Optical storage system auxiliary service collaborative optimization method, system, equipment and medium

Abstract

The invention belongs to the technical field of power systems, and discloses an optical storage system auxiliary service collaborative optimization method, system, device and medium, and the method comprises the steps: analyzing adjustable resource multi-dimensional response characteristics of various types of energy storage virtual power plants based on resource response characteristics, response capability and economy; according to the adjustable resource multi-dimensional response characteristics of the virtual power plant, constructing a response model of each distributed resource in the virtual power plant; and establishing a virtual power plant operation decision optimization model by taking benefit maximization of the virtual power plant as a target, and solving by using an open source solver to obtain a collaborative optimization result. From the perspective of the virtual power plant, the collaborative optimization operation strategy that the user side distributed optical storage system participates in the auxiliary service market is analyzed, so that the user side distributed optical storage system can obtain the collaborative operation strategy which is more comprehensive in consideration, more accurate in scheduling and more intelligent in management and control by accessing the virtual power plant.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and in particular to a method, system, device and medium for collaborative optimization of auxiliary services of a photovoltaic storage system. Background Art

[0002] The photovoltaic storage system is an important component of the virtual power plant. The virtual power plant is a distributed energy aggregation and coordinated optimization management system implemented through advanced information and communication technology and software systems, which aims to participate in the electricity market and power grid operation as a special power plant. The electricity generated by the photovoltaic storage system through photovoltaic power generation can be supplied to the virtual power plant, and the energy storage system can release the stored electricity during peak electricity demand to help balance the grid load. By connecting to the virtual power plant, the user-side distributed photovoltaic storage system obtains a more comprehensive, more precise scheduling, and more intelligent management and control collaborative operation strategy. Therefore, from the perspective of the virtual power plant, the present invention analyzes the collaborative optimization operation strategy of the user-side distributed photovoltaic storage system participating in the auxiliary service market, and provides a photovoltaic storage system auxiliary service collaborative optimization method, system, device and medium. Summary of the invention

[0003] The embodiments of the present invention provide a method, system, device and medium for collaborative optimization of auxiliary services of a photovoltaic storage system to solve the above-mentioned technical problems in the prior art.

[0004] According to a first aspect of an embodiment of the present invention, a method for collaborative optimization of auxiliary services of a photovoltaic storage system is provided.

[0005] In one embodiment, the photovoltaic storage system auxiliary service collaborative optimization method includes:

[0006] Based on resource response characteristics, response capabilities and economic efficiency, the multi-dimensional response characteristics of adjustable resources of various types of energy storage virtual power plants are analyzed;

[0007] According to the multi-dimensional response characteristics of the adjustable resources of the virtual power plant, the response model of each distributed resource in the virtual power plant is constructed;

[0008] With the goal of maximizing the benefits of virtual power plants, a virtual power plant operation decision optimization model is constructed, and the open source solver is used to solve it to obtain collaborative optimization results.

[0009] In one embodiment, the analysis of the multi-dimensional response characteristics of adjustable resources of various types of energy storage virtual power plants based on resource response characteristics, response capabilities and economy includes:

[0010] Analyze the load response characteristics of industrial and commercial power users based on industry production characteristics, production cycle, load energy consumption ratio, compressible capacity and response speed, and provide safety regulation suggestions;

[0011] Based on historical charging data records, analyze the spatiotemporal distribution characteristics of electric vehicle charging demand and the charging scenario response characteristics of electric vehicles;

[0012] Construct an economic analysis model for energy storage system participation in regulation technology, and use the economic analysis model to analyze the response characteristics of the photovoltaic storage system.

[0013] In one embodiment, the industrial and commercial power users include industrial and commercial power users in high energy consumption industries, industrial and commercial power users in continuous production industries, industrial and commercial power users in discontinuous production industries, and industrial and commercial power users in other industries.

[0014] In one embodiment, the spatiotemporal distribution characteristics include charging duration distribution characteristics and charging start time distribution characteristics, and the charging scenarios include public charging scenarios, community charging scenarios, and highway charging scenarios.

[0015] According to a second aspect of an embodiment of the present invention, a system for collaborative optimization of auxiliary services of a photovoltaic storage system is provided.

[0016] In one embodiment, the photovoltaic storage system auxiliary service collaborative optimization system includes:

[0017] The response characteristics analysis module is used to analyze the multi-dimensional response characteristics of adjustable resources of various types of energy storage virtual power plants based on resource response characteristics, response capabilities and economic efficiency;

[0018] A response model building module is used to build a response model of each distributed resource in the virtual power plant according to the multi-dimensional response characteristics of the adjustable resources of the virtual power plant;

[0019] The optimization model solving module is used to build a virtual power plant operation decision optimization model with the goal of maximizing the virtual power plant's revenue, and use an open source solver to solve it and obtain collaborative optimization results.

[0020] In one embodiment, when analyzing the multi-dimensional response characteristics of adjustable resources of various types of energy storage virtual power plants based on resource response characteristics, response capabilities and economy, the response characteristic analysis module analyzes the load response characteristics of industrial and commercial power users based on industry production characteristics, production cycle, load energy consumption ratio, compressible capacity and response speed, and gives safety regulation suggestions; analyzes the spatiotemporal distribution characteristics of electric vehicle charging demand and the charging scenario response characteristics of electric vehicles based on historical charging data records; constructs an economic analysis model of energy storage system participation in regulation technology, and uses the economic analysis model to analyze the response characteristics of the photovoltaic storage system.

[0021] In one embodiment, the industrial and commercial power users include industrial and commercial power users in high energy consumption industries, industrial and commercial power users in continuous production industries, industrial and commercial power users in discontinuous production industries, and industrial and commercial power users in other industries.

[0022] In one embodiment, the spatiotemporal distribution characteristics include charging duration distribution characteristics and charging start time distribution characteristics, and the charging scenarios include public charging scenarios, community charging scenarios, and highway charging scenarios.

[0023] According to a third aspect of an embodiment of the present invention, a computer device is provided.

[0024] In some embodiments, the computer device includes a memory and a processor, the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.

[0025] According to a fourth aspect of embodiments of the present invention, a computer-readable storage medium is provided.

[0026] In one embodiment, the computer-readable storage medium stores a computer program, and the computer program implements the steps of the above method when executed by a processor.

[0027] The technical solution provided by the embodiment of the present invention may have the following beneficial effects:

[0028] The present invention analyzes the collaborative optimization operation strategy of the user-side distributed photovoltaic storage system participating in the auxiliary service market from the perspective of the virtual power plant, and provides a method for collaborative optimization of auxiliary services of the photovoltaic storage system, so that the user-side distributed photovoltaic storage system can obtain a more comprehensive, more precise scheduling, and more intelligent management and control collaborative operation strategy by connecting to the virtual power plant.

[0029] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0031] Figure 1 is a flow chart of a method for collaborative optimization of auxiliary services of a photovoltaic storage system according to an exemplary embodiment;

[0032] Figure 2 is a structural block diagram of a photovoltaic storage system auxiliary service collaborative optimization system according to an exemplary embodiment;

[0033] Figure 3 The figure is a schematic diagram showing the structure of a computer device according to an exemplary embodiment. DETAILED DESCRIPTION

[0034] Figure 1A method for collaborative optimization of auxiliary services of a photovoltaic storage system of the present invention is shown.

[0035] In this optional embodiment, the photovoltaic storage system auxiliary service collaborative optimization method includes:

[0036] Step S101, analyzing the multi-dimensional response characteristics of adjustable resources of various types of energy storage virtual power plants based on resource response characteristics, response capabilities and economy;

[0037] Step S103, constructing a response model of each distributed resource in the virtual power plant according to the multi-dimensional response characteristics of the adjustable resources of the virtual power plant;

[0038] Step S105, with the goal of maximizing the profit of the virtual power plant, construct a virtual power plant operation decision optimization model, and use an open source solver to solve it to obtain a collaborative optimization result.

[0039] Figure 2 An embodiment of a photovoltaic storage system auxiliary service collaborative optimization system of the present invention is shown.

[0040] In this optional embodiment, the photovoltaic storage system auxiliary service collaborative optimization system includes:

[0041] The response characteristic analysis module 201 is used to analyze the multi-dimensional response characteristics of adjustable resources of various types of energy storage virtual power plants based on resource response characteristics, response capabilities and economy;

[0042] The response model building module 203 is used to build a response model of each distributed resource in the virtual power plant according to the multi-dimensional response characteristics of the adjustable resources of the virtual power plant;

[0043] The optimization model solving module 205 is used to construct a virtual power plant operation decision optimization model with the goal of maximizing the virtual power plant's revenue, and use an open source solver to solve it to obtain a collaborative optimization result.

