A source network load storage multi-objective coordinated regulation method and system

Abstract

The application discloses a kind of source network load storage multi-objective collaborative regulation method and system, belong to district energy system optimization regulation technical field, the method is: constructing electricity carbon objective function and setting constraint condition;Based on optimization algorithm and constraint condition, electricity carbon objective function is solved, initial solution is generated and iteration updating process is executed, if iteration number does not satisfy iteration requirement, then calculate nonlinear attenuation factor and obtain first random number and second random number, to match search mode;Then initial solution is updated to search until iteration number satisfies preset iteration requirement, obtain multi-objective optimization solution set, based on multi-objective optimization solution set generation power dispatching strategy, to regulate source network load storage.Therefore, by implementing the present application, low-carbon targets can be integrated into optimization algorithms to improve the interpretability and decision support capabilities of the algorithms. In addition, the application provides a variety of low-carbon alternative solutions through customized search mechanisms, thereby improving the reliability of power regulation strategies.

Description

Technical Field

[0001] This invention relates to the field of regional energy system optimization and regulation technology, and in particular to a multi-objective coordinated regulation method and system for energy sources, grids, loads and storage. Background Technology

[0002] Regional energy systems, such as integrated energy parks, urban microgrids, and smart distribution networks, serve as core carriers of energy consumption and carbon emission sources. Their optimized regulation has become a crucial link in achieving the low-carbon transformation of the power system, and the power dispatch field is facing a shift from traditional economic optimization to a low-carbon-economic synergistic optimization. However, in the actual operation of regional energy systems, low-carbon goals and operating cost goals often conflict. How to effectively control carbon emissions while ensuring the economic efficiency of system operation through reliable resource regulation strategies has become a key challenge that needs further research.

[0003] Currently, some methods for coordinated optimization and control of electricity and carbon emissions have been proposed in existing technologies. These methods calculate grid emission factors, define economic and environmental objective functions, and use particle swarm optimization (PSO) algorithms for multi-objective optimization. However, the emission factors in these existing technologies are usually based on static historical data, which cannot reflect the time-varying characteristics of carbon emission intensity in real time. Furthermore, their optimization models often use linear weighting to handle economic and environmental objectives, making it difficult to effectively adapt to real-time price fluctuations and dynamic changes in carbon intensity in the carbon market. More importantly, existing algorithms are mostly designed for single-objective continuous unconstrained optimization problems, lacking customized improvements for the multi-objective trade-offs, strong real-time constraints, and dynamic uncertainties of regional energy systems. When dealing with the dual objectives of cost and carbon emissions optimization, they fail to establish an effective low-carbon-oriented search mechanism, resulting in uneven distribution of the generated Pareto fronts and insufficient solution density in low-carbon regions. This fails to provide high-quality decision-making solutions that balance low-carbon priority and economic feasibility. Therefore, existing technologies suffer from problems such as insufficient integration of low-carbon objectives into algorithms, poor adaptability to the coupling characteristics of electricity and carbon emissions, and poor reliability of power regulation strategies. Summary of the Invention

[0004] This invention proposes a multi-objective coordinated regulation method and system for power generation, grid, load, and storage, which can solve the problem of poor reliability of power regulation strategies caused by the insufficient integration of low-carbon objectives into existing algorithms and poor adaptability to the coupling characteristics of electricity and carbon. This invention integrates low-carbon objectives into the optimization algorithm, improving its interpretability and decision support capabilities. Furthermore, it provides rich low-carbon alternatives through a customized search mechanism, thereby enhancing the reliability of power regulation strategies.

[0005] To achieve the above objectives, embodiments of the present invention provide a multi-objective coordinated regulation method for power generation, grid, load, and storage, comprising: constructing an electric carbon objective function based on pre-acquired power equipment operation data and dynamic electric carbon factor data, and setting constraints on the electric carbon objective function; solving the electric carbon objective function based on a preset optimization algorithm and the constraints to generate an initial solution; performing an iterative update process on the initial solution, and if the number of iterations does not meet a preset iteration requirement, calculating a nonlinear decay factor and obtaining a first random number and a second random number, and matching several search patterns based on the first random number and the second random number; searching and updating the initial solution based on several search patterns and the nonlinear decay factor until the number of iterations meets the preset iteration requirement, obtaining a multi-objective optimized solution set, and ending the iterative update process; and generating a power dispatch strategy based on the multi-objective optimized solution set to regulate power generation, grid, load, and storage.

[0006] This invention proposes a multi-objective coordinated regulation method for power generation, grid, load, and storage. It deeply integrates dynamic carbon dioxide factors into the optimization objectives, constructing a carbon dioxide objective function. Simultaneously, it iteratively solves the objective function using a pre-defined optimization algorithm, establishing an adaptive search mechanism based on random number matching of multiple search modes during the iteration process. This effectively enhances the algorithm's adaptability to the coupling characteristics of carbon dioxide and electricity. Based on this, it generates a multi-objective optimization solution set and outputs a power dispatch strategy, providing decision-makers with diverse low-carbon dispatch schemes and improving the reliability and interpretability of power regulation strategies. Thus, it integrates low-carbon objectives into the optimization algorithm, improving its interpretability and decision support capabilities, and provides rich low-carbon alternatives through a customized search mechanism, thereby enhancing the reliability of power regulation strategies.

[0007] Furthermore, the step of constructing an electric carbon objective function based on pre-acquired power equipment operation data and dynamic electric carbon factor data, and setting constraints on the electric carbon objective function, includes: acquiring power equipment operation data, wherein the power equipment operation data includes gas turbine output and main grid exchange power; acquiring dynamic electric carbon factor data, wherein the dynamic electric carbon factor data includes carbon price, electricity sales price, and electricity purchase price; constructing a fuel cost function based on the gas turbine output and a preset fuel cost coefficient; constructing a grid interaction cost function based on the main grid exchange power, electricity sales price, and electricity purchase price; constructing a carbon emission cost function based on the gas turbine output, main grid exchange power, carbon price, preset carbon emission coefficient, and preset implicit carbon emission coefficient; summing the fuel cost function, the grid interaction cost function, and the carbon emission cost function to obtain the electric carbon objective function; and setting power balance constraints, battery energy state constraints, and gas turbine ramp rate constraints on the electric carbon objective function.

[0008] In the above scheme, the construction methods of fuel cost function, grid interaction cost function and carbon emission cost function are defined, and dynamic electric carbon factor data are transformed into quantifiable cost items, providing a data foundation for the accurate expression of electric carbon objective function. Then, the constructed fuel cost function, grid interaction cost function and carbon emission cost function are summed to obtain electric carbon objective function. At the same time, power balance constraints, battery energy state constraints and gas turbine ramp rate constraints are set for electric carbon objective function to constrain electric carbon objective function. This not only comprehensively reflects the synergistic relationship between the economic cost and carbon emission cost of system operation, but also enables the optimization process to respond in real time to carbon price fluctuations and changes in carbon emission intensity, avoiding the problem of low-carbon target and cost target being disconnected in traditional methods, and helping to improve the reliability of power regulation strategy.

[0009] Furthermore, the step of solving the electrocarbon objective function based on the preset optimization algorithm and the constraints to generate an initial solution includes: determining the decision space dimension, population size, and decision variable boundary based on the constraints; initializing the slime mold population using a random generation strategy through the preset optimization algorithm based on the decision space dimension, the population size, and the decision variable boundary, wherein the position vector of each slime mold individual corresponds to a set of scheduling schemes, and calculating the initial fitness value of each slime mold individual, and taking individuals whose initial fitness values ​​meet the preset fitness requirements as the initial solution.

[0010] In the above scheme, the decision space dimension, population size, and decision variable boundaries are first determined, and a random generation strategy is used to initialize the slime mold population to ensure that the initial population is uniformly distributed in the search space, providing a high-quality initial solution for subsequent iterative searches. In addition, by initializing the slime mold population through a random generation strategy, the position vector of each slime mold individual corresponds to a set of scheduling schemes, which can cover more potential optimal solution regions, effectively avoiding the problems of low initial population quality and insufficient search space coverage, and helping to improve the reliability of power regulation strategies.

