Mine integrated energy system scheduling two-stage multi-form differential evolution method
By constructing alternative tasks and a two-stage optimization framework, combined with neighborhood double mutation and knowledge transfer strategies, the uncertainty problem in CMIES scheduling is solved, and efficient and accurate energy utilization and scheduling optimization are achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHINA UNIV OF MINING & TECH
- Filing Date
- 2025-01-07
- Publication Date
- 2026-06-23
AI Technical Summary
Uncertainty exists in the scheduling of integrated mine energy systems (CMIES), making it difficult for scheduling schemes to accurately match actual needs, thus affecting the accuracy and effectiveness of optimization. Existing methods suffer from high computational complexity and inaccurate results when dealing with diverse energy types and complex conversion processes.
A two-stage, multi-form differential evolution method is adopted. By constructing two alternative tasks, namely electric-heat and electric-cold, and combining neighborhood double mutation strategy and knowledge transfer strategy, the complexity of the problem is reduced and the efficiency of locating feasible solutions is improved. Furthermore, the convergence performance of the algorithm is enhanced by improving interval constraint processing technology.
It effectively solves the uncertainty problem in CMIES scheduling, improves the accuracy and energy efficiency of the scheduling scheme, reduces computational complexity, and improves the algorithm's performance in terms of diversity and convergence.
Smart Images
Figure CN119831105B_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of intelligent scheduling technology, and specifically relates to a two-stage multi-form differential evolution method for scheduling integrated energy systems in mines. Background Technology
[0002] In recent years, Integrated Mining Ecosystems (CMIES) have attracted much attention due to their ability to effectively alleviate the high energy consumption and pollution problems in mines. CMIES integrates traditional integrated energy systems (IESs) such as wind power, photovoltaics, gas turbines, and energy storage devices, while fully utilizing associated energy sources in coal mines, such as air heat, exhaust air, mine water, and geothermal energy. These associated energy resources are abundant, and direct discharge would result in waste and even negative environmental impacts. Therefore, how to efficiently utilize these associated energy sources is an important issue in promoting the transformation of the mine's energy structure and reducing carbon emissions.
[0003] Compared to traditional energy sources (IESs), CMIES involve more diverse energy types and more complex energy conversion processes, increasing the difficulty of system management and optimization. Particularly in scheduling optimization, the stable operation of CMIES relies on reasonable scheduling strategies, which not only improve the system's economic efficiency but also enhance the utilization efficiency of associated energy sources in the mine. However, system scheduling must also address the uncertainties in the output of many associated energy sources. For example, air heat and exhaust gas volume fluctuate with mine ventilation and production capacity; mine water flow is affected by climate and geological conditions; and geothermal resources exhibit seasonal variations. These uncertainties make it difficult for scheduling schemes to accurately match actual needs, significantly impacting the accuracy and effectiveness of optimization. Therefore, further consideration of uncertainties in CMIES scheduling is necessary, but it also significantly increases the complexity of optimization.
[0004] To address the uncertainties in scheduling optimization, various methods have been developed in recent years, including stochastic programming, fuzzy programming, robust optimization, and interval optimization. Stochastic programming describes uncertain variables through probability distributions, aiming to obtain the optimal solution under the expected conditions; however, it relies on a large amount of historical data, has high computational complexity, and its solutions are often overly optimistic. Fuzzy programming uses fuzzy numbers to handle uncertainty, allowing for some constraint relaxation, but its results are limited by the definition of membership functions and are prone to large errors. In contrast, robust optimization and interval optimization do not rely on large amounts of data or pre-assumed parameter distributions, making them more advantageous in application. Robust optimization focuses on worst-case optimization and is a conservative decision-making method; interval optimization focuses on the overall performance of the target interval, often transforming uncertain problems into deterministic problems through interval order relations or minimax regret criteria. However, this model transformation may lose some uncertainty information, and different transformation models may lead to the same problem being transformed into different deterministic problems, affecting the accuracy of the optimization results and the stability of the system.
[0005] Therefore, it is particularly important to develop a direct method that does not rely on model transformation to handle the uncertainty problem in CMIES scheduling, retain more uncertainty information, and improve the accuracy of the scheduling scheme. Summary of the Invention
[0006] The problem to be solved by this invention is to provide a two-stage, multi-form differential evolution method for scheduling integrated energy systems in mines, which comprehensively considers the multiple uncertainties of associated energy and load in mines, thereby improving the economic benefits and energy utilization efficiency of the system.
[0007] The present invention adopts the following technical solution: a two-stage multi-form differential evolution method for scheduling a mine integrated energy system, comprising the following steps: Step 1, constructing alternative tasks: under the multi-form optimization framework, for the original model of the mine integrated energy system, construct two alternative tasks, electric-heat and electric-cool, to solve the feasible region location problem caused by strong coupling constraints in the interval.
[0008] Step 2: Construct a knowledge transfer strategy based on two alternative tasks. Each time, transfer the coupled variable part of the feasible solution in the two alternative tasks to improve the efficiency of locating feasible solutions.
[0009] Step 3: Construct a two-stage optimization framework: the first stage is used to search multiple feasible domains, and the second stage is used to perform a fine search on the original task.