[0044] In order to facilitate understanding of the above technical solutions of the present invention, the above technical solutions of the present invention are further described below from the perspective of architecture and principle, as follows:

[0045] (1) Analysis of multi-dimensional response characteristics of adjustable resources of virtual power plants including multiple types of energy storage

[0046] This section analyzes the control mode and external response characteristics of virtual power plants. It is necessary to investigate and analyze the internal resources such as distributed photovoltaics and energy storage systems, analyze the operation mechanism and production rules of various resources of virtual power plants, and conduct detailed classification and analysis of the response characteristics of flexible resources. Since virtual power plants often have different forms in pilot projects in various places, and the types of aggregated resources are also different, this section first selects various resources according to industry classification, and analyzes the response of resources with adjustable potential by industry and equipment type in combination with their industry production and operation mechanisms and control methods. The adjustable capacity of virtual power plant resources is the basis for virtual power plants to obtain profits, and the response characteristics of virtual power plants are also the basis for virtual power plant operation decisions. Therefore, the response characteristics, response capabilities and economic efficiency of various resources will affect the results of the optimization strategy of the virtual power plant operation decision-making process.

[0047] 1) Internal and external characteristics of virtual power plants

[0048] ① Analysis of external characteristics of virtual power plants

[0049] In general, virtual power plants have the following external characteristics compared to traditional power plants:

[0050] Environmentally friendly construction and operation: Virtual power plants use control technologies and mechanisms to achieve complementary and coordinated dispatch of new energy sources to reduce wind and solar power abandonment, thereby improving the utilization rate of new energy while ensuring the stability and security of the power grid;

[0051] Geographical dispersion: Virtual power plants centrally manage equipment distributed in different regions, integrate various power sources and power loads, and achieve efficient control of power generation and power consumption units;

[0052] Flexibility in market participation: Virtual power plants can not only aggregate a large number of distributed renewable energy generation to participate in the market as power generation entities, but also aggregate loads to participate in the market as power sales companies;

[0053] Virtual power plant assets are virtual: the internal resources of a virtual power plant do not necessarily belong to the virtual power plant operator. The relationship between the operator and the assets is mainly energy scheduling, value distribution and information sharing. The virtual power plant does not necessarily have ownership of the access resources.

[0054] ② Internal resource characteristics of virtual power plants

[0055] Output characteristics of distributed power sources within the virtual power plant: During peak load periods, it is difficult for distributed photovoltaics to provide sufficient output, but during low load periods, distributed photovoltaics experience output peaks, which have anti-peak regulation characteristics. In addition, wind power generation will be affected by factors such as weather and seasons, and is random and unstable.

[0056] Typical adjustable load characteristics within the virtual power plant: Charging load: The charging of electric buses is affected by travel schedules and is mostly charged at night, which is a transferable resource; with the popularization of electric vehicles, the layout of charging piles is relatively scattered, with many points and a wide range; in addition, the charging regulation response capability is strong. With the improvement of the flexible adjustable capabilities of some equipment, the peak charging load can be transferred to the off-peak period while keeping the total charging capacity unchanged, which is a transferable load.

[0057] Photovoltaic storage system: It is a production and marketing resource with decentralized access points in space and strong climbing and response capabilities. Energy storage needs to maintain a balance between total charging and discharging power.

[0058] Thermal storage electric heating: It has seasonal characteristics and mainly operates by utilizing wind abandonment at night and low electricity prices. It has obvious peak and valley characteristics and can meet potential system ramp-up needs by reserving additional flexibility capacity.

[0059] The layout of charging load resources is dispersed, and they are shiftable and transferable with strong response capabilities. The photovoltaic storage system is a production and sales resource with strong climbing ability. Thermal storage electric heating is a seasonal resource with obvious peak and valley characteristics and strong regulation ability. In contrast, conventional units aim to maximize power generation revenue, while virtual power plants focus on providing balancing regulation services to the power grid, with the goal of maximizing the total revenue taking into account the electricity energy transaction and balancing service response revenue.

[0060] 2) Analysis of response characteristics of typical industrial and commercial users

[0061] In fact, the types of industrial and commercial users connected to virtual power plants are diverse, and the typical load curves of users are closely related to the types of industries. Adjustable loads should be screened by industry, and industrial and commercial users with high single load capacity and high digital level of power load management should be screened to access virtual power plants to participate in power trading and dispatch response. The load response characteristics analysis of industrial and commercial power users starts from the two aspects of power load characteristics and response capabilities. Combined with the load control requirements of the "two high" industries, load safety characteristics analysis and evaluation and other specific contents, this research content conducts research and investigation on industrial and commercial users by industry, and conducts detailed analysis according to the dimensions of industry production characteristics, production cycle, load energy consumption ratio, compressible capacity, response speed, etc., and gives safety regulation suggestions.

[0062] ①High energy consumption industries

[0063] Priority is given to typical high-energy-consuming enterprises for analysis, such industries mainly include chemical raw materials and product manufacturing, non-metallic mineral products, etc. Such industries generally use 35kV and above high-voltage power supply, with high power load, and generally three-shift continuous production operations. Due to the wide variety of production equipment, the main production load accounts for a high proportion, and adjusting the load will bring greater safety risks and economic losses. Within the scope of equipment safety permission, the load can be adjusted in different proportions.

[0064] The following analysis takes the chlor-alkali industry (chemical raw materials and chemical products manufacturing industry) as an example. The alkali industry is a typical chemical raw materials and chemical products manufacturing industry. Enterprises produce continuously for 24 hours. The main products involved include caustic soda, chlorine, hydrochloric acid, and hydrogen. It is one of the important basic chemical industries. The diaphragm method is the main production process in the chlor-alkali industry. The brine solution is electrolyzed by a vertical diaphragm electrolyzer.

[0065] Load response capability analysis: The core production equipment of chlor-alkali production enterprises is the electrolytic cell, and its load capacity accounts for more than 75% of the total load of the whole plant. Under normal circumstances, the electrolytic cell has been operating stably at a constant power. Frequent load fluctuations and equipment start and stop will cause great damage to the equipment and diaphragm, greatly reduce the life and performance of the equipment, and also bring certain safety hazards. Chlor-alkali enterprises can reduce part of the load capacity by reducing production, but it will have a certain impact on the life and performance of the equipment. Interruption of power supply during the electrolysis process will lead to major safety accidents such as toxic gas leakage and equipment explosion. It takes about 3 days for the main production equipment in the industry to be shut down in an orderly manner. You can communicate with the company in advance and make production plan adjustments according to the power demand situation, electricity prices and power supply gap forecasts. Direct control cannot be used for adjustment to prevent major accidents. The load round-by-round safety regulation suggestions are shown in Table 1:

[0066] Table 1 Analysis of various load characteristics

[0067]

[0068] Load response capability analysis: The core production equipment of chlor-alkali production enterprises is the electrolytic cell, and its load capacity accounts for more than 75% of the total load of the whole plant. Under normal circumstances, the electrolytic cell has been operating stably at a constant power. Frequent load fluctuations and equipment start-up and shutdown will cause great damage to the equipment and diaphragm, greatly reduce the life and performance of the equipment, and also bring certain safety hazards. Chlor-alkali enterprises can reduce part of the load capacity by reducing production, but it will have a certain impact on the life and performance of the equipment. Interruption of power supply during the electrolysis process will lead to major safety accidents such as toxic gas leakage and equipment explosion. It takes about 3 days for the main production equipment in the industry to be shut down in an orderly manner. You can communicate with the company in advance and make production plan adjustments according to the power demand situation, electricity prices and power supply gap forecasts. Direct control cannot be used for adjustment to prevent major accidents. The load round-by-round safety regulation suggestions are shown in Table 2:

[0069] Table 2 Recommendations for safe load regulation in phases

[0070]

[0071] Safety regulation suggestions: In the first round, non-productive loads such as non-important office lighting, office electrical appliances, air conditioning, factory lighting, etc. are regulated. Depending on the customer scale, the expected load limit of about 1% can be achieved. In the second round, some auxiliary production loads such as fans, circulating water pumps, centrifugal pumps, dissolution tanks, etc. are regulated. Depending on the customer scale, the expected load limit of about 8% can be achieved. In the third round of regulation, part of the production capacity of the electrolyzer can be compressed according to the customer situation. Depending on the customer scale, the expected load limit of about 20% can be achieved.

[0072] ②Continuous production industry

[0073] Continuous production industries mainly include beer manufacturing, rubber manufacturing, tobacco manufacturing, automobile manufacturing, etc., which are generally powered by 10kV substations and 0.4kV. The nature of production is generally three-shift continuous operation. Due to the wide variety of production equipment in this industry, most of which are assembly line production, adjusting the load will bring greater safety risks to the equipment and cause certain economic losses. The direct control response capability is limited, and it is mainly carried out by arranging and adjusting production plans and maintenance plans. Below, the automobile manufacturing industry is taken as an example for analysis. The product types of the automobile manufacturing industry are relatively wide, including various parts manufacturers and vehicle processing companies. The process standard requirements of different companies in this industry are also different. The daily average load range is large, ranging from tens of kilowatts to 10,000 kilowatts.

