Method and system for dispatching comprehensive energy of power distribution network based on myxobacteria algorithm

By introducing chaotic reverse learning and adaptive strategies into the slimming mold algorithm, combining random search and Bayesian optimization algorithm to optimize the LightGBM model hyperparameters, the problem of local optimization and insufficient accuracy of the slimming mold algorithm in complex energy systems is solved. Significantly improves the accuracy and stability of energy scheduling.

CN119204541APending Publication Date: 2024-12-27GUIZHOU POWER GRID CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411282744.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-13
Publication Date
2024-12-27

AI Technical Summary

Technical Problem

Traditional slime mold algorithms are prone to fall into local optimization in the optimization of complex energy systems, and the model accuracy is insufficient, resulting in low convergence accuracy of energy scheduling.

Method used

By combining chaotic reverse learning and adaptive strategies to optimize the basic slime mold algorithm, population diversity and exploration capabilities are enhanced, and the LightGBM model hyperparameters are optimized using random search and Bayesian optimization algorithm to build a target function for comprehensive energy scheduling. to improve algorithm performance.

Benefits of technology

It effectively avoids the problem of slime mold algorithm falling into local optimality, improves accuracy and stability, and improves the economy, safety, reliability and stability of the comprehensive energy system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119204541A_ABST
    Figure CN119204541A_ABST
Patent Text Reader

Abstract

The invention discloses a power distribution network integrated energy scheduling method and system based on a myxobacteria algorithm, and relates to the technical field of system scheduling, and the method comprises the steps: obtaining a first optimization algorithm which is obtained through the optimization of a basic myxobacteria algorithm through a first learning strategy and a second learning strategy, the first optimization algorithm comprises a first sub-step and a second sub-step, and each sub-step corresponds to one optimization operation; and according to the information of the first optimization algorithm, constructing an objective function of integrated energy scheduling by using a second optimization algorithm, and optimizing the first sub-step and the second sub-step based on the objective function. According to the method, the comprehensive energy system is optimized by using the improved mucilage bacteria algorithm and the LightGBM model, so that the economical efficiency, the safety, the reliability and the stability of the comprehensive energy system can be improved, the optimal target value of the controllable parameter can be obtained through calculation, and adjustment operation guidance can be provided for operators.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of system dispatch optimization, and in particular to a method and system for comprehensive energy dispatching of a distribution network based on a slime mold algorithm. Background Art

[0002] The integrated energy dispatching system can achieve the cascade utilization of energy and the organic coordination among energy systems through the unified planning and coordinated operation of multiple energy systems, improve energy utilization, promote the consumption of renewable energy, and achieve the goals of increasing energy supply and saving energy.

[0003] At present, optimizing the energy dispatch of distribution networks through intelligent optimization algorithms is an important guarantee for the distribution networks to operate in the most economical and environmentally friendly state. However, different intelligent optimization algorithms will produce different results when applied to the energy dispatch of distribution networks. However, the current mainstream intelligent algorithms have different defects, such as the performance of particle swarm optimization algorithms will be affected by parameters, different problems require different parameter controls, lack of dynamic adjustment of speed during the search process, resulting in low convergence accuracy; genetic algorithms are prone to fall into local optimality. Therefore, it is necessary to improve the existing algorithms to optimize the operation of the integrated energy system, thereby improving its economy, safety, reliability and stability. Summary of the invention

[0004] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0005] In view of the technical problems that traditional slime mold algorithms are prone to fall into local optimality in the optimization of complex energy systems and the insufficient accuracy of existing models, the present invention is proposed.

[0006] Therefore, the present invention provides a distribution network comprehensive energy scheduling method and system based on the slime mold algorithm, which can solve the problems mentioned in the background technology.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0008] In a first aspect, the present invention provides a method for comprehensive energy dispatching of a distribution network based on a slime mold algorithm, comprising: obtaining a first optimization algorithm, wherein the first optimization algorithm is obtained by optimizing a basic slime mold algorithm through a first learning strategy and a second learning strategy, wherein the first optimization algorithm comprises a first sub-step and a second sub-step, and each sub-step corresponds to an optimization operation;

[0009] According to the information of the first optimization algorithm, the objective function of the comprehensive energy scheduling is constructed using the second optimization algorithm, and the first sub-step and the second sub-step are optimized based on the objective function.