[0011] Furthermore, the search modes include a random diffusion mode, a directional fluctuation mode, and a local contraction mode. An iterative update process is performed on the initial solution. If the number of iterations does not meet a preset iteration requirement, a nonlinear decay factor is calculated, and a first random number and a second random number are obtained. Based on the first and second random numbers, several search modes are matched, including: if the number of iterations does not meet the preset iteration requirement, a nonlinear decay factor is calculated based on the current iteration number and the target iteration number, and a first random number and a second random number are obtained; when the first random number is less than a preset first exploration threshold, a random diffusion mode is matched; when the first random number is greater than or equal to the preset first exploration threshold and the second random number is less than a preset second exploration threshold, a directional fluctuation mode is matched; when the first random number is greater than or equal to the preset first exploration threshold and the second random number is greater than or equal to the preset second exploration threshold, a local contraction mode is matched.

[0012] In the above scheme, a nonlinear decay factor and a random number matching mechanism are introduced to dynamically determine the switching between three search modes: random diffusion, directional fluctuation, and local contraction. This achieves a dynamic balance between global exploration and local development. The nonlinear decay factor is calculated based on the current iteration number and the maximum iteration number. When it is large in the early stage of iteration, it corresponds to a high-activity state and tends to explore a large area. As the iteration progresses, it gradually decreases and approaches zero, corresponding to a decrease in activity and a shift to fine-grained search. Thus, the comparison between the first random number and the preset exploration threshold determines whether to perform completely random exploration, and the comparison between the second random number and the preset division threshold divides directional fluctuation and local contraction. This energy-driven dynamic mechanism provides a quantitative basis for the switching of movement modes, thereby helping to improve the reliability of power regulation strategies.

[0013] Furthermore, based on several search modes and the nonlinear decay factor, the initial solution is searched and updated until the number of iterations meets a preset iteration requirement to obtain a multi-objective optimization solution set, including: when the search mode is a random diffusion mode, a new position is randomly generated in the decision space according to the decision variable boundary to obtain a first search solution; when the search mode is a directional fluctuation mode, the position is updated according to the initial solution, random individual difference, the nonlinear decay factor, and weight coefficients to obtain a second search solution, wherein the weight coefficients are calculated based on fitness values, and the random individual difference is obtained by subtracting the position vectors of two randomly selected individuals; when the search mode is a local contraction mode, a random vector is determined according to the nonlinear decay factor, and the position is updated within a preset range corresponding to the current individual position vector using the random vector to obtain a third search solution; based on the first search solution, the second search solution, or the third search solution, the initial solution is searched and updated until the number of iterations meets a preset iteration requirement to obtain a multi-objective optimization solution set.

[0014] In the above scheme, position update rules are defined under three search modes. The directional fluctuation mode introduces weight coefficients to guide the search direction, while the local contraction mode uses oscillation vectors for refined searching, effectively improving the convergence accuracy and global search capability of the algorithm. The random diffusion mode randomly generates new positions in the decision space based on the upper and lower boundaries of the decision variables, ensuring comprehensive coverage of the search space. The directional fluctuation mode updates positions based on the current global optimum, random individual differences, and weight coefficients calculated according to fitness ranking, maintaining both the convergence to the optimum and the population diversity through random perturbation. The local contraction mode determines random vectors based on nonlinear decay factors and performs small-range position updates near the current individual, which helps improve the accuracy of local optima. Thus, a rich set of low-carbon alternatives is provided through a customized search mechanism, improving the reliability of power regulation strategies.

[0015] Furthermore, based on the first search solution, the second search solution, or the third search solution, the initial solution is searched and updated until the number of iterations meets a preset iteration requirement to obtain a multi-objective optimized solution set. This includes: based on the decision variable boundary, performing boundary checks on the first search solution, the second search solution, or the third search solution generated in each iteration; correcting the first search solution, the second search solution, or the third search solution that does not meet the boundary check requirement to obtain the corresponding optimized first search solution, optimized second search solution, or optimized third search solution; calculating the search fitness value corresponding to the optimized first search solution, optimized second search solution, or optimized third search solution, and comparing the initial fitness value corresponding to the initial solution with the search fitness value to obtain the target search solution and the corresponding target fitness value; updating the target search solution and the corresponding target fitness value until the number of iterations meets a preset iteration requirement to obtain a multi-objective optimized solution set.

[0016] In the above scheme, boundary checks and fitness comparison mechanisms are adopted to ensure that the search solution is always within the feasible region. Greedy selection is used to retain better solutions and drive the population to evolve towards the global optimum. Boundary checks are performed on the new positions generated in each iteration to correct components that exceed the upper and lower limits back to the feasible region, avoiding solutions that violate physical constraints due to search movement and ensuring the engineering feasibility of the scheduling scheme. The fitness value of the new position is calculated and compared with the fitness value of the original position to retain better solutions. At the same time, the global optimum and the optimal fitness value are updated in real time to ensure that the optimal fitness of the population continues to decrease after each iteration, driving the population to gradually converge to the global optimum scheduling scheme and improving the reliability of the power regulation strategy.

[0017] Furthermore, a power dispatch strategy is generated based on the multi-objective optimization solution set to regulate the power generation, grid, load, and storage system. This includes: extracting individuals whose total operating cost meets preset dispatch requirements from the multi-objective optimization solution set as target cost solutions, and generating corresponding economic dispatch strategies based on the target cost solutions; extracting individuals whose total carbon emissions meet preset dispatch requirements from the multi-objective optimization solution set as target carbon emission solutions, and generating corresponding carbon emission dispatch strategies; selecting knee-point compromise solutions based on the economic dispatch strategies and the carbon emission dispatch strategies to obtain the power dispatch strategy; decoding the power dispatch strategy into a sequence of decision variables, and generating output curves for each device, a state-of-load change diagram for energy storage, a power grid interaction diagram, and a cost composition diagram based on the sequence of decision variables and the power-carbon objective function; and sending the output curves for each device, the state-of-load change diagram for energy storage, the power grid interaction diagram, and the cost composition diagram to regional energy management equipment to regulate the power generation, grid, load, and storage system.

[0018] In the above scheme, target cost solutions, target carbon emission solutions, and knee-point compromise solutions that meet preset scheduling requirements are extracted from the multi-objective optimization solution set. This generates diverse scheduling strategies, and the visualization output significantly enhances the interpretability and decision support capabilities of the optimization results. The target cost solution that meets the preset scheduling requirements corresponds to the most economically optimal scheduling strategy, while the target carbon emission solution represents the extreme low-carbon priority scheme. The knee-point compromise solution, by normalizing the frontier target value and calculating the distance to the ideal point for each solution, combined with curvature analysis, selects the individual with the best balance between cost and carbon emissions as the compromise scheme. Finally, the extracted scheduling schemes are decoded into a sequence of decision variables, generating output curves for each device, a state-of-load diagram for energy storage, a power grid interaction diagram, and a cost composition diagram. These are then transmitted to the regional energy system's energy management system for source-grid-load-storage regulation. Thus, the low-carbon objective is integrated into the optimization algorithm, improving its interpretability and decision support capabilities. Furthermore, a customized search mechanism provides rich low-carbon alternatives, enhancing the reliability of power regulation strategies.

[0019] This invention also provides a multi-objective coordinated control system for power generation, grid, load, and storage, including a data processing module, an objective solving module, a pattern matching module, an iterative update module, and a control module. The data processing module constructs an electric carbon objective function based on pre-acquired power equipment operation data and dynamic electric carbon factor data, and sets constraints on the electric carbon objective function. The objective solving module solves the electric carbon objective function based on a preset optimization algorithm and the constraints to generate an initial solution. The pattern matching module performs an iterative update process on the initial solution; if the number of iterations does not meet a preset iteration requirement, it calculates a nonlinear decay factor and obtains a first random number and a second random number, matching several search patterns based on the first and second random numbers. The iterative update module searches and updates the initial solution based on several search patterns and the nonlinear decay factor until the number of iterations meets the preset iteration requirement, obtaining a multi-objective optimized solution set and ending the iterative update process. The control module generates a power dispatch strategy based on the multi-objective optimized solution set to control power generation, grid, load, and storage.

[0020] This invention proposes a multi-objective coordinated control system for power generation, grid, load, and storage. It deeply integrates dynamic carbon dioxide factors into the optimization objectives, constructing a carbon dioxide objective function. Simultaneously, it iteratively solves the objective function using a pre-defined optimization algorithm, establishing an adaptive search mechanism based on random number matching of multiple search modes during the iteration process. This effectively enhances the algorithm's adaptability to the coupling characteristics of carbon dioxide and electricity. Based on this, it generates a multi-objective optimization solution set and outputs power dispatch strategies, providing decision-makers with diverse low-carbon dispatch schemes and improving the reliability and interpretability of power control strategies. Thus, it integrates low-carbon objectives into the optimization algorithm, improving its interpretability and decision support capabilities, and provides rich low-carbon alternatives through a customized search mechanism, thereby enhancing the reliability of power control strategies.

