Electricity-carbon responsibility allocation method based on matching of power generation capacity and user side load

Abstract

The invention is suitable for the technical field of carbon emission, and provides a power-carbon responsibility allocation method based on matching of power generation capacity and user side load, which comprises the following steps: collecting weather information, a thermal power generating unit, and power generation side operation conditions of a wind power and photovoltaic new energy station, and analyzing the power generation capacity of a regional power system in the next N days according to the weather information; collecting power utilization behavior data of different power distribution and utilization clients, and performing load prediction on a user side based on a particle swarm improved LSTM neural network to obtain a load prediction result; based on the power generation capacity and a load prediction result, a multi-target scheduling strategy optimization method is used to analyze a carbon reduction benefit and economic cost relationship between power generation side output peak regulation and user side load response; and determining an electricity-carbon responsibility allocation mechanism based on the relationship between the carbon reduction benefit and the economic cost. According to the method, an electricity-carbon responsibility allocation mechanism is designed based on the carbon reduction cost of each party, the power generation side output peak regulation and the power utilization side load response are excited, and the balance of carbon emission reduction and cost is realized.

Description

Technical Field

[0001] The present invention relates to the field of carbon emission technology, and in particular to an electricity-carbon responsibility sharing method based on matching power generation capacity with user-side load. Background Art

[0002] Carbon accounting is the current method of quantifying carbon emissions data. Especially in the power industry, which accounts for about 42% of the energy industry's carbon emissions, the measurement of direct and indirect carbon emissions from electricity is particularly important. Although thermal power units are the main source of carbon emissions in the power system, electricity consumption on the load side can bring different carbon emissions, and therefore become the responsible party for direct carbon emissions; in the construction of low-carbon electricity, it is urgent to explore the method of allocating carbon responsibilities among all parties in the power system, and to stimulate the motivation for emission reduction by adjusting the size of the responsibilities of each participant.

[0003] The development of the carbon emission flow theory of power systems provides a new perspective for carbon responsibility accounting of power systems. This theory attaches the carbon emission flow implied in the power use process to the virtual network flow of the power flow until it finally reaches the load side. Through the carbon flow matrix algorithm, the node carbon potential vector, branch carbon flow rate distribution matrix, etc. can be calculated, providing a static carbon emission responsibility accounting model for power grids for carbon emission accounting and carbon responsibility sharing of electricity customers.

[0004] The carbon responsibility allocation of the power system based on the carbon emission flow theory only calculates the implicit carbon emissions of each entity, ignoring the power generation capacity and load response potential of power market participants, making it difficult to stimulate source-load coordination, thereby restricting the consumption of new energy and the overall carbon emission reduction effect of the system. Therefore, it is necessary to provide an electricity-carbon responsibility allocation method based on the matching of power generation capacity and user-side load to solve the above problems. Summary of the invention

[0005] In view of the deficiencies in the prior art, the purpose of the present invention is to provide an electricity-carbon responsibility sharing method based on matching power generation capacity with user-side load, so as to solve the problems existing in the above-mentioned background technology.

[0006] The present invention is implemented as follows: a method for allocating electricity-carbon responsibility based on matching power generation capacity with user-side load, the method comprising the following steps:

[0007] Collect weather information, operating conditions of thermal power units, wind power and photovoltaic new energy stations, and analyze the power generation capacity of the regional power system in the next N days based on weather information, where N is a fixed value;

[0008] Collect electricity consumption behavior data of different power distribution customers, perform user-side load forecasting based on LSTM neural network improved by particle swarm, and obtain load forecasting results;

[0009] Based on the power generation capacity and load forecast results, a multi-objective dispatch strategy optimization method is used to analyze the relationship between the carbon reduction benefits and economic costs of peak load regulation on the power generation side and load response on the user side.

[0010] Determine the electricity-carbon responsibility sharing mechanism based on the relationship between carbon reduction benefits and economic costs.