[0074] Analysis of load safety response capability: The automotive manufacturing industry requires uninterruptible power supply for high-frequency furnaces, sintering furnaces and other equipment. Power outages may cause fires. The industry's production equipment shuts down quickly, and some equipment can be shut down with 4 hours' notice. The automotive manufacturing industry has a complex variety of sub-industries, and the scale and production process of enterprises vary. When implementing load control, you should understand the specific situation of the enterprise in advance, communicate with customers in advance, and clarify the load control method under the premise of ensuring safety. The recommendations for load safety regulation in rounds are shown in Table 3:

[0075] Table 3 Recommendations for safe load regulation in phases

[0076]

[0077] Safety regulation suggestions for the automotive manufacturing industry: The first round is implemented for non-productive loads such as office lighting and public electricity, air-conditioning factory lighting in non-important production areas. Depending on the customer situation, a response of about 2% of the total load can be achieved. The second round is implemented for some auxiliary production loads such as fans, conveyor belts, refrigerators, and electric welding. Depending on the customer scale, a response of about 5% of the total load can be achieved. Expected regulation: The third round can be implemented for all assembly lines in some factory areas or for part of the main productive loads such as welding machines, paint sprayers, electroplating machines, lathes, production lines, assembly lines, etc. in the factory area.

[0078] ③ Non-continuous production industry

[0079] Non-continuous production industries mainly include information service industry, medical equipment and equipment manufacturing, printing and recording media duplication industry, instrument manufacturing industry, food manufacturing industry, construction industry, etc. Analysis of the production characteristics of this enterprise shows that the load adjustment potential is large, and load adjustment can be carried out through rotation.

[0080] Take the information service industry as an example: the information service industry mainly includes information transmission, software and information technology services. Some customers in the education industry and postal industry are greatly affected by time periods, seasons and climate. The electricity consumption time period of this industry is fixed, the electricity consumption elasticity coefficient is relatively stable, and the adjustable load is small. Adjusting the load for some customers will bring greater safety risks and economic losses, and the adjustable load potential and flexibility are small. The software industry is a high-tech industry, and the main functional areas are: software industry center, R&D center, training center, reception and exhibition center, closed R&D center, logistics center, catering center, etc.

[0081] Analysis of load safety response capability: Sudden power outage in this industry will cause the temperature of the computer room to rise, which may cause damage to data exchange equipment. Firefighting facilities, communication systems, and emergency lighting in related buildings cannot be powered off. The load safety control recommendations are shown in Table 4:

[0082] Table 4 Recommendations for safe load regulation in phases

[0083]

[0084] Suggestions for regulation and control of information transmission, software and information technology service industries: The first round will be implemented for non-productive loads such as non-important corridor lighting, exterior wall lighting, billboards, etc., and depending on the customer's situation, the expected limit of about 2% of the total load can generally be achieved; the second round will be implemented for some auxiliary production loads such as office lighting, central air-conditioning, split air-conditioning, ceiling air-conditioning, etc., and depending on the customer's scale, the expected limit of about 10% of the total load can be achieved: the third round can be based on customer communication to regulate all assembly lines in some factory areas or part of the main productive loads such as network rooms, laboratory equipment, switches, computers in all factory areas, and according to the customer's scale, the expected response of about 50% of the total load can be achieved.

[0085] ⑤Other industries

[0086] Other industries, such as general equipment manufacturing users, can also adjust their loads by adjusting production plans and taking turns to rest, and have a large potential for adjustable loads. Other types of industries that use electricity for guaranteed purposes generally have low electricity loads, are not affected by seasons, climate, or time periods, and use electricity mainly at night. They have fast adjustable load response speeds, but their total loads are low and their response capabilities are limited.

[0087] 3) Analysis of electric vehicle response characteristics

[0088] As the number of electric vehicles increases, electric vehicles may become one of the largest adjustable loads in the future power grid. The randomness of load characteristics will bring new demands and challenges to the operation of the power grid. Studies have shown that as the number of electric vehicles increases, their random and disordered charging has a significant impact on the peak-to-valley difference of the power grid. The larger the number of electric vehicles, the greater the peak-to-valley difference. Taking Shanghai as an example, under normal development conditions, there will be about 1.55 million electric vehicles in Shanghai in 2030; under high-speed development conditions, the number of electric vehicles will reach 2.45 million, with a peak at around 19:00, and the peak load of the power grid will reach 37.93GW. Taking the data of Beijing, Hangzhou, and Hefei as examples, as the penetration rate of electric vehicles increases, the impact of charging on the peak-to-valley difference of urban loads will become increasingly greater.

[0089] ① Temporal and spatial distribution characteristics of electric vehicle charging demand

[0090] The temporal and spatial distribution characteristics of social vehicle charging are analyzed through historical charging data records. The data taken are randomly selected charging station information under the jurisdiction of State Grid Beijing Electric Power Company, including the geographical location information of these stations, the number of charging piles, charging transaction records, etc. In the data processing stage, records with charging time less than 15 minutes are considered invalid records and are not included in the analysis scope. After appropriate processing and analysis, the following characteristics of electric vehicle charging behavior are obtained.

[0091] Charging time distribution: The transaction records completed at the charging stations under the name of State Grid Beijing Electric Power Company in July 2022 were summarized and the charging time was counted. There are two peaks in the frequency histogram of charging time, one centered on approximately half an hour and the other centered on approximately 10 hours, corresponding to the "fast charging" type and "slow charging" type respectively.

[0092] Distribution of charging start time: Similar to the above method, a statistical analysis of the effective charging records of the above charging stations in July 2022 is made. The charging start time is concentrated around 22:00-24:00. This is partly because the electricity price is also lower at night, and there are also other reasons such as fewer electric vehicle travel activities at night. For these electric vehicles that start charging between 22:00 and 24:00, the pick-up time is generally the next morning, so the charging time is generally longer. The following scenario analysis selects the self-operated electric vehicle charging stations connected to the State Grid Internet of Vehicles platform for analysis.

[0093] ②Analysis of the response characteristics of electric vehicles in different scenarios

[0094] Public charging scenario: The public charging stations mainly serve private cars and taxis / online ride-hailing vehicles. This is related to the large number of private cars, high usage intensity of taxis / online ride-hailing vehicles, and the fact that most of them do not have private charging piles. In terms of charging time, public charging stations form peaks at 8:00-9:00, 11:00-13:00, and 18:00-19:00. The slow charging peaks are at 8:00-9:00 and 18:00-19:00, while the fast charging peaks are at 11:00-13:00. Charging time distribution is an important reference indicator for vehicle-grid interaction. Vehicles that are charging during low-peak electricity consumption and have long parking times are of greater value to vehicle-grid interaction and have high response potential.

[0095] Community charging scenario: From the perspective of service objects, community charging stations are mainly for private cars, and the proportion of taxis / online ride-hailing cars is less than 10%. Community charging piles are mainly private piles, and the owners are mainly private car users. In terms of charging time, charging at community charging stations is mainly concentrated between 17:00 and 24:00. After 17:00, the charging demand rises rapidly and reaches the highest point at 19:00. After 19:00, the charging demand gradually declines, which provides the possibility for vehicle-network interaction during the evening peak period and has a large potential for translational response.

[0096] Highway charging scenario: From the perspective of service objects, the proportion of private cars at highway charging stations is relatively high, followed by taxis / online ride-hailing vehicles. Buses account for 14.7%, which is higher than the proportion in dedicated charging stations, but the total number of high-speed charging stations is small, and the number of buses charged in dedicated charging stations is 2.1 times that of high-speed charging stations. In terms of charging time, high-speed charging stations form multiple peaks at 8-9, 11-13, and 18-19. At the same time, the proportion of use at 1-3 in the morning is also relatively high. High-speed charging stations have charging peaks in multiple time periods, providing multiple possibilities for vehicle-grid interaction.

[0097] 4) Analysis of Response Characteristics of Solar Storage System

[0098] ① Classification and characteristics of photovoltaic storage systems

[0099] The industry generally recognizes that energy storage is a high-quality adjustable resource. There are many studies on adjustable response capabilities, which need not be elaborated. For virtual power plants, energy storage resources can effectively support various response requirements in terms of response duration and response adjustment speed. Generally speaking, large-scale independent energy storage can be directly connected to the grid to participate in the market, and distributed energy storage is more suitable for connecting to virtual power plants to participate in market support system regulation. However, the investment and construction costs of energy storage are relatively high, and the focus of regulating energy storage is to consider the economic feasibility of its investment, construction and operation throughout its life cycle. Therefore, the content of this study focuses on analyzing the economic feasibility of energy storage. At present, key technical indicators such as the number of cycles and unit cost of energy storage batteries initially meet the commercial needs of energy storage. With the rapid advancement of core technologies such as battery bodies and energy management systems, key indicators such as the cycle life, response time, and energy storage efficiency of energy storage systems have rapidly improved. At present, distributed user-side photovoltaic storage systems are mainly based on electrochemical energy storage, and other energy storage methods are still in the pilot demonstration stage or even the initial research stage. Most of them cannot meet the user-side energy storage needs in terms of reliability, service life, manufacturing cost, application capabilities, etc.: Among electrochemical energy storage, lithium-ion batteries and lead-carbon batteries are currently the most widely used user-side energy storage technologies.

