Intelligent evaluation method for collaborative peak regulation effect of micro-grid group

Abstract

The invention discloses an intelligent evaluation method for the collaborative peak regulation effect of a micro-grid group. The method comprises the following steps: step 1, establishing an evaluation index system for the collaborative peak regulation effect of the micro-grid group; step 2, according to the evaluation index system of the collaborative peak regulation effect of the micro-grid group established in the step 1, constructing a comprehensive evaluation model of the collaborative peak regulation effect of the micro-grid group based on a fuzzy evidence fusion framework; and step 3, obtaining an evaluation result through the micro-grid group collaborative peak regulation effect comprehensive evaluation model. According to the method, the stability and economical efficiency of the micro-grid group in actual operation are enhanced, and powerful support is provided for development and application of the micro-grid technology.

Description

Technical Field

[0001] The present invention relates to the field of microgrid evaluation methods, and in particular to an intelligent evaluation method for the coordinated peak-shaving effect of a microgrid group. Background Art

[0002] In modern power systems, with the rapid popularization of renewable energy, microgrids, as an important distributed energy management and power supply method, play an increasingly critical role. Microgrid groups can not only effectively balance the fluctuations between load and power generation through coordinated peak load regulation, but also improve the stability and economy of the power grid. Therefore, accurately evaluating the coordinated peak load regulation effect of microgrid groups and ensuring their stable operation are of great strategic significance for the overall optimization of the power system. At the same time, with the transformation of energy structure and the advancement of carbon neutrality goals, optimizing the peak load regulation effect evaluation method has also become an important topic in the current research field.

[0003] However, the existing methods for evaluating the peak load regulation of microgrid groups have obvious limitations. Due to the uncertainty of the output of renewable energy sources such as wind power and photovoltaics within the microgrid, coupled with the volatility of the adjustable load, traditional evaluation methods often cannot fully consider the complex influence of these factors, resulting in evaluation results that deviate from the actual situation. In addition, many methods have problems such as low computational efficiency and inaccurate results when dealing with multi-source heterogeneous data and complex system uncertainties. These shortcomings limit the effect of microgrid group collaborative peak load regulation in practical applications and also restrict the further promotion and development of related technologies. Summary of the invention

[0004] The present invention provides an intelligent evaluation method for the collaborative peak-shaving effect of a microgrid group, which introduces a fuzzy evidence fusion algorithm and combines it with a reinforcement learning algorithm to solve the limitations of existing evaluation methods in dealing with uncertainty and multi-source information fusion.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is:

[0006] A method for intelligently evaluating the coordinated peak-shaving effect of a microgrid group comprises the following steps:

[0007] Step 1: Establish an evaluation index system for the coordinated peak-shaving effect of microgrid groups;

[0008] Step 2: Based on the microgrid group coordinated peak-shaving effect evaluation index system established in step 1, a microgrid group coordinated peak-shaving effect comprehensive evaluation model is constructed based on the fuzzy evidence fusion framework. The process is as follows:

[0009] S201. Based on the microgrid group coordinated peak-shaving effect evaluation index system obtained in step 1, a microgrid group coordinated peak-shaving effect comprehensive evaluation index set X applicable to the fuzzy evidence fusion framework is established. u , Xu The elements in are the identification evaluation index fuzzy mapping thresholds of each evaluation index in the microgrid group collaborative peak-shaving effect evaluation index system;

[0010] Map each evaluation index in the microgrid group coordinated peak-shaving effect evaluation index system obtained in step 1 to the microgrid group coordinated peak-shaving effect comprehensive evaluation index set X u , when the evaluation index in the microgrid group collaborative peak-shaving effect evaluation index system reaches the corresponding mapping threshold, the evaluation index value is mapped to the evidence evaluation effect corresponding to the mapping threshold;

[0011] S202. Calculate the comprehensive evaluation index set X of the coordinated peak-shaving effect of the microgrid group u The distance between the evidence characteristics of each piece of evidence;