[0010] As a preferred solution of the comprehensive energy dispatching method for distribution network based on slime mold algorithm described in the present invention, wherein: the first optimization algorithm also includes:

[0011] Initialize the parameters, set the number of slime mold individuals N, the problem dimension D, and the maximum number of iterations t max , search boundaries ub and lb;

[0012] The basic slime mold algorithm is optimized through the first learning strategy;

[0013] Record the best fitness bF and the worst fitness wF among slime mold individuals;

[0014] Update control parameters p, vb, vc and weight coefficient W;

[0015] The basic slime mold algorithm is optimized through the second learning strategy;

[0016] Update slime mold position;

[0017] Update the current optimal solution, calculate the fitness values ​​before and after the update, retain the best individual, iterate, and after reaching the maximum number of times, output the global optimal solution and fitness value.

[0018] As a preferred solution of the integrated energy dispatching method for distribution network based on slime mold algorithm described in the present invention, wherein: the first learning strategy includes enhancing the diversity of the basic slime mold algorithm population through chaotic reverse learning; the second learning strategy includes balancing the exploration ability of the basic slime mold algorithm through an adaptive strategy.

[0019] As a preferred solution of the comprehensive energy dispatching method for distribution network based on slime mold algorithm described in the present invention, wherein: the information of the first optimization algorithm includes a first parameter and a second parameter; the first parameter is related to the number of iterations of the first optimization algorithm, and the second parameter is related to the control parameter of the first optimization algorithm.

[0020] As a preferred solution of the distribution network integrated energy dispatching method based on the slime mold algorithm of the present invention, wherein: the objective function of constructing the integrated energy dispatching includes:

[0021] Determine the influencing factors of the integrated energy system; collect modeling data and preprocess it; optimize the hyperparameters of the second optimization algorithm; and establish the objective function of the integrated energy system.

[0022] As a preferred solution of the comprehensive energy dispatching method for distribution network based on slime mold algorithm described in the present invention, wherein: the acquisition of modeling data and preprocessing includes:

[0023] The SGF filter is used to perform noise reduction on the collected modeling data; the noise-reduced data is converted to the [0, 1] interval using the normalization method;

[0024] The calculation formula of SG filter is as follows:

[0025]

[0026] Among them, x is the data to be fitted, b k are the coefficients of the polynomial, y is the output data after fitting, and b is the parameter to be solved.

[0027] As a preferred solution of the comprehensive energy dispatching method for distribution network based on slime mold algorithm described in the present invention, wherein: the optimization of the hyperparameters of the second optimization algorithm includes:

[0028] A first optimization algorithm is used to optimize a first set of hyperparameters, where the first optimization algorithm is a random search algorithm, and the first set of hyperparameters includes at least one of max_depth, num_leaves, bagging_fraction, min_data_in_leaf, and feature_fraction_seed;

[0029] A second optimization algorithm is used to optimize a second set of hyperparameters, where the second optimization algorithm is a Bayesian optimization algorithm, and the second set of hyperparameters includes at least one of feature_fraction, min_data_in_leaf, lambda_l1, and lambda_l2;

[0030] The selection of the first optimization algorithm and the second optimization algorithm is based on the following principle: for hyperparameters taking integer values, a random search algorithm is used for optimization; for hyperparameters taking continuous values, a Bayesian optimization algorithm is used for optimization.

[0031] In a second aspect, the present invention provides a distribution network integrated energy dispatching system based on a slime mold algorithm, comprising: an algorithm optimization module, used to obtain a first optimization algorithm, wherein the first optimization algorithm is obtained by optimizing a basic slime mold algorithm through a first learning strategy and a second learning strategy, wherein the first optimization algorithm comprises a first sub-step and a second sub-step, each sub-step corresponding to an optimization operation;

[0032] The integrated energy scheduling module is used to construct an objective function of integrated energy scheduling using a second optimization algorithm according to information of the first optimization algorithm, and optimize the first sub-step and the second sub-step based on the objective function.

[0033] In a third aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the processor executes the computer program, it implements the steps of a distribution network comprehensive energy scheduling method based on a slime mold algorithm.

[0034] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, the steps of a distribution network integrated energy scheduling method based on a slime mold algorithm are implemented.