[0021] Furthermore, the data processing module includes a power data acquisition unit, an electric carbon data acquisition unit, a first cost function construction unit, a second cost function construction unit, a third cost function construction unit, an objective function construction unit, and a constraint setting unit; wherein: the power data acquisition unit is used to acquire power equipment operation data, including gas turbine output and main grid switching power; the electric carbon data acquisition unit is used to acquire dynamic electric carbon factor data, including carbon price, electricity sales price, and electricity purchase price; the first cost function construction unit is used to construct the electric carbon function based on the gas turbine output and a preset fuel cost coefficient. The system employs a first cost function construction unit to construct a fuel cost function; a second cost function construction unit to construct a grid interaction cost function based on the main grid exchange power, electricity sales price, and electricity purchase price; a third cost function construction unit to construct a carbon emission cost function based on the gas turbine output, main grid exchange power, carbon price, preset carbon emission coefficient, and preset implicit carbon emission coefficient; a target function construction unit to sum the fuel cost function, the grid interaction cost function, and the carbon emission cost function to obtain an electric carbon target function; and a constraint setting unit to set power balance constraints, battery energy state constraints, and gas turbine ramp rate constraints on the electric carbon target function.

[0022] Furthermore, the objective solving module includes a parameter determination unit and an optimization strategy solving unit; wherein: the parameter determination unit is used to determine the decision space dimension, population size, and decision variable boundary based on the constraints; the optimization strategy solving unit is used to initialize the slime mold population using a random generation strategy through the preset optimization algorithm based on the decision space dimension, the population size, and the decision variable boundary, where the position vector of each slime mold individual corresponds to a set of scheduling schemes, and calculates the initial fitness value of each slime mold individual, and uses individuals whose initial fitness values ​​meet the preset fitness requirements as the initial solution. Attached Figure Description

[0023] To more clearly illustrate the technical solution of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0024] Figure 1 A flowchart illustrating the steps of a multi-objective coordinated regulation method for source-grid-load-storage provided in a certain embodiment of the present invention; Figure 2 A schematic diagram of the equipment output curve of a multi-objective coordinated regulation method for source-grid-load-storage provided in a certain embodiment of the present invention; Figure 3A schematic diagram of the state of charge change of energy storage in a multi-objective coordinated regulation method for source-grid-load-storage provided in a certain embodiment of the present invention; Figure 4 A schematic diagram of grid interaction power for a multi-objective coordinated regulation method of source, grid, load and storage provided in a certain embodiment of the present invention; Figure 5 A schematic diagram illustrating the cost structure of a multi-objective coordinated regulation method for source-grid-load-storage provided in a certain embodiment of the present invention; Figure 6 This is a schematic diagram of the module structure of a multi-objective coordinated control system for source, grid, load and storage provided in a certain embodiment of the present invention. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0026] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application; the terms “comprising” and “having”, and any variations thereof, in the specification, claims, and foregoing description of the drawings are intended to cover non-exclusive inclusion.

[0027] In the description of the embodiments of this application, technical terms such as "first" and "second" are used only to distinguish different objects and should not be construed as indicating or implying relative importance or implicitly specifying the number, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, "multiple" means two or more, unless otherwise explicitly defined.

[0028] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0029] In the description of the embodiments in this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this document generally indicates that the preceding and following related objects have an "or" relationship.

[0030] In the description of the embodiments of this application, the term "multiple" refers to two or more (including two), similarly, "multiple sets" refers to two or more (including two sets), and "multiple pieces" refers to two or more (including two pieces).

[0031] In the description of the embodiments of this application, unless otherwise expressly specified and limited, technical terms such as "installation," "connection," "joining," and "fixing" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. For those skilled in the art, the specific meaning of the above terms in the embodiments of this application can be understood according to the specific circumstances.

[0032] To address the issue of poor reliability in power regulation strategies caused by the insufficient integration of low-carbon targets into existing algorithms and poor adaptability to the coupling characteristics of electricity and carbon, see [reference needed]. Figure 1 , Figure 1 This is a schematic flowchart illustrating the steps of a multi-objective coordinated regulation method for source-grid-load-storage provided in a certain embodiment of the present invention. Figure 1 As shown in the figure, this embodiment of the invention proposes a multi-objective coordinated regulation method for source-grid-load-storage, including steps 101 to 105, each step of which is as follows: Step 101: Based on the pre-acquired power equipment operation data and dynamic electric carbon factor data, construct an electric carbon objective function and set constraints on the electric carbon objective function; Step 102: Solve the objective function of the electrocarbon based on the preset optimization algorithm and the constraints to generate an initial solution; Step 103: Perform an iterative update process on the initial solution. If the number of iterations does not meet the preset iteration requirements, calculate the nonlinear decay factor and obtain the first random number and the second random number. Match several search patterns based on the first random number and the second random number. Step 104: Based on several search modes and the nonlinear decay factor, search and update the initial solution until the number of iterations meets the preset iteration requirements, obtain a multi-objective optimization solution set, and end the iterative update process; Step 105: Generate a power dispatch strategy based on the multi-objective optimization solution set to regulate the power generation, grid, load, and storage systems.

[0033] One possible implementation involves first acquiring operational data of the target region's energy system within a scheduling cycle, typically set to 24 hours, with hours as the time granularity. This operational data includes load forecast sequences, photovoltaic output forecast sequences, and wind power output forecast sequences obtained from historical databases or prediction models. In this embodiment, gas turbine output and grid exchange power are also interpreted as operational data, along with time-of-use (TOU) and time-of-use (TOU) electricity purchase price sequences and TOU electricity sales price sequences obtained from the electricity market. Simultaneously, carbon price data is acquired from the carbon trading market or carbon tax policy; this carbon price data can be a fixed value or a dynamic sequence that changes over time. Furthermore, the fuel cost coefficient and carbon emission coefficient of the gas turbine are obtained based on equipment nameplate parameters or operating manuals. The implicit carbon emission coefficient for purchasing electricity from the external grid is determined based on the average emission factor or marginal emission factor published by the grid. All data are then aligned according to the time series to form a complete input dataset.

[0034] Next, an electric carbon objective function is constructed based on the acquired data. This objective function aims to minimize the total operating cost of the system. For each time period within the scheduling cycle, the fuel cost term is first calculated based on the product of the gas turbine output and the fuel cost coefficient. Then, the grid interaction cost term is processed according to the power purchase and sale status. Specifically, when the exchange power with the external main grid is positive, it indicates power purchase. The power purchase cost is obtained by multiplying the exchange power by the power purchase price of the corresponding time period. When the exchange power is negative, it indicates power sale. The power sale revenue is obtained by multiplying the absolute value of the exchange power by the power sale price of the corresponding time period. The algebraic sum of the power purchase cost and the power sale revenue constitutes the grid interaction cost term. Then, the direct carbon emissions are obtained by multiplying the gas turbine output by the gas turbine carbon emission coefficient, and the indirect carbon emissions are obtained by multiplying the power purchased from the external main grid by the implicit carbon emission coefficient. The direct and indirect carbon emissions are summed and multiplied by the carbon price to obtain the internalized carbon cost term. Finally, the fuel cost term, grid interaction cost term, and internalized carbon cost term for all time periods within the scheduling cycle are summed to obtain the electric carbon objective function.

[0035] Subsequently, to ensure the engineering feasibility of the dispatching scheme, multiple operational constraints were set. These included: a power balance constraint requiring that the sum of gas turbine output, renewable energy output, energy storage discharge power, and power purchased from the external grid in each time period equal the sum of load demand, energy storage charging power, and power sold to the external grid in that period. The energy storage charging and discharging process incorporates charging and discharging efficiency to reflect energy conversion losses. Energy storage state of charge (SOC) constraints were also set, including a recursive relationship for SOC calculation, which calculates the SOC at the end of the current time period based on the SOC at the end of the previous time period and the charging and discharging power and efficiency of the current time period. The SOC was also set to remain within a preset safety range throughout the dispatching cycle, typically set to 20% to 90% of maximum capacity. A terminal constraint was set to ensure the SOC is close to the initial SOC to guarantee the sustainability of the dispatching. Finally, a gas turbine ramp rate constraint was set, requiring that the absolute value of the change in gas turbine output between adjacent time periods not exceed a preset maximum ramp rate to avoid sudden changes in unit output causing mechanical stress and system frequency fluctuations.