[0010] Step 4: Construct a differential evolution strategy based on neighborhood double mutation: Combine the DE / rand / 1 and DE / current-to-pbest / 1 operators to construct a double mutation mechanism that integrates the neighborhood mechanism. Optimize the corresponding tasks in the two stages respectively. The mutation strategy in the first stage is a diversity-driven neighborhood double mutation strategy, and the mutation strategy in the second stage is a convergence-driven neighborhood double mutation strategy.
[0011] Step 5: Perform two-stage multi-form differential evolution for the scheduling of the integrated energy system in the mine: Based on the alternative tasks constructed in Step 1 and the two-stage optimization framework built in Step 3, in the first stage, the populations corresponding to the two alternative tasks are evolved in parallel through the diversity-driven neighborhood double mutation strategy. Knowledge is transferred between the two tasks through the knowledge transfer strategy. The population obtained at the end of the evolution is used as the initial population for the second stage. The original task is optimized through the convergence-driven neighborhood double mutation strategy to obtain the optimal feasible solution set.
[0012] Step 6: Analyze the time complexity to demonstrate the rationality of the proposed method for optimizing the scheduling of the integrated energy system in mines.
[0013] Specifically, due to the strong coupling and uncertainty of constraints, the original CMIES scheduling problem is difficult to solve directly. To address this challenge, this invention draws on the idea of multi-form optimization, reducing the difficulty of solving the problem and achieving efficient optimization by formulating alternative tasks for the original task. In fact, methods such as changing constraints, altering the objective function structure, and reducing the number of control variables have achieved good results in formulating alternative tasks for multi-form optimization. The core of these methods lies in simplifying the task to facilitate solving, rather than blindly pursuing a high-quality solution.
[0014] Preferably, in step 1, based on domain knowledge, two alternative tasks are constructed by decoupling the coupling relationships in the electrical, cold, and heat equation constraints. Specifically, the coupling space of electrical, cold, and heat variables is divided into two low-dimensional subspaces: the electrical-heat subspace and the electrical-cold subspace, thereby forming two alternative tasks: the electrical-heat task and the electrical-cold task.
[0015] An electrothermal task is represented as follows: .
[0016] The electric-cooling task is represented as follows: Among them, optimization variables Variables are divided into three categories, including those related to electricity. Variables related to heat and variables related to cold ;function Represent the objective function; function Represents the economic operating cost target; function This indicates the abandonment of renewable energy and associated energy cost targets in mining; Indicates the search space; express 3D real space.
[0017] In the electric-thermal task, variables and constraints related to cold are ignored; in the electric-cold task, variables and constraints related to heat are ignored. By decoupling the relationship between electricity, heat, and cold, the variable dimension and constraint coupling strength of the original task are reduced.
[0018] Preferably, in step 2, a feasibility-driven knowledge transfer strategy is constructed to exchange feasible solutions between the two alternative tasks and determine the intersecting feasible regions. The feasibility-driven knowledge transfer strategy performs knowledge transfer only when feasible solutions are generated in both tasks, maximizing the utilization of feasible information from both sides to explore the intersecting feasible regions. Specifically, in the early stages of iteration, the algorithm first searches for feasible regions under each task. Since the feasible regions of the two alternative tasks are larger than those of the original tasks, the algorithm can quickly converge to the feasible solution space, and then transfer the feasible solutions.
[0019] However, it's important to note that these two alternative tasks are heterogeneous because they involve different decision variables, making direct knowledge transfer impossible. Nevertheless, both tasks contain electricity-related decision variables. Considering that the electrical variables in the intersecting feasible regions are the same, this invention only performs knowledge transfer on the electrical variables in the feasible solutions and concatenates them with the hot or cold variables from each task, thereby achieving information exchange. By contributing electrical variables, the two tasks can mutually transfer effective information from their feasible regions, enhancing the ability to find intersecting feasible regions and accelerating the location of feasible solutions.
[0020] Specifically, in the optimization process of evolutionary algorithms, single-stage optimization is prone to getting trapped in local optima, which is even more pronounced when dealing with the scheduling problem of integrated energy systems in mines. Therefore, this invention reduces the risk of getting trapped in local optima by constructing a two-stage optimization strategy that combines extensive exploration and refined search.
[0021] Preferably, in step 3, a two-stage optimization framework is constructed as follows: In the first stage, the algorithm simultaneously optimizes two alternative tasks (electric-thermal task and electric-cold task) to fully explore their respective feasible spaces and obtain the intersecting feasible regions of the two alternative tasks. Then, the second stage focuses on optimizing the original task, emphasizing a refined search of the intersecting feasible regions found in the first stage to obtain the optimal feasible Pareto front. This strategy reduces the complexity of the problem by decoupling the original task in the first stage and effectively avoids the possibility of early convergence to local optima. In the second stage, deep optimization within the confirmed feasible region ensures the globality and convergence of the solution.
[0022] To better allocate computing resources, this invention sets a stage transition condition: when the number of iterations reaches 70% of the maximum number of iterations, the algorithm jumps from the first stage to the second stage.