[0100] Lithium-ion batteries: Currently, lithium-ion batteries are the most widely used and fastest growing type of electrochemical energy storage in global power storage applications. The cost of lithium-ion batteries is about 0.56 yuan / kWh / time, and their lifespan and cost can basically achieve commercial promotion. Lithium-ion batteries have also been most widely used in China. At present, the cycle number of lithium iron phosphate batteries is 6,000 times and 1.5 yuan / wh. Lithium-ion batteries have poor overcharge / discharge resistance; the use of organic electrolytes has major safety hazards and safety needs to be improved; the battery grouping and protection circuits are complex, and the cost is relatively high compared to traditional batteries such as lead-acid batteries. Compared with electric vehicles, the application environment of customer-side energy storage is better, so there is a lot of room for improvement in the economic efficiency of lithium-ion battery technology used for user-side energy storage.

[0101] Lead-carbon battery: Lead-carbon battery retains the advantages of lead-acid battery, such as safety, reliability, high recycling rate and low cost, but has relatively low cycle life, low power density and low charge and discharge rate. At present, the cycle life of lead-carbon battery is about 3,000 times (50% D0D), the cost is about 1.0 yuan / Wh, and the comprehensive cost per kilowatt-hour is 0.56 yuan-0.6 yuan / kWh. At present, lead-carbon battery is widely used in Jiangsu and other regions of my country, and is suitable for providing 1-4 hours peak load regulation service or power demand side energy management service for peak load of various industrial and commercial users.

[0102] Second-life batteries: The current comprehensive utilization cost of retired power batteries is basically the same as that of new batteries. From the exploration and engineering demonstration of second-life technologies at home and abroad, past research content mainly includes retired battery performance evaluation, sorting and reorganization, energy storage system integration, application demonstration and other technical links.

[0103] With the application of big data analysis technology for state assessment and rapid sorting technology based on characteristic parameters, the sorting efficiency of second-life batteries has been greatly improved, effectively promoting the cost reduction of second-life batteries. At present, the cost of second-life batteries is 40% of that of new batteries, and the number of cycles is one-third of that of new batteries. If the retired batteries are replaced three times according to the use cycle of new batteries, the cost of electricity is about 0.5-0.6 yuan / kWh·time, which is equivalent to using new batteries.

[0104] ②Analysis of Response Characteristics of Solar Storage System

[0105] The application of energy storage is mainly concentrated in peak shaving and valley filling, tracking planned output, frequency regulation and smoothing of wind and solar output. Under my country's current policy system and electricity price mechanism, energy storage may form a better profit model in the two application scenarios of peak shaving and valley filling and tracking planned output.

[0106] The technical and economic analysis model for building an energy storage system to participate in regulation can be established as follows:

[0107]

[0108] In the formula, IRR is the internal rate of return of energy storage, n is the calculation period of the energy storage power station project, CI is the cash inflow of the energy storage project, CO is the cash outflow of the energy storage project, P is the present value, that is, the value of future cash flow discounted to the current time point, F is the future value, that is, the value of cash flow or assets at a certain moment in the future, t is time, and P is the current value of the future cash flow. ts is the static investment payback period of energy storage (years), T is the maximum time, i c is the benchmark rate of return for energy storage projects, P td is the dynamic investment payback period of the energy storage project (years),

[0109] T + It is the cumulative net cash flow of energy storage in each year.

[0110] The energy storage used to coordinate the output regulation of the photovoltaic power station within the virtual power plant is selected for economic analysis. Taking various photovoltaic storage systems as examples, the economic performance is estimated when the cost is reduced by 5%, 10%, 15%, and 20%. The benchmark rate of return is selected as 8% when calculating the technical economic performance of the above photovoltaic storage system, and the energy release price is calculated according to the photovoltaic grid-connected benchmark price of 0.98 yuan / kWh in the resource area. Through economic calculations, it can be obtained that under different degrees of cost reduction, the investment payback period of various energy storage technologies has decreased, and the internal rate of return and net present value have increased.

[0111] (2) Coordinated optimization operation strategy of virtual power plants participating in the auxiliary market taking into account various distributed resources

[0112] Virtual power plants participate in power trading by aggregating distributed resources without the need to transform the power grid, and can provide auxiliary services for the operation of large power grids. The main costs of virtual power plant operators may include: investment and construction costs, operation and maintenance costs, external power purchase costs, internal entity management costs, internal entity sharing costs, etc.; the benefits that can be obtained include: power generation income, external power sales income, internal power sales income, subsidy income, etc. The model of virtual power plants participating in the auxiliary service market, medium- and long-term contract market, and electric energy market.

[0113] 1) Virtual power plants participate in market transactions

[0114] In the near future, virtual power plants will mainly participate in electricity balance regulation market transactions, and can gradually participate in medium- and long-term, spot and other electricity energy markets and auxiliary service markets with higher technical requirements.

[0115] At present, virtual power plants can mainly participate in the balancing and regulating electricity market (market-based demand response, peak-shaving auxiliary service market). The competition between virtual power plants and thermal power units depends on the comparison between the regulation cost of virtual power plants and the peak-shaving cost of thermal power units.

[0116] In the medium and long term, market transactions that virtual power plants can participate in include medium- and long-term transactions, power generation rights (contract transfer) transactions, spot and more complex ancillary service markets.

[0117] In terms of medium- and long-term transactions, compared with other market players, the advantages of virtual power plants in participating in the medium- and long-term market are: first, virtual power plants can form some transactions in advance before participating in medium- and long-term market transactions. When the transaction does not meet the power purchase and sale needs of virtual power plants, the incremental part is traded in the power market; second, through the resource aggregation and response capabilities of virtual power plants, uncertainty is reduced, thereby reducing the risk constraints of virtual power plants in the power market: third, the response capabilities of virtual power plants can make execution deviations smaller: fourth, virtual power plants have more flexibility in arranging output than other market players, and better serve the spot market trading space and spot settlement. The details are shown in Table 5.

[0118] Table 5 Recommendations for virtual power plant power trading products

[0119]

[0120] In terms of power generation rights trading, virtual power plants, as the transferors of power generation rights, purchase electricity from the transferees of power generation rights. Coal-fired power plants are sellers of power generation rights, while virtual power plants are buyers of power generation rights. The subject of the transaction between the buyer and seller is the traded electricity volume. Compared with renewable energy power plants, virtual power plants have low uncertainty and high regulation capabilities, so they have a time advantage in trading. In terms of spot and ancillary service transactions, the advantage of virtual power plants in participating in the electricity spot market lies in their regulatory response performance, which enables virtual power plants to flexibly adjust the declared quantity and price strategies based on expected market prices in a short period of time. On the other hand, compared with power sales companies, virtual power plants have regulatory means, which makes it more certain to formulate electricity packages for customers, and there is more room for profit.

[0121] 2) Virtual power plant pre-revenue assessment

[0122] The method for estimating the benefits of virtual power plants participating in the electricity market is as follows: first, establish the operating cost and output model of each distributed energy source in the virtual power plant, further derive the cost model and benefit model of the aggregated virtual power plant, and then use the cost-benefit method to calculate the maximum benefit of the virtual power plant based on the market electricity price simulation and renewable energy output forecast data. The power generation costs of virtual power plants participating in the electricity energy market include the operating costs of distributed power sources, energy storage equipment, and the response costs of controllable loads. The distributed power sources aggregated by virtual power plants are generally divided into two categories: controllable distributed power sources (such as micro-gas turbines) and uncontrollable distributed power sources (such as wind power and photovoltaics). Among them, the operating costs of uncontrollable distributed power sources are negligible compared to controllable units because they use natural resources such as wind and light to generate electricity, and the marginal cost of distributed renewable energy generation is considered to be zero. The benefits of virtual power plants participating in different types of ancillary service markets vary greatly. The balancing principles and market rules of virtual power plants participating in the standby ancillary service, peak-shaving and frequency-regulating ancillary service markets, as well as the newly opened ancillary service markets such as rotational inertia and ramp climbing in the future, are different. In addition, the balancing requirements and rules of regional power grids and provinces are also different. Virtual power plants need to classify and aggregate various internal resources to meet the access, trading and dispatching requirements of different markets.

[0123] Since there is a coupling mechanism between the participation of virtual power plants in the electricity energy market, ancillary service market and demand response, and no unified national standards and rules have been formed. The rules of the power market are also being updated with the times. For example, with the development of the power spot pilot and the normalized long-term continuous trial operation, the role of the power peak-shaving ancillary service market may be replaced by the spot market; for example, in the power market where power demand response is not normalized and uninterrupted, the total year-end duration of virtual power plants participating in demand response is insufficient, and the total revenue cannot support the construction of virtual power plants. This study does not traverse the transaction mechanisms, rules and business processes of each market that virtual power plants may participate in, but only selects the main transaction varieties for detailed analysis. This study focuses on the study of virtual power plants participating in the power market, and solves the problem of maximizing the benefits of virtual power plants under the premise of given external market conditions, considering the response characteristics of virtual power plants, that is, benchmarking traditional new energy power plants and power sales companies with more timely response and adjustment capabilities.

[0124] 3) Response model of distributed resources

[0125] Virtual power plants aggregate various types of distributed resources, such as distributed photovoltaic, distributed wind power and other distributed new energy units, energy storage equipment and controllable loads. The specific expressions of cost functions, utility functions and output models of various distributed energy sources are as follows.

