An Optimal Scheduling Method for Electric Energy Storage Load in a Virtual Power Plant for Deep Peak Regulation

Abstract

The present invention discloses a method for optimizing the scheduling of electrical storage loads in a virtual power plant for deep peak shaving, belonging to the technical field of optimizing the scheduling of electrical storage loads. The present invention obtains historical peak shaving information, constructs a deep peak shaving prediction model using machine learning based on the historical peak shaving information, and calculates the preliminary allocation information of the distributed energy types of the virtual power plant with the minimum cost of the virtual power plant as the objective based on the distributed energy information of the virtual power plant. Furthermore, in the case where the preliminary allocation information of the distributed energy types of the virtual power plant does not meet the preset resource consumption index, the preliminary allocation information of the distributed energy types of the virtual power plant is input into the feasible region corresponding to the preset resource consumption index for optimization, and the optimal allocation of the distributed energy types of the virtual power plant that meets the preset resource consumption index is obtained. The present invention can implement green and low-carbon while achieving a better energy allocation effect and improve energy utilization efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of optimized scheduling of electrical storage loads, and specifically provides an optimized scheduling method for electrical storage loads in a virtual power plant for deep peak shaving. Background Art

[0002] In recent years, new energy sources such as solar energy and wind energy have been gradually developed. To meet the demand for the full and precise utilization of new energy sources, there is a huge demand for optimized scheduling methods during the energy supply process of new energy sources; especially with the proposal and popularization of virtual power plants, intelligent microgrids, and intelligent microgrid clusters, the simple cooperation of photovoltaic and energy storage can no longer meet the increasing demand for new energy grid connection; in the future, how to effectively, efficiently, and accurately perform deep peak shaving during the process of new energy participating in the energy supply process has become a technical problem that urgently needs to be solved;

[0003] In the prior art, a Chinese patent with the patent application number "202310618101.4" discloses that according to energy scheduling indicators, parts that do not meet the preset carbon emission indicators can be obtained. The energy consumption law is represented by energy scheduling characteristics, and the energy scheduling indicators are disassembled based on the energy scheduling characteristics. The obtained energy scheduling results can make the integrated energy system meet the preset carbon emission indicators, so as to achieve optimized scheduling with the specified carbon emission as the optimization goal, obtain the optimal operating state of each device in the integrated energy system, thereby reducing the carbon emissions of the integrated energy system, improving its environmental protection benefits, and realizing the optimization of the integrated energy system. However, the problem of economic benefits is not considered during this process. But in the actual optimization process, most use the multi-objective optimization method to optimize the electrical storage loads in the virtual power plant with economic benefits as the optimal, and do not comprehensively consider economic benefits and carbon emission indicators. Therefore, when actually optimizing the electrical storage loads in the virtual power plant, the optimization should not only be based on cost and economic value as the goal, but also consider resource and carbon-saving constraints. Summary of the Invention

[0004] The purpose of the present invention is to provide an optimized scheduling method for electrical storage loads in a virtual power plant for deep peak shaving, so as to solve the problem of considering resource and carbon emission constraints during the process of optimizing the electrical storage loads in the virtual power plant mentioned in the above background art.

[0005] To solve the above technical problems, the present invention provides the following technical solution. An optimized scheduling method for electrical storage loads in a virtual power plant for deep peak shaving, and the steps of the optimized scheduling method include:

[0006] Obtain historical peak shaving information, and use machine learning to construct a deep peak shaving prediction model based on the historical peak shaving information;

[0007] Based on the distributed energy information of the virtual power plant, with the minimum cost of the virtual power plant as the goal, calculate the preliminary allocation information of the types of distributed energy in the virtual power plant;

[0008] In the case where the preliminary allocation information of the types of distributed energy in the virtual power plant does not meet the preset resource consumption index, input the preliminary allocation information of the types of distributed energy in the virtual power plant into the feasible region corresponding to the preset resource consumption index for optimization, and obtain the optimal allocation of the types of distributed energy in the virtual power plant that meets the preset resource consumption index.