[0035] Compared with the prior art, the beneficial effect of the present invention is that the improved slime mold algorithm contains a variety of improved algorithms, which can effectively avoid the problem of the slime mold algorithm falling into local optimality. The improved slime mold algorithm has higher accuracy and stronger stability, and is more suitable for the optimization of integrated energy systems; the random search algorithm is combined with the Bayesian optimization algorithm to optimize the hyperparameters of the LightGBM model, and different optimization algorithms are selected for different hyperparameters, which are more targeted and can effectively overcome the limitations of the hyperparameters of the LightGBM model, and can improve the accuracy and stability of the LightGBM model; the integrated energy system is optimized using the improved slime mold algorithm and the LightGBM model to improve its economy, safety, reliability and stability, and the optimal target values ​​of the controllable parameters are calculated, which can provide adjustment guidance for operators. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0037] Figure 1 A method flow chart of a method and system for comprehensive energy dispatching of a distribution network based on a slime mold algorithm provided by an embodiment of the present invention;

[0038] Figure 2 An improved slime mold algorithm flow chart of a distribution network integrated energy dispatching method and system based on a slime mold algorithm provided by an embodiment of the present invention;

[0039] Figure 3A flow chart of an objective function of building an integrated energy system for a distribution network integrated energy dispatching method and system based on a slime mold algorithm provided by an embodiment of the present invention;

[0040] Figure 4 An internal structure diagram of a computer device of a distribution network integrated energy dispatching method and system based on a slime mold algorithm provided in one embodiment of the present invention. DETAILED DESCRIPTION

[0041] In order to make the above-mentioned purposes, features and advantages of the present invention more understandable, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0042] The terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and need not be used to describe a specific order or sequential order. It should be understood that the terms used in this way can be interchangeable under appropriate circumstances, which is only to describe the distinction mode adopted by the objects of the same attributes when describing in the embodiments of the present application. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, so that the process, method, system, product or equipment comprising a series of units need not be limited to those units, but may include other units that are not clearly listed or inherent to these processes, methods, products or equipment.

[0043] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0044] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0045] Example 1

[0046] Reference Figure 1-Figure 4 This is the first embodiment of the present invention, which provides a comprehensive energy scheduling method for distribution network based on slime mold algorithm. In order to make this application clearer, some concepts and processing procedures mentioned in this application are first briefly introduced.

[0047] Slime Mould Algorithm (SMA): A bionic intelligent optimization algorithm that simulates the behavioral characteristics of fungi. It solves complex optimization problems by simulating the movement, division and fusion behaviors of slime molds in the process of searching for food.

[0048] Chaotic Opposition-Based Learning (COBL): A strategy to enhance population diversity by introducing chaotic mapping and reverse learning mechanisms to generate new solutions relative to the current solution to expand the search space.

[0049] LightGBM: An efficient gradient boosting framework based on decision trees, with the characteristics of fast training speed and small memory usage. It is often used to solve machine learning problems such as classification and regression.

[0050] Random Search Algorithm: A simple and effective hyperparameter optimization method that searches for the optimal parameter combination by randomly sampling in a predefined parameter space.

[0051] Bayesian Optimization Algorithm: A global optimization method based on a probability model. It guides the search process by constructing a probability model of the objective function. It is suitable for optimizing black-box functions with high computational costs.

[0052] SG filter (Savitzky-Golay Filter): A digital filtering and smoothing technique that achieves data smoothing and noise removal by performing local polynomial regression on the data while maintaining the high-order moments of the signal.

[0053] Normalization: The process of converting data of different ranges to a uniform interval (such as [0,1]) helps to eliminate the dimensional differences between different features and improve the training effect of the model.

[0054] The present application provides a method that can effectively solve the above-mentioned problems. Next, multiple embodiments will be combined to explain in detail how to implement the comprehensive energy dispatching method for distribution network based on slime mold algorithm;

[0055] Figure 1 A method flow chart of a distribution network integrated energy dispatching method and system based on a slime mold algorithm is shown, including:

[0056] S1: Obtain a first optimization algorithm, where the first optimization algorithm is obtained by optimizing a basic slime mold algorithm through a first learning strategy and a second learning strategy, and the first optimization algorithm includes a first sub-step and a second sub-step, and each sub-step corresponds to an optimization operation;

[0057] Further, see Figure 3 , the first optimization algorithm also includes,

[0058] Initialize the parameters, set the number of slime mold individuals N, the problem dimension D, and the maximum number of iterations t max , search boundaries ub and lb;

[0059] The basic slime mold algorithm is optimized through the first learning strategy;

[0060] Record the best fitness bF and the worst fitness wF among slime mold individuals;

[0061] Update control parameters p, vb, vc and weight coefficient W;

[0062] The update formula of the control parameter p is as follows:

[0063] p=tanh|S(i)-DF|

[0064] Where S(i) represents the fitness value of X, i = 1, 2, ..., N, and DF is the best fitness of each iteration;

[0065] The update formula of the control parameter vb is as follows:

[0066] vb=[-a,a]

[0067] Among them, a is the control strategy parameter;

[0068] The control parameter vc is adaptively adjusted. The adaptive adjustment formula of the control parameter vc is as follows:

[0069]

[0070] Where t is the number of iterations, t max is the maximum number of iterations.