[0011] As a further solution of the present invention: before analyzing the power generation capacity of the regional power system in the next N days according to weather information, a power generation capacity prediction model for the power generation side is constructed, and the power generation capacity prediction model for the power generation side includes a photovoltaic new energy site model, a wind new energy site model and a thermal power unit model.

[0012] As a further solution of the present invention: the photovoltaic new energy site model is: P1=S×C×η, wherein P1 is the photovoltaic power generation power, S is the light intensity, C is the photovoltaic installed capacity of the new energy site, and η is the power generation efficiency of the entire system.

[0013] As a further solution of the present invention: the wind power new energy station model is: Where P2 is the output power of the wind turbine at wind speed v, ρ is the air density, A is the swept area of ​​the wind rotor, v is the wind speed, and C p is the power factor, and k is the power generation efficiency of the wind turbine.

[0014] As a further solution of the present invention: the thermal power unit model is: P3 = P rated ×F Demand ×B×f, where P3 is the actual power generation of the thermal power unit, P rated is the rated power of the thermal power unit, F Demand is the grid load factor, B is the equipment availability factor, and f is the fossil fuel combustion and conversion efficiency of the thermal power unit.

[0015] As a further solution of the present invention: the step of performing load forecasting on the user side based on the LSTM neural network improved by the particle swarm to obtain the load forecasting result specifically includes:

[0016] Collect electricity load data, weather and climate data, and historical data sets of electricity consumption behavior of different power distribution and utilities customers, identify and correct distorted data, and use Lagrange interpolation method to process and generate missing data;

[0017] Conduct PSO-LSTM neural network model training, initialize the initial position and speed of each particle in the particle swarm, and set the value range of the optimization parameter; import the divided training set data and parameter values ​​into the PSO-LSTM neural network model for training and prediction, and use the mean absolute percentage error as the formula for calculating the fitness value of the prediction result; obtain the individual extreme value and global extreme value of the population, and update the speed and position of the next generation of particles; import the current best parameter value into the PSO-LSTM neural network model again to calculate the new fitting error; after multiple cycles, until the termination condition is met, import the optimal parameter value into the PSO-LSTM neural network model, predict the load data and output the result;

[0018] The user-side load forecasting is performed based on the trained and optimized PSO-LSTM neural network model.

[0019] As a further solution of the present invention: the step of analyzing the relationship between the carbon reduction benefits and economic costs of the peak load regulation on the power generation side and the load response on the user side specifically includes:

[0020] Determine power balance constraints, wind and photovoltaic power generation constraints, thermal power constraints, virtual power plant load response capability constraints, and capacity constraints during virtual power plant load response;

[0021] Based on constraint information and linear programming optimization algorithm, the relationship between carbon reduction benefits and economic costs is determined.

[0022] As a further solution of the present invention: the step of determining the electricity-carbon responsibility sharing mechanism based on the relationship between carbon reduction benefits and economic costs specifically includes: constructing a step-type demand response compensation cost model based on transferable loads and curtailable loads, wherein the demand response compensation cost is proportional to the user's demand response amount; wherein the transferable load is expressed as follows:

[0023]

[0024] In the formula, is the load that can be transferred during period t; are the loads that can be transferred in and out during period t respectively; are state variables, Indicates that the load is transferred in during period t, Indicates that the load is transferred out during period t; are the upper limits of load that can be transferred in and out during period t respectively;

[0025] The load that can be reduced is expressed as follows: Where: The load that can be reduced during period t; is the state variable, It means that there is load reduction during period t; is the upper limit of load reduction in segment t;

[0026] The total load participating in demand response in a dispatch cycle It can be expressed as:

[0027]

[0028] Where: is the total load participating in demand response during period t;

[0029] Based on the total power of demand response, a step-by-step demand response compensation cost model is established:

[0030]

[0031] In the formula, C demand is the total cost of step-type demand response compensation; χ is the demand response compensation base price; δ is the compensation price growth rate; H is the interval length of the response load.