[0009] According to the above technical solution, the distributed energy information includes the types of distributed energy, the cost corresponding to each type of distributed energy, the power generation information and electricity storage information corresponding to each type of new distributed energy, and the types of distributed energy include wind power, photovoltaic, controllable negative charge, gas turbines, etc.

[0010] According to the above technical solution, the historical peak shaving information includes the historical peak shaving time period, the total amount of electricity storage load scheduling during historical peak shaving, and the historical peak shaving impact factor; the deep peak shaving prediction model is used to predict the peak shaving time period and the total amount of electricity storage load scheduling during historical peak shaving, and the historical peak shaving impact factor includes seasons, weather, electricity consumption information, temperature, etc.

[0011] According to the above technical solution, the calculation steps of the preliminary allocation information of the types of distributed energy in the virtual power plant include:

[0012] Obtain the peak shaving impact factor in real time, and use the deep peak shaving prediction model to calculate the peak shaving prediction time period and the total predicted amount of electricity storage load scheduling during peak shaving;

[0013] Based on the distributed energy information of the virtual power plant and the total predicted amount of electricity storage load scheduling during peak shaving, construct an optimization scheduling objective function for the electricity storage load; on this basis, constraints are imposed on each type of distributed energy in the virtual power plant itself, such as constraints on the upper and lower limits of the energy storage device capacity and the upper limits of charging and discharging, and the gas turbine output meets the upper and lower limit constraints, etc.;

[0014] Taking the total predicted amount of electricity storage load scheduling during peak shaving and the power balance of the virtual power plant as the total constraint conditions for the optimization scheduling of the electricity storage load, use the particle swarm optimization method to obtain the preliminary allocation of the types of distributed energy in the virtual power plant.

[0015] According to the above technical solution, the resource consumption index includes the total consumption index of each type of distributed energy and the carbon emission index.

[0016] According to the above technical solution, input the preliminary allocation information of the distributed energy types of the virtual power plant into the selection layer of the feasible region, calculate the consumption difference between each distributed energy consumption in the preliminary allocation information of the distributed energy types of the virtual power plant and the corresponding resource consumption index by using the difference formula, and determine whether there is any distributed energy consumption exceeding the standard in the preliminary allocation information of the distributed energy types of the virtual power plant;

[0017] Calculate the total carbon emissions produced according to the preliminary allocation information of the distributed energy types of the virtual power plant, calculate the carbon emission difference from the carbon emission index in the resource consumption index by using the difference calculation formula, and determine whether the corresponding total carbon emissions in the preliminary allocation information of the distributed energy types of the virtual power plant exceed the standard;

[0018] Determine the types of optimization guiding vectors according to the consumption difference and the carbon emission difference, and guide the preliminary allocation information of the distributed energy types of the virtual power plant to be input into the corresponding optimization layer according to the optimization guiding vectors to re-optimize the preliminary allocation information of the distributed energy types of the virtual power plant.

[0019] According to the above technical solution, when the total carbon emissions corresponding to the preliminary allocation information of the distributed energy types of the virtual power plant reach the standard and there is a certain distributed energy consumption exceeding the standard in the preliminary allocation information of the distributed energy types of the virtual power plant, the type of the optimization guiding vector is the consumption optimization guiding vector;

[0020] When the total carbon emissions corresponding to the preliminary allocation information of the distributed energy types of the virtual power plant exceed the standard and all the distributed energy consumptions in the preliminary allocation information of the distributed energy types of the virtual power plant exceed the standard, the type of the optimization guiding vector is the carbon emission optimization guiding vector;

[0021] When the total carbon emissions corresponding to the preliminary allocation information of the distributed energy types of the virtual power plant exceed the standard and there is a certain distributed energy consumption exceeding the standard in the preliminary allocation information of the distributed energy types of the virtual power plant, the type of the optimization guiding vector is the overall optimization guiding vector.