[0071] The update formula of the weight coefficient W is as follows:

[0072]

[0073] SmellIndex = sort(S)

[0074] Among them, r is a random number between [0,1], bF is the best fitness value of the current individual, wF is the worst fitness value of the current individual, condition is the individuals with fitness values ​​in the first half of the current population, others is the individuals with fitness values ​​in the second half of the current population, and SmellIndex is the descending order of fitness values;

[0075] The basic slime mold algorithm is optimized through the second learning strategy;

[0076] Update the slime mold position, the calculation formula is as follows:

[0077]

[0078] Among them, X * is the updated position of the slime mold, X b (t) is the individual with the highest fitness in the tth generation, X(t) is the position of the tth generation slime mold, and X A (t) and X B (t) are two random individuals in the tth generation, rand and r are random values ​​between [0, 1], and z is a custom parameter;

[0079] Update the current optimal solution, calculate the fitness values ​​before and after the update, retain the best individual, iterate, and after reaching the maximum number of times, output the global optimal solution and fitness value.

[0080] Furthermore, the first learning strategy includes enhancing the diversity of the basic slime mold algorithm population through chaotic reverse learning; the second learning strategy includes balancing the exploration ability of the basic slime mold algorithm through an adaptive strategy.

[0081] Specifically, the optimization of the slime mold algorithm includes enhancing the diversity of the population based on chaotic reverse learning and balancing the exploration ability of the algorithm through adaptive strategies. The specific calculation formula is as follows:

[0082]

[0083] in, is the chaotic reverse solution corresponding to the i-th individual in the population, lb and ub are the boundary ranges; λ i is the corresponding chaotic map value, is the current solution.

[0084] The exploration capability of the adaptive strategy balancing algorithm is to optimize the parameter a. The optimized parameter a is:

[0085]

[0086] Furthermore, the information of the first optimization algorithm includes a first parameter and a second parameter; the first parameter is related to the number of iterations of the first optimization algorithm, and the second parameter is related to a control parameter of the first optimization algorithm.

[0087] It should be noted that when improving the slime mold algorithm, optimization strategies such as the spiral search strategy / lens imaging reverse learning strategy can also be used to expand the search range and better improve the global exploration performance.

[0088] S2: Based on the information of the first optimization algorithm, a second optimization algorithm is used to construct an objective function of comprehensive energy scheduling, and the first sub-step and the second sub-step are optimized based on the objective function.

[0089] Furthermore, the objective function of building comprehensive energy dispatch includes:

[0090] Determine the influencing factors of the integrated energy system; collect modeling data and preprocess it; optimize the hyperparameters of the second optimization algorithm; and establish the objective function of the integrated energy system.

[0091] Furthermore, the influencing factors vary due to the different integrated energy systems. Therefore, in the actual process, the influencing factors need to be determined in combination with the actual integrated energy system, and the present invention does not make specific limitations.

[0092] Furthermore, collecting modeling data and preprocessing includes:

[0093] The SGF filter is used to perform noise reduction on the collected modeling data; the noise-reduced data is converted to the [0, 1] interval using the normalization method;

[0094] The calculation formula of SG filter is as follows:

[0095]

[0096] Among them, x is the data to be fitted, b k are the coefficients of the polynomial, y is the output data after fitting, and b is the parameter to be solved.

[0097] Furthermore, the normalized calculation formula is as follows:

[0098]

[0099] Among them, x is the collected modeling data, y is the normalized data, and x max and x min are the maximum and minimum values ​​of the collected modeling data, y max and min are the normalized maximum and minimum values, respectively.

[0100] Furthermore, optimizing the hyperparameters of the second optimization algorithm includes:

[0101] A first optimization algorithm is used to optimize a first set of hyperparameters, where the first optimization algorithm is a random search algorithm, and the first set of hyperparameters includes at least one of max_depth, num_leaves, bagging_fraction, min_data_in_leaf, and feature_fraction_seed;

[0102] A second optimization algorithm is used to optimize a second set of hyperparameters, where the second optimization algorithm is a Bayesian optimization algorithm, and the second set of hyperparameters includes at least one of feature_fraction, min_data_in_leaf, lambda_l1, and lambda_l2;

[0103] The selection of the first optimization algorithm and the second optimization algorithm is based on the following principle: for hyperparameters taking integer values, a random search algorithm is used for optimization; for hyperparameters taking continuous values, a Bayesian optimization algorithm is used for optimization.

