage out-of-limit, reverse power flow and other problems, aggravating the operation pressure of the power grid. At the same time, the simple "surplus power on-grid" mode has limited economic efficiency, and the temporal and spatial value of photovoltaic power is not fully utilized.

[0003] In order to suppress photovoltaic fluctuation and improve consumption capacity, battery energy storage system is introduced into distribution area. Energy storage system has flexible charging and discharging characteristics, and is the key equipment to realize "peak clipping and valley filling" and energy time shift. However, the existing energy storage control strategy is mostly simple, or only oriented to a single economic target, lacking comprehensive consideration of system operation stability (such as load fluctuation). Common fixed charging and discharging threshold strategy or control method based on simple rules is difficult to achieve optimal overall benefit under complex time-of-use price environment and random fluctuation of source and load conditions.

[0004] In addition, the operation optimization of distribution network is essentially a multi-objective problem, and there is often a contradiction between economic targets (such as minimum operation cost and maximum benefit) and technical targets (such as smoothest load curve and most stable voltage). The traditional weighted sum method converts multi-objective into single objective, and the selection of weights depends on prior knowledge, and it is difficult to obtain a set of balanced solutions for decision makers to choose. Therefore, how to design an intelligent control method that can automatically coordinate multiple conflicting targets and give a Pareto optimal decision set in an uncertain environment has become a technical problem to be solved in the field of operation control of distribution area. SUMMARY

[0005] The present application aims to provide a multi-objective coordinated control method of distributed energy storage and photovoltaic, which solves the problem that the existing energy storage control strategy is mostly simple or only oriented to a single economic target, lacking comprehensive consideration of system operation stability (such as load fluctuation).

[0006] The present application is implemented by the following technical solutions:

[0007] A multi-objective coordinated control method of distributed energy storage and photovoltaic, comprising:

[0008] establishing an output probability model of a photovoltaic power generation system and a charging and discharging dynamic model of a distributed energy storage system;

[0009] A coordinated control strategy for distributed energy storage is generated; wherein, the coordinated control strategy refers to making decisions on the charging and discharging of the distributed energy storage system based on the power difference between photovoltaic output and demand load during different electricity price periods;

[0010] A multi-objective optimization model for the distribution network is constructed; wherein, the multi-objective optimization model for the distribution network includes a first objective function that maximizes the daily net profit of the distribution network and a second objective function that minimizes the variance of the load curve;

[0011] Based on the output probability model and the charging and discharging dynamic model, the multi-objective optimization model of the distribution network is solved using a decomposition-based multi-objective evolutionary algorithm to obtain the Pareto optimal solution set. The final solution is selected from the Pareto optimal solution set, and the charging and discharging behavior is controlled according to the coordinated control strategy, and the charging and discharging power of the energy storage system is controlled according to the final solution.

[0012] In one possible implementation, the output probability model of the photovoltaic power generation system is as follows:

[0013]

[0014] in, This represents the output probability model of a photovoltaic power generation system. The first shape parameter of the Beta distribution is represented. This represents the second shape parameter of the Beta distribution. Represents the Gamma function. Represents the normalization parameter. This indicates the actual output power of the photovoltaic power generation system. This represents the maximum solar radiation intensity within the statistical period.

[0015] In one possible implementation, the charging and discharging dynamic model of the distributed energy storage system is as follows:

[0016] Charging process:

[0017]

[0018] Discharge process:

[0019]

[0020] in, express Energy storage capacity at any given time express Energy storage capacity at any given time This indicates the remaining power loss rate of the energy storage system. Indicating distributed energy storage systems in Charging power at any time Indicating distributed energy storage systems in Discharge power at any given time Indicates time interval, Indicates the capacity of the distributed energy storage system. This indicates the charging efficiency parameter. This represents the discharge efficiency parameter.

[0021] In one possible implementation, a coordinated control strategy for distributed energy storage is generated, including...

[0022] Obtain the power difference between photovoltaic output and demand load;

[0023] When the power difference is less than zero, and the photovoltaic output is determined to be less than the demand load, the coordinated control strategy for distributed energy storage is as follows:

[0024] During peak electricity price periods, if the distributed energy storage system has surplus power, it will discharge to the off-grid energy storage capacity; otherwise, the distributed energy storage system will not charge or discharge.

[0025] During off-peak electricity prices, if the energy storage is not fully charged, the distribution network will charge the energy storage while supplying power to the load; otherwise, the distributed energy storage system will not be charged or discharged.