[0022] According to the above technical solution, according to the consumption optimization guiding vector, input the preliminary allocation information of the distributed energy types of the virtual power plant into the corresponding consumption optimization layer. The consumption optimization layer determines the types of distributed energy with consumption exceeding the standard and the consumption difference, calculates the carbon emissions corresponding to the distributed energy consumption differences in all the preliminary allocation information and adds them to the carbon emission difference to obtain the carbon emission constraint condition;

[0023] Based on the total electric charge storage capacity corresponding to the distributed energy consumption differences in all the preliminary allocation information, re-confirm the distributed energy types for re-optimization without considering the distributed energy types with consumption exceeding the standard of the virtual power plant;

[0024] Based on the carbon emission constraint conditions and the re-optimized distributed energy types, reconstruct the minimum cost objective function, and use the particle swarm optimization method to obtain a partial allocation result of the distributed energy types of the virtual power plant;

[0025] Overlay the partial allocation result of the distributed energy types of the virtual power plant with the preliminary allocation result of the distributed energy types of the virtual power plant, and subtract the corresponding consumption difference of the distributed energy in the preliminary allocation information to obtain the optimized allocation of the distributed energy types of the virtual power plant.

[0026] According to the above technical solution, input the preliminary allocation information of the distributed energy types of the virtual power plant into the corresponding carbon emission optimization layer according to the carbon emission optimization vector. The carbon emission optimization layer calculates the consumption required for the carbon emission difference generated by each distributed new energy, and reuses the consumption required for the carbon emission difference of each distributed energy as the constraint condition for re-optimization;

[0027] Based on the carbon emission difference, use the calculation method of carbon emissions and the conversion formula with electricity to determine the corresponding total electric storage load, and then reconstruct the minimum cost objective function;

[0028] Based on the re-optimized constraint conditions and the reconstructed minimum cost objective function, use the particle swarm optimization method to obtain a partial allocation result of the distributed energy types of the virtual power plant;

[0029] Overlay the partial allocation result of the distributed energy types of the virtual power plant with the preliminary allocation result of the distributed energy types of the virtual power plant to determine the optimized allocation of the distributed energy types of the virtual power plant.

[0030] According to the above technical solution, input the preliminary allocation information of the distributed energy types of the virtual power plant into the consumption optimization layer and the carbon emission optimization layer for optimization respectively according to the overall optimization vector. After taking the average value of the optimization results of the consumption optimization layer and the carbon emission optimization layer, input it into the selection layer of the feasible region to judge whether further optimization is needed. If further optimization is needed, re-select the optimization layer for optimization according to the type of the optimization vector until the allocation result of the distributed energy types of the optimized virtual power plant meets the resource consumption index standard.

[0031] Compared with the prior art, the beneficial effects achieved by the present invention are:

[0032] 1. The present invention can obtain the peak shaving impact factor in real time, and calculate the peak shaving prediction time period and the total predicted amount of electric storage load scheduling during peak shaving through the deep peak shaving prediction model, providing accurate data support for the preliminary allocation of the distributed energy types of the virtual power plant.

[0033] 2. The present invention constructs an optimal scheduling objective function for the electrical storage load, restricts the types of distributed energy sources in the virtual power plant, and ensures the stability and reliability of the scheduling process.

[0034] 3. The present invention uses the particle swarm optimization method to obtain the preliminary allocation of the types of distributed energy sources in the virtual power plant, improving the efficiency and accuracy of the allocation.