[0126] ① Controllable distributed generator set: The operating cost of the controllable distributed generator set is mainly the startup cost C z,sn and power generation cost C(P z′ ), the power generation cost can be expressed by a quadratic function, and the specific expression is as follows:

[0127]

[0128] In the formula, C(P g,t ) is the power generation cost of controllable distributed generator sets, a s 、b s and c s are the quadratic coefficient, linear coefficient and constant term of the controllable distributed generation power generation cost, P g,t is the active power of the controllable distributed power source g in time period t;

[0129] ② Distributed new energy units: The cost of distributed renewable power sources is mainly concentrated in the initial investment and construction, that is, the fixed cost. Since it relies on natural resources to generate electricity, the operating cost is negligible compared to conventional units, and the power generation operating cost is considered to be zero.

[0130] C(P q,t )=0

[0131] In the formula, C(P q,t) is the cost of distributed renewable power generation, P q,t is the predicted output of distributed renewable power source q in time period t;

[0132] ③ Energy storage equipment: Energy storage equipment also considers operating costs, and the specific expression is as follows:

[0133] C(P s,t )=π s |P s,t |

[0134] In the formula, C(P s,J ) is the operating cost of energy storage equipment, π s is the unit operating cost coefficient of energy storage equipment, P s,t is the charge and discharge power of the energy storage device s in time period t, a positive value indicates the discharge power of the energy storage device, and a negative value indicates the charge power of the energy storage device;

[0135] ④ Controllable load: Controllable load includes shiftable load, transferable load, interruptible load and flexible adjustable load, etc. Interruptible load includes building air conditioning load, etc. Transferable load includes electric vehicles, etc. Shiftable load includes industrial and commercial load, etc. The utility function of controllable load can be expressed by quadratic function like conventional units. The specific expression is as follows:

[0136]

[0137] In the formula, E(P l,t ) is the controllable load utility, a l 、b l and c l are the quadratic coefficient, linear coefficient and constant term of the controllable load utility function, P l,t is the active power of controllable load l in time period t;

[0138] The specific expression of the control cost function of the controllable load is as follows:

[0139] C(P l,t )=α l P l,t +β l

[0140] In the formula, C(P l,t ) is the control cost of controllable load, α l , β l are the linear coefficient and constant term of the cost of controllable load participating in the electricity market respectively;

[0141] The revenue model of virtual power plants participating in the electricity market is as follows:

[0142]

[0143] In the formula, E ene is the revenue of the virtual power plant participating in the electricity market, T is the maximum time, λ ene,t is the electricity price in the electricity market at time period t, P ene,t is the amount of electricity that the virtual power plant participates in the electricity market in time period t, and Δt is the time period interval;

[0144] The power of the virtual power plant in time period t is P ene,t is the sum of the power of all internal resources. The specific expression is as follows:

[0145]

[0146] Where N g 、N q 、N s 、N l are the total number of controllable distributed generators, distributed new energy generators, energy storage equipment and controllable loads, respectively. s,t,d is the discharge power of energy storage device s in time period t, P s,t,c is the charging power of energy storage device s in time period t;

[0147] The revenue model of virtual power plants participating in the ancillary service peak load regulation market is as follows:

[0148]

[0149] In the formula, E pr is the revenue of the virtual power plant participating in the ancillary service peak load regulation market, λ pr,t is the price of ancillary services provided by the virtual power plant in period t, F pr,t is the auxiliary service peak load of the virtual power plant in time period t, λ res,t is the market price at time period t, P pr,a,t , P pr,t are the actual peak load and bid amount of the virtual power plant in time period t, β is the penalty threshold set by the peak load market, ranging from 0 to l;

[0150] The revenue model of virtual power plants participating in the reserve market is as follows:

[0151]

[0152] R t =R g,t +R s,t +R l,t

[0153] In the formula, E res is the revenue of the virtual power plant participating in the reserve market, R t is the reserve capacity of the virtual power plant in period t, R g,t , Rs,t , R l,t They are the spare capacities that can be provided by the distributed controllable generator set g, energy storage equipment s, and controllable load l respectively.

[0154] When participating in the electricity market, virtual power plants can choose to participate in one or several electricity trading products based on the characteristics of the aggregated internal flexibility resources and response characteristics. The trading product of the electric energy market is electric energy. After self-dispatching, virtual power plants interact with the large power grid to generate electricity and purchase and sell electricity; the trading product of the peak-shaving market is deep peak-shaving services. Virtual power plants can provide deep peak-shaving services by adjusting load power or power generation power. They can adjust the controllable load power consumption plan, arrange the energy storage equipment charging and discharging plan, and appropriately adjust the power generation unit output to achieve deep peak-shaving services. The trading products of the reserve market are upper reserve and lower reserve auxiliary services. When arranging the internal unit power consumption plan, virtual power plants reserve power space to form a reserve. This part of the reserved power can participate in the reserve market.

[0155] 4) Virtual power plants participate in the electricity market

[0156] The electric energy market includes the medium- and long-term electricity market, the day-ahead spot market, the intraday and real-time markets. When a virtual power plant participates in the forward market in the energy market, it first needs to estimate the operating status of the virtual power plant based on the forecast of wind speed and irradiation and historical operating information, and then sign a medium- and long-term transaction contract through bilateral transactions or listing transactions. The bilateral contract must clearly state the amount of electricity traded, the date of delivery, the transaction price, the payment date, the penalty information, etc. During formal operation, the deviation electricity quantity must also be assessed and settled. Signing a bilateral contract can effectively avoid the risk of income brought about by the uncertainty of electricity prices in the electricity spot market. In order to further avoid transaction risks, transactions.

[0157] The form of a difference contract is often adopted. Considering the bilateral open electricity market, in this model, both power generators and electricity users can participate in the electricity market clearing price and can be settled according to the marginal node electricity price. The source-grid-load-storage integrated virtual power plant participates in the electricity spot market by quoting quantity and quotation, but settles according to the unified clearing price of the entire network. Based on the fact that the capacity of virtual power plants is generally not large and generally there are no independent physical nodes, virtual power plants can be considered to participate in the electricity market as price takers, that is, their quotations will not affect the market electricity price. At this time, the virtual power plant only needs to optimize its bidding capacity and actual operation curve according to the predicted electricity market price.

[0158] Before the day-ahead market opens, virtual power plants can sign medium- and long-term contracts with other institutions. The day-ahead market is open every 4 hours one day before the trading day, and virtual power plants can modify their power generation and consumption plans for the next 4-8 hours. Wind power and photovoltaic output are updated and re-forecasted in real time. At this time, the day-ahead market electricity price and bid power are known quantities, and wind power and photovoltaic output forecasts are more accurate. During the real-time control execution phase, the virtual power plant can control the operation of each aggregation unit according to the submitted bid power to complete the electricity transaction. If there is a deviation between the average output of the virtual power plant in the real-time market and the corresponding intraday report, the corresponding deviation needs to be regulated.

[0159] 5) Decision-making model for virtual power plants participating in spot market operations

[0160] Since virtual power plants participate in many types of market transactions, but there are few studies on virtual power plants participating in the spot market. This study takes the spot scenario as an example to model the operation decision model and analyze the case. The driving force for the response of adjustable resources is the direct price difference between the spot market price and the medium- and long-term contracts, which is also the source of the response income of the virtual power plant. Similarly, when participating in other power markets, some models and algorithms of the example can be reused to make operational decisions.

[0161] ① Virtual power plant participation in the day-ahead spot decision-making model

[0162] This study is based on the operation rules of virtual power plants, and takes the "source-grid-load-storage integration" virtual power plant as an example to build a model. In the electricity spot market, the "source-grid-load-storage integration" virtual power plant participates in the market as a combination of power generation and consumption in the form of "reporting quantity and quoting price". Under the condition that the reported power load and the reported power generation load meet certain constraints, it independently decides on the quantity-price curve under the reported load state and the quantity-price curve under the power generation state, as well as the upper limit of power generation operation output and the upper limit of load operation power consumption, and uses this as the basis for clearing the spot market throughout the day. In the scenario of the day-ahead spot market, this model is based on the user's adjustable load, and achieves the role of "peak shaving and valley filling" by configuring a reasonable energy storage system and adjustable load ratio to maximize settlement benefits.