[0032] Compared with the prior art, the present invention has the following beneficial effects:

[0033] By collecting the electricity consumption behavior data of different power distribution and power consumption customers, the load forecasting on the user side is carried out based on the LSTM neural network improved by the particle swarm, the load forecasting results are obtained, the carbon reduction benefits and economic cost relationship between the peak load regulation on the power generation side and the load response on the user side are analyzed, and the electricity-carbon responsibility sharing mechanism is determined. Based on the carbon reduction costs of all parties, the present invention designs an electricity-carbon responsibility sharing mechanism, thereby stimulating the peak load regulation on the power generation side and the load response on the power consumption side, promoting the maximization of new energy consumption through the optimal matching of the power system source and load, and achieving the trade-off between the overall carbon emission reduction and cost of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 The figure is a flow chart of a method for allocating electricity-carbon responsibility based on matching power generation capacity with user-side load.

[0035] Figure 2 This is a schematic diagram of the prediction results of the regional power system's power generation capacity.

[0036] Figure 3 This is a schematic diagram of the prediction results of power generation and load matching in the regional power system. DETAILED DESCRIPTION

[0037] In order to make the purpose, technical solution and advantages of the present invention clearer, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0038] The specific implementation of the present invention is described in detail below in conjunction with specific embodiments.

[0039] like Figure 1 As shown, an embodiment of the present invention provides an electricity-carbon responsibility allocation method based on matching power generation capacity with user-side load, and the method includes the following steps:

[0040] S100, collects weather information, operating conditions of thermal power units, wind power and photovoltaic new energy stations, and analyzes the power generation capacity of the regional power system in the next N days based on the weather information, where N is a fixed value;

[0041] S200, collecting electricity consumption behavior data of different power distribution and power consumption customers, and performing user-side load forecasting based on the LSTM neural network improved by particle swarm, to obtain load forecasting results;

[0042] S300, based on the power generation capacity and load forecast results, uses a multi-objective scheduling strategy optimization method to analyze the carbon reduction benefits and economic cost relationship between power generation side output peak regulation and user side load response;

[0043] S400, determine the electricity-carbon responsibility sharing mechanism based on the relationship between carbon reduction benefits and economic costs.

[0044] In an embodiment of the present invention, before analyzing the power generation capacity of the regional power system for the next N days based on weather information, preferably N is 7, it is necessary to construct a power generation capacity prediction model on the power generation side, and the power generation capacity prediction model on the power generation side includes a photovoltaic new energy station model, a wind power new energy station model and a thermal power unit model. The photovoltaic new energy station model is: P1=S×C×η, where P1 is the photovoltaic power generation power, S is the light intensity, C is the photovoltaic installed capacity of the new energy station, and η is the power generation efficiency of the whole system, which depends on operating conditions, temperature and other factors. The wind power new energy station model is: Where P2 is the output power of the wind turbine at wind speed v, ρ is the air density, A is the swept area of ​​the wind rotor, v is the wind speed, and C p is the power coefficient, k is the power generation efficiency of the wind turbine; further consider the power curve of the wind power station and wind speed, including the starting wind speed, rated wind speed, shutdown wind speed, etc., to more accurately evaluate the power generation capacity of the wind power station. The thermal power unit model is: P3 = P rated ×F Demand ×B×f, where P3 is the actual power generation of the thermal power unit, P rated is the rated power of the thermal power unit, F Demandis the grid load factor, which depends on the actual power load and security requirements of the power system dispatch, B is the equipment availability coefficient, and f is the fossil fuel combustion and conversion efficiency of the thermal power unit. The operating conditions of the power generation side such as thermal power units, wind power and photovoltaic new energy stations are collected, and the power generation capacity of all power plants in the regional power system in the next week is analyzed based on weather conditions.