[0012] S203, based on the obtained evidence feature distance, determine the similarity between the evidences in the comprehensive evaluation index set of the microgrid group coordinated peak-shaving effect;

[0013] S204, based on the obtained similarity between each piece of evidence, determine the support level of each piece of evidence in the comprehensive evaluation index set of microgrid group coordinated peak-shaving effect;

[0014] S205. Determine the evaluation weight of the corresponding evidence on the coordinated peak-shaving effect of the microgrid group based on the support degree, and establish a comprehensive evaluation model for the coordinated peak-shaving effect of the microgrid group according to the evaluation weight;

[0015] Step 3: Obtain the evaluation results through the comprehensive evaluation model of the coordinated peak-shaving effect of the microgrid group.

[0016] Furthermore, in step 1, the evaluation indicators in the microgrid group collaborative peak-shaving effect evaluation index system include a new energy peak-shaving uncertainty index, a load peak-shaving uncertainty index, and an energy storage response effect index.

[0017] Furthermore, the new energy peak-shaving uncertainty index includes a new energy peak-shaving response delay uncertainty index, a new energy peak-shaving response capacity uncertainty index, a new energy peak-shaving response speed uncertainty index, and a new energy peak-shaving response cost uncertainty index.

[0018] Furthermore, step 2 S201 also includes:

[0019] The power grid operation state model S suitable for reinforcement learning algorithm is established under the current identification and evaluation index fuzzy mapping threshold. grid ;

[0020] The fuzzy mapping threshold of the identification evaluation index is used as the action model A of the reinforcement learning algorithm grid ;

[0021] Use reinforcement learning algorithm to continuously iterate and optimize action model A grid Specific variable values ​​to realize action model A grid Optimal, that is, achieving the optimal threshold of fuzzy mapping of identification and evaluation indicators;

[0022] Then, the optimal identification evaluation index fuzzy mapping threshold is used as the comprehensive evaluation index set X of the microgrid group collaborative peak load regulation effect. u The elements in .

[0023] The present invention provides an intelligent evaluation method for the coordinated peak-shaving effect of a microgrid group. By introducing a fuzzy evidence fusion algorithm, the peak-shaving effect of a microgrid group is comprehensively evaluated, thereby improving the accuracy and reliability of the evaluation. At the same time, a reinforcement learning algorithm is used to optimize the semantic recognition framework of the fuzzy evidence fusion algorithm to further improve the computational efficiency and adaptability of the evaluation process.

[0024] The present invention enhances the stability and economy of microgrid groups in actual operation, which not only improves the evaluation accuracy and robustness, but also can better adapt to the complexity and dynamics of microgrid groups in actual operation, further promotes the development and application of microgrid technology, and provides strong support for the development and application of microgrid technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a schematic diagram of the method of an embodiment of the present invention.

[0026] Figure 2 It is a performance simulation analysis result diagram of two different methods in the experiment of the embodiment of the present invention.

[0027] Figure 3 It is a convergence curve diagram of the method proposed in the embodiment of the present invention.

[0028] Figure 4 It is the convergence curve diagram of the genetic algorithm used in the experiment. DETAILED DESCRIPTION

[0029] The present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0030] like Figure 1 As shown, this embodiment discloses a method for intelligently evaluating the coordinated peak-shaving effect of a microgrid group, comprising the following steps:

[0031] Step 1: Establish an evaluation index system for the coordinated peak-shaving effect of microgrid groups.

[0032] In this embodiment, a new energy peak-shaving uncertainty index, a load peak-shaving uncertainty index, and an energy storage response effect index are obtained, thereby constructing a microgrid group collaborative peak-shaving effect evaluation index system.

[0033] Specifically, the uncertainty index of new energy peak regulation includes the uncertainty index of new energy peak regulation response delay Uncertainty index of new energy peak load response capacity Uncertainty index of new energy peak load response speed Uncertainty index of new energy peak load response cost.