[0036] Then, the slime mold optimization algorithm improved by elite back learning and quadratic interpolation is used as the preset optimization algorithm for explanation. First, the decision space dimension is determined based on the physical upper and lower limits of the decision variables. The decision variables include the gas turbine output, energy storage charging and discharging power, and exchange power with the external main grid for each time period. There are a total of 72 continuous variables for a 24-hour scheduling cycle. Then, the slime mold population size is set, usually set to 50 to 200 individuals depending on the complexity of the problem. Next, the slime mold population is initialized using a random generation strategy. The position vector of each slime mold individual corresponds to a complete scheduling scheme, that is, a solution vector containing the specific values ​​of the decision variables for all time periods. Finally, the fitness value of each slime mold individual is calculated. This fitness value is the calculated value of the electrocarbon objective function. The individual with the smallest fitness value is recorded as the initial global optimal solution, and the population composed of all current individuals is used as the initial solution.

[0037] Subsequently, a maximum number of iterations is set, and in each iteration, it is first determined whether the current number of iterations meets the preset iteration requirements. If the maximum number of iterations has not been reached, the update process continues. Then, a nonlinear decay factor is calculated based on the current number of iterations and the maximum number of iterations. The nonlinear decay factor has a larger value in the early stage of the iteration to enhance the global exploration capability, and gradually approaches zero in the later stage of the iteration to focus on local development. At the same time, a first random number and a second random number are obtained. Both random numbers are independent random numbers that are uniformly distributed in the interval between 0 and 1. Then, based on the first random number and the second random number, one of three search modes is matched. The search modes include: random diffusion mode, directional fluctuation mode, and local contraction mode.

[0038] Next, position updates are performed according to different search patterns, and the iterative update process is repeated until the number of iterations reaches the preset maximum number of iterations, at which point the iterative update process ends and a multi-objective optimization solution set is obtained.

[0039] Finally, three representative scheduling schemes are extracted from the multi-objective optimization solution set, and the selected scheduling schemes are decoded into a sequence of decision variables, that is, the specific values ​​of gas turbine output, energy storage charging and discharging power, and power exchange with the external main grid are obtained for each time period. These values ​​are then substituted back into the electric carbon objective function and carbon emission accounting model to verify their feasibility, and a visualization output is generated, including the output curves of each device, the energy storage state of charge change diagram, the grid interaction power diagram, and the cost composition diagram. The generated scheduling strategy and visualization results are transmitted to the regional energy management equipment to execute the day-ahead scheduling plan, adjust the intraday real-time operation strategy, or collaboratively optimize the interaction power with the external main grid, ultimately realizing the regulation of source-grid-load-storage equipment.

[0040] This invention proposes a multi-objective coordinated regulation method for power generation, grid, load, and storage. It deeply integrates dynamic carbon dioxide factors into the optimization objectives, constructing a carbon dioxide objective function. Simultaneously, it iteratively solves the objective function using a pre-defined optimization algorithm, establishing an adaptive search mechanism based on random number matching of multiple search modes during the iteration process. This effectively enhances the algorithm's adaptability to the coupling characteristics of carbon dioxide and electricity. Based on this, it generates a multi-objective optimization solution set and outputs a power dispatch strategy, providing decision-makers with diverse low-carbon dispatch schemes and improving the reliability and interpretability of power regulation strategies. Thus, it integrates low-carbon objectives into the optimization algorithm, improving its interpretability and decision support capabilities, and provides rich low-carbon alternatives through a customized search mechanism, thereby enhancing the reliability of power regulation strategies.

[0041] In a preferred embodiment, the step of constructing an electric carbon objective function based on pre-acquired power equipment operation data and dynamic electric carbon factor data, and setting constraints on the electric carbon objective function, includes: acquiring power equipment operation data, wherein the power equipment operation data includes gas turbine output and main grid exchange power; acquiring dynamic electric carbon factor data, wherein the dynamic electric carbon factor data includes carbon price, electricity sales price, and electricity purchase price; constructing a fuel cost function based on the gas turbine output and a preset fuel cost coefficient; constructing a grid interaction cost function based on the main grid exchange power, electricity sales price, and electricity purchase price; constructing a carbon emission cost function based on the gas turbine output, main grid exchange power, carbon price, preset carbon emission coefficient, and preset implicit carbon emission coefficient; summing the fuel cost function, the grid interaction cost function, and the carbon emission cost function to obtain the electric carbon objective function; and setting power balance constraints, battery energy state constraints, and gas turbine ramp rate constraints on the electric carbon objective function.

[0042] For example, firstly, a typical daily load curve is obtained. In this embodiment, the decision variables are set as the hourly output of the gas turbine and the power exchanged with the main grid, and time-of-use electricity price data, carbon market trading prices, or carbon price data determined by carbon tax policies are obtained. The fuel cost coefficient of the gas turbine, the carbon emission coefficient of the gas turbine, and the implicit carbon emission coefficient of purchasing electricity from the external main grid are also obtained. Then, a fuel cost function is constructed based on the gas turbine output and fuel cost coefficient. , is represented as: ; In the formula, This is the fuel cost coefficient for gas turbines, expressed in $ / kWh, reflecting the fuel cost per unit of electricity generated. The power output of the gas turbine is measured in kW and is a continuous variable. Construct a grid interaction cost function based on main grid switching power, electricity sales price, and electricity purchase price. , is represented as: ; In the formula, The power exchanged with the main grid is measured in kW and is a continuous variable. A positive value indicates the purchase of electricity, and a negative value indicates the sale of electricity. The electricity purchase price for time period t is expressed in $ / kWh and is usually a time-of-use price, published by the electricity market or grid company. This is the electricity sales price for time period t, expressed in $ / kWh. It is generally lower than the purchase price and reflects the revenue of the microgrid from selling electricity to the main grid. A carbon emission cost function is constructed based on gas turbine output, main grid switching power, carbon price, carbon emission factor, and implicit carbon emission factor. , is represented as: ; In the formula, The power output of the gas turbine is measured in kW and is a continuous variable. The power exchanged with the main grid is measured in kW and is a continuous variable. A positive value indicates the purchase of electricity, and a negative value indicates the sale of electricity. The carbon emission factor of a gas turbine is expressed in kgCO2 / kWh, representing the direct carbon emissions generated per unit of electricity generated. The implicit carbon emission factor for purchasing electricity from the main grid, expressed in kgCO2 / kWh, reflects the indirect carbon emission intensity corresponding to purchasing electricity from the main grid. It is usually determined based on the grid average emission factor or marginal emission factor. The carbon price, expressed in $ / kgCO2, is the unit price that converts carbon emissions into economic costs. It can be determined by carbon market trading prices or carbon tax policies.

[0043] The objective function for carbon dioxide emission is obtained by summing the three cost functions mentioned above. Objective function of carbon electrolysis To minimize the total system operating cost F, it is expressed as: ; Simultaneously, power balance constraints, battery energy state constraints, and gas turbine ramp rate constraints are set for the objective function of the electric carbon project. Among them, the power balance constraint requires that the sum of power generation equipment output, renewable energy output, energy storage discharge power, and purchased electricity power in each time period equals the sum of load demand, energy storage charging power, and sold electricity power, expressed as: ; In the formula, Contribute to photovoltaic forecasting; For load forecasting; battery discharge power is The battery charging power is ,in The charging and discharging power of the battery, measured in kW, is a continuous variable; a positive value indicates charging, and a negative value indicates discharging. (Purchasing power) Electricity sales volume ;in, and These refer to the battery charging and discharging efficiency, respectively.

[0044] Battery energy state constraints include the recursive relationship of state of charge, safety range constraints, and terminal regression constraints, expressed as: ; In the formula, the initial SOC is given as ; These are the model input parameters, set according to the actual measurement or test scenario before optimization begins. The algorithm is only responsible for planning the subsequent charging and discharging strategy under a given initial state. In the simulation, a medium state of charge (SOC) of 100 is set, representing the system in a normal, bidirectionally adjustable operating point; the SOC must be kept within a safe range, i.e. Furthermore, to ensure repeatability of scheduling, it is typically required that the State of Charge (SOC) be close to its initial value at the end of the scheduling cycle. ; The gas turbine ramp rate constraint requires that the output variation between adjacent time periods not exceed a preset maximum value, expressed as: ; In the formula, This represents the output of the gas turbine during time period t, expressed in kW. For the gas turbine in the previous period contribution; This represents the maximum allowable grade rate of the gas turbine, expressed in kW / h, which is the upper limit of the output variation between adjacent time periods; for any adjacent time periods The absolute value of the change in gas turbine output does not exceed This constraint ensures stable unit operation and avoids excessive mechanical stress, shortened lifespan, or system frequency fluctuations caused by rapid changes in output. It is an important condition for ensuring the safe and stable operation of equipment in power system dispatching.