[0023] Specifically, differential evolution is a simple and efficient evolutionary algorithm that has been widely applied to various optimization problems, especially in integrated energy system scheduling. Based on its superior performance in related fields, this invention selects differential evolution as the core algorithm for CMIES scheduling optimization. However, the original differential evolution is prone to getting trapped in local optima when dealing with high-dimensional and strongly constrained problems.
[0024] Preferably, in step 4, a neighborhood-based double mutation strategy (NDM) is proposed to improve the performance of differential evolution. It combines neighborhood technology and DE / rand / 1 strategy to increase the diversity of the algorithm, and introduces DE / current-to-pbest / 1 strategy to improve the feasibility and convergence speed of the algorithm.
[0025] In the two-stage optimization, the first stage focuses on diversity to explore more feasible domains, while the second stage focuses on feasibility and convergence to obtain a better Pareto front.
[0026] Furthermore, the present invention controls parameters To adjust the performance of the double mutation strategy at different stages, thereby maximizing the overall effectiveness of the algorithm, the specific mutation strategy is as follows:
[0027] Phase 1 mutation strategy: Diversity-driven neighborhood double mutation strategy (NDM_div)
[0028] ;
[0029] Second-stage mutation strategy: Convergence-driven neighborhood double mutation strategy (NDM_con)
[0030] ;
[0031] in, Represents a random number between 0 and 1; Indicates from Two individuals are randomly selected from the neighborhood formed, and ; This represents two individuals randomly selected from the current population, and ; This represents the best individual in the current population; Indicates the top of the population One of the outstanding individuals, Set to 0.1; from Randomly selected from; It is a probability value used to control the performance emphasis of the mutation strategy in the two stages.
[0032] Specifically, after completing the mutation operation, selecting superior individuals to enter the next generation is also one of the key steps in differential evolution. However, due to the uncertainty of the problem in this invention, and since both the target value and the constraint value are interval numbers, traditional Pareto dominance relations and crowding distances cannot effectively determine the dominance relations and crowding levels between interval numbers.
[0033] Furthermore, this invention uses conditional interval dominance relations and interval crowding distance to evaluate individual performance. To address the problem of handling interval constraints, it introduces interval-based feasibility rules and combines these rules with constraint dominance principles and improved epsilon to propose an improved interval environment selection strategy to select better individuals for the next generation.
[0034] Specifically, knowledge transfer occurs only in the first stage, namely on two alternative tasks, to help locate the feasible solution space of the original task. These two alternative tasks have different optimization preferences, and the main objective is to explore the feasible spaces of the two alternative tasks and find the intersecting feasible region between them.
[0035] Preferably, in step 5, a two-stage multi-form differential evolution is performed for the scheduling of the mine's integrated energy system. The evolution process is divided into two stages, with different optimization objectives achieved in each stage.
[0036] First, in the first phase, the populations corresponding to the two alternative tasks of electric-heat and electric-cool were analyzed. and They were subjected to parallel evolution through a diversity-driven neighborhood double mutation strategy, and individual evolution was evaluated.
[0037] Then, knowledge is transferred between the two alternative tasks through a knowledge transfer strategy. Based on the different goals and constraints of the two alternative tasks, the goals and constraints of the transferred individuals are re-evaluated. An improved interval epsilon constraint handling mechanism is adopted to relax the constraints in the early stage of the iteration and select individuals with potential to enter the next generation.
[0038] Next, to fully utilize the optimization results of the first stage, the population obtained at the end of the first stage of evolution is used as the initial population for the second stage. .
[0039] Finally, in the second phase, the initial population... The original task is evaluated, and the original task is optimized using a convergence-driven neighborhood double mutation strategy to obtain the optimal feasible solution set.
[0040] Preferably, in step 6, the time complexity is calculated as follows:
[0041] In the first phase, two alternative tasks evolve in parallel. The evolutionary process of each population includes selection, crossover, and mutation operations. The population size is assumed to be... The number of objective functions is The maximum number of iterations is In each iteration, all individuals need to be evaluated and selected; therefore, the time complexity of a single iteration is approximately O(n log n). .
[0042] Since the two populations evolve independently and in parallel during the first stage, the total time complexity is O(n). = .
[0043] Furthermore, the knowledge transfer operation is performed once per generation, involving the selection and re-evaluation of feasible individuals. Since this process only affects a subset of individuals, its time complexity is [time complexity per generation]. Therefore, the total complexity of knowledge transfer is This additional operation does not significantly increase the overall time complexity.
[0044] In the second stage, the population The evolution is based on the optimization results of the first stage, and the total time complexity of the second stage is O(n). .
[0045] In addition, two key calculations are involved: the dominance relationship based on the condition interval and the interval crowding distance, which are respectively based on... and The time complexity is calculated.
[0046] The environment selection mechanism involves applying the interval constraint dominance principle and an improved interval epsilon constraint handling mechanism. The complexity of this process is... .
[0047] Combining the two phases and considering all operations, the overall time complexity of the algorithm is expressed as follows:
[0048] .
[0049] The present invention also provides: an electronic device, comprising:
[0050] One or more processors;
[0051] A storage device on which one or more programs are stored;
[0052] When the one or more programs are executed by the one or more processors, the one or more processors implement the two-stage multi-form differential evolution method for scheduling integrated energy systems in mines as described above.