[0026] During periods of normal electricity prices, if the next period is a peak period and the distributed energy storage system is not fully charged, the distribution network will charge the distributed energy storage system; if the next period is a peak period and the distributed energy storage system is fully charged, the distributed energy storage system will discharge without charging; if the next period is a valley period and the distributed energy storage system has surplus power, the distributed energy storage system will discharge; if the next period is a valley period and the distributed energy storage system has no surplus power, the distributed energy storage system will neither charge nor discharge.

[0027] When the power difference is greater than or equal to zero, and the photovoltaic output is determined to be greater than or equal to zero demand load, the coordinated control strategy for distributed energy storage is as follows:

[0028] During peak electricity price periods, if the distributed energy storage system is not fully charged, it will be charged by photovoltaic power; if the distributed energy storage system is fully charged, it will not be charged or discharged.

[0029] During off-peak electricity prices, if the distributed energy storage system is not fully charged, it will be charged by photovoltaic power; if the distributed energy storage system is fully charged, it will not be charged or discharged.

[0030] During periods of normal electricity prices, if the next period is a peak period and the energy storage is not fully charged, the energy storage will be charged through photovoltaic power. If the next period is a peak period and the distributed energy storage system is fully charged, the distributed energy storage system will not be charged or discharged. If the next period is a valley period and the energy storage is not fully charged, the distributed energy storage system will be charged through photovoltaic power. If the next period is a valley period and the energy storage is fully charged, the distributed energy storage system will not be charged or discharged.

[0031] In one possible implementation, the first objective function for maximizing the daily net profit of the distribution network is:

[0032]

[0033] in, This indicates finding the maximum value. This represents the daily net profit of the power distribution network. Indicates electricity price revenue, Indicates environmental benefits. Indicates energy-saving benefits. Indicates the cost of energy storage dispatch. Indicates the cost of photovoltaic power generation. This indicates the cost of penalties for abandoning light. This indicates the cost of generating electricity using a micro gas turbine. This indicates the start-stop cost of a micro gas turbine.

[0034] In one possible implementation, the second objective function for minimizing the variance of the load curve is:

[0035]

[0036] in, This indicates finding the minimum value. Indicates the variance of the load curve. This represents the total number of time slots contained within a complete optimized scheduling cycle. express The load during the time period is considered after the coordination and control strategy is implemented.

[0037] In one possible implementation, a decomposition-based multi-objective evolutionary algorithm is used as the MOEA / D algorithm, and the Chebyshev decomposition method is used to decompose the multi-objective problem into a series of single-objective sub-problems.

[0038] In one possible implementation, the calculation formula for the Chebyshev decomposition method is as follows:

[0039]

[0040] in, The number of objective functions; This is the region of feasible solutions; As the reference point matrix, These represent the 1st, ..., mth reference points, respectively. For the weight vector, These represent the 1st, ..., mth weights, respectively.

[0041] This represents the Chebyshev scalarization function. Represents decision variables, Represents the weight vector The i-th component, Represents the reference point matrix. Let i represent the i-th objective function. The symbol represents the constraint condition, and T represents the transpose.

[0042] In one possible implementation, the process of solving the multi-objective optimization model of the distribution network using a decomposition-based multi-objective evolutionary algorithm further includes:

[0043] Constraints are imposed using power flow balance constraints, upper and lower limit constraints on transmission line power, generator and node voltage constraints, adjustable transformer turns ratio constraints, generator output power and ramp rate constraints, and / or energy storage-related constraints.

[0044] In one possible implementation, the process of solving the multi-objective optimization model of the distribution network using a decomposition-based multi-objective evolutionary algorithm further includes:

[0045] Constraints are applied using the distribution network reliability spinning reserve constraint.

[0046] The reliability spinning reserve constraint of the distribution network is as follows:

[0047]

[0048] in, This indicates the system's spinning reserve capacity requirement. Indicates the first The start-stop status of the micro gas turbine during time period t. This indicates the total number of micro-engines. Indicates the first The maximum output of the micro gas turbine This represents the reserve capacity set up during time period t to address load forecasting deviations. This represents the reserve capacity set up during time period t to address deviations in photovoltaic power output forecasts. The confidence level coefficient represents the photovoltaic output. This represents the load forecast value for time period t. Indicates the confidence level.