[0035] 4. The present invention takes into account the resource consumption index and the carbon emission index to achieve the overall optimization of the distributed energy consumption in the virtual power plant. By calculating the consumption difference and the carbon emission difference through the difference formula, the rationality of the allocation scheme is judged, further improving the accuracy of the optimization scheme. According to the optimization guiding vector, the preliminary allocation information of the types of distributed energy sources in the virtual power plant is optimized again to ensure the rationality and compliance of the allocation scheme. In practical applications, the types and calculation methods of the optimization guiding vector can be adjusted according to specific situations to achieve a better energy allocation effect. The present invention can effectively reduce the consumption and carbon emissions of the types of distributed energy sources in the virtual power plant on the premise of meeting the total carbon emission standard, improve the energy utilization efficiency, and contribute to the realization of the green and low-carbon development goal. In addition, the present invention is applicable to various combinations of distributed energy sources and has strong generality and practicability. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The drawings are used to provide a further understanding of the present invention, and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention, and do not constitute a limitation to the present invention. In the drawings:

[0037] Figure 1 is the step flow chart of an optimal scheduling method for the electrical storage load in a virtual power plant for deep peak shaving according to the present invention;

[0038] Figure 2 is the step flow chart of the preliminary allocation of the types of distributed energy sources in the virtual power plant in this embodiment;

[0039] Figure 3 is the step flow chart of the re-optimization of the preliminary allocation information of the types of distributed energy sources in the virtual power plant in this embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0040] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0041] The present invention provides a technical solution, an optimized scheduling method for electrical storage load in a virtual power plant for deep peak shaving, and its steps include:

[0042] S1. Obtain historical peak shaving information, and use machine learning to construct a deep peak shaving prediction model according to the historical peak shaving information; the historical peak shaving information includes the historical peak shaving time period, the total amount of electrical storage load scheduling during historical peak shaving, and the historical peak shaving impact factor; the deep peak shaving prediction model is used to predict the peak shaving time period and the total amount of electrical storage load scheduling during historical peak shaving. The machine learning model can use deep learning models, Bayesian network models, LSTM network models, neural network models, etc.

[0043] S2. Based on the distributed energy information of the virtual power plant, the distributed energy information includes the types of distributed energy, the cost corresponding to each type of distributed energy, the power generation information and electricity storage information corresponding to each type of distributed new energy; taking the minimum cost of the virtual power plant as the goal, calculate the preliminary allocation information of the types of distributed energy in the virtual power plant, specifically:

[0044] S201. Obtain the peak shaving impact factor in real time, and use the deep peak shaving prediction model to calculate the peak shaving prediction time period and the total predicted amount of electrical storage load scheduling during peak shaving;

[0045] S202. Based on the distributed energy information of the virtual power plant and the total predicted amount of electrical storage load scheduling during peak shaving, construct an optimized scheduling objective function for electrical storage load;

[0046] S203. Taking the total predicted amount of electrical storage load scheduling during peak shaving and the power balance of the virtual power plant as the total constraint conditions for the optimized scheduling of electrical storage load, use the particle swarm optimization method to obtain the preliminary allocation of the types of distributed energy in the virtual power plant.

[0047] S3. In the case that the preliminary allocation information of the types of distributed energy in the virtual power plant does not meet the preset resource consumption index, input the preliminary allocation information of the types of distributed energy in the virtual power plant into the feasible region corresponding to the preset resource consumption index for optimization, and obtain the optimal allocation of the types of distributed energy in the virtual power plant that meets the preset resource consumption index, specifically:

[0048] The resource consumption index includes the total consumption index of each type of distributed energy and the carbon emission index;

[0049] S301. Input the preliminary allocation information of the types of distributed energy in the virtual power plant into the selection layer of the feasible region, use the difference formula to calculate the consumption difference between the consumption of each distributed energy in the preliminary allocation information of the types of distributed energy in the virtual power plant and the corresponding resource consumption index, and judge whether there is an excessive consumption of distributed energy in the preliminary allocation information of the types of distributed energy in the virtual power plant;

[0050] S302. Calculate the total carbon emissions produced based on the preliminary allocation information of the distributed energy types of the virtual power plant. Use the difference calculation formula to calculate the carbon emission difference from the carbon emission index in the resource consumption index. Determine whether the corresponding total carbon emissions in the preliminary allocation information of the distributed energy types of the virtual power plant exceed the standard. Determine the type of optimization guiding vector based on the consumption difference and the carbon emission difference;