[0045] In the above scheme, the construction methods of fuel cost function, grid interaction cost function and carbon emission cost function are defined, and dynamic electric carbon factor data are transformed into quantifiable cost items, providing a data foundation for the accurate expression of electric carbon objective function. Then, the constructed fuel cost function, grid interaction cost function and carbon emission cost function are summed to obtain electric carbon objective function. At the same time, power balance constraints, battery energy state constraints and gas turbine ramp rate constraints are set for electric carbon objective function to constrain electric carbon objective function. This not only comprehensively reflects the synergistic relationship between the economic cost and carbon emission cost of system operation, but also enables the optimization process to respond in real time to carbon price fluctuations and changes in carbon emission intensity, avoiding the problem of low-carbon target and cost target being disconnected in traditional methods, and helping to improve the reliability of power regulation strategy.

[0046] In a preferred embodiment, the step of solving the electrocarbon objective function based on a preset optimization algorithm and the constraints to generate an initial solution includes: determining the decision space dimension, population size, and decision variable boundary based on the constraints; initializing the slime mold population using a random generation strategy through the preset optimization algorithm based on the decision space dimension, the population size, and the decision variable boundary, wherein the position vector of each slime mold individual corresponds to a set of scheduling schemes, and calculating the initial fitness value of each slime mold individual, and using individuals whose initial fitness values ​​meet the preset fitness requirements as the initial solution.

[0047] For example, firstly, the decision space dimension, population size, and upper and lower boundaries of decision variables are determined based on constraints. Decision variables include the gas turbine output, battery charging and discharging power, and power exchanged with the external main grid for each time period, totaling 72 continuous variables. Then, based on the decision space dimension, population size, and upper and lower boundaries of decision variables, a pre-defined optimization algorithm is used to initialize the slime mold population using a random generation strategy. The position vector of each slime mold individual corresponds to a set of scheduling schemes, and the initial fitness value of each slime mold individual is calculated. The individual with the smallest fitness value is recorded as the initial global optimal solution, and the current population is used as the initial solution.

[0048] In this embodiment, the preset optimization algorithm is explained using the Improved Slime Mould Algorithm (ISMA) with elite back learning and quadratic interpolation. By simulating the exploration and information transmission mechanism of slime mold in the environment, the algorithm gradually approaches the optimal solution by continuously adjusting the search path and the quality of the solution. The initialization phase is the basic preparatory step of the ISMA algorithm. First, the decision space dimension, slime mold population size, and upper and lower boundaries of decision variables of the optimization problem are defined to ensure that the generation of candidate solutions conforms to the physical constraints of the microgrid scheduling problem. These physical constraints include the output range of the gas turbine and the charging and discharging power limit of the battery. Subsequently, the algorithm initializes the position of each slime mold through a random generation strategy. Each position vector corresponds to a set of scheduling schemes. In this embodiment, the decision space dimension is 72-dimensional decision variables to ensure that the initial population is evenly distributed in the search space, covering more potential optimal solution areas. After initialization, the algorithm calculates the fitness value of each slime mold, i.e., the electrocarbon objective function value, to evaluate the quality of candidate solutions. The smaller the fitness value, the closer the position is to the optimal scheduling scheme, providing a basis for subsequent weight allocation and search strategy selection.

[0049] One explanation is that during algorithm initialization, N individuals, representing slime mold locations, are generated, with each individual representing a scheduling scheme, and the fitness value of each individual is calculated. The objective function value of electrocarbon is the total operating cost. The individual with the lowest fitness in the current population is denoted as the initial global optimum. The corresponding fitness value is and the initial global optimal solution and the corresponding fitness value As the initial solution.

[0050] In the above scheme, the decision space dimension, population size, and decision variable boundaries are first determined, and a random generation strategy is used to initialize the slime mold population to ensure that the initial population is uniformly distributed in the search space, providing a high-quality initial solution for subsequent iterative searches. In addition, by initializing the slime mold population through a random generation strategy, the position vector of each slime mold individual corresponds to a set of scheduling schemes, which can cover more potential optimal solution regions, effectively avoiding the problems of low initial population quality and insufficient search space coverage, and helping to improve the reliability of power regulation strategies.

[0051] In a preferred embodiment, the search modes include a random diffusion mode, a directional fluctuation mode, and a local contraction mode. An iterative update process is performed on the initial solution. If the number of iterations does not meet a preset iteration requirement, a nonlinear decay factor is calculated, and a first random number and a second random number are obtained. Based on the first and second random numbers, several search modes are matched, including: if the number of iterations does not meet the preset iteration requirement, a nonlinear decay factor is calculated based on the current iteration number and the target iteration number, and a first random number and a second random number are obtained; when the first random number is less than a preset first exploration threshold, a random diffusion mode is matched; when the first random number is greater than or equal to the preset first exploration threshold and the second random number is less than a preset second exploration threshold, a directional fluctuation mode is matched; when the first random number is greater than or equal to the preset first exploration threshold and the second random number is greater than or equal to the preset second exploration threshold, a local contraction mode is matched.

[0052] For example, first, based on the current iteration number With maximum number of iterations The nonlinear attenuation factor α is calculated using the following formula: ; The nonlinear decay factor 'a' is relatively large in the early stages of iteration, approaching positive infinity, corresponding to the high viability state of slime molds, indicating a tendency for large-scale exploration; as the iteration progresses... As the count gradually decreases and approaches zero, the slime mold activity decreases, its behavior becomes more cautious, and it focuses more on a refined search within known high-quality areas.

[0053] Simultaneously, a first random number z and a second random number p are obtained, both z and p being uniformly distributed random numbers within the interval [0,1]. The preset first exploration threshold is typically 0.03, and the preset second exploration threshold is typically 0.5. When the first random number z is less than 0.03, a random diffusion mode is matched; when the first random number z is greater than or equal to 0.03 and the second random number p is less than 0.5, a directional fluctuation mode is matched; when the first random number z is greater than or equal to 0.03 and the second random number p is greater than or equal to 0.5, a local contraction mode is matched. A p value of 0.5 means that both modes are assigned equal selection probabilities. This helps to dynamically balance global exploration and local development during the iteration process, allowing individuals to converge quickly using optimal solution information while retaining a certain probability of random perturbation to maintain population vitality and avoid falling into local extrema due to over-development.

[0054] In the above scheme, a nonlinear decay factor and a random number matching mechanism are introduced to dynamically determine the switching between three search modes: random diffusion, directional fluctuation, and local contraction. This achieves a dynamic balance between global exploration and local development. The nonlinear decay factor is calculated based on the current iteration number and the maximum iteration number. When it is large in the early stage of iteration, it corresponds to a high-activity state and tends to explore a large area. As the iteration progresses, it gradually decreases and approaches zero, corresponding to a decrease in activity and a shift to fine-grained search. Thus, the comparison between the first random number and the preset exploration threshold determines whether to perform completely random exploration, and the comparison between the second random number and the preset division threshold divides directional fluctuation and local contraction. This energy-driven dynamic mechanism provides a quantitative basis for the switching of movement modes, thereby helping to improve the reliability of power regulation strategies.

[0055] A preferred embodiment involves searching and updating the initial solution based on several search modes and the nonlinear decay factor until the number of iterations meets a preset iteration requirement, thereby obtaining a multi-objective optimization solution set. This includes: when the search mode is a random diffusion mode, randomly generating new positions within the decision space according to the decision variable boundaries to obtain a first search solution; when the search mode is a directional fluctuation mode, updating the positions based on the initial solution, random individual differences, the nonlinear decay factor, and weight coefficients to obtain a second search solution, wherein the weight coefficients are calculated based on fitness values, and the random individual differences are obtained by subtracting the position vectors of two randomly selected individuals; when the search mode is a local contraction mode, determining a random vector based on the nonlinear decay factor, and updating the position within a preset range corresponding to the current individual's position vector using the random vector to obtain a third search solution; and updating the initial solution based on the first, second, or third search solution until the number of iterations meets a preset iteration requirement, thereby obtaining a multi-objective optimization solution set.