[0034] New energy peak load response delay uncertainty index It can be expressed as the following formula:

[0035]

[0036] in: is the uncertainty index of the peak-shaving response delay of renewable energy; n is the number of microgrids; N is the total number of microgrids; k is the number of renewable energy in the microgrid; K is the total number of renewable energy in the microgrid; p() is the probability function; τ is the response delay time of renewable energy.

[0037] Uncertainty index of new energy peak load response capacity It can be expressed as the following formula:

[0038]

[0039] in: is the uncertainty index of the peak-shaving response capacity of new energy; V is the expected response capacity of new energy; and v is the actual response capacity of new energy.

[0040] Uncertainty index of new energy peak load response speed It can be expressed as the following formula:

[0041]

[0042] in: is the uncertainty index of the peak load response speed of new energy; T e The time it takes for the new energy to reach the expected response capacity; t a It is the time for the new energy to reach the actual response capacity.

[0043] The uncertainty index of the new energy peak load response cost can be expressed as follows:

[0044]

[0045] in: is the uncertainty index of the peak load response cost of renewable energy; C e is the expected response cost of new energy; C a The actual response cost of new energy.

[0046] In this embodiment, the load peak regulation uncertainty index FLoad It can be expressed as follows:

[0047]

[0048] Among them: F Load is the load peak regulation uncertainty index; l is the number of loads in the microgrid; L is the total number of loads in the microgrid; L up is the upward peak regulation deviation of the load in the microgrid; L down It is the downward peak regulation deviation of the load in the microgrid.

[0049] Energy storage response effect index F ES It can be expressed as a prime number as follows:

[0050]

[0051] Among them: F ES is the energy storage response effect index; e is the number of energy storage in the microgrid; E is the total number of energy storage in the microgrid; E a is the actual regulation value of energy storage in the microgrid; E v It is the adjustable capacity of energy storage in the microgrid.

[0052] Step 2: Based on the microgrid group coordinated peak-shaving effect evaluation index system established in step 1, a microgrid group coordinated peak-shaving effect comprehensive evaluation model is constructed based on the fuzzy evidence fusion framework. The process is as follows:

[0053] S201. Based on the microgrid group coordinated peak-shaving effect evaluation index system obtained in step 1, a microgrid group coordinated peak-shaving effect comprehensive evaluation index threshold set X applicable to the fuzzy evidence fusion framework is established. u , X u The elements in are the identification evaluation index fuzzy mapping thresholds of each evaluation index in the microgrid group collaborative peak-shaving effect evaluation index system. The threshold set of comprehensive evaluation indexes for the microgrid group collaborative peak-shaving effect is shown in the following formula:

[0054] X u ={x1,x2,x3,x4,x5}

[0055] Among them: x1, x2, x3, x4, x5 are the identification evaluation index fuzzy mapping thresholds of each evaluation index in the microgrid group collaborative peak-shaving effect evaluation index system, and the new energy peak-shaving response delay uncertainty index Uncertainty index of new energy peak load response capacity Uncertainty index of new energy peak load response speed New energy peak load response cost uncertainty index, load peak load uncertainty index F Load , Energy storage response effect index F ESThere are six types of indicators in total, each of which has an indicator threshold set X u For example, for a certain type of indicator, its indicator threshold set X u It is divided into five levels: {poor, general, medium, good, excellent}, then x1, x2, x3, x4, x5 correspond to poor, general, medium, good, excellent respectively.