[0163] Virtual power plant operation decision optimization model: Operation decision optimization aims to maximize the profit of virtual power plant operators:

[0164] J2=max(C jiesuan_income +C joint_user_income +C discrele_r_income +C rent_income -C adjust_cost -C storage_construction_cost -C loss_cost -C storage_use_cost -C examine_cost )

[0168] Where J2 is the operator revenue of the virtual power plant, C jiesuan_income For settlement income (in the CFD settlement method, the contract electricity fee expenditure needs to be deducted), C joint_user_income is the aggregated user electricity fee income (for CFD settlement, the contract electricity fee expenditure needs to be deducted), C discrele_r_income is the discrete user income, C rent_income is the energy storage device rental fee (income), C adjust_cost is the load response adjustment cost (response revenue share), C storage_construction_cost The daily average cost of construction and operation and maintenance of energy storage throughout its life cycle, C storage_use_cost is the energy storage call cost, C loss_cost is the energy loss cost of the energy storage device, C examine_cost The cost of calling for energy storage;

[0169] The settlement income formula is:

[0170]

[0171] In the formula, T is the total time, is the power reported at time t (power generation is income, power purchase is expenditure), P riqian,t is the predicted value of the electricity price of the whole network at the tth moment (calculated as the actual electricity price afterwards), Δt is the time interval (the default value is 0.25h), is the total power generated by the generator set at the i-th moment, is the electricity load of all users at the i-th moment, is the battery charging power and discharging power at the i-th moment, and K is the number of users;

[0172] The electricity revenue formula for aggregated users is:

[0173]

[0174] In the formula, is the adjusted power load value of the kth user at the i-th moment, is the total power load of all users after adjustment at time t, is the contract electricity price corresponding to the kth user at the i-th moment;

[0175] The formula for load response regulation cost (response revenue sharing) is:

[0176]

[0177] In the formula, ε k is the share ratio of the kth user, Q′ l L is the adjustment load size at the first moment, is the adjusted load size of the kth user at the lth moment, P user_contract,t is the user’s contract electricity price at time t, P riqian,t is the predicted value of the electricity price of the entire network at the tth moment (calculated as the actual electricity price afterwards);

[0178] The energy storage leasing income formula is:

[0179] C rent_income =P×P storage

[0180] Where P is the rated power of the energy storage device, P storage is the unit power rental of energy storage device (yuan / 1kW·year);

[0181] The formula for the cost of electric energy loss of energy storage device is:

[0182]

[0183] Where η e , η d is the charging efficiency and discharging efficiency of the energy storage device, is the average spot price on the previous day, P transmission is the transmission and distribution price, P government For government funds and additions;

[0184] The formula for the energy storage device call fee is:

[0185]

[0186] In the formula, C one The cost of one cycle of energy storage equipment, C i is the ratio of the charge and discharge amount to the rated capacity within a certain time, S is the rated capacity of the energy storage device;

[0187] Constraints of the operation decision optimization model: First, the output constraints of the winning bid under the power generation state. Then use the following constraints to generate the power generation state The minimum and maximum power generation output declaration constraints are:

[0188] Q0≥Q min

[0189] Q n ≤βQ max

[0190] Where n is the number of output stages under power generation, Q0 is the lower limit of the first stage output under power generation, Q n is the upper limit of the last output in the power generation state, β is the power generation ratio coefficient (the default is 50%), Q minThe minimum output size in the power generation state (the default is 0MW), Q max is the total installed capacity;

[0191] The output interval constraints before and after the declaration are:

[0192]

[0193] In the formula, α is the proportional coefficient in the power generation state (the default value is 0.5), Q i is the upper limit of the output of the i-th stage under the power generation state (also the lower limit of the output of the i+1th stage), Q i+1 is the upper limit of the output of the i+1th stage under the power generation state (and the lower limit of the output of the i+2th stage); secondly, the load constraint of the winning bid under the load state. Then use the following constraints to generate the load state The minimum and maximum power load declaration constraints are:

[0194] Q0′≥Q′ min

[0195] Q n ′≤β′Q′ max

[0196] In the formula, Q′0 is the lower limit of the first section power load under load state, Q′ n is the upper limit of the last section of power load under load state, β′ is the load ratio coefficient (the default is 50%), Q′ min is the minimum power load (the default is 0MW), Q′ max The maximum value of the power load (the extreme value of the user load added together in different time periods);

[0197] The output interval constraints before and after the declaration are:

[0198]

[0199] In the formula, α′ is the proportional coefficient under load (the default value is 0.5), Q′ i is the upper limit of the power load of the i-th section under load state (also the lower limit of the power load of the i+1th section), Q′ i+1 The upper limit of the power load of the i+1th section under load state (also the lower limit of the power load of the i+2th section);

[0200] Then there is the adjustable load response capability constraint:

[0201]

[0202] In the formula, α k,l is the adjustment ratio of the kth user at the lth moment, is the maximum ratio that the k-th user can increase at the t+i-th time, is the maximum ratio that the k-th user can reduce at the t+i time, Whether the load is increased at time t, a value of 1 indicates yes, and a value of 0 indicates no. is whether the load is lowered at time t, with a value of 1 for yes and 0 for no. i is the number of future time points, ranging from 0 to 7, indicating that the adjustment direction in the next two hours is consistent. k,t Adjust the scale for the kth user at time t, Q k,t+i is the power load value of the kth user before adjustment at time t+i, is the adjusted power load value of the kth user at time t+i;

[0203] Finally, there are energy storage device constraints, including capacity constraints, charging and discharging power constraints, power constraints, rated capacity and rated power constraints.

[0204] The capacity constraint of the energy storage device is:

[0205] E min ≤E t ≤E max

[0206] In the formula, E min is the minimum energy storage capacity of the energy storage device (temporarily set as 0.05 times the rated capacity), E max is the maximum energy storage capacity of the energy storage device (tentatively set as 0.95 times the rated capacity), E t is the current power of the energy storage device in time period t;

[0207] The charge and discharge power constraints of the energy storage device are:

[0208]

[0209] In the formula, is the charging state and discharging state of the energy storage device in time period i;

[0210] Energy storage device power constraints:

[0211]

[0212] In the formula, E t is the current power of the energy storage device in time period t, η e , η d is the charging efficiency and discharging efficiency of the energy storage device (η e ×η d =0.83);

[0213] Rated capacity and rated power constraints of energy storage devices:

[0214]

[0215] In the formula, is the ratio of the rated capacity to the rated power of the energy storage device (tentatively set at 2, adjustable);

[0216] ②Model solving algorithm and solver selection

[0217] Through analysis, most of the variables in the model are restricted to integers, and the optimal solution of the objective function is transformed into a typical mixed integer programming problem.

[0218] Commonly used integer programming problem solutions include branch and bound method (can be used to solve pure or mixed integer linear programming), cutting plane method (can be used to solve pure or mixed integer linear programming), implicit method: used to solve 0-1 integer programming, there are filtering method and branching method, Hungarian method (solve assignment problems), Monte Carlo method (solve various types of programming), specifically:

[0219] The branch and bound method is a typical mathematical programming or search algorithm that finds the optimal solution by decomposing the problem and applying linear programming to each sub-problem. The algorithm decomposes the problem into a tree, selects an integer variable to branch each time, and then uses linear programming to solve the remaining problem.

[0220] The pruning method is an improvement on the branch and bound method. It reduces the number of subproblems through appropriate pruning strategies, thereby effectively reducing the computational time.

[0221] The mixed integer linear programming algorithm is a typical computer science algorithm. The algorithm first uses linear programming to solve the original problem, and then rounds the solution of the linear programming to the nearest integer value to obtain an integer solution. This method is more efficient than the traditional branch and bound method and pruning method, but it requires a computer program to solve. This study uses a computer software solver for solving. Integer constraint programming solvers are divided into commercial solvers and open source solvers. Commonly used commercial solvers include Gurobi, COPT, Matlab, etc. Commonly used open source solvers include SCIP, GLPK, LPSOLVE, etc. If the purpose is commercial, you must purchase a commercial version license. The effect of commercial solvers is better than that of open source solvers. Considering that the model parameters and data involved in this study are not complicated, the open source solver SCIP (Solving Constraint Integer Programs) is used for solution.

[0222] The integer constraint programming solver SCIP has a mature mixed integer programming (MILP) framework and supports various modules in the custom search tree, including the function of adding variables in the branch and bound process. The SCIP core library is written in C and has corresponding C++ wrapper functions. SCIP is a relatively complete framework, and it is more difficult to get started compared to commercial software such as CPLEX and GUROBI.

[0223] SCIP is named as Integer Constraint Programming Solver because it combines the techniques of Integer Linear Programming and Constraint Programming. SCIP has been maintained by many developers and has rich documentation and functions.

[0224] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 3 As shown. The computer device includes a processor, a memory and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store static information and dynamic information data. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the steps in the above method embodiment are implemented.

[0225] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0226] In addition, the present invention also provides a computer device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiment when executing the computer program.

[0227] In addition, the present invention further provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps in the above method embodiment are implemented.

[0228] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided by the present invention can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0229] The present invention is not limited to the structures which have been described above and shown in the drawings, and various modifications and changes may be made without departing from the scope thereof. The scope of the present invention is limited only by the appended claims.

Claims

1. A method for collaborative optimization of auxiliary services of a photovoltaic storage system, characterized in that: include: Based on resource response characteristics, response capabilities and economic efficiency, the multi-dimensional response characteristics of adjustable resources of various types of energy storage virtual power plants are analyzed; According to the multi-dimensional response characteristics of the adjustable resources of the virtual power plant, the response model of each distributed resource in the virtual power plant is constructed; With the goal of maximizing the benefits of virtual power plants, a virtual power plant operation decision optimization model is constructed, and the open source solver is used to solve it to obtain collaborative optimization results.