[0051] (1) When the total carbon emissions corresponding to the preliminary allocation information of the distributed energy types of the virtual power plant meet the standard, and there is a distributed energy consumption that exceeds the standard in the preliminary allocation information of the distributed energy types of the virtual power plant, the type of the optimization guiding vector is the consumption optimization guiding vector;

[0052] (2) When the total carbon emissions corresponding to the preliminary allocation information of the distributed energy types of the virtual power plant exceed the standard, and all the distributed energy consumptions in the preliminary allocation information of the distributed energy types of the virtual power plant meet the standard, the type of the optimization guiding vector is the carbon emission optimization guiding vector;

[0053] (3) When the total carbon emissions corresponding to the preliminary allocation information of the distributed energy types of the virtual power plant exceed the standard, and there is a distributed energy consumption that exceeds the standard in the preliminary allocation information of the distributed energy types of the virtual power plant, the type of the optimization guiding vector is the overall optimization guiding vector.

[0054] S303. Guide the preliminary allocation information of the distributed energy types of the virtual power plant according to the optimization guiding vector and input it into the corresponding optimization layer to re-optimize the preliminary allocation information of the distributed energy types of the virtual power plant;

[0055] (1) According to the consumption optimization guiding vector, input the preliminary allocation information of the distributed energy types of the virtual power plant into the corresponding consumption optimization layer. The consumption optimization layer determines the types of distributed energy with consumption exceeding the standard and the consumption difference, calculates the carbon emissions corresponding to the distributed energy consumption differences in all preliminary allocation information and adds them to the carbon emission difference to obtain the carbon emission constraint condition;

[0056] Based on the total electrical storage load corresponding to the distributed energy consumption differences in all preliminary allocation information, re-confirm the types of distributed energy for re-optimization without considering the distributed energy types with consumption exceeding the standard of the virtual power plant;

[0057] Based on the carbon emission constraint condition and the types of distributed energy for re-optimization, re-construct the minimum cost objective function, and use the particle swarm optimization method to obtain the partial allocation results of the distributed energy types of the virtual power plant;

[0058] Superimpose the partial allocation results of the distributed energy types of the virtual power plant on the preliminary allocation results of the distributed energy types of the virtual power plant, and subtract the corresponding consumption differences of the distributed energy in the preliminary allocation information to obtain the optimized allocation of the distributed energy types of the virtual power plant.

[0059] (2) According to the carbon emission optimization guiding vector, input the preliminary allocation information of the distributed energy types of the virtual power plant into the corresponding carbon emission optimization layer. The carbon emission optimization layer calculates the consumption required to generate the carbon emission difference for each distributed new energy, and reuses the consumption required for the carbon emission difference of each distributed energy type as the constraint condition for re-optimization;

[0060] Based on the carbon emission difference, use the calculation method of carbon emissions and the conversion formula with electricity to determine the total electrical storage load, and then reconstruct the minimum cost objective function; the calculation method of carbon emissions and the conversion formula with electricity belong to the prior art. Among them, saving 1 degree of electricity is equivalent to saving 0.4 kilograms of standard coal and reducing 0.997 kilograms of carbon dioxide at the same time. Therefore, the total electrical storage load can be determined according to the conversion relationship between various new energies and standard coal.

[0061] Based on the re-optimized constraint conditions and the reconstructed minimum cost objective function, use the particle swarm optimization method to obtain the partial allocation results of the distributed energy types of the virtual power plant;

[0062] Superimpose the partial allocation results of the distributed energy types of the virtual power plant on the preliminary allocation results of the distributed energy types of the virtual power plant to determine the optimized allocation of the distributed energy types of the virtual power plant.