[0049] Compared with the prior art, this application has the following advantages and beneficial effects:

[0050] This application provides a multi-objective coordinated control method for distributed energy storage and photovoltaics. First, a coordinated control strategy for distributed energy storage is generated and implemented based on the output probability model and the charging and discharging dynamic model. Then, a multi-objective optimization model of the distribution network that maximizes the daily net profit of the distribution network and minimizes the variance of the load curve is used for optimization. The final solution is used to control the charging and discharging behavior of the energy storage system, overcoming the shortcomings of single-objective optimization in balancing multiple operating indicators. The proposed method can effectively smooth the load curve and reduce the peak-valley difference. At the same time, under the time-of-use pricing mechanism, it significantly improves the system's operating revenue by optimizing the charging and discharging behavior of energy storage, enhancing the operational reliability and economy of distribution substations with a high proportion of photovoltaic access, and has obvious advantages. Attached Figure Description

[0051] To more clearly illustrate the technical solutions of the exemplary embodiments of this application, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of this application and should not be considered as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort. In the drawings:

[0052] Figure 1 A flowchart illustrating a multi-objective coordinated control method for distributed energy storage and photovoltaics provided in this application embodiment;

[0053] Figure 2 A schematic diagram illustrating the coordination control strategy provided in an embodiment of this application;

[0054] Figure 3 A basic flowchart of the MOEA / D algorithm provided in the embodiments of this application;

[0055] Figure 4 A schematic diagram of the improved IEEE 33-node power system provided for embodiments of this application;

[0056] Figure 5 A schematic diagram illustrating the energy storage output in typical scenario 2 provided in this application embodiment;

[0057] Figure 6 This is a power balance diagram for a typical scenario 2 provided in the embodiments of this application;

[0058] Figure 7 The solution results of the decomposition-based multi-objective evolutionary algorithm provided in the embodiments of this application are obtained by solving the model. Detailed Implementation

[0059] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the embodiments and accompanying drawings. The illustrative embodiments and descriptions of this application are only for explaining this application and are not intended to limit this application.

[0060] like Figure 1 As shown in the figure, this application provides a multi-objective coordinated control method for distributed energy storage and photovoltaics, including:

[0061] S101. Establish the output probability model of the photovoltaic power generation system and the charging and discharging dynamic model of the distributed energy storage system.

[0062] S102. Generate a coordinated control strategy for distributed energy storage; wherein, the coordinated control strategy refers to making decisions on the charging and discharging of the distributed energy storage system based on the power difference between photovoltaic output and demand load during different electricity price periods.

[0063] S103. Construct a multi-objective optimization model for the distribution network; wherein, the multi-objective optimization model for the distribution network includes a first objective function that maximizes the daily net profit of the distribution network and a second objective function that minimizes the variance of the load curve;

[0064] S104. Based on the output probability model and the charging and discharging dynamic model, and using a decomposition-based multi-objective evolutionary algorithm to solve the multi-objective optimization model of the distribution network, a Pareto optimal solution set is obtained; the final solution is selected from the Pareto optimal solution set, and the charging and discharging behavior is controlled according to the coordinated control strategy, and the charging and discharging power of the energy storage system is controlled according to the final solution.

[0065] To facilitate understanding of the technical solutions described in the embodiments of this application by those skilled in the art, the embodiments of this application first introduce the overall technical ideas, which may include: (1) rule strategies provide a safety boundary for the optimization algorithm. (2) probabilistic models and dynamic models are embedded in the optimization model as core constraints. (3) multi-objective evolutionary algorithms are the engine for solving this complex model. (4) rolling optimization mechanisms are the key closed-loop process for achieving the final "coordinated control".

[0066] ① Photovoltaic system model.

[0067] Photovoltaic power generation is a technology that directly converts light energy into electrical energy using the photovoltaic effect at the semiconductor interface. Its photoelectric conversion efficiency is around 15%, and the relationship between its power generation and light intensity is as follows:

[0068]

[0069] In the formula, The output power of photovoltaic power generation; Photoelectric conversion efficiency; For the efficiency of the maximum power point tracking controller; Photovoltaic area; The intensity of solar radiation; The angle of incidence of the sun.

[0070] The output of a photovoltaic system can be considered to approximately follow a Beta distribution, and its probability distribution can be expressed as:

[0071]

[0072] in, This represents the output probability model of a photovoltaic power generation system. The first shape parameter of the Beta distribution is represented. This represents the second shape parameter of the Beta distribution. The Gamma function is an extension of the factorial function over the real and complex numbers. This indicates the actual output power of the photovoltaic power generation system. This represents the maximum solar radiation intensity within the statistical period. This parameter is used to normalize the actual output of photovoltaic power, mapping it to the interval [0,1] to meet the domain requirements of the Beta distribution.