[0045] In the embodiment of the present invention, the step of performing load forecasting on the user side based on the LSTM neural network improved by the particle swarm to obtain the load forecasting result specifically includes:

[0046] Data collection and preprocessing: Collect electricity load data, weather and climate data such as temperature, relative humidity, rainfall, and historical data on electricity consumption such as time tags from different power distribution and power consumption customers, and then use threshold setting and periodic longitudinal comparison methods to identify and correct distorted data. Use Lagrange interpolation to process and generate missing data to improve the overall quality of the data set for training the PSO-LSTM neural network model.

[0047] Conduct PSO-LSTM neural network model training: initialize the initial position and speed of each particle in the particle swarm, and set the value range of the optimization parameter; import the divided training set data and parameter values ​​into the PSO-LSTM neural network model for training and prediction, and use the mean absolute percentage error as the formula for calculating the fitness value of the prediction result; obtain the individual extreme value and global extreme value of the population, and update the speed and position of the next generation of particles; import the current best parameter value into the PSO-LSTM neural network model again to calculate the new fitting error; after multiple cycles, until the termination condition is met, import the optimal parameter value into the PSO-LSTM neural network model, predict the load data and output the result;

[0048] Power load forecasting: After completing the training and optimization of the PSO-LSTM neural network model, the load size of different power consumption themes in the regional power system in the next week is analyzed based on weather and climate conditions and period characteristics.

[0049] In the embodiment of the present invention, when estimating the cost of peak-shaving output on the power generation side and load response on the user side, it is necessary to determine the peak-shaving output cost on the power generation side and the load response cost on the user side.

[0050] The peak-shaving cost on the power generation side includes direct and indirect costs, which can be further divided into: Fuel cost: additional fuel is required to heat the boiler and increase the unit to a stable state during startup; Maintenance cost: frequent starts and stops may cause increased equipment wear and tear, and increase maintenance and overhaul costs; Labor cost: additional manpower is required to monitor and operate the start-stop process; Opportunity cost: due to long unit startup time or slow response time, it may not be able to provide sufficient electricity during peak power demand.

[0051] The user-side load response cost realizes multi-energy complementarity through the coordinated control behavior of distributed energy by the virtual power plant on the user side of the power system, increases the consumption of renewable energy such as wind power, and reduces the overall operating cost and carbon emissions of the power system. The coordinated operation strategy between the output peak-shaving capacity of the power generation side of the power system and the load response capacity of the virtual power plant on the user side is used to match power generation and load. On the user side, based on the participation of the virtual power plant in the consumption of wind power, solar energy and thermal power, a multi-objective scheduling strategy optimization model is established. The goal is to find a suitable dynamic electricity price through optimization methods to achieve the minimum operating cost and the lowest carbon emissions. While reducing the system operating costs, it also implements the responsibility of carbon emission reduction.

[0052] The objective function aims to minimize the total cost, including fuel cost, operation and maintenance cost, and carbon emission cost, while maximizing the use of renewable energy, as follows:

[0053]

[0054] Among them C fuel (t): fuel cost of thermal power; C om (t): operation and maintenance cost; C emission (t): Carbon emission cost; C wind (t), C pv (t), C vpp (t): revenue from wind power and photovoltaic power generation, and cost of virtual power plants; C penalty (t): Penalty cost for power shortage or excess. In the optimization of multi-objective scheduling strategy, in order to ensure the matching of power generation and load, the constraints set include:

[0055] 1. Power balance constraints:

[0056] P load (t) = P wind (t)+P pv (t)+P thermal (t)+P vpp (t), where: P load (t): Load demand at time t. wind (t), P pv (t), P thermal (t), P vpp (t): are the output powers of wind power, photovoltaic power generation, thermal power and virtual power plant energy storage system respectively.

[0057] 2. Wind power and photovoltaic power generation power constraints: in and are the maximum power of wind power and photovoltaic power generation at time t, respectively, which are calculated according to wind speed and solar radiation intensity.

[0058] 3. Thermal power constraints: in and They are the minimum and maximum output power of thermal power respectively.