[0104] It should be noted that the reason for using different algorithms to optimize hyperparameters is that the Bayesian optimization algorithm cannot be set to an integer value, and the values ​​of hyperparameters such as max_depth, Num_leaves, and Bagging_fraction can only be integers. If the Bayesian algorithm is used, it will incur huge computational overhead, so the random search algorithm is used to optimize it, saving computational overhead and achieving the best effect.

[0105] Furthermore, the objective function of the integrated energy system is established, where the objective function is:

[0106]

[0107] In the formula, T is the total number of time periods, is the total electricity purchase cost at time t, is the total gas purchase cost at time t, is the total wind curtailment cost at time t, is the total abandoned light cost at time t, C C is the total carbon transaction cost, N W is the number of nodes connected to wind power in the power system, N PV is the number of photovoltaic nodes connected to the power system, is the cost coefficient of unit electricity purchase at time t, is the cost coefficient of unit gas purchase at time t, is the unit wind curtailment cost coefficient of wind power generation node i at time t, is the unit cost coefficient of photovoltaic power generation node j at time t, P t E is the amount of electricity purchased by the distribution network from its superior at time t, P t G is the amount of gas purchased by the distribution network from its superior at time t, is the amount of wind abandoned at wind power generation node i at time t, is the amount of abandoned light at photovoltaic power generation node j at time t, p is the unit carbon trading price, E and D are carbon emissions and carbon quotas respectively.

[0108] Example 2

[0109] According to an embodiment of the present invention, an energy consumption analysis and optimization management system for a distributed photovoltaic power station is provided, comprising: an algorithm optimization module, for obtaining a first optimization algorithm, wherein the first optimization algorithm is obtained by optimizing a basic slime mold algorithm through a first learning strategy and a second learning strategy, wherein the first optimization algorithm comprises a first sub-step and a second sub-step, and each sub-step corresponds to an optimization operation;

[0110] The integrated energy scheduling module is used to construct an objective function of integrated energy scheduling using a second optimization algorithm according to information of the first optimization algorithm, and optimize the first sub-step and the second sub-step based on the objective function.

[0111] Example 3

[0112] Reference Figure 4 , which is the third embodiment of the present invention. Based on the first two embodiments, the embodiment of the present invention provides a computer device, which may be a server, and its internal structure diagram may be as shown in FIG. Figure 4 As shown. The computer device includes a processor, a memory and a network interface connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium.

[0113] The database of the computer device is used to store action detection data. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the steps in any of the above-mentioned sparse tensor operation acceleration method embodiments are implemented.

[0114] Those skilled in the art will understand that Figure 4 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0115] In one embodiment, an embodiment of the present invention provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps in any of the above-mentioned sparse tensor operation acceleration method embodiments are implemented.

[0116] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0117] Example 4

[0118] Reference Figure 1 - Figure 2 , which is the second embodiment of the present invention, and this embodiment provides a comprehensive energy dispatching method for distribution network based on slime mold algorithm. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0119] The following is a comparative analysis of the method proposed in the present invention, the slime mold algorithm, and the differential evolution algorithm. Different algorithms are set to run 1000 iterations, using a variety of different initial conditions and a population size of 110.

[0120] To avoid algorithm randomness problems, the three algorithms are independently executed 50 times and the best result is selected. Table 1 shows the statistical results of the three different algorithms for the above test examples.

[0121] Table 1 Statistics of fitness results of three different algorithms

[0122] algorithm average value Standard Deviation Differential Evolution Algorithm 124234.12 442.11 Slime Mold Algorithm 122654.25 38.48 The method of the present invention 120131.24 6.52

[0123] In addition, it is worth noting that the contents not elaborated in detail in the present invention are all prior arts and are well known to those skilled in the art.

[0124] Therefore, the present invention adopts the above-mentioned integrated energy dispatching method for distribution network based on slime mold algorithm, takes LightGBM model as the main body, and combines the improved slime mold algorithm to optimize the operation of the integrated energy system, which can improve the economy, safety, reliability and stability of the integrated energy system.