[0073] ② Energy storage system model.

[0074] The charging and discharging power model of energy storage devices is divided into two parts: the charging process and the discharging process, as shown in the following formulas.

[0075] Charging process:

[0076]

[0077] Discharge process:

[0078]

[0079] in, and They are respectively , The amount of energy stored at any given moment; For time intervals; The remaining energy loss rate of the energy storage system is expressed as % / h. and For BESS The charging and discharging power at any given time , These refer to the charge and discharge efficiency of BESS (Battery Energy Storage System). The charge and discharge efficiency of an energy storage system is a physical property of its own hardware and is generally a known condition. The efficiency of energy storage systems from different manufacturers varies slightly, generally between 95% and 98%. This refers to the capacity of the energy storage system.

[0080] like Figure 2 As shown, considering peak-valley time-of-use pricing, the control strategy for energy storage under this condition is as follows: Figure 1 As shown, the coordinated control strategy for energy storage is divided into two cases: when the output of distributed power sources is less than the load; and when the output of distributed power sources is greater than the load.

[0081] (1) When the output of distributed power sources (i.e., photovoltaics) is less than the load, the load first obtains power from the distributed power sources, and the insufficient part is supplemented by the distribution network: ① During peak electricity price periods, if the energy storage has surplus power, it will discharge to the off-grid energy storage capacity; otherwise, the energy storage will not charge or discharge. ② During off-peak electricity price periods, if the energy storage is not fully charged, the distribution network will charge the energy storage in addition to supplying power to the load. Otherwise, the energy storage will not charge or discharge. ③ During normal electricity price periods, it is determined whether the next period will be a peak period. If it is a peak period, and the energy storage is not fully charged, the distribution network will charge the energy storage. If the energy storage is fully charged, the energy storage will be charged without charging or discharging; if the next period is an off-peak period, and the energy storage has surplus power, it will discharge. If the energy storage has no surplus power, the energy storage will not charge or discharge.

[0082] (2) When the output of the distributed power source exceeds the load, and the load only obtains power from the distributed power source: ① During peak electricity price periods, if the energy storage is not fully charged, the net photovoltaic (PV) power source charges the energy storage. If the energy storage is fully charged, there is no charging or discharging. ② During off-peak electricity price periods, if the energy storage is not fully charged, the net PV power source charges the energy storage. If the energy storage is fully charged, there is no charging or discharging. ③ During normal electricity price periods, if the next period is a peak period, and the energy storage is not fully charged, the net PV power source charges the energy storage. If the energy storage is fully charged, there is no charging or discharging; if the next period is an off-peak period, and the energy storage is not fully charged, the net PV power source charges the energy storage. If the energy storage is fully charged, there is no charging or discharging.

[0083] Based on historical photovoltaic and load data, a multi-objective coordinated control strategy is proposed. This includes the daily operating cost of the power grid (energy storage dispatch cost). Cost of photovoltaic power generation The cost of abandoning light Cost of generating electricity from micro gas turbines and the start-stop cost of micro gas turbines ) and daily grid revenue (electricity price revenue) Environmental benefits and energy-saving benefits ).

[0084] The daily operating costs of the power grid are as follows:

[0085] (1) Energy storage dispatch cost.

[0086] Energy storage dispatch cost of distribution network for:

[0087]

[0088] in, Unit dispatch cost for energy storage systems; and They are respectively Time period The charging and discharging power of an energy storage system It represents the reciprocal of charging efficiency or the charging loss coefficient. Indicates discharge efficiency. Indicates a time interval.

[0089] (2) Cost of photovoltaic power generation.

[0090] Photovoltaic power generation cost of power distribution network for:

[0091]

[0092] in, The cost price per unit of photovoltaic power. For the first The amount of photovoltaic power consumed during the period.

[0093] (3) Cost of abandoning light penalty.

[0094] Cost of curtailment of solar power in distribution networks for:

[0095]

[0096] in, For photovoltaic feed-in tariffs; The total amount of photovoltaic power generation absorbed. Let t be the actual output value of photovoltaic power generation during time period t. This represents the total number of time periods contained within a complete optimized scheduling cycle.

[0097] (4) Cost of generating electricity from micro gas turbines.