[0063] (3) According to the overall optimization guiding vector, input the preliminary allocation information of the distributed energy types of the virtual power plant into the consumption optimization layer and the carbon emission optimization layer for optimization respectively. After taking the average of the optimization results of the consumption optimization layer and the carbon emission optimization layer, input them into the selection layer of the feasible region to judge whether further optimization is needed. If further optimization is needed, reselect the optimization layer according to the type of the optimization guiding vector for optimization until the allocation result of the distributed energy types of the optimized virtual power plant meets the resource consumption index standard.

[0064] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "including", "comprising" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or device.

[0065] Finally, it should be noted that the above are only the preferred embodiments of the present invention and are not used to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or perform equivalent replacements for some of the technical features. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for optimizing the dispatching of electrical storage loads in a virtual power plant for deep peak shaving, characterized in that, Including: Obtain historical peak shaving information, and construct a deep peak shaving prediction model using machine learning based on the historical peak shaving information; Based on the distributed energy information of the virtual power plant, with the minimum cost of the virtual power plant as the goal, calculate the preliminary allocation information of the types of distributed energy of the virtual power plant; In the case that the preliminary allocation information of the types of distributed energy of the virtual power plant does not meet the preset resource consumption index, input the preliminary allocation information of the types of distributed energy of the virtual power plant into the feasible region corresponding to the preset resource consumption index for optimization, and obtain the optimal allocation of the types of distributed energy of the virtual power plant that meets the preset resource consumption index; The calculation steps of the preliminary allocation information of the types of distributed energy of the virtual power plant include: Obtain the peak shaving impact factor in real time, and use the deep peak shaving prediction model to calculate the peak shaving prediction time period and the total predicted amount of electricity storage load scheduling during peak shaving; Based on the distributed energy information of the virtual power plant and the total predicted amount of electricity storage load scheduling during peak shaving, construct an optimization objective function for electricity storage load scheduling; Taking the total predicted amount of electricity storage load scheduling during peak shaving and the power balance of the virtual power plant as the total constraint conditions for the optimization of electricity storage load scheduling, use the particle swarm optimization method to obtain the preliminary allocation of the types of distributed energy of the virtual power plant; Input the preliminary allocation information of the types of distributed energy of the virtual power plant into the selection layer of the feasible region, use the difference formula to calculate the consumption difference between each distributed energy consumption in the preliminary allocation information of the types of distributed energy of the virtual power plant and the corresponding resource consumption index, and judge whether there is an excessive consumption of distributed energy in the preliminary allocation information of the types of distributed energy of the virtual power plant; Calculate the total carbon emissions produced according to the preliminary allocation information of the types of distributed energy of the virtual power plant, use the difference calculation formula to calculate the carbon emission difference from the carbon emission index in the resource consumption index, and judge whether the corresponding total carbon emissions in the preliminary allocation information of the types of distributed energy of the virtual power plant exceed the standard; Determine the type of the optimization guiding vector according to the consumption difference and the carbon emission difference, and guide the preliminary allocation information of the types of distributed energy of the virtual power plant to be input into the corresponding optimization layer according to the optimization guiding vector to re-optimize the preliminary allocation information of the types of distributed energy of the virtual power plant.

2. The optimized scheduling method for electric energy storage load in a virtual power plant for deep peak shaving according to claim 1, wherein The distributed energy information includes the types of distributed energy, the cost corresponding to each type of distributed energy, the power generation information and electricity storage information corresponding to each type of new distributed energy.

3. The optimized scheduling method for electric energy storage load in a virtual power plant for deep peak shaving according to claim 1, characterized in that The historical peak shaving information includes the historical peak shaving time period, the total amount of electricity storage load scheduling during historical peak shaving, and the historical peak shaving impact factor; the deep peak shaving prediction model is used to predict the peak shaving time period and the total amount of electricity storage load scheduling during historical peak shaving.

4. The optimized scheduling method for the electrical storage load in a virtual power plant for deep peak shaving according to claim 1, wherein The resource consumption index includes the total consumption index of each type of distributed energy and the carbon emission index.