[0053] The present invention also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps in any of the above-mentioned two-stage multi-form differential evolution methods for scheduling integrated energy systems in mines.
[0054] Compared with the prior art, the present invention, employing the above technical solution, has the following technical effects:
[0055] 1. The method of this invention comprehensively considers the multiple uncertainties of associated energy and load in mines. Based on domain knowledge, it proposes two alternative tasks, namely electric-heat and electric-cooling, which solves the problem of locating the feasible region caused by strong coupling constraints between intervals and effectively reduces the complexity of the problem.
[0056] 2. The method of the present invention adopts a two-stage optimization framework. The first stage searches multiple feasible domains, and the second stage performs a fine search on the original task, thereby improving the diversity of knowledge and the distribution of Pareto fronts.
[0057] 3. The method of the present invention constructs a knowledge transfer strategy based on two alternative tasks, and transfers only the coupled variable part of the feasible solution in the two tasks each time, which effectively improves the efficiency of the algorithm in locating feasible solutions in complex scheduling problems.
[0058] 4. The method of this invention combines the DE / rand / 1 and DE / current-to-pbest / 1 operators to propose a double mutation mechanism that integrates the neighborhood mechanism, which improves the convergence performance of the algorithm in uncertain problems. In addition, it improves the constraint handling technology by combining the interval dominance relationship and the interval constraint handling rules, and proposes a new interval-based environment selection strategy that can directly deal with interval problems and significantly improve the effectiveness of interval variable optimization. Attached Figure Description
[0059] Figure 1 This is a flowchart of the two-stage multi-form differential evolution method for scheduling integrated energy systems in mines, as described in this invention.
[0060] Figure 2 This is a framework diagram of the integrated energy system for mines used in the embodiments of the present invention;
[0061] Figure 3 The figures show a comparison of Pareto front results obtained by the constrained multi-objective optimization algorithm and the method of this invention, respectively, in embodiments of the present invention.
[0062] Figure 4 The figures show a comparison of Pareto front results obtained using the interval multi-objective optimization algorithm and the method of this invention, respectively, in embodiments of the present invention.
[0063] Figure 5 The diagram shows a comparison of the Pareto front results obtained by the interval-constrained multi-objective optimization algorithm and the method of this invention, respectively, in the embodiments of this invention. Detailed Implementation
[0064] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of the application will be further described in detail below with reference to the accompanying drawings. The described embodiments are only a part of the embodiments involved in this invention. All non-innovative embodiments based on these embodiments by other researchers in the art are within the protection scope of this invention. Furthermore, the step numbers in the embodiments of this invention are only set for ease of explanation and do not limit the order of the steps. The execution order of each step in the embodiments can be adaptively adjusted according to the understanding of those skilled in the art.
[0065] This invention proposes a two-stage multi-form differential evolution method for scheduling integrated energy systems in mines. By introducing core technologies such as alternative task construction, two-stage optimization strategy, feasibility-driven knowledge transfer strategy, neighborhood-based double mutation differential evolution strategy, and improved interval constraint processing technology into the two-stage multi-form differential evolution method, the problem can be solved efficiently.
[0066] The evolutionary process of this invention is divided into two stages, such as... Figure 1 As shown, different optimization objectives are achieved at different stages.
[0067] The first phase aims to optimize two alternative construction tasks: an electro-thermal task and an electro-cooling task. The population corresponding to each task... and Parallel evolution continues until stage transition conditions are met. During evolution, the two populations generate offspring through a diversity-driven neighborhood double mutation strategy (NDM_div) and transfer effective information to each other using a feasibility-driven knowledge transfer mechanism. To preserve individuals with potential, an improved interval environment selection mechanism combining interval dominance relationships and constraint handling mechanisms is employed.
[0068] The goal of the second phase is to analyze the population corresponding to the original task. The process continues until the algorithm's termination condition is met. To fully utilize the optimization results of the first stage, the final population from the first stage is used as the initial population for the second stage. Offspring are generated through a convergence-driven neighborhood double mutation strategy (NDM_con), and an improved interval environmental selection mechanism is employed to ensure that promising individuals enter the next generation.
[0069] The pseudocode for the two-stage multi-form differential evolution framework of this invention is as follows:
[0070]
[0071] The input population size is Maximum number of iterations Stage transition parameters ,in Set it to 0.7.
[0072] First, the populations for the two alternative tasks are initialized and evaluated under their respective tasks. Then, the algorithm enters the main loop, iterating through the number of iterations. Controlled evolution. When If the condition is met, the algorithm is in the first stage; otherwise, it proceeds to the second stage.
[0073] Obtaining the optimal feasible solution set includes the following sub-steps:
[0074] Step 5.1, Descendant Generation: In the first stage, and Descendants are generated using the NDM_div strategy respectively; in the second stage, Offspring are generated using the NDM_con strategy.
[0075] Step 5.2, Individual Assessment: The first stage of assessment... and Calculate the objective and constraint violation values under their respective tasks; in the second phase, The evaluation was conducted under the original task.