[0098] The cost of micro gas turbine power generation in the power distribution network for:

[0099]

[0100] in, This represents the total number of generating units; This indicates the start / stop status of the unit, with 0 representing shutdown and 1 representing operation. The quadratic coefficient determines the curvature of the cost curve; The coefficient of the first-order term determines the slope of the cost curve. The constant term coefficient represents the fuel consumption cost of the unit when it is running at no load, which is the first constant. The fuel consumption coefficient of the generator set; For the first Taiwanese crew Active output at all times.

[0101] (5) Start-up and shutdown costs of micro gas turbines.

[0102] Start-up and shutdown costs of micro gas turbines in power distribution networks for:

[0103]

[0104] in, For the first Start-up and shutdown costs of a conventional generating unit This represents the duration during which the l-th unit has been continuously shut down in time period t-1.

[0105] The typical daily revenue of the power grid is as follows:

[0106] (1) Electricity price revenue.

[0107] Distribution network electricity price revenue for:

[0108]

[0109] in, Time-of-use electricity pricing for users; Subsidized electricity prices for photovoltaic power. The price of electricity delivered to the higher-level power grid; For user load power, For photovoltaic grid connection power, Contribute to photovoltaic power.

[0110] (2) Environmental benefits.

[0111] This article measures environmental benefits by comparing the environmental losses from reducing pollutant emissions compared to coal-fired power generation when producing the same amount of electricity. Environmental benefits of power distribution networks. for:

[0112]

[0113] in, The number of photovoltaic cells; The types of pollutants; For the first The environmental value of reducing pollutant emissions; For the first thermal power unit The amount of each pollutant emitted; For the first The first photovoltaic The amount of pollutants emitted.

[0114] (3) Energy saving benefits.

[0115] This article measures energy-saving benefits by using the reduction in fossil fuel consumption through photovoltaic power generation. Energy-saving benefits of distribution networks. for:

[0116]

[0117] in, The amount of coal consumed by a coal-fired power unit in a production unit; For coal prices; For the first The power generation of a single photovoltaic unit.

[0118] For the multi-objective coordinated control of adding energy storage systems to the distribution network, the objectives are as follows.

[0119] Maximize the typical daily net profit of photovoltaic power distribution networks for:

[0120]

[0121] The minimum variance of the load curve is:

[0122]

[0123] in, for The time period considers the load after the energy storage charging and discharging strategy. This represents the total number of time periods contained within a complete optimized scheduling cycle.

[0124] The constraints are as follows:

[0125] It must satisfy the power flow balance of the power grid (including active power balance and reactive power balance), the upper and lower limits of power of transmission lines, the voltage constraints of generators and nodes, the turns ratio constraints of adjustable transformers, the output power and ramp rate constraints of generators, as well as the constraints related to energy storage.

[0126] In addition, the following constraints must be met regarding the reliability of the distribution network's spinning reserve:

[0127]

[0128] in, This indicates the system's spinning reserve capacity requirement. Indicates the first The start-stop status of the micro gas turbine during time period t is a 0 / 1 variable, where 1 represents running and 0 represents shutting down. This indicates the total number of micro-engines. Indicates the first The maximum output of the micro gas turbine, i.e. its rated power; This represents the reserve capacity set up during time period t to address load forecasting deviations. This represents the reserve capacity set up during time period t to address deviations in photovoltaic power output forecasts. The confidence level coefficient for photovoltaic output is a weighting or discounting factor used to consider the correlation between photovoltaic output and load or the confidence level of photovoltaic output when calculating reserve demand. Its value is between 0 and 1. This represents the load forecast value for time period t. Indicates the confidence level. For example... =0.95 means that the system is required to have a 95% probability of not experiencing power deficit under all possible photovoltaic output and load fluctuation scenarios.

[0129] In the established multi-objective programming model, there are certain constraints, and even contradictions, between the objectives. Improving any one objective function cannot guarantee that at least one other objective function will not change. This means that there is no absolutely optimal solution that simultaneously optimizes all objective functions, thus introducing Pareto optimality. Taking a maximization problem as an example, for any two decision variables... , ,have At this time, it is called Pareto Dominance (or) Dominate ), if and only if the following expression is satisfied:

[0130]

[0131] in, The number of objective functions. Let be the i-th objective function.

[0132] say For the entire solution space A Pareto optimal solution, or non-dominated solution, is obtained if and only if:

[0133]

[0134] in, As decision variables, To solve the space, This is the Pareto optimal solution to be determined.