5. A method for optimizing the scheduling of electrical storage loads in a virtual power plant for deep peak shaving according to claim 1, characterized in that When the total carbon emissions corresponding to the preliminary allocation information of the types of distributed energy of the virtual power plant meet the standard, and there is an excessive consumption of a certain distributed energy in the preliminary allocation information of the types of distributed energy of the virtual power plant, the type of the optimization guiding vector is the consumption optimization guiding vector; When the total carbon emissions corresponding to the preliminary allocation information of the distributed energy types in the virtual power plant exceed the standard and the consumption of all distributed energy in the preliminary allocation information of the distributed energy types in the virtual power plant meets the standard, the type of the optimization guiding vector is the carbon emission optimization guiding vector; When the total carbon emissions corresponding to the preliminary allocation information of the distributed energy types in the virtual power plant exceed the standard and there is a certain distributed energy consumption exceeding the standard in the preliminary allocation information of the distributed energy types in the virtual power plant, the type of the optimization guiding vector is the overall optimization guiding vector.

6. The optimal scheduling method for the electrical storage load in a virtual power plant for deep peak shaving according to claim 5, characterized in that According to the consumption optimization guiding vector, the preliminary allocation information of the distributed energy types in the virtual power plant is input into the corresponding consumption optimization layer. The consumption optimization layer is used to determine the types of distributed energy with consumption exceeding the standard and the consumption difference, calculate the carbon emissions corresponding to the distributed energy consumption difference in all the preliminary allocation information and add them to the carbon emission difference to obtain the carbon emission constraint condition; Based on the total electrical storage load corresponding to the distributed energy consumption difference in all the preliminary allocation information, reconfirm the types of distributed energy for re-optimization without considering the distributed energy types with consumption exceeding the standard in the virtual power plant; Based on the carbon emission constraint condition and the re-optimized distributed energy types, reconstruct the minimum cost objective function, and use the particle swarm optimization method to obtain a partial allocation result of the distributed energy types in the virtual power plant; Overlay the partial allocation result of the distributed energy types in the virtual power plant with the preliminary allocation result of the distributed energy types in the virtual power plant and subtract the corresponding consumption difference of the distributed energy in the preliminary allocation information to obtain the optimized allocation of the distributed energy types in the virtual power plant.

7. A method for optimizing the dispatching of electric energy storage loads in a virtual power plant for deep peak shaving according to claim 5, characterized in that According to the carbon emission optimization guiding vector, the preliminary allocation information of the distributed energy types in the virtual power plant is input into the corresponding carbon emission optimization layer. The carbon emission optimization layer calculates the consumption required for each distributed new energy to generate a carbon emission difference, and reuses the consumption required for the carbon emission difference of each distributed energy as the constraint condition for re-optimization; Based on the carbon emission difference, determine the corresponding total electrical storage load using the calculation method of carbon emissions and the conversion formula with electricity, and then reconstruct the minimum cost objective function; Based on the re-optimized constraint condition and the reconstructed minimum cost objective function, use the particle swarm optimization method to obtain a partial allocation result of the distributed energy types in the virtual power plant; Overlay the partial allocation result of the distributed energy types in the virtual power plant with the preliminary allocation result of the distributed energy types in the virtual power plant to determine the optimized allocation of the distributed energy types in the virtual power plant.

8. A method for optimizing the dispatching of the electrical storage load in a virtual power plant for deep peak shaving according to claim 5, characterized in that According to the overall optimization guiding vector, the preliminary allocation information of the distributed energy types in the virtual power plant is respectively input into the consumption optimization layer and the carbon emission optimization layer for optimization. After taking the average of the optimization results of the consumption optimization layer and the carbon emission optimization layer, input them into the feasible region selection layer to judge whether further optimization is needed. If further optimization is needed, reselect the optimization layer for optimization according to the type of the optimization guiding vector until the allocation result of the distributed energy types in the optimized virtual power plant meets the resource consumption index standard.

Images