[0056] In this embodiment, the comprehensive evaluation index threshold set X of the microgrid group coordinated peak load regulation effect is realized by strengthening the student algorithm. u The identification and fuzzy mapping thresholds x1, x2, x3, x4, x5 of each evaluation index are optimized, and the optimal identification and fuzzy mapping thresholds x1, x2, x3, x4, x5 are used as the comprehensive evaluation index set X of the coordinated peak-shaving effect of the microgrid group. u The optimization method for the elements in is as follows:

[0057] (S2011), establish a power grid operation state model S suitable for reinforcement learning algorithm under the current identification evaluation index fuzzy mapping threshold x1, x2, x3, x4, x5 grid ; The specific model is:

[0058]

[0059] Among them, S grid It is the grid operation status model; It is the current operating state vector of the power grid, including the power grid generation side state, power grid network side state, power grid load side state, and power grid peak regulation demand state. is the evaluation index of the coordinated peak-shaving effect of each microgrid group established in step 1, and i is {1, 2, 3, 4}; X u is a comprehensive evaluation index threshold set of coordinated peak-shaving effect of microgrid group composed of fuzzy mapping thresholds x1, x2, x3, x4, and x5 of current identification and evaluation index; Indicates the evaluation result under the current evaluation index and evaluation threshold conditions.

[0060] (S2012), the action model contains the evaluation index fuzzy mapping threshold x1, x2, x3, x4, x5, and the action model can affect the state model through the value of the action (adjustment) parameter, and then the action model is continuously iterated and optimized through the feedback of the state model, that is, the evaluation index fuzzy mapping threshold x1, x2, x3, x4, x5 is optimized. The identification evaluation index fuzzy mapping threshold x1, x2, x3, x4, x5 is used as the action model A of the reinforcement learning algorithm grid , as shown in the following formula:

[0061]

[0062] Combined with S2011, it can be seen that in the reinforcement learning framework, the action model A grid The elements {x1, x2, x3, x4, x5} in can take any values ​​under the constraints, and different values ​​will affect the evaluation results of the current operating status of the power grid.

[0063] (S2013) In order to make the current evaluation result of the current operation status of the power grid effectively reflect the current operation status of the power grid, the action model A is continuously optimized by using the reinforcement learning algorithm. grid The specific variable value of the action model A is realized by continuously iterating the evaluation index fuzzy mapping thresholds x1, x2, x3, x4, x5, so that the evaluation result caused by the current threshold division method is closer to the actual state. grid Optimal, that is, to achieve the optimal fuzzy mapping thresholds x1, x2, x3, x4, x5 for the identification evaluation index. The optimization reward function during iteration is shown in the following formula:

[0064]

[0065] Where: R is the iterative optimization reward function; r λ , r λ-1 are the iteration rounds λ and the actual peak-shaving effect of the microgrid group when λ-1; Γ is the maximum value of the iteration rounds.

[0066] Then, the optimal identification evaluation index fuzzy mapping threshold is used as the comprehensive evaluation index set X of the microgrid group collaborative peak load regulation effect. u The elements in .

[0067] The comprehensive evaluation index set X of the coordinated peak-shaving effect of the microgrid group is obtained, which is composed of the optimal identification evaluation index fuzzy mapping threshold. u Then, each evaluation index in the microgrid group collaborative peak-shaving effect evaluation index system obtained in step 1, namely, the new energy peak-shaving response delay uncertainty index Uncertainty index of new energy peak load response capacity Uncertainty index of new energy peak load response speed New energy peak load response cost uncertainty index, load peak load uncertainty index F Load , Energy storage response effect index F ES , respectively mapped to the comprehensive evaluation index set X of the coordinated peak-shaving effect of the microgrid group uThe corresponding identification and evaluation index fuzzy mapping threshold in the mapping, when the evaluation index in the microgrid group collaborative peak-shaving effect evaluation index system reaches the corresponding identification and evaluation index fuzzy mapping threshold, the evaluation index value is mapped to the evidence evaluation effect corresponding to the identification and evaluation index fuzzy mapping threshold. The evidence evaluation effects corresponding to x1, x2, x3, x4, and x5 are:

[0068] {Excellent response, good response, average response, poor response, substandard response}

[0069] S202. Calculate the comprehensive evaluation index set X of the coordinated peak-shaving effect of the microgrid group u The evidence feature distance between each piece of evidence is as shown in the following formula:

[0070]

[0071] Among them: F ds (·,·) is the characteristic distance between the evidences of the coordinated peak-shaving effect of the microgrid group; si is evidence i; sj is evidence j. Evidence i is the evaluation result of the current evaluation threshold condition of the evaluation index i when the evaluation index i in the microgrid group coordinated peak-shaving effect evaluation index system reaches the corresponding identification evaluation index fuzzy mapping threshold. Mapped to evidence i.