[0135] The Pareto optimal front (PF) is the surface formed by the objective vectors corresponding to all Pareto optimal solutions. Solving multi-objective optimization problems is essentially about finding as many Pareto optimal solutions as possible and distributing the corresponding objective vectors as evenly as possible on the optimal front.

[0136] The aforementioned nonlinear multi-objective programming model includes equality and inequality constraints. Because it is often difficult to find suitable weights for selecting multiple objectives when transforming a multi-objective problem into a single-objective problem, this paper uses the MOEA / D algorithm to solve the constructed multi-objective optimization model with the optimization objectives of maximizing the net profit of the distribution network and minimizing the load variance. The MOEA / D algorithm is superior to the MOGLS and NSGA-II algorithms. Figure 3 As shown, the MOEA / D algorithm decomposes a multi-objective optimization problem into a certain number of single-objective optimization subproblems. Then, using information from neighboring problems, it simultaneously optimizes the decomposed single-objective subproblems using an evolutionary algorithm. Because the solutions on the Pareto front correspond one-to-one with the optimal solutions of the single-objective optimization subproblems, a set of Pareto optimal solutions can be obtained. Due to the decomposition operation, this method has a significant advantage in preserving the distribution of solutions. Furthermore, optimizing by analyzing information from neighboring problems effectively avoids getting trapped in local optima. The basic flowchart of the MOEA / D algorithm is shown below. Figure 2 As shown.

[0137] Because the Chebyshev decomposition method is insensitive to the shape of the optimal front and can approximate the optimal front of convex and non-convex problems well, this paper adopts the Chebyshev decomposition method for decomposition, and its mathematical expression is as follows:

[0138]

[0139] in, The number of objective functions; This is the region of feasible solutions; As the reference point matrix, These represent the 1st, ..., mth reference points, respectively. For the weight vector, These represent the 1st, ..., mth weights, respectively. The Chebyshev scalarization function, representing the core of the entire formula, is a function that transforms a multi-objective optimization problem into a single-objective optimization problem. This indicates that the decision variables include all dispatchable variables such as energy storage charging and discharging power and micro-turbine output for each time period within the scheduling cycle. It represents a weight vector. The i-th component represents the weight of the i-th objective function. Represents the reference point matrix. Let i represent the i-th objective function. "Subject to" is an abbreviation for "subject to," meaning "satisfied with" or "constrained by." Here, "T" (as a superscript) represents the transpose operator in mathematics. The reference point matrix satisfies the following conditions:

[0140]

[0141] in, Let i be the reference point. Let i be the objective function. This is the region of feasible solutions.

[0142] Assume there is For each sub-problem, the weight vector have They are evenly distributed. and One-to-one correspondence, if there is and If they are similar in size, then , The optimal solutions are also similar. (If the two weight vectors) and If the distance in Euclidean space is sufficiently small, meaning the preference directions they represent are similar, then the corresponding subproblem obtained through Chebyshev scalarization function decomposition is... and (Approximate) optimal solution and It has a small distance in the decision space, and its image and They are also adjacent to each other on the Pareto optimal front in the target space. The weight neighborhood O is defined as... The number of similar weighted solutions is crucial for finding the optimal solution to a subproblem and for iterative optimization. This information is indispensable.

[0143] The improved IEEE 33-node power system was selected. Figure 4 This is a diagram of an IEEE 33-node distribution network system. The established energy storage charging and discharging model and multi-objective coordinated control strategy are simulated using the MATLAB platform to obtain the load demand changes before and after energy storage grid connection, thus verifying the role of energy storage grid connection. Simultaneously, a decomposition-based multi-objective evolutionary algorithm is used to solve the established multi-objective coordinated control strategy, and the corresponding Pareto front is calculated. Load data from a photovoltaic array and the distribution network are selected as example data.

[0144] The simulation results for the load demand and energy storage output after grid connection in scenario 2 are shown in Table 1 and 2. Figure 5 .from Figure 5 , Figure 6 The diagram clearly shows the charging and discharging status and SOC (State of Charge) of the energy storage system over 24 hours after grid connection. The proposed energy storage control strategy reduces the load's need to purchase electricity from the distribution network, resulting in significant economic benefits. Furthermore, grid connection of energy storage alters the original peak load time, further increasing the spatiotemporal freedom of distributed energy resources in the distribution network and indirectly reducing the network's reserve capacity. In summary, by controlling the charging and discharging strategy of the energy storage system, substantial benefits can be obtained while significantly reducing load volatility. The energy storage and photovoltaic coordinated output diagram under scenario 2 is shown below, with the left vertical axis representing energy storage output and the right vertical axis representing the SOC status of the energy storage device.