[0076] Step 5.3, Knowledge Transfer: Knowledge transfer only occurs in the first stage, and in each generation, it occurs in feasible individuals. and The process can be carried out between the two tasks, but because the goals and constraints of the two tasks are different, the individual after the transfer needs to re-evaluate the goals and constraints.
[0077] Step 5.4, Environment Selection: The first stage adopts an improved interval epsilon constraint processing mechanism, which relaxes the constraints in the early stage of iteration and selects individuals with potential to enter the next generation; the second stage adopts the interval constraint dominance principle.
[0078] Step 5.5, Stage Transition: Once the next generation of individuals is generated, based on the number of iterations... To determine whether to proceed to the second stage, when At that time, evolution entered the second stage.
[0079] Step 5.6, Final Solution Set: The Algorithm Continues to Evolve This continues until the termination condition is met, ultimately yielding the non-dominated feasible solution set for the interval. .
[0080] In one embodiment of the present invention, a source-load uncertain mine integrated energy system is constructed, such as... Figure 2 As shown, the system provides heat, electricity, and cooling energy through a variety of energy devices.
[0081] Specifically, thermal energy is provided by associated energy devices in the mine and gas turbines, electrical energy is provided by the power grid, wind and solar power units, gas turbines and energy storage devices, and cold energy is provided by electric chillers and absorption chillers.
[0082] During mining operations, associated energy sources such as exhaust air, air heat, mine water, and geothermal energy are used to extract heat energy through equipment such as exhaust air thermal storage oxidation devices, air source heat pumps, water source heat pumps, and ground source heat pumps. Furthermore, these associated energy devices in mines are not only suppliers of heat energy but also consumers of electricity.
[0083] System loads are mainly divided into electrical loads, thermal loads, and cooling loads. Electrical loads include rigid electrical loads and indeterminate electrical loads; thermal loads are divided into fixed thermal loads and indeterminate thermal loads.
[0084] Currently, the prediction of uncertain parameters mainly includes probabilistic prediction and interval prediction. Probabilistic prediction requires a large amount of data to obtain the probability density function, while interval prediction only needs to determine the range of values for the uncertain parameters and does not require a precise probability distribution model. Therefore, this embodiment uses interval numbers to describe the uncertainties of renewable energy, associated energy from mining, electrical load, and heat load.
[0085] Furthermore, the effectiveness of the two-stage multi-form differential evolution method of this invention for solving scheduling problems is verified by constructing a source-load uncertain integrated energy system for mines.
[0086] The scheduling problem model solved in this embodiment mainly includes two parts: objective function and constraints.
[0087] The objective function is as follows:
[0088] (1) Economic operating costs
[0089] The economic operating cost of a mine integrated energy system is This includes: the electricity purchase cost of the power grid. Gas purchase cost of gas network Operation and maintenance costs of renewable energy devices Operation and maintenance costs of associated energy devices The charging and discharging cost of energy storage devices and the operation and maintenance costs of cooling units .
[0090] ;
[0091] ;
[0092] ;
[0093] ;
[0094] ;
[0095] ;
[0096] ;
[0097] in, This is the cost coefficient; These represent the electrical, thermal, and cooling output power of each device, respectively. These represent the electrical and thermal output power ranges for each device; subscript express time; and These represent the output and input states of the electrical storage device, respectively; subscripts The power purchasing device corresponding to the power grid; subscript The corresponding gas purchasing device of the gas network; subscript Corresponding renewable energy devices; subscript These correspond to exhaust gas oxidation unit, water source heat pump unit, gas source heat pump unit, and ground source heat pump unit, respectively; subscript These correspond to the output and input of the electrical storage device, respectively; subscript These represent electric chillers and absorption chillers, respectively.
[0098] (2) Costs of abandoning renewable energy and associated energy in mining
[0099] To fully utilize renewable and associated energy sources and reduce the pollution caused by electricity purchases and associated energy emissions from the power grid, the objective function is to minimize the cost of discarding renewable energy sources and associated energy sources from mining.
[0100] ;
[0101] in, The energy waste penalty cost coefficient; Predicted power ranges for electricity and heat from renewable energy sources and associated energy from mining; The range of actual output power for renewable energy and associated energy from mining, including electricity and heat.
[0102] The constraints are as follows:
[0103] 1) Power balance constraint
[0104] Electrical loads include fixed electrical loads and uncertain electrical loads.
[0105] ;
[0106] In the formula, for Constantly fixed electrical load power; for The predicted power range of electrical load is uncertain at any given time.
[0107] 2) Thermal power balance constraint
[0108] ;
[0109] In the formula, for Constantly maintain a fixed heat load power; for The predicted power range for heat load is uncertain at any given time.
[0110] 3) Cold power balance constraint
[0111] ;
[0112] In the formula, for Constant cooling load power.
[0113] 4) Renewable energy device constraints
[0114] The predicted power range for renewable energy can be expressed as:
[0115] ;
[0116] In the formula, superscript These represent the upper and lower limits of the prediction interval, respectively.
[0117] The actual renewable energy consumption range can be expressed as:
[0118] ;
[0119] Both must satisfy the following constraints:
[0120] ;
[0121] 5) Constraints on associated energy devices
[0122] The predicted power range of associated energy sources can be expressed as:
[0123] ;
[0124] The actual power range of associated energy consumption can be expressed as:
[0125] ;
[0126] Both must satisfy the following constraints:
[0127] ;
[0128] ;
[0129] In the formula, It is the predicted power range of associated energy in mines; This refers to the actual power range of associated energy consumption in mines; The electrothermal conversion coefficient of associated energy equipment in mines.