[0056] For example, in the random diffusion model, new positions are randomly generated within the decision space based on the upper and lower boundaries of the decision variables, allowing individuals to completely leave their current regions to explore the unknown solution space and obtain the first search solution, represented as: ; In the formula, , For the upper and lower bounds of the decision variables, for The algorithm uses a uniformly distributed random vector, and this completely random update method can effectively avoid getting stuck in local optima in the early stages of the algorithm, ensuring full coverage of the search space. In the directional fluctuation mode, the position is updated based on the current global optimal solution, the random individual difference, and the weight coefficients calculated according to fitness ranking. Specifically, the random individual difference is obtained by subtracting the position vectors of two different individuals. The weight coefficients are calculated based on the individual's fitness ranking in the population, with individuals with higher fitness receiving greater weights to guide the search direction. Combined with the fluctuation amplitude determined by the nonlinear decay factor, the new position maintains random perturbation while moving closer to the optimal solution, resulting in the second search solution, expressed as: ; In the formula, This is the current globally optimal solution; In order to be in A uniformly distributed random vector controls the fluctuation amplitude; The weighting coefficients are calculated based on fitness ranking, with high-quality individuals receiving greater weights to guide the search direction; , These are the position vectors of two randomly selected different individuals.

[0057] In the local contraction mode, the oscillation vector is determined based on the nonlinear decay factor, and a small-range position update is performed near the current individual position vector. This allows the individual to perform a refined search within the high-quality region, resulting in the third search solution, expressed as: ; In the formula, In order to be in A random vector that is uniformly distributed within the vector, but with... different, The value of is more focused on oscillation around the current individual, simulating the behavior of slime molds contracting pseudopodia for fine local detection after discovering a food source. Let i be the position vector of the current individual i, representing a set of scheduling schemes, which are 72-dimensional decision variables in this embodiment. This update method allows the individual to search within a small range near its current position, which helps to improve the accuracy of local optima.

[0058] Finally, based on the first search solution, the second search solution, or the third search solution, the initial solution is searched and updated until the number of iterations meets the preset iteration requirements, thereby obtaining a multi-objective optimization solution set.

[0059] In the above scheme, position update rules are defined under three search modes. The directional fluctuation mode introduces weight coefficients to guide the search direction, while the local contraction mode uses oscillation vectors for refined searching, effectively improving the convergence accuracy and global search capability of the algorithm. The random diffusion mode randomly generates new positions in the decision space based on the upper and lower boundaries of the decision variables, ensuring comprehensive coverage of the search space. The directional fluctuation mode updates positions based on the current global optimum, random individual differences, and weight coefficients calculated according to fitness ranking, maintaining both the convergence to the optimum and the population diversity through random perturbation. The local contraction mode determines random vectors based on nonlinear decay factors and performs small-range position updates near the current individual, which helps improve the accuracy of local optima. Thus, a rich set of low-carbon alternatives is provided through a customized search mechanism, improving the reliability of power regulation strategies.

[0060] A preferred embodiment involves updating the initial solution based on the first, second, or third search solution until the number of iterations meets a preset iteration requirement, thereby obtaining a multi-objective optimized solution set. This includes: performing boundary checks on the first, second, or third search solutions generated in each iteration based on the decision variable boundaries; correcting the first, second, or third search solutions that do not meet the boundary check requirements to obtain corresponding optimized first, second, or third search solutions; calculating the search fitness value corresponding to the optimized first, second, or third search solution; comparing the initial fitness value corresponding to the initial solution with the search fitness value to obtain the target search solution and its corresponding target fitness value; and updating the target search solution and its corresponding target fitness value until the number of iterations meets a preset iteration requirement, thereby obtaining a multi-objective optimized solution set.

[0061] For example, boundary checks are performed on the new positions generated in each iteration, by... The operation is corrected; if it exceeds the upper limit, it is corrected to the upper limit value; if it exceeds the lower limit, it is corrected to the lower limit value. Boundary correction ensures that all scheduling schemes meet the physical constraints of the equipment, and the fitness value of the corrected new location is calculated. and the fitness value at the original position. Compare the fitness values ​​of the new location. If the fitness value is smaller, replace the original position with the new position; otherwise, retain the original position. Additionally, if the fitness value of the new position is smaller... Fitness value better than the current global optimum Then Update the global optimal solution When the number of iterations meets the preset iteration requirement, that is, the maximum number of iterations is met. Thus, a multi-objective optimization solution set is obtained.

[0062] It is worth mentioning that an external archive set is maintained during the iteration process. After each generation update, the current population and the archive set are combined, and Pareto front solutions are retained by non-dominated sorting and crowding distance filtering. In this embodiment, Pareto front solutions can be characterized as solutions in the multi-objective optimization solution set. The external archive set is used to store all non-dominated solutions, i.e. Pareto front solutions, discovered during the iteration process. After each generation update, the current population and the external archive set are merged, and the solutions are divided into different front levels by non-dominated sorting. Then, the distribution density of solutions within the same front level is calculated by crowding distance, and Pareto front solutions with uniform distribution are selected and stored in the archive set. Solutions that are dominated or have excessively high crowding are deleted from the archive set.

[0063] In the above scheme, boundary checks and fitness comparison mechanisms are adopted to ensure that the search solution is always within the feasible region. Greedy selection is used to retain better solutions and drive the population to evolve towards the global optimum. Boundary checks are performed on the new positions generated in each iteration to correct components that exceed the upper and lower limits back to the feasible region, avoiding solutions that violate physical constraints due to search movement and ensuring the engineering feasibility of the scheduling scheme. The fitness value of the new position is calculated and compared with the fitness value of the original position to retain better solutions. At the same time, the global optimum and the optimal fitness value are updated in real time to ensure that the optimal fitness of the population continues to decrease after each iteration, driving the population to gradually converge to the global optimum scheduling scheme and improving the reliability of the power regulation strategy.

[0064] A preferred embodiment generates a power dispatch strategy based on the multi-objective optimization solution set to regulate the power generation, grid, load, and storage system. This includes: extracting individuals from the multi-objective optimization solution set whose total operating cost meets preset dispatch requirements as target cost solutions, and generating corresponding economic dispatch strategies based on these target cost solutions; extracting individuals from the multi-objective optimization solution set whose total carbon emissions meet preset dispatch requirements as target carbon emission solutions, and generating corresponding carbon emission dispatch strategies; selecting a knee-point compromise solution based on the economic dispatch strategy and the carbon emission dispatch strategy to obtain a power dispatch strategy; decoding the power dispatch strategy into a sequence of decision variables, and generating output curves for each device, a state-of-load (SVR) graph for energy storage, a grid interaction power graph, and a cost composition graph based on the SVR and the carbon emission objective function; and sending these output curves, SVR graphs, grid interaction power graphs, and cost composition graphs to regional energy management equipment to regulate the power generation, grid, load, and storage system.

[0065] For example, the solution with the lowest total operating cost is extracted as the minimum cost solution, corresponding to the most economically efficient dispatch strategy. The solution with the lowest total carbon emissions is extracted as the minimum carbon emission solution, corresponding to a low-carbon priority carbon emission dispatch strategy. The Pareto front objective value is normalized, the Euclidean distance from each solution to the ideal point is calculated, and curvature analysis is combined to select the solution with the best balance between cost and carbon emissions as the knee compromise solution. This solution corresponds to a balanced dispatch strategy that considers both economic efficiency and low carbon emissions, i.e., a power dispatch strategy. The power dispatch strategy is decoded into a 72-dimensional sequence of decision variables and substituted back into the objective function and carbon emission accounting model to verify its feasibility. Simultaneously, visualization results such as power output curves for each device, energy storage state-of-charge change diagrams, grid interaction power diagrams, and cost composition diagrams are generated. (See [link to relevant documentation]). Figure 2 , Figure 3 , Figure 4 and Figure 5 , Figure 2 A schematic diagram of the equipment output curve of a multi-objective coordinated regulation method for source-grid-load-storage provided in a certain embodiment of the present invention; Figure 3 A schematic diagram of the state of charge change of energy storage in a multi-objective coordinated regulation method for source-grid-load-storage provided in a certain embodiment of the present invention; Figure 4 A schematic diagram of grid interaction power for a multi-objective coordinated regulation method of source, grid, load and storage provided in a certain embodiment of the present invention; Figure 5 This is a schematic diagram illustrating the cost structure of a multi-objective coordinated regulation method for source-grid-load-storage provided in a certain embodiment of the present invention; as shown below. Figure 2 As shown, Figure 2 The load curve is represented by a solid black line, the photovoltaic output curve by a solid yellow line, the gas turbine output curve by a solid red line, and the battery discharge power curve by a dashed blue line. Figure 2 It allows for a direct view of the power supply and demand matching situation of the system at different time periods, verifying the degree to which power balance constraints are met; such as Figure 3 As shown, Figure 3 The state of charge (SOC) change curve of the battery over 24 hours is represented by the solid green line. Figure 3 The text also indicates that the SOC safety range is represented by a green shaded area, and the initial SOC level is represented by a red dashed line. Figure 2 We can observe whether the energy storage charging and discharging strategy conforms to the upper and lower limits of SOC, and whether the terminal SOC returns to near its initial value; for example Figure 4 As shown, Figure 4 The main grid interaction power graph shows the power exchange between the microgrid and the external main grid over a 4-hour period. Orange bars represent purchased electricity, and purple bars represent sold electricity. This graph allows analysis of the system's implementation of the strategy of utilizing time-of-use pricing for low-storage and high-generation electricity generation at different times. Figure 5 As shown, Figure 5This is a cost breakdown diagram, showing the total cost of the scheduling plan and the cost of each component.