[0145] Table 1 Load demand and energy storage output after grid connection of energy storage

[0146]

[0147] Table 2 shows the peak-valley load characteristics after energy storage is integrated into the distribution network in Scenario 2. As can be seen from Table 2, by controlling the charging and discharging of the energy storage, the load during peak hours decreased by 0.185MW after grid connection, while the load during normal and valley hours increased by 0.0138MW and 0.637MW, respectively. After grid connection, the peak-valley difference decreased from 1.867MW to 1.054MW, and the peak-valley ratio also decreased from 1.59 to 1.24, effectively achieving peak shaving and valley filling.

[0148] Table 2 Peak-valley load characteristics before and after energy storage grid connection

[0149]

[0150] The model is solved using a decomposition-based multi-objective evolutionary algorithm to obtain the following results: Figure 7 As shown.

[0151] When the daily revenue is 7208.3 yuan, grid-connected energy storage reduces the load variance from 0.91 to 0.875; when the load variance is 0.92, grid-connected energy storage increases the daily revenue from 6640.5 yuan to 8168.9 yuan. Therefore, it can be concluded that integrating energy storage into the distribution network increases profits while reducing load variance, thus reducing load fluctuations and significantly improving the reliability of the distribution network, thereby achieving optimal overall revenue planning in this typical scenario.

[0152] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of this application. It should be understood that the above description is only a specific embodiment of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A multi-objective coordinated control method for distributed energy storage and photovoltaics, characterized in that, include: Establish a power output probability model for a photovoltaic power generation system and a charging and discharging dynamic model for a distributed energy storage system; A coordinated control strategy for distributed energy storage is generated; wherein, the coordinated control strategy refers to making decisions on the charging and discharging of the distributed energy storage system based on the power difference between photovoltaic output and demand load during different electricity price periods; A multi-objective optimization model for the distribution network is constructed; wherein, the multi-objective optimization model for the distribution network includes a first objective function that maximizes the daily net profit of the distribution network and a second objective function that minimizes the variance of the load curve; Based on the output probability model and the charging and discharging dynamic model, the multi-objective optimization model of the distribution network is solved using a decomposition-based multi-objective evolutionary algorithm to obtain the Pareto optimal solution set. The final solution is selected from the Pareto optimal solution set, and the charging and discharging behavior is controlled according to the coordinated control strategy, and the charging and discharging power of the energy storage system is controlled according to the final solution.

2. The multi-objective coordinated control method for distributed energy storage and photovoltaics according to claim 1, characterized in that, The output probability model of the photovoltaic power generation system is as follows: ； in, This represents the output probability model of a photovoltaic power generation system. The first shape parameter of the Beta distribution is represented. This represents the second shape parameter of the Beta distribution. Represents the Gamma function. Represents the normalization parameter. This indicates the actual output power of the photovoltaic power generation system. This represents the maximum solar radiation intensity within the statistical period.

3. The multi-objective coordinated control method for distributed energy storage and photovoltaics according to claim 2, characterized in that, The charging and discharging dynamic model of the distributed energy storage system is as follows: Charging process: ； Discharge process: ； in, express Energy storage capacity at any given time express Energy storage capacity at any given time This indicates the remaining power loss rate of the energy storage system. Indicating distributed energy storage systems in Charging power at any time Indicating distributed energy storage systems in Discharge power at any given time Indicates time interval, Indicates the capacity of the distributed energy storage system. This indicates the charging efficiency parameter. This represents the discharge efficiency parameter.