[0130] 6) Gas turbine constraints
[0131] ;
[0132] In the formula, For gas turbine Maximum output at any given moment; The electrothermal conversion coefficient of the gas turbine; and These represent the upper and lower limits of the gas turbine's ramp rate, respectively.
[0133] 7) Constraints of electric energy storage devices
[0134] ;
[0135] In the formula, for The charging and discharging state variables of the energy storage device at all times. =1 indicates that the energy storage device is in the energy storage state. =0 indicates that the energy storage device is in a discharging state; and These are the upper limits for charging and discharging power of the energy storage device, respectively. and These are the upper and lower limits of the amount of electricity that an energy storage device can store.
[0136] 8) Output constraints of refrigeration equipment
[0137] ;
[0138] In the formula, for The maximum output power of the constant cooling equipment; This is the electro-cooling conversion coefficient.
[0139] 9) Power grid output constraints
[0140] ;
[0141] In the formula, for The maximum output power of the power grid at any given time.
[0142] To verify the effectiveness of the proposed two-stage multi-form differential evolution method, it was simultaneously applied, along with three types of comparison algorithms, to the scheduling problem of an integrated energy system in a mine with uncertain source-load conditions. The specific descriptions of the three types of comparison algorithms are as follows:
[0143] Category 1: Six popular constrained multi-objective optimization algorithms, including Co-evolutionary Constrained Multi-objective Algorithm (CCMO), Push-Pull Search Algorithm (PPS), Evolutionary Multi-Task Constrained Multi-objective Algorithm (EMCMO), Evolutionary Multi-Task Algorithm Based on Dynamic Auxiliary Task (MTCMO), Improved MTCMO Algorithm (IMTCMO), and Multiform Optimization Algorithm Based on SPEA2 (MFOSPEA2).
[0144] The second category is a type of interval multi-objective optimization algorithm, namely the interval multi-objective optimization algorithm based on the elite genetic strategy (EG-IMOEA).
[0145] The third category is a type of interval-constrained multi-objective optimization algorithm, namely the interval-constrained multi-objective evolutionary algorithm guided by the degree of interval constraint violation (DIC-MOEA / D).
[0146] It should be noted that the first type of algorithm lacks an interval comparison mechanism, and the second type of algorithm lacks a constraint handling mechanism. Therefore, these two types of algorithms cannot directly solve the problem in this embodiment. To address this, this embodiment combines an interval-based comparison strategy with the first type of algorithm, and simultaneously combines the constraint dominance principle with the second type of algorithm, in order to overcome these shortcomings.
[0147] Regarding parameter settings, the population size for the first two comparison algorithms is set to... Number of iterations However, the third type of algorithm, under these parameter settings, failed to find a feasible solution after iterating to the end. Therefore, for the third type of algorithm, the population size was adjusted to... The number of iterations was adjusted to Furthermore, the parameters of all other algorithms remained consistent. To ensure fairness in the comparison, all algorithms were run independently 20 times, and the results were statistically analyzed.
[0148] In this embodiment, the following three performance metrics are used to evaluate the performance of each algorithm in terms of convergence, diversity, and uncertainty:
[0149] (1) Inverse Generation Distance (IGD):
[0150] IGD is used to measure the proximity between the Pareto front generated by the algorithm and the optimal Pareto front. It is designed to evaluate the convergence and distribution of the algorithm. The smaller the IGD value, the better the algorithm performs.
[0151] (2) Interval hypervolume (HV):
[0152] HV is another commonly used metric for evaluating population diversity and convergence. It is calculated based on the reference point (1.1,1.1) in the normalized target space. The higher the HV value, the better the performance of the algorithm.
[0153] (3) Uncertainty (IM):
[0154] This metric reflects the population's performance under uncertainty, and is measured by the sum of the interval lengths of the objective values of all individuals in the Pareto optimal solution set. The smaller the uncertainty metric, the better the algorithm's performance in handling uncertainty.
[0155] Experiment 1: Comparison and Analysis of Results between the First Type of Comparison Algorithm and the Algorithm of this Invention
[0156] This experiment compares the performance of TMFDE with that of the first type of algorithm to further verify its effectiveness. The Pareto front comparison results for all algorithms are as follows: Figure 3 As shown. Figure 3 This demonstrates that, compared to other comparison algorithms, the method of this invention obtains a more uniform and extensive Pareto front, which can better solve the scheduling problem of integrated energy systems in mines with uncertain source-load.
[0157] Experiment 2: Comparison and Analysis of Results between the Second Type of Comparison Algorithm and the Algorithm of this Invention
[0158] This experiment compares the performance of TMFDE with that of the second type of algorithm. The Pareto front comparison results are as follows: Figure 4 As shown in the figure, the method of the present invention achieves better uniformity and wider coverage.