[0066] Finally, the output curves of each device, the energy storage state of charge change diagram, the grid interaction power diagram, and the cost composition diagram are sent to the regional energy management equipment to regulate the source, grid, load, and storage.

[0067] In the above scheme, target cost solutions, target carbon emission solutions, and knee-point compromise solutions that meet preset scheduling requirements are extracted from the multi-objective optimization solution set. This generates diverse scheduling strategies, and the visualization output significantly enhances the interpretability and decision support capabilities of the optimization results. The target cost solution that meets the preset scheduling requirements corresponds to the most economically optimal scheduling strategy, while the target carbon emission solution represents the extreme low-carbon priority scheme. The knee-point compromise solution, by normalizing the frontier target value and calculating the distance to the ideal point for each solution, combined with curvature analysis, selects the individual with the best balance between cost and carbon emissions as the compromise scheme. Finally, the extracted scheduling schemes are decoded into a sequence of decision variables, generating output curves for each device, a state-of-load diagram for energy storage, a power grid interaction diagram, and a cost composition diagram. These are then transmitted to the regional energy system's energy management system for source-grid-load-storage regulation. Thus, the low-carbon objective is integrated into the optimization algorithm, improving its interpretability and decision support capabilities. Furthermore, a customized search mechanism provides rich low-carbon alternatives, enhancing the reliability of power regulation strategies.

[0068] Based on the above method embodiments, corresponding apparatus embodiments are provided; see [link to apparatus embodiments]. Figure 6 , Figure 6 This is a schematic diagram of the module structure of a multi-objective coordinated control system for source-grid-load-storage provided in a certain embodiment of the present invention. Figure 6As shown, this embodiment of the invention also provides a multi-objective coordinated control system for power generation, grid, load, and storage, including a data processing module 201, an objective solving module 202, a pattern matching module 203, an iterative update module 204, and a control module 205; wherein: the data processing module 201 is used to construct an electric carbon objective function based on pre-acquired power equipment operation data and dynamic electric carbon factor data, and set constraints on the electric carbon objective function; the objective solving module 202 is used to solve the electric carbon objective function based on a preset optimization algorithm and the constraints to generate an initial solution; the pattern matching module 203... The iterative update module 204 is used to perform an iterative update process on the initial solution. If the number of iterations does not meet the preset iteration requirements, a nonlinear decay factor is calculated and a first random number and a second random number are obtained. Several search patterns are matched based on the first random number and the second random number. The iterative update module 204 is used to search and update the initial solution based on several search patterns and the nonlinear decay factor until the number of iterations meets the preset iteration requirements, thereby obtaining a multi-objective optimized solution set and ending the iterative update process. The control module 205 is used to generate a power dispatch strategy based on the multi-objective optimized solution set to control the power generation, grid, load, and storage.

[0069] This invention proposes a multi-objective coordinated control system for power generation, grid, load, and storage. It deeply integrates dynamic carbon dioxide factors into the optimization objectives, constructing a carbon dioxide objective function. Simultaneously, it iteratively solves the objective function using a pre-defined optimization algorithm, establishing an adaptive search mechanism based on random number matching of multiple search modes during the iteration process. This effectively enhances the algorithm's adaptability to the coupling characteristics of carbon dioxide and electricity. Based on this, it generates a multi-objective optimization solution set and outputs power dispatch strategies, providing decision-makers with diverse low-carbon dispatch schemes and improving the reliability and interpretability of power control strategies. Thus, it integrates low-carbon objectives into the optimization algorithm, improving its interpretability and decision support capabilities, and provides rich low-carbon alternatives through a customized search mechanism, thereby enhancing the reliability of power control strategies.

[0070] Furthermore, the data processing module 201 includes a power data acquisition unit 301, an electric carbon data acquisition unit 302, a first cost function construction unit 303, a second cost function construction unit 304, a third cost function construction unit 305, an objective function construction unit 306, and a constraint setting unit 307; wherein: the power data acquisition unit 301 is used to acquire power equipment operation data, wherein the power equipment operation data includes gas turbine output and main grid switching power; the electric carbon data acquisition unit 302 is used to acquire dynamic electric carbon factor data, wherein the dynamic electric carbon factor data includes carbon price, electricity sales price, and electricity purchase price; the first cost function construction unit 303 is used to acquire the gas turbine output... The first unit constructs a fuel cost function based on the power output and a preset fuel cost coefficient; the second cost function construction unit 304 constructs a grid interaction cost function based on the main grid exchange power, electricity sales price, and electricity purchase price; the third cost function construction unit 305 constructs a carbon emission cost function based on the gas turbine output, main grid exchange power, carbon price, preset carbon emission coefficient, and preset implicit carbon emission coefficient; the objective function construction unit 306 sums the fuel cost function, the grid interaction cost function, and the carbon emission cost function to obtain an electric carbon objective function; and the constraint setting unit 307 sets power balance constraints, battery energy state constraints, and gas turbine ramp rate constraints on the electric carbon objective function.

[0071] Furthermore, the objective solving module 202 includes a parameter determination unit 401 and an optimization strategy solving unit 402; wherein: the parameter determination unit 401 is used to determine the decision space dimension, population size, and decision variable boundary based on the constraints; the optimization strategy solving unit 402 is used to initialize the slime mold population using a random generation strategy through the preset optimization algorithm based on the decision space dimension, the population size, and the decision variable boundary, where the position vector of each slime mold individual corresponds to a set of scheduling schemes, and calculates the initial fitness value of each slime mold individual, and uses individuals whose initial fitness values ​​meet the preset fitness requirements as the initial solution.

[0072] It is understood that the above-described device embodiments correspond to the method embodiments of the present invention, and can realize the multi-objective coordinated regulation method of source, grid, load and storage provided by any of the above-described method embodiments of the present invention.

[0073] It should be noted that the device embodiments described above are merely illustrative, and some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the device embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can specifically be implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0074] Based on the above-described embodiment of a multi-objective coordinated regulation method for source, grid, load, and storage, another embodiment of the present invention provides a terminal device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements a multi-objective coordinated regulation method for source, grid, load, and storage according to any embodiment of the present invention.

[0075] For example, in this embodiment, the computer program can be divided into one or more modules, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the terminal device.

[0076] The terminal device may be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0077] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the terminal device, connecting all parts of the terminal device via various interfaces and lines.

[0078] Based on the above-described method embodiments, another embodiment of the present invention provides a computer-readable storage medium, including a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to execute the multi-objective coordinated regulation method of source-grid-load-storage described in any of the above-described method embodiments of the present invention.

[0079] The modules / units integrated in the device / terminal equipment, if implemented as software functional units and sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0080] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for multi-objective collaborative regulation of source, network, load and storage, characterized in that, include: Based on the pre-acquired power equipment operation data and dynamic electric carbon factor data, an electric carbon objective function is constructed and constraints are set on the electric carbon objective function; The objective function of the electrocarbon is solved based on the preset optimization algorithm and the constraints to generate an initial solution; An iterative update process is performed on the initial solution. If the number of iterations does not meet the preset iteration requirements, a nonlinear decay factor is calculated and a first random number and a second random number are obtained. Several search patterns are matched based on the first random number and the second random number. Based on several search modes and the nonlinear decay factor, the initial solution is searched and updated until the number of iterations meets the preset iteration requirements, thereby obtaining a multi-objective optimization solution set and ending the iterative update process. Based on the multi-objective optimization solution set, a power dispatch strategy is generated to regulate the power generation, grid, load, and storage systems.