[0072] S203. Based on the obtained evidence feature distance, the similarity between the evidences in the comprehensive evaluation index set of the coordinated peak-shaving effect of the microgrid group is determined, as shown in the following formula:

[0073]

[0074] in, It is the similarity between the evidences of the coordinated peak-shaving effect of microgrid groups.

[0075] S204. Based on the similarity between the evidences, the support level of each evidence in the comprehensive evaluation index set of the coordinated peak-shaving effect of the microgrid group is determined, as shown in the following formula:

[0076]

[0077] in, is the degree of support for evidence i; n is the number of evidences.

[0078] S205. Based on the support level, the corresponding evidence is judged to evaluate the effect of microgrid group coordinated peak load regulation, as shown in the following formula:

[0079]

[0080] Among them, ξ iCalculates function for evaluation weights.

[0081] A comprehensive evaluation model for the coordinated peak-shaving effect of microgrid groups is established based on the evaluation weights, as shown in the following formula:

[0082]

[0083] Among them, F total It is a comprehensive evaluation model for the coordinated peak-shaving effect of microgrid groups.

[0084] Step 3: Obtain the evaluation results through the comprehensive evaluation model of the coordinated peak-shaving effect of the microgrid group.

[0085] In order to make the purpose, technical solution and advantages of this embodiment more clear, this embodiment is further described in detail below in combination with experiments.

[0086] A case study of a microgrid group peak load regulation scenario in a certain region in eastern my country in 2023 was selected for analysis. The microgrid group peak load regulation scenario includes three microgrids, each of which includes wind power, photovoltaic power, electric vehicles, central air conditioning, and energy storage. The computer hardware for reinforcement learning training includes: Intel I9 CPU, 32GB memory, GTX 4070TI GPU. The number of reinforcement learning iterations is set to 500. The peak load regulation scenarios are aggregated into 50 scenarios.

[0087] In order to compare the performance of the method proposed in this embodiment, two different methods were set up for simulation analysis in the case analysis. The results are as follows: Figure 2 As shown. Method 1: The method proposed in this embodiment (M1 curve in the figure); Method 2: Based on the traditional hierarchical analysis method (M2 curve in the figure). The true curve in the figure is the actual peak-shaving effect of the microgrid group. The horizontal axis in the figure represents the peak-shaving scenario, and the vertical axis represents the collaborative peak-shaving effect of the microgrid group under each peak-shaving scenario. It can be seen that the evaluation effect proposed in this embodiment can effectively evaluate the actual peak-shaving effect, and the maximum evaluation error is about 5%.

[0088] In order to compare the optimization performance of the reinforcement learning proposed in this embodiment, two different methods were set up for simulation analysis in the case analysis. The results are as follows: Figure 3 and Figure 4 shown. Figure 3 is the convergence curve of the method proposed in this embodiment, Figure 4 It is the convergence curve of the genetic algorithm. It can be seen that the optimization speed of the method proposed in this embodiment is faster, thereby improving the optimization ability of the proposed method.

[0089] The preferred embodiments of the present invention are described in detail above in conjunction with the accompanying drawings. The embodiments described in the present invention are merely descriptions of the preferred embodiments of the present invention, and do not limit the concept and scope of the present invention. The various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction, and such combinations should also be regarded as the contents disclosed in the present disclosure as long as they do not violate the concept of the present invention. In order to avoid unnecessary repetition, the present invention will not further describe various possible combinations.