[0159] Experiment 3: Comparison and Analysis of Results between the Third Type of Comparison Algorithm and the Algorithm of this Invention
[0160] This experiment compares the performance of TMFDE with that of the third type of algorithm. The Pareto front comparison results are as follows: Figure 5 As shown in the figure, the method of the present invention achieves better uniformity and wider coverage.
[0161] It is evident that the two-stage multi-form differential evolution method of this invention can not only effectively address the limitations of existing optimization methods in dealing with strong coupling constraints and multiple uncertainties, but also outperforms existing comparative algorithms in all three performance indicators, demonstrating its superior performance in terms of economy, energy utilization, and scheduling stability.
[0162] In this embodiment of the invention, an electronic device is also provided, comprising: one or more processors; a storage device storing one or more programs thereon; when the one or more programs are executed by the one or more processors, the one or more processors implement the two-stage multi-form differential evolution method for scheduling integrated energy systems in mines described in any of the above embodiments.
[0163] In this embodiment of the invention, a computer-readable storage medium is also provided, on which a computer program is stored. When the program is executed by a processor, it implements the steps of any of the two-stage multi-form differential evolution methods for scheduling integrated energy systems in mines described in the above embodiments.
[0164] The above-described embodiments are merely preferred embodiments of the present invention, and the scope of protection of the present invention is not limited thereto. Any simple changes or equivalent substitutions of the technical solutions that can be obviously obtained by those skilled in the art within the scope of the technology disclosed in the present invention shall fall within the scope of protection of the present invention.
Claims
1. A two-stage, multi-form differential evolution method for scheduling integrated energy systems in mines, characterized in that, Includes the following steps: Step 1: Construct alternative tasks: Under the multi-form optimization framework, for the original model of the integrated energy system of the mine, construct two alternative tasks, namely electric-heat and electric-cooling, to solve the feasible region location problem caused by strong coupling constraints between intervals. Step 2: Construct a knowledge transfer strategy based on two alternative tasks, transferring the coupled variable parts of the feasible solution in each of the two alternative tasks to improve the efficiency of locating feasible solutions; The knowledge transfer strategy described above is used to improve the efficiency of transferring effective information between two alternative tasks, as follows: Step 2.1: Knowledge transfer occurs in the first stage and is carried out on two alternative tasks to help locate the feasible solution space of the original task; different optimization preferences are set for the two alternative tasks, with the goal of finding the intersecting feasible solution space of the two alternative tasks. Step 2.2: Swap feasible solutions in the two alternative tasks to accelerate the efficiency of finding intersecting feasible regions; Step 2.3: Using a feasibility-driven knowledge transfer strategy, knowledge transfer is performed when feasible solutions are generated in two alternative tasks. Both alternative tasks contain decision variables related to electricity. The electrical variables in the intersecting feasible domains are the same. Knowledge transfer is performed on the electrical variables in the feasible solutions and they are spliced with the hot or cold variables under their respective tasks to achieve information interaction. Step 3: Construct a two-stage optimization framework: the first stage is used to search multiple feasible domains, and the second stage is used to perform a fine search on the original task; Step 4: Construct a differential evolution strategy based on neighborhood double mutation: Combine the DE / rand / 1 and DE / current-to-pbest / 1 operators to construct a double mutation mechanism that integrates the neighborhood mechanism, and optimize the corresponding tasks in the two stages respectively. The mutation strategy in the first stage is a diversity-driven neighborhood double mutation strategy, and the mutation strategy in the second stage is a convergence-driven neighborhood double mutation strategy. The proposed differential evolution strategy is based on neighborhood double mutation. It combines neighborhood methods and the DE / rand / 1 operator to increase diversity, and introduces the DE / current-to-pbest / 1 operator to improve feasibility and convergence speed. This is achieved by controlling parameters. Adjusting the performance of the neighborhood double mutation strategy at different stages; The diversity-driven neighborhood double mutation strategy is expressed as follows: ； The convergence-driven neighborhood double mutation strategy is expressed as follows: ； in, Represents a random number between 0 and 1; Indicates from Two individuals are randomly selected from the neighborhood formed, and ; This represents two individuals randomly selected from the current population, and ; This represents the best individual in the current population; Indicates the top of the current population One of the outstanding individuals, Set to 0.1; From A value randomly selected from the data; It is a probability value used to control the performance emphasis of the mutation strategy in the two phases; Step 5: Perform two-stage multi-form differential evolution for the scheduling of the integrated energy system of the mine: Based on the alternative tasks constructed in Step 1 and the two-stage optimization framework constructed in Step 3, in the first stage, the populations corresponding to the two alternative tasks are evolved in parallel through the diversity-driven neighborhood double mutation strategy. Knowledge is transferred between the two tasks through the knowledge transfer strategy. The population obtained at the end of the evolution is used as the initial population for the second stage. The original task is optimized through the convergence-driven neighborhood double mutation strategy to obtain the optimal feasible solution set. Step 6: Analyze the time complexity to demonstrate the rationality of the proposed method for optimizing the scheduling of the integrated energy system in mines.