[0059] 4. Virtual power plant load response capacity constraints:

[0060]

[0061] S min ≤S(t)≤S max

[0062] S min =d1N VPP S VPP,max

[0063] S max =d2N VPP S VPP,max

[0064] In the formula, Indicates the minimum charging power; Indicates the maximum discharge power; S max and S min They represent the maximum and minimum values ​​of the virtual power plant dispatch resources respectively; S VPP,max is the maximum value of the average capacity of a single virtual power plant; d1 and d2 are the minimum and maximum resource proportions that the virtual power plant should retain, respectively, and the value range should be between 10% and 90%.

[0065] 5. Capacity constraint of virtual power plant load response: S(t)=S(t-1)+[η c P char (t)-P dis (t) / η d ]Δt, where S(t) represents the capacity of the virtual power plant dispatching resources, which is equal to the capacity at the previous moment plus the current charging or discharging capacity; η c and η d Represents charging and discharging efficiency respectively; P char (t) and P dis (t) represent the electric and discharge power in time period t, respectively.

[0066] Finally, a linear programming optimization algorithm is used to solve the problem and obtain the optimal power distribution at each moment. Dynamic electricity prices are formulated, and time-of-use electricity price strategies are formulated based on the output power of wind power and photovoltaic power generation:

[0067]

[0068] Ultimately, based on the cost drive of dynamic electricity prices, electricity users are encouraged to actively respond to loads, achieve new energy consumption and reduce carbon emissions.

[0069] In the embodiment of the present invention, the step of determining the electricity-carbon responsibility sharing mechanism based on the relationship between carbon reduction benefits and economic costs specifically includes: constructing a step-type demand response compensation cost model based on transferable loads and curtailable loads, wherein the demand response compensation cost is proportional to the user's demand response amount; wherein the transferable load is expressed as follows:

[0070]

[0071] In the formula, is the load that can be transferred during period t; are the loads that can be transferred in and out during period t respectively; are state variables, Indicates that the load is transferred in during period t, Indicates that the load is transferred out during period t; are the upper limits of load that can be transferred in and out during period t respectively;

[0072] The load that can be reduced is expressed as follows: Where: The load that can be reduced during period t; is the state variable, It means that there is load reduction during period t; is the upper limit of load reduction in segment t;

[0073] The total load participating in demand response in a dispatch cycle It can be expressed as:

[0074]

[0075] Where: is the total load participating in demand response during period t;

[0076] Based on the total power of demand response, a step-by-step demand response compensation cost model is established:

[0077]

[0078] In the formula, C demand is the total cost of step-type demand response compensation; χ is the demand response compensation base price; δ is the compensation price growth rate; H is the interval length of the response load.

[0079] When the distribution and use customers respond to the load, the load response cost is considered, that is, there are related costs when transferring electricity. Once the carbon reduction cost of the load response is greater than the carbon reduction cost of the peak load regulation on the power generation side, the carbon responsibility sharing principle should be proportionally allocated according to the carbon emissions of the actual electricity consumption; conversely, since the carbon reduction cost of the load response on the power consumption side is much lower than the carbon reduction cost of the peak load regulation on the power generation side, the carbon responsibility on the power generation side should be less than that on the power consumption side. The demand response cost can be introduced to further reflect the difference in carbon emission responsibilities of the distribution and use customers: In the formula, It is an adjustment coefficient for carbon emission responsibility sharing, which is used to encourage users to actively assume carbon emission reduction responsibilities and realize a carbon responsibility sharing mechanism for carbon emission reduction in the power system based on cost considerations through load response and new energy consumption.

[0080] See also Figure 2 and Figure 3 ,The following example analysis is conducted: Taking a city area in Inner Mongolia as an example, in which the photovoltaic installed capacity is 4 million kilowatts, the wind power installed capacity is 2.5 million kilowatts, and the thermal power installed capacity is 16 million kilowatts, the conventional hybrid power generation output scenario is collected, and the typical meteorological data of a certain year is collected, and the power generation capacity of the city area is predicted.