2. The multi-objective coordinated regulation method for source-grid-load-storage as described in claim 1, characterized in that, The process of constructing an electric carbon objective function based on pre-acquired power equipment operation data and dynamic electric carbon factor data, and setting constraints on the electric carbon objective function, includes: Acquire power equipment operation data, wherein the power equipment operation data includes gas turbine output and main grid switching power; Acquire dynamic carbon factor data, wherein the dynamic carbon factor data includes carbon price, electricity sales price and electricity purchase price; Based on the gas turbine output and the preset fuel cost coefficient, a fuel cost function is constructed; Based on the main grid switching power, electricity sales price, and electricity purchase price, a grid interaction cost function is constructed; Based on the gas turbine output, main grid exchange power, carbon price, preset carbon emission coefficient and preset implicit carbon emission coefficient, a carbon emission cost function is constructed. The fuel cost function, the power grid interaction cost function, and the carbon emission cost function are summed to obtain the electricity carbon objective function; The objective function for the electric carbon is subject to power balance constraints, battery energy state constraints, and gas turbine ramp rate constraints.

3. The multi-objective coordinated regulation method for source-grid-load-storage as described in claim 1, characterized in that, The step of solving the objective function of the electrocarbon based on the preset optimization algorithm and the constraints to generate an initial solution includes: Based on the aforementioned constraints, the decision space dimension, population size, and decision variable boundaries are determined. Based on the decision space dimension, the population size, and the decision variable boundary, the slime mold population is initialized using a random generation strategy through the preset optimization algorithm. The position vector of each slime mold individual corresponds to a set of scheduling schemes, and the initial fitness value of each slime mold individual is calculated. Individuals whose initial fitness values ​​meet the preset fitness requirements are used as the initial solutions.

4. The multi-objective coordinated regulation method for source-grid-load-storage as described in claim 3, characterized in that, The search modes include random diffusion mode, directional fluctuation mode, and local contraction mode. An iterative update process is performed on the initial solution. If the number of iterations does not meet a preset iteration requirement, a nonlinear decay factor is calculated, and a first random number and a second random number are obtained. Based on the first random number and the second random number, several search modes are matched, including: If the number of iterations does not meet the preset iteration requirements, calculate the nonlinear decay factor based on the current number of iterations and the target number of iterations, and obtain the first random number and the second random number. When the first random number is less than a preset first exploration threshold, the random diffusion pattern is matched; When the first random number is greater than or equal to a preset first exploration threshold and the second random number is less than a preset second exploration threshold, the directional fluctuation mode is matched. When the first random number is greater than or equal to a preset first exploration threshold and the second random number is greater than or equal to a preset second exploration threshold, the local contraction mode is matched.

5. The multi-objective coordinated regulation method for source-grid-load-storage as described in claim 4, characterized in that, Based on several of the aforementioned search patterns and the nonlinear decay factor, the initial solution is searched and updated until the number of iterations meets a preset iteration requirement, resulting in a multi-objective optimization solution set, including: When the search mode is a random diffusion mode, a new position is randomly generated in the decision space according to the boundary of the decision variables to obtain the first search solution; When the search mode is a directional fluctuation mode, the position is updated according to the initial solution, random individual difference, nonlinear decay factor and weight coefficient to obtain the second search solution. The weight coefficient is calculated according to the fitness value sorting, and the random individual difference is obtained by subtracting the position vectors of two randomly selected individuals. When the search mode is a local contraction mode, a random vector is determined according to the nonlinear decay factor, and the position is updated by the random vector within a preset range corresponding to the current individual position vector to obtain the third search solution. Based on the first search solution, the second search solution, or the third search solution, the initial solution is searched and updated until the number of iterations meets the preset iteration requirements, thereby obtaining a multi-objective optimization solution set.

6. The multi-objective coordinated regulation method for source-grid-load-storage as described in claim 5, characterized in that, Based on the first search solution, the second search solution, or the third search solution, the initial solution is searched and updated until the number of iterations meets a preset iteration requirement, resulting in a multi-objective optimization solution set, including: Based on the decision variable boundaries, the first search solution, second search solution, or third search solution generated in each iteration is subjected to boundary checks. The first search solution, second search solution, or third search solution that does not meet the boundary check requirements is corrected to obtain the corresponding optimized first search solution, optimized second search solution, or optimized third search solution. Calculate the search fitness value corresponding to the optimized first search solution, optimized second search solution, or optimized third search solution, and compare the initial fitness value corresponding to the initial solution with the search fitness value to obtain the target search solution and the corresponding target fitness value; Update the target search solution and the corresponding target fitness value until the number of iterations meets the preset iteration requirements to obtain a multi-objective optimization solution set.

7. The multi-objective coordinated regulation method for source-grid-load-storage as described in claim 1, characterized in that, Based on the aforementioned multi-objective optimization solution set, a power dispatch strategy is generated to regulate the power generation, grid, load, and storage systems, including: Extract individuals whose total operating cost meets the preset scheduling requirements from the multi-objective optimization solution set as target cost solutions, and generate corresponding economic scheduling strategies based on the target cost solutions; Extract individuals whose total carbon emissions meet the preset scheduling requirements from the multi-objective optimization solution set as target carbon emission solutions, and generate corresponding carbon emission scheduling strategies. Based on the economic dispatch strategy and the carbon emission dispatch strategy, a knee-point compromise solution is selected to obtain the power dispatch strategy; The power dispatch strategy is decoded into a sequence of decision variables. Based on the sequence of decision variables and the carbon emission objective function, output curves of each device, energy storage state of charge change diagram, power grid interaction diagram and cost composition diagram are generated. The output curves of each device, the energy storage state of charge change diagram, the grid interaction power diagram, and the cost composition diagram are sent to the regional energy management equipment to regulate the source, grid, load, and storage.

8. A multi-objective coordinated control system for source-grid-load-storage, characterized in that, It includes a data processing module, an objective solution module, a pattern matching module, an iterative update module, and a control module; among which: The data processing module is used to construct an electric carbon objective function and set constraints on the electric carbon objective function based on pre-acquired power equipment operation data and dynamic electric carbon factor data. The objective solving module is used to solve the objective function of the electrocarbon based on a preset optimization algorithm and the constraints, and generate an initial solution; The pattern matching module is used to perform an iterative update process on the initial solution. If the number of iterations does not meet the preset iteration requirements, a nonlinear decay factor is calculated and a first random number and a second random number are obtained. Several search patterns are matched based on the first random number and the second random number. The iterative update module is used to search and update the initial solution based on several search modes and the nonlinear decay factor until the number of iterations meets the preset iteration requirements, thereby obtaining a multi-objective optimization solution set and ending the iterative update process. The control module is used to generate a power dispatch strategy based on the multi-objective optimization solution set, so as to regulate the power generation, grid, load and storage.

9. The multi-objective coordinated control system for source-grid-load-storage as described in claim 8, characterized in that, The data processing module includes a power data acquisition unit, an electric carbon data acquisition unit, a first cost function construction unit, a second cost function construction unit, a third cost function construction unit, an objective function construction unit, and a constraint setting unit; wherein: The power data acquisition unit is used to acquire power equipment operation data, which includes gas turbine output and main grid switching power. The carbon data acquisition unit is used to acquire dynamic carbon factor data, wherein the dynamic carbon factor data includes carbon price, electricity sales price and electricity purchase price; The first cost function construction unit is used to construct a fuel cost function based on the gas turbine output and a preset fuel cost coefficient; The second cost function construction unit is used to construct a power grid interaction cost function based on the main grid switching power, electricity sales price, and electricity purchase price; The third cost function construction unit is used to construct a carbon emission cost function based on the gas turbine output, main grid exchange power, carbon price, preset carbon emission coefficient and preset implicit carbon emission coefficient; The objective function construction unit is used to sum the fuel cost function, the power grid interaction cost function, and the carbon emission cost function to obtain the electricity carbon objective function; The constraint setting unit is used to set power balance constraints, battery energy state constraints, and gas turbine ramp rate constraints for the electric carbon objective function.

10. The multi-objective coordinated control system for source-grid-load-storage as described in claim 8, characterized in that, The objective solving module includes a parameter determination unit and an optimization strategy solving unit; wherein: The parameter determination unit is used to determine the decision space dimension, population size, and decision variable boundary based on the constraints. The optimization strategy solving unit is used to initialize the slime mold population using a random generation strategy through the preset optimization algorithm based on the decision space dimension, the population size, and the decision variable boundary. The position vector of each slime mold individual corresponds to a set of scheduling schemes, and the unit calculates the initial fitness value of each slime mold individual. Individuals whose initial fitness values ​​meet the preset fitness requirements are used as the initial solutions.

Images