2. The two-stage, multi-form differential evolution method for scheduling integrated energy systems in mines according to claim 1, characterized in that, In step 1, alternative tasks are constructed. By decoupling the coupling relationships in the electric, cold, and hot equation constraints, the coupling space of electric, cold, and hot variables is divided into two low-dimensional subspaces: the electric-hot subspace and the electric-cold subspace, forming two alternative tasks: the electric-hot task and the electric-cold task, thereby reducing the variable dimension and constraint coupling strength of the original task. The electro-thermal task, ignoring cold-related variables and constraints, is represented as follows: ； The electric-cooling task, ignoring heat-related variables and constraints, is represented as follows: ； Among them, optimization variables They are divided into three categories: variables related to electricity. Variables related to heat and variables related to cold ;function Represent the objective function; function Represents the economic operating cost target; function This indicates the abandonment of renewable energy and associated energy cost targets in mining; Indicates the search space; express 3D real space.
3. The two-stage, multi-form differential evolution method for scheduling integrated energy systems in mines according to claim 1, characterized in that, Step 3 constructs a two-stage optimization framework to improve the diversity of understanding and the distribution of the Pareto front, specifically including the following sub-steps: Step 3.1: In the first stage, simultaneously optimize the two alternative tasks, explore their respective feasible spaces, and obtain the intersecting feasible region of the two alternative tasks; Step 3.2: Set stage transition conditions. When the number of iterations reaches the preset maximum number of iterations, jump from the first stage to the second stage. Step 3.3: In the second stage, optimize the original task by performing a refined search on the intersecting feasible regions found in the first stage to obtain the optimal feasible Pareto front.
4. The two-stage, multi-form differential evolution method for scheduling integrated energy systems in mines according to claim 1, characterized in that, In step 4, after the mutation operation, the individual performance is evaluated using conditional interval dominance and interval crowding distance. An improved interval epsilon environment selection strategy is used to select superior individuals to enter the next generation.
5. The two-stage, multi-form differential evolution method for scheduling integrated energy systems in mines according to claim 1, characterized in that, In step 5, a two-stage multi-form differential evolution is performed for the scheduling of the mine's integrated energy system, with an input population size of [missing information]. The maximum number of iterations is The stage transition parameters are: Through the number of iterations Controlled evolution, when If the evolution is in the first stage, it proceeds to the second stage to obtain the optimal feasible solution set. The specific method is as follows: Step 5.1, Offspring Generation: In the first stage, the electric-heat task and the electric-cool task correspond to the population respectively. and The two alternative tasks generate offspring through a diversity-driven neighborhood double mutation strategy. Step 5.2, Individual Assessment: The first stage of assessment... and Calculate the objective and constraint violation values for each task, and use the optimization results of the first stage as the initial population for the second stage. In the second phase The evaluation was conducted under the original task; Step 5.3, Knowledge Transfer: Knowledge transfer only occurs in the first stage, and it happens in each generation of the population. and Corresponding feasible individuals and The process is conducted between the two alternative tasks, and the goals and constraints of the transferred individuals are re-evaluated based on the differences in the goals and constraints of the two alternative tasks. Step 5.4, Environment Selection: In the first stage, an improved interval epsilon constraint processing mechanism is adopted. In the early stage of iteration, the constraints are relaxed to select individuals with potential to enter the next generation. In the second stage, the interval constraint dominance principle is adopted. Step 5.5, Stage Transition: After generating the next generation of individuals, based on the number of iterations... The state determines whether to enter the second stage. At that time, evolution entered the second stage; Step 5.6, Final Solution Set: In the second stage, the initial population Offspring are generated through a convergence-driven neighborhood double mutation strategy, allowing for continuous evolution. This continues until the termination condition is met, ultimately yielding the non-dominated feasible solution set for the interval. .
6. The two-stage, multi-form differential evolution method for scheduling integrated energy systems in mines according to claim 5, characterized in that, Step 6 analyzes the time complexity, including the following sub-steps: Step 6.1: In each iteration, all individuals are evaluated and selected. The time complexity of a single iteration is O(n log n). Since the two populations evolve independently and in parallel during the first stage, the total time complexity is O(n log n). = ; Step 6.2: The knowledge transfer operation is performed once per generation, involving the selection and re-evaluation of feasible individuals. The total complexity of knowledge transfer is... This will not significantly increase the overall time complexity; Step 6.3, in the second stage, the initial population The evolution is based on the optimization results of the first stage, and the total time complexity of the second stage is O(n). ; Step 6.4: Evolution also involves dominance relationships based on conditional intervals and interval crowding distances, with time complexities of... and ; Step 6.5, the environment selection mechanism involves applying the interval constraint dominance principle and an improved interval epsilon constraint handling mechanism, with a complexity of O(n). ; Step 6.5: Considering all operations in both phases, the overall time complexity of the algorithm is: ； in, The number of objective functions, Population size; This indicates the maximum number of iterations.
7. An electronic device, characterized in that, include: One or more processors; A storage device on which one or more programs are stored; When the one or more programs are executed by the one or more processors, the one or more processors implement the two-stage multi-form differential evolution method for scheduling integrated energy systems in mines as described in any one of claims 1-6.
8. A computer-readable storage medium, characterized in that, It stores a computer program, which, when executed by a processor, implements the steps of the two-stage multi-form differential evolution method for scheduling integrated energy systems in mines, as described in any one of claims 1-6.