[0081] In user load forecasting, the predicted value using the PSO-LSTM neural network model has a good fitting effect with the true value, indicating that it has good forecasting ability, and its root mean square error RMSE can be reduced to 14%.

[0082] Based on the power generation capacity and load forecast results, a multi-objective scheduling strategy optimization method is used to allocate the charging time of the user-side virtual power plant to the period of high wind power generation and low load after scheduling, so that the operating cost of the system can be effectively reduced and new energy can be absorbed. In addition, during the peak load period, the virtual power plant can reversely transmit power to the power system through the energy storage system to alleviate the power supply pressure of the system. The final predicted power load is as follows Figure 3 .

[0083] After implementing the load response strategy, the total cost of the power system has been reduced and the carbon emissions have been significantly reduced compared to the unregulated situation. After implementing the load response strategy, the total cost of the power system has been reduced by 2.43% and the carbon emissions have been reduced by 3.84%, as shown in the following table:

[0084] Table 1 Comparison of costs and carbon emissions before and after user-side load response

[0085]

[0086] As the transferable and curtailable load capacity increases, the total cost of electricity consumption is rising, while the carbon emission cost is falling, and the carbon emissions are also showing a certain decreasing trend, as shown in Table 2. When the load response reaches a certain critical value, the carbon reduction cost of the load response is greater than the carbon reduction cost of the peak load regulation on the power generation side. At this time, due to the economic cost, the consumption of new energy and carbon emission reduction on the load side will reduce the overall benefits of the power system, so the carbon emissions of actual electricity consumption are apportioned in proportion. Before the critical value, carbon emission reduction through load response on the user side is the least costly way, and the user side should bear a greater share of carbon responsibility sharing according to the demand response cost.

[0087] Table 2 Comparison of electricity consumption, carbon emission costs and carbon emissions under different load response capacities

[0088]

[0089] This example adopts an electricity-carbon responsibility sharing method based on power generation capacity forecast and user-side load matching forecast. By predicting the power generation capacity of the regional power system and the user-side load matching in the next week, it evaluates the implementation cost and carbon reduction benefits of power generation side output peak regulation and user-side load response. Based on the carbon reduction costs of all parties, a reasonable and fair carbon responsibility sharing mechanism is designed to encourage all participants to actively consume new energy and reduce the overall carbon emissions of the system, thereby achieving the carbon reduction target.

[0090] The above only describes in detail the preferred embodiments of the present invention, which is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

[0091] It should be understood that, although each step in the flow chart of each embodiment of the present invention is shown in sequence according to the indication of the arrow, these steps are not necessarily performed in sequence according to the order indicated by the arrow. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.

[0092] 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 program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0093] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the disclosure in the specification and examples. This application is intended to cover any variations, uses or adaptations of the present disclosure, which follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present disclosure are indicated by the claims.

Claims

1. The electricity-carbon responsibility sharing method based on matching power generation capacity with user-side load is characterized by: The method comprises the following steps: Collect weather information, operating conditions of thermal power units, wind power and photovoltaic new energy stations, and analyze the power generation capacity of the regional power system in the next N days based on weather information, where N is a fixed value; Collect electricity consumption behavior data of different power distribution customers, perform user-side load forecasting based on LSTM neural network improved by particle swarm, and obtain load forecasting results; Based on the power generation capacity and load forecast results, a multi-objective dispatch strategy optimization method is used to analyze the relationship between the carbon reduction benefits and economic costs of peak load regulation on the power generation side and load response on the user side. Determine the electricity-carbon responsibility sharing mechanism based on the relationship between carbon reduction benefits and economic costs.

2. The electricity-carbon responsibility sharing method based on matching power generation capacity with user-side load according to claim 1 is characterized in that: Before analyzing the power generation capacity of the regional power system in the next N days according to weather information, a power generation capacity prediction model on the power generation side is constructed. The power generation capacity prediction model on the power generation side includes a photovoltaic new energy site model, a wind new energy site model and a thermal power unit model.