2. The method for collaborative optimization of auxiliary services of a photovoltaic storage system according to claim 1, characterized in that: The analysis of the multi-dimensional response characteristics of adjustable resources of various types of energy storage virtual power plants based on resource response characteristics, response capabilities and economy includes: Analyze the load response characteristics of industrial and commercial power users based on industry production characteristics, production cycle, load energy consumption ratio, compressible capacity and response speed, and provide safety regulation suggestions; Based on historical charging data records, analyze the spatiotemporal distribution characteristics of electric vehicle charging demand and the charging scenario response characteristics of electric vehicles; Construct an economic analysis model for energy storage system participation in regulation technology, and use the economic analysis model to analyze the response characteristics of the photovoltaic storage system.

3. The method for collaborative optimization of auxiliary services of a photovoltaic storage system according to claim 2, characterized in that: The industrial and commercial electricity users include industrial and commercial electricity users in high-energy-consuming industries, industrial and commercial electricity users in continuous production industries, industrial and commercial electricity users in non-continuous production industries, and industrial and commercial electricity users in other industries.

4. The method for collaborative optimization of auxiliary services of a photovoltaic storage system according to claim 2, characterized in that: The spatiotemporal distribution characteristics include charging duration distribution characteristics and charging start time distribution characteristics, and the charging scenarios include public charging scenarios, community charging scenarios, and highway charging scenarios.

5. The method for collaborative optimization of auxiliary services of a photovoltaic storage system according to claim 2, characterized in that: The economic analysis model is expressed as: In the formula, IRR is the internal rate of return of energy storage, n is the calculation period of the energy storage power station project, CI is the cash inflow of the energy storage project, CO is the cash outflow of the energy storage project, P is the present value, F is the future value, t is the time, P ts is the static investment payback period of energy storage, T is the maximum time, i c is the benchmark rate of return for energy storage projects, P td is the dynamic investment payback period of the energy storage project, T + It is the cumulative net cash flow of energy storage in each year.

6. The method for collaborative optimization of auxiliary services of a photovoltaic storage system according to claim 1, characterized in that: The response model of distributed resources includes: The expression of the power generation cost of controllable distributed generator sets is: The expression for the cost of distributed renewable power generation is: C(P q,t )=0 The expression of the operating cost of energy storage equipment is: C(P s,t )=π s |P s,t | The expression of the controllable load utility function is: The expression of the control cost function of the controllable load is: C(P l,t )=a l P l,t +b l The expression of the revenue model of virtual power plant participating in the electric energy market is: The expression of the revenue model of virtual power plants participating in the ancillary service peak load regulation market is: The expression of the revenue model of virtual power plant participating in the reserve market is: R t =R g,t +R s,t +R l,t In the formula, C(P g,t ) is the power generation cost of the controllable distributed generator set, as, bs and cs are the quadratic coefficient, linear coefficient and constant term of the power generation cost of the controllable distributed power source, respectively. g,t is the active power of the controllable distributed generation g in time period t, C(P q,t ) is the cost of distributed renewable power, P q,t is the predicted output of distributed renewable energy source q in time period t, C(P s,J ) is the operating cost of energy storage equipment, π s is the unit operating cost coefficient of energy storage equipment, P s,t is the charging and discharging power of energy storage device s in time period t, E(P l,t ) is the controllable load utility, a l , b l and c l are the quadratic coefficient, linear coefficient and constant term of the controllable load utility function, P l,t is the active power of the controllable load l in time period t, C(P l,t ) is the control cost of controllable load, α l , β l are the linear coefficient and constant term of the cost of controllable load participating in the electricity market, E ene is the revenue of the virtual power plant participating in the electric energy market, T is the maximum time, λ ene,t is the electricity price in the electricity market at time period t, P ene,t is the amount of electricity that the virtual power plant participates in the electricity market during period t, Δt is the period interval, N g 、N q 、N s 、N l are the total number of controllable distributed generators, distributed new energy generators, energy storage equipment and controllable loads, respectively. s,t,d is the discharge power of energy storage device s in time period t, P s,t,c is the charging power of energy storage device s in time period t, E pr is the revenue of the virtual power plant participating in the ancillary service peak load regulation market, λ pr,t is the price of ancillary services provided by the virtual power plant in period t, F pr,t is the auxiliary service peak load of the virtual power plant in time period t, λ res,t is the market price at time period t, P pr,a,t , P pr,t are the actual peak load and bid volume of the virtual power plant in time period t, β is the penalty threshold set by the peak load market, and E res is the revenue of the virtual power plant participating in the reserve market, R t is the reserve capacity of the virtual power plant in period t, R g,t , R s,t , R l,t They are the spare capacities that can be provided by the distributed controllable generator set g, energy storage equipment s, and controllable load l respectively.

7. The method for collaborative optimization of auxiliary services of a photovoltaic storage system according to claim 1, characterized in that: The expression of the virtual power plant operation decision optimization model is: J2=max(C jiesuan_income +C joint_user_income +C discrele_r_income +C rent_income -C adjust_cost -C storage_construction_cost -C loss_cost -C storage_use_cost -C examine_cost ) C rent_income =P×P storage The constraints of the virtual power plant operation decision optimization model include: when When , the minimum and maximum power generation output declaration constraints are: Q0≥Q min Q n ≤βQ max The output interval constraints before and after the declaration are: when When , the minimum and maximum power generation output declaration constraints are: Q0′≥Q′ min Q n ′≤β′Q′ max The output interval constraints before and after the declaration are: Adjustable load response capability constraints: The capacity constraint of the energy storage device is: AND min ≤E t ≤E max The charge and discharge power constraints of the energy storage device are: The power constraint of the energy storage device is: The rated capacity and rated power constraints of the energy storage device are: Where J2 is the operator revenue of the virtual power plant, C jiesuan_income is the settlement income, C joint_user_income is the aggregated user electricity revenue, C discrele_r_income is the discrete user income, C rent_income is the energy storage device rental fee, C adjust_cost is the load response adjustment cost, C storage_construction_cost The daily average cost of construction and operation and maintenance of energy storage throughout its life cycle, C storage_use_cost is the energy storage call cost, C loss_cost is the energy loss cost of the energy storage device, C examine_cost is the energy storage call cost, T is the total time, is the power reported at time t, P riqian,t is the predicted value of the day-ahead electricity price of the entire network at time t, Δt is the time interval, is the total power generated by the generator set at the i-th moment, is the electricity load of all users at the i-th moment, is the battery charging power and discharging power at the i-th moment, K is the number of users, is the adjusted power load value of the kth user at the i-th moment, is the total power load of all users after adjustment at time t, is the contract electricity price corresponding to the kth user at the i-th moment, ε k is the share ratio of the kth user, is the adjustment load size at the first moment, is the adjusted load size of the kth user at the lth moment, P user_contract,t is the user’s contract electricity price at time t, P riqian,t is the predicted value of the day-ahead electricity price of the entire network at time t, P is the rated power of the energy storage device, and P storage is the unit power rental of the energy storage device, η e , η d is the charging efficiency and discharging efficiency of the energy storage device, is the average spot price on the previous day, P transmission is the transmission and distribution price, P government For government funds and additional one The cost of one cycle of energy storage equipment, C i is the ratio of the charge and discharge amount to the rated capacity within a certain time, S is the rated capacity of the energy storage device, n is the number of output segments under the power generation state, Q0 is the lower limit of the output of the first segment under the power generation state, Q n is the upper limit of the last stage of power generation, β is the power generation ratio coefficient, Q min is the minimum output size under power generation state, Q max is the total installed capacity, α is the proportional coefficient under power generation, Q i is the upper limit of the output of the i-th stage under the power generation state, Q i+1 is the upper limit of the output of the i+1th stage under the power generation state, Q′0 is the lower limit of the power load of the first stage under the load state, and Q′ n is the upper limit of the last section of power load under load state, β′ is the load proportion coefficient, Q′ min is the minimum value of power load, Q′ max is the maximum value of power load, α′ is the proportional coefficient under load condition, Q i ′ is the upper limit of the power load in the i-th section under load state, Q i ' +1 is the upper limit of the power load of the i+1th section under load state, α k,l is the adjustment ratio of the kth user at the lth moment, is the maximum ratio that the k-th user can increase at the t+i-th time, is the maximum ratio that the k-th user can reduce at the t+i time, is whether the load is increased at time t, is whether the load is reduced at time t, i is the number of future moments, α k,t Adjust the scale for the kth user at time t, Q k,t+i is the power load value of the kth user before adjustment at time t+i, is the adjusted power load value of the kth user at time t+i, E min is the minimum energy storage capacity of the energy storage device, E max is the maximum energy storage capacity of the energy storage device, E t is the current power of the energy storage device in time period t, is the charging state and discharging state of the energy storage device in time period i, It is the ratio of the rated capacity to the rated power of the energy storage device.

8. A photovoltaic storage system auxiliary service collaborative optimization system, characterized in that: include: The response characteristics analysis module is used to analyze the multi-dimensional response characteristics of adjustable resources of various types of energy storage virtual power plants based on resource response characteristics, response capabilities and economic efficiency; A response model building module is used to build a response model of each distributed resource in the virtual power plant according to the multi-dimensional response characteristics of the adjustable resources of the virtual power plant; The optimization model solving module is used to build a virtual power plant operation decision optimization model with the goal of maximizing the virtual power plant's revenue, and use an open source solver to solve it and obtain collaborative optimization results.