[0090] The present invention is not limited to the specific details of the above-mentioned embodiments. Within the scope of the technical concept of the present invention and without departing from the design concept of the present invention, various modifications and improvements made to the technical solution of the present invention by technical personnel in this field should fall within the protection scope of the present invention. The technical contents for which protection is sought in the present invention have been fully recorded in the claims.

Claims

1. An intelligent evaluation method for the coordinated peak-shaving effect of a microgrid group, characterized in that: The following steps are involved: Step 1: Establish an evaluation index system for the coordinated peak-shaving effect of microgrid groups; Step 2: Based on the microgrid group coordinated peak-shaving effect evaluation index system established in step 1, a microgrid group coordinated peak-shaving effect comprehensive evaluation model is constructed based on the fuzzy evidence fusion framework. The process is as follows: S201. Based on the microgrid group coordinated peak-shaving effect evaluation index system obtained in step 1, a microgrid group coordinated peak-shaving effect comprehensive evaluation index set X applicable to the fuzzy evidence fusion framework is established. u , X u The elements in are the identification evaluation index fuzzy mapping thresholds of each evaluation index in the microgrid group collaborative peak-shaving effect evaluation index system; Map each evaluation index in the microgrid group coordinated peak-shaving effect evaluation index system obtained in step 1 to the microgrid group coordinated peak-shaving effect comprehensive evaluation index set X u , when the evaluation index in the microgrid group collaborative peak-shaving effect evaluation index system reaches the corresponding mapping threshold, the evaluation index value is mapped to the evidence evaluation effect corresponding to the mapping threshold; S202. Calculate the comprehensive evaluation index set X of the coordinated peak-shaving effect of the microgrid group u The distance between the evidence characteristics of each piece of evidence; S203, based on the obtained evidence feature distance, determine the similarity between the evidences in the comprehensive evaluation index set of the microgrid group coordinated peak-shaving effect; S204, based on the obtained similarity between each piece of evidence, determine the support level of each piece of evidence in the comprehensive evaluation index set of microgrid group coordinated peak-shaving effect; S205. Determine the evaluation weight of the corresponding evidence on the coordinated peak-shaving effect of the microgrid group based on the support degree, and establish a comprehensive evaluation model for the coordinated peak-shaving effect of the microgrid group according to the evaluation weight; Step 3: Obtain the evaluation results through the comprehensive evaluation model of the coordinated peak-shaving effect of the microgrid group.

2. According to claim 1, a method for intelligently evaluating the coordinated peak-shaving effect of a microgrid group is characterized in that: In step 1, the evaluation indicators in the microgrid group collaborative peak-shaving effect evaluation index system include a new energy peak-shaving uncertainty index, a load peak-shaving uncertainty index, and an energy storage response effect index.

3. According to claim 2, a method for intelligently evaluating the coordinated peak-shaving effect of a microgrid group is characterized in that: The new energy peak-shaving uncertainty indicators include a new energy peak-shaving response delay uncertainty indicator, a new energy peak-shaving response capacity uncertainty indicator, a new energy peak-shaving response speed uncertainty indicator, and a new energy peak-shaving response cost uncertainty indicator.

4. According to claim 1, a method for intelligently evaluating the coordinated peak-shaving effect of a microgrid group is characterized in that: Step 2 S201 also includes: The power grid operation state model S suitable for reinforcement learning algorithm is established under the current identification and evaluation index fuzzy mapping threshold. grid ; The fuzzy mapping threshold of the identification evaluation index is used as the action model A of the reinforcement learning algorithm grid ; Use reinforcement learning algorithm to continuously iterate and optimize action model A grid Specific variable values ​​to realize action model A grid Optimal, that is, achieving the optimal threshold of fuzzy mapping of identification and evaluation indicators; Then, the optimal identification evaluation index fuzzy mapping threshold is used as the comprehensive evaluation index set X of the microgrid group collaborative peak regulation effect. u The elements in .

Images