3. The electricity-carbon responsibility sharing method based on matching power generation capacity with user-side load according to claim 2 is characterized in that: The photovoltaic new energy station model is: P1=S×C×η, where P1 is the photovoltaic power generation power, S is the light intensity, C is the photovoltaic installed capacity of the new energy station, and η is the power generation efficiency of the entire system.

4. The method for allocating electricity-carbon responsibility based on matching power generation capacity with user-side load according to claim 2 is characterized in that: The wind power new energy station model is: Where P2 is the output power of the wind turbine at wind speed v, ρ is the air density, A is the swept area of ​​the wind rotor, v is the wind speed, and C p is the power factor, and k is the power generation efficiency of the wind turbine.

5. The electricity-carbon responsibility sharing method based on matching power generation capacity with user-side load according to claim 2 is characterized in that: The thermal power unit model is: P3 = P rated ×F Demand ×B×f, where P3 is the actual power generation of the thermal power unit, P rated is the rated power of the thermal power unit, F Demand is the grid load factor, B is the equipment availability factor, and f is the fossil fuel combustion and conversion efficiency of the thermal power unit.

6. The electricity-carbon responsibility sharing method based on matching power generation capacity with user-side load according to claim 1 is characterized in that: The step of performing user-side load forecasting based on the LSTM neural network improved by the particle swarm to obtain the load forecast result specifically includes: Collect electricity load data, weather and climate data, and historical data sets of electricity consumption behavior of different power distribution and utilities customers, identify and correct distorted data, and use Lagrange interpolation method to process and generate missing data; Conduct PSO-LSTM neural network model training, initialize the initial position and speed of each particle in the particle swarm, and set the value range of the optimization parameter; import the divided training set data and parameter values ​​into the PSO-LSTM neural network model for training and prediction, and use the mean absolute percentage error as the formula for calculating the fitness value of the prediction result; obtain the individual extreme value and global extreme value of the population, and update the speed and position of the next generation of particles; import the current best parameter value into the PSO-LSTM neural network model again to calculate the new fitting error; after multiple cycles, until the termination condition is met, import the optimal parameter value into the PSO-LSTM neural network model, predict the load data and output the result; The user-side load forecasting is performed based on the trained and optimized PSO-LSTM neural network model.

7. The method for allocating electricity-carbon responsibility based on matching power generation capacity with user-side load according to claim 1 is characterized in that: The step of analyzing the relationship between the carbon reduction benefits and economic costs of the peak load regulation on the power generation side and the load response on the user side specifically includes: Determine power balance constraints, wind and photovoltaic power generation constraints, thermal power constraints, virtual power plant load response capability constraints, and capacity constraints during virtual power plant load response; Based on constraint information and linear programming optimization algorithm, the relationship between carbon reduction benefits and economic costs is determined.

8. The method for allocating electricity-carbon responsibility based on matching power generation capacity with user-side load according to claim 1 is characterized in that: The step of determining the electricity-carbon responsibility sharing mechanism based on the relationship between carbon reduction benefits and economic costs specifically includes: constructing a step-type demand response compensation cost model based on transferable loads and curtailable loads, wherein the demand response compensation cost is proportional to the user's demand response amount; wherein the transferable load is expressed as follows: In the formula, is the load that can be transferred during period t; are the loads that can be transferred in and out during period t respectively; are state variables, Indicates that the load is transferred in during period t, Indicates that the load is transferred out during period t; are the upper limits of load that can be transferred in and out during period t respectively; The load that can be reduced is expressed as follows: Where: The load that can be reduced during period t; is the state variable, It means that there is load reduction during period t; is the upper limit of load reduction in segment t; The total load participating in demand response in a dispatch cycle It can be expressed as: Where: is the total load participating in demand response during period t; Based on the total power of demand response, a step-by-step demand response compensation cost model is established: In the formula, C demand is the total cost of step-type demand response compensation; χ is the demand response compensation base price; δ is the compensation price growth rate; H is the interval length of the response load.

Images