[0125] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A comprehensive energy dispatching method for distribution network based on slime mold algorithm, characterized by: include, Obtaining a first optimization algorithm, where the first optimization algorithm is obtained by optimizing a basic slime mold algorithm through a first learning strategy and a second learning strategy, and the first optimization algorithm includes a first sub-step and a second sub-step, and each sub-step corresponds to an optimization operation; According to the information of the first optimization algorithm, the objective function of the comprehensive energy scheduling is constructed using the second optimization algorithm, and the first sub-step and the second sub-step are optimized based on the objective function.

2. The method for comprehensive energy dispatching of distribution network based on slime mold algorithm according to claim 1, characterized in that: The first optimization algorithm also includes: Initialize the parameters, set the number of slime mold individuals N, the problem dimension D, and the maximum number of iterations t max , search boundaries ub and lb; The basic slime mold algorithm is optimized through the first learning strategy; Record the best fitness bF and the worst fitness wF among slime mold individuals; Update control parameters p, vb, vc and weight coefficient W; The basic slime mold algorithm is optimized through the second learning strategy; Update slime mold position; Update the current optimal solution, calculate the fitness values ​​before and after the update, retain the best individual, iterate, and after reaching the maximum number of times, output the global optimal solution and fitness value.

3. The method for comprehensive energy dispatching of distribution network based on slime mold algorithm according to claim 2, characterized in that: The first learning strategy includes enhancing the diversity of the basic slime mold algorithm population through chaotic reverse learning; the second learning strategy includes balancing the exploration ability of the basic slime mold algorithm through an adaptive strategy.

4. The method for comprehensive energy dispatching of distribution network based on slime mold algorithm as claimed in claim 3, characterized in that: The information of the first optimization algorithm includes a first parameter and a second parameter; the first parameter is related to the number of iterations of the first optimization algorithm, and the second parameter is related to the control parameter of the first optimization algorithm.

5. The method for comprehensive energy dispatching of distribution network based on slime mold algorithm according to claim 4, characterized in that: The objective function of constructing comprehensive energy scheduling includes: Identify factors affecting integrated energy systems; Collect modeling data and preprocess them; Optimizing the hyperparameters of the second optimization algorithm; Establish the objective function of the integrated energy system.

6. The method for comprehensive energy dispatching of distribution network based on slime mold algorithm according to claim 5, characterized in that: The collecting and preprocessing of modeling data includes: The SGF filter is used to perform noise reduction on the collected modeling data; the noise-reduced data is converted to the [0, 1] interval using the normalization method; The calculation formula of SG filter is as follows: Among them, x is the data to be fitted, b k are the coefficients of the polynomial, y is the output data after fitting, and b is the parameter to be solved.

7. The method for comprehensive energy dispatching of distribution network based on slime mold algorithm according to claim 6, characterized in that: Optimizing the hyper parameters of the second optimization algorithm includes: A first optimization algorithm is used to optimize a first set of hyperparameters, where the first optimization algorithm is a random search algorithm, and the first set of hyperparameters includes at least one of max_depth, num_leaves, bagging_fraction, min_data_in_leaf, and feature_fraction_seed; A second optimization algorithm is used to optimize a second set of hyperparameters, where the second optimization algorithm is a Bayesian optimization algorithm, and the second set of hyperparameters includes at least one of feature_fraction, min_data_in_leaf, lambda_l1, and lambda_l2; The selection of the first optimization algorithm and the second optimization algorithm is based on the following principle: for hyperparameters taking integer values, a random search algorithm is used for optimization; for hyperparameters taking continuous values, a Bayesian optimization algorithm is used for optimization.

8. A distribution network integrated energy dispatching system based on a slime mold algorithm, based on the distribution network integrated energy dispatching method based on a slime mold algorithm according to any one of claims 1 to 7, characterized in that: include, An algorithm optimization module, used to obtain a first optimization algorithm, wherein the first optimization algorithm is obtained by optimizing a basic slime mold algorithm through a first learning strategy and a second learning strategy, wherein the first optimization algorithm includes a first sub-step and a second sub-step, and each sub-step corresponds to an optimization operation; The integrated energy scheduling module is used to construct an objective function of integrated energy scheduling using a second optimization algorithm according to information of the first optimization algorithm, and optimize the first sub-step and the second sub-step based on the objective function.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the distribution network comprehensive energy scheduling method based on the slime mold algorithm according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the comprehensive energy dispatching method for distribution network based on slime mold algorithm described in any one of claims 1 to 7 are implemented.