4. The multi-objective coordinated control method for distributed energy storage and photovoltaics according to claim 1, characterized in that, Generate a coordinated control strategy for distributed energy storage, including Obtain the power difference between photovoltaic output and demand load; When the power difference is less than zero, and the photovoltaic output is determined to be less than the demand load, the coordinated control strategy for distributed energy storage is as follows: During peak electricity price periods, if the distributed energy storage system has surplus power, it will discharge to the off-grid energy storage capacity; otherwise, the distributed energy storage system will not charge or discharge. During off-peak electricity prices, if the energy storage is not fully charged, the distribution network will charge the energy storage in addition to supplying power to the load; otherwise, the distributed energy storage system will not charge or discharge. During periods of normal electricity prices, if the next period is a peak period and the distributed energy storage system is not fully charged, the distribution network will charge the distributed energy storage system; if the next period is a peak period and the distributed energy storage system is fully charged, the distributed energy storage system will discharge without charging; if the next period is a valley period and the distributed energy storage system has surplus power, the distributed energy storage system will discharge; if the next period is a valley period and the distributed energy storage system has no surplus power, the distributed energy storage system will neither charge nor discharge. When the power difference is greater than or equal to zero, and the photovoltaic output is determined to be greater than or equal to zero demand load, the coordinated control strategy for distributed energy storage is as follows: During peak electricity price periods, if the distributed energy storage system is not fully charged, it will be charged by photovoltaic power; if the distributed energy storage system is fully charged, it will not be charged or discharged. During off-peak electricity prices, if the distributed energy storage system is not fully charged, it will be charged by photovoltaic power; if the distributed energy storage system is fully charged, it will not be charged or discharged. During periods of normal electricity prices, if the next period is a peak period and the energy storage is not fully charged, the energy storage will be charged by photovoltaic power; if the next period is a peak period and the distributed energy storage system is fully charged, the distributed energy storage system will not be charged or discharged. If the next time period is a valley period and the energy storage is not fully charged, the distributed energy storage system will be charged through photovoltaic power; if the next time period is a valley period and the energy storage is fully charged, there will be no charging or discharging.

5. The multi-objective coordinated control method for distributed energy storage and photovoltaics according to claim 1, characterized in that, The first objective function for maximizing the daily net profit of the distribution network is: ； in, This indicates finding the maximum value. This represents the daily net profit of the power distribution network. Indicates electricity price revenue, Indicates environmental benefits. Indicates energy-saving benefits. Indicates the cost of energy storage dispatch. Indicates the cost of photovoltaic power generation. This indicates the cost of penalties for abandoning light. This indicates the cost of generating electricity using a micro gas turbine. This indicates the start-stop cost of a micro gas turbine.

6. The multi-objective coordinated control method for distributed energy storage and photovoltaics according to claim 1, characterized in that, The second objective function for minimizing the variance of the load curve is: ； in, This indicates finding the minimum value. Indicates the variance of the load curve. This represents the total number of time slots contained within a complete optimized scheduling cycle. express The load during the time period is considered after the coordination and control strategy is implemented.

7. The multi-objective coordinated control method for distributed energy storage and photovoltaics according to claim 1, characterized in that, The multi-objective evolutionary algorithm based on decomposition is adopted as the MOEA / D algorithm, and the Chebyshev decomposition method is used to decompose the multi-objective problem into a series of single-objective sub-problems.

8. The multi-objective coordinated control method for distributed energy storage and photovoltaics according to claim 7, characterized in that, The calculation formula for the Chebyshev decomposition method is as follows: ； in, The number of objective functions; This is the region of feasible solutions; As the reference point matrix, These represent the 1st, ..., mth reference points, respectively. For the weight vector, These represent the 1st, ..., mth weights, respectively. This represents the Chebyshev scalarization function. Represents decision variables, Represents the weight vector The i-th component, Represents the reference point matrix. Let i represent the i-th objective function. The symbol represents the constraint condition, and T represents the transpose.

9. The multi-objective coordinated control method for distributed energy storage and photovoltaics according to claim 1, characterized in that, The process of solving the multi-objective optimization model of the distribution network using a decomposition-based multi-objective evolutionary algorithm also includes: Constraints are imposed using power flow balance constraints, upper and lower limit constraints on transmission line power, generator and node voltage constraints, adjustable transformer turns ratio constraints, generator output power and ramp rate constraints, and / or energy storage-related constraints.

10. The multi-objective coordinated control method for distributed energy storage and photovoltaics according to claim 9, characterized in that, The process of solving the multi-objective optimization model of the distribution network using a decomposition-based multi-objective evolutionary algorithm also includes: Constraints are applied using the distribution network reliability spinning reserve constraint. The reliability spinning reserve constraint of the distribution network is as follows: ； in, This indicates the system's spinning reserve capacity requirement. Indicates the first The start-stop status of the micro gas turbine during time period t. This indicates the total number of micro-engines. Indicates the first The maximum output of the micro gas turbine This represents the reserve capacity set up during time period t to address load forecasting deviations. This represents the reserve capacity set up during time period t to address deviations in photovoltaic power output forecasts. The confidence level coefficient represents the photovoltaic output. This represents the load forecast value for time period t. Indicates the confidence level.

Images