9. The photovoltaic storage system auxiliary service collaborative optimization system according to claim 8, characterized in that: The response characteristic analysis module analyzes the multi-dimensional response characteristics of adjustable resources of various types of energy storage virtual power plants based on resource response characteristics, response capabilities and economy. It also analyzes the load response characteristics of industrial and commercial power users based on industry production characteristics, production cycles, load energy usage ratios, compressible capacity and response speed, and gives safety regulation suggestions; based on historical charging data records, it analyzes the spatiotemporal distribution characteristics of electric vehicle charging demand and the charging scenario response characteristics of electric vehicles; Construct an economic analysis model for energy storage system participation in regulation technology, and use the economic analysis model to analyze the response characteristics of the photovoltaic storage system.

10. The photovoltaic storage system auxiliary service collaborative optimization system according to claim 9, characterized in that: The industrial and commercial electricity users include industrial and commercial electricity users in high-energy-consuming industries, industrial and commercial electricity users in continuous production industries, industrial and commercial electricity users in non-continuous production industries, and industrial and commercial electricity users in other industries.

11. The photovoltaic storage system auxiliary service collaborative optimization system according to claim 9, characterized in that: The spatiotemporal distribution characteristics include charging duration distribution characteristics and charging start time distribution characteristics, and the charging scenarios include public charging scenarios, community charging scenarios, and highway charging scenarios.

12. The photovoltaic storage system auxiliary service collaborative optimization system according to claim 8, characterized in that: The economic analysis model is expressed as: In the formula, IRR is the internal rate of return of energy storage, n is the calculation period of the energy storage power station project, CI is the cash inflow of the energy storage project, CO is the cash outflow of the energy storage project, P is the present value, F is the future value, t is the time, P ts is the static investment payback period of energy storage, T is the maximum time, i c is the benchmark rate of return for energy storage projects, P td is the dynamic investment payback period of the energy storage project, T + It is the cumulative net cash flow of energy storage in each year.

13. The photovoltaic storage system auxiliary service collaborative optimization system according to claim 8, characterized in that: The response model of distributed resources includes: The expression of the power generation cost of controllable distributed generator sets is: The expression for the cost of distributed renewable power generation is: C(P q,t )=0 The expression of the operating cost of energy storage equipment is: C(P s,t )=π s |P s,t | The expression of the controllable load utility function is: The expression of the control cost function of the controllable load is: C(P l,t )=a l P l,t +b l The expression of the revenue model of virtual power plant participating in the electric energy market is: The expression of the revenue model of virtual power plants participating in the ancillary service peak load regulation market is: The expression of the revenue model of virtual power plant participating in the reserve market is: R t =R g,t +R s,t +R l,t In the formula, C(P g,t ) is the power generation cost of the controllable distributed generator set, as, bs and cs are the quadratic coefficient, linear coefficient and constant term of the power generation cost of the controllable distributed power source, respectively. g,t is the active power of the controllable distributed generation g in time period t, C(P q,t ) is the cost of distributed renewable power, P q,t is the predicted output of distributed renewable energy source q in time period t, C(P s,J ) is the operating cost of energy storage equipment, π s is the unit operating cost coefficient of energy storage equipment, P s,t is the charging and discharging power of energy storage device s in time period t, E(P l,t ) is the controllable load utility, a l , b l and c l are the quadratic coefficient, linear coefficient and constant term of the controllable load utility function, P l,t is the active power of the controllable load l in time period t, C(P l,t ) is the control cost of controllable load, α l , β l are the linear coefficient and constant term of the cost of controllable load participating in the electricity market, E ene is the revenue of the virtual power plant participating in the electric energy market, T is the maximum time, λ ene,t is the electricity price in the electricity market at time period t, P ene,t is the amount of electricity that the virtual power plant participates in the electricity market during period t, Δt is the period interval, N g 、N q 、N s 、N l are the total number of controllable distributed generators, distributed new energy generators, energy storage equipment and controllable loads, respectively. s,t,d is the discharge power of energy storage device s in time period t, P s,t,c is the charging power of energy storage device s in time period t, E pr is the revenue of the virtual power plant participating in the ancillary service peak load regulation market, λ pr,t is the price of ancillary services provided by the virtual power plant in period t, F pr,t is the auxiliary service peak load of the virtual power plant in time period t, λ res,t is the market price at time period t, P pr,a,t , P pr,t are the actual peak load and bid volume of the virtual power plant in time period t, β is the penalty threshold set by the peak load market, and E res is the revenue of the virtual power plant participating in the reserve market, R t is the reserve capacity of the virtual power plant in period t, R g,t , R s,t , R l,t They are the spare capacities that can be provided by the distributed controllable generator set g, energy storage equipment s, and controllable load l respectively.

14. The photovoltaic storage system auxiliary service collaborative optimization system according to claim 8, characterized in that: The expression of the virtual power plant operation decision optimization model is: J2=max(C jiesuan_income +C joint_user_income +C discrete_user_income +C rent_income -C adjust_cost -C storage_construction_cost -C loss_cost -C storage_use_cost -C examine_cost ) C rent_ncome =P×P storage The constraints of the virtual power plant operation decision optimization model include: when When , the minimum and maximum power generation output declaration constraints are: Q0≥Q min Q n ≤βQ max The output interval constraints before and after the declaration are: when When , the minimum and maximum power generation output declaration constraints are: Q0′≥Q′ min Q n ′≤β′Q′ max The output interval constraints before and after the declaration are: Adjustable load response capability constraints: The capacity constraint of the energy storage device is: AND min ≤E t ≤E max The charge and discharge power constraints of the energy storage device are: The power constraint of the energy storage device is: The rated capacity and rated power constraints of the energy storage device are: Where J2 is the operator revenue of the virtual power plant, C jiesuan_income is the settlement income, C joint_user_income is the aggregated user electricity revenue, C discrete_user_income is the discrete user income, C rent_income is the energy storage device rental fee, C adjust_cost is the load response adjustment cost, C storage_construction_cost The daily average cost of construction and operation and maintenance of energy storage throughout its life cycle, C storage_use_cost is the energy storage call cost, C loss_cost is the energy loss cost of the energy storage device, C examine_cost is the energy storage call cost, T is the total time, is the power reported at time t, P riqian,t is the predicted value of the day-ahead electricity price of the entire network at time t, Δt is the time interval, is the total power generated by the generator set at the i-th moment, is the electricity load of all users at the i-th moment, is the battery charging power and discharging power at the i-th moment, K is the number of users, is the adjusted power load value of the kth user at the i-th moment, is the total power load of all users after adjustment at time t, is the contract electricity price corresponding to the kth user at the i-th moment, ε k is the share ratio of the kth user, is the adjustment load size at the first moment, is the adjusted load size of the kth user at the lth moment, P user_contract,t is the user’s contract electricity price at time t, P riqian,t is the predicted value of the day-ahead electricity price of the entire network at time t, P is the rated power of the energy storage device, and P storage is the unit power rental of the energy storage device, η e , η d is the charging efficiency and discharging efficiency of the energy storage device, is the average spot price on the previous day, P transmission is the transmission and distribution price, P government For government funds and additional one The cost of one cycle of energy storage equipment, C i is the ratio of the charge and discharge amount to the rated capacity within a certain time, S is the rated capacity of the energy storage device, n is the number of output segments under the power generation state, Q0 is the lower limit of the output of the first segment under the power generation state, Q n is the upper limit of the last stage of power generation, β is the power generation ratio coefficient, Q min is the minimum output size under power generation state, Q max is the total installed capacity, α is the proportional coefficient under power generation, Q i is the upper limit of the output of the i-th stage under the power generation state, Q i+1 is the upper limit of the output of the i+1th stage under the power generation state, Q′0 is the lower limit of the power load of the first stage under the load state, and Q′ n is the upper limit of the last section of power load under load state, β′ is the load proportion coefficient, Q′ min is the minimum value of power load, Q′ max is the maximum value of the power load, α′ is the proportional coefficient under load conditions, Q′ i is the upper limit of the power load in the i-th section under load state, Q′ i+1 is the upper limit of the power load of the i+1th section under load state, α k,l is the adjustment ratio of the kth user at the lth moment, is the maximum ratio that the k-th user can increase at the t+i-th time, is the maximum ratio that the k-th user can reduce at the t+i time, is whether the load is increased at time t, is whether the load is reduced at time t, i is the number of future moments, α k,t Adjust the scale for the kth user at time t, Q k,t+i is the power load value of the kth user before adjustment at time t+i, is the adjusted power load value of the kth user at time t+i, E min is the minimum energy storage capacity of the energy storage device, E max is the maximum energy storage capacity of the energy storage device, E t is the current power of the energy storage device in time period t, is the charging state and discharging state of the energy storage device in time period i, It is the ratio of the rated capacity to the rated power of the energy storage device.

15. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

16. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images