Multi-energy virtual power plant low-carbon optimization method and system considering carbon transaction mechanism

Abstract

The invention relates to a multi-energy virtual power plant low-carbon optimization method and system considering a carbon transaction mechanism, and the method comprises the steps: building a multi-energy virtual power plant model through considering the characteristics and mutual constraints of all distributed energy conversion devices; based on the multi-energy virtual power plant model, considering a carbon transaction mechanism, and respectively establishing a day-ahead planning scheduling model and an intra-day rolling optimization scheduling model; solving the day-ahead planning scheduling model by adopting a multi-target particle swarm algorithm to obtain a day-ahead planning scheme under a first time scale; and based on the day-ahead planning scheme under the first time scale, solving the intra-day rolling optimization scheduling model by adopting the multi-target particle swarm algorithm to obtain an intra-day planning scheme under a second time scale, and performing rolling optimization to complete a low-carbon optimization process. Compared with the prior art, the method has the advantages that the influence of prediction deviation on actual scheduling is reduced, and the problems of low efficiency and low precision caused by model nonlinearity are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of virtual power plants, and in particular to a low-carbon optimization method and system for multi-energy virtual power plants taking into account a carbon trading mechanism. Background Art

[0002] Renewable energy generation has received increasing attention. Virtual power plants are able to manage and control distributed loads and energy storage equipment, thereby coordinating a variety of different energy sources, thereby improving the flexibility of the power grid and achieving dynamic balance regulation of the power grid, and therefore have received increasing attention. However, when high-penetration wind and solar power generation is connected to virtual power plants, it will have a huge impact on the accuracy of system optimization operations. At the same time, how to better and stably transmit electricity generated by renewable energy to virtual power plants has always been a problem, because their power generation will fluctuate dramatically with changes in the weather. Therefore, how to perfectly combine renewable energy and traditional energy generation to form an accurate scheduling plan is the current research focus.

[0003] The paper "Microgrid Optimal Dispatch Based on Multi-Objective Particle Swarm Optimization" uses MOPSO to solve the unified dispatch problem of isolated wind turbines, photovoltaic cells and traditional generators. In order to encourage renewable energy generation, the paper "Optimal Operation of Integrated Energy System Considering Demand Response under Carbon Trading Mechanism" introduces a carbon trading mechanism to reduce carbon emissions by increasing the power generation cost of traditional energy. The paper "Cooperative Optimization of Electrical Integrated Energy System Considering Uncertainty of New Energy Output" optimizes the dispatch of renewable energy generation through power flow calculation under the power system network, greatly enhancing the practicality of the research. The paper "Dynamic Modeling and Stability Control Strategy of Integrated Energy System in Multi-Time Scales" proposes a discrete model predictive control (DMPC) to unify the stability control of different energy forms. The paper "A Multi-Time Scale Scheduling Method for Wind-PV-PumpedStorage-Electrochemical Energy Storage Integrated Generation System" uses the cplex solver to solve the model, which has day-ahead scheduling, intraday rolling optimization and real-time scheduling to integrate wind power generation, photovoltaic power generation and heat pumps. The paper "Optimal Dispatch of Regional Integrated Energy SystemBased on a Generalized Energy Storage Model" proposes an optimal dispatching method for regional integrated energy systems and uses the cplex solver to solve it. The paper "Fast Decomposed Energy Flow in Large-Scale Integrated Electricity–Gas–Heat Energy Systems" proposes a renewable decomposition strategy to solve the power flow problem in large-scale multi-energy carrier (MEC) systems, which has high efficiency and accuracy. The paper "A Multi Time-Scale and Multi Energy-Type Coordinated MicrogridScheduling Solution—Part I: Model and Methodology" applies improved particle swarm optimization to solve the combined cooling, heating and power (CCHP) model, which is also divided into day-ahead and intraday parts.The paper "Cooperative Planning of Active Distribution System With Renewable Energy Sources and Energy Storage Systems" proposed a multi-objective, multi-level model for high penetration of renewable energy sources (RESs) and energy storage systems (ESS), which was solved by K-means and PSO.

[0004] However, previous studies often only focus on some simplified models, without highly simulating the real situation, and rarely conduct day-ahead dispatch and intraday rolling optimization of the power system network, which makes it difficult to apply in actual production practice. Most previous studies use the cplex solver to solve the model, which requires a strict and accurate linear mathematical model. When dealing with some practical problems, it is difficult to find a mathematical model with a high matching degree, resulting in low solution efficiency and low accuracy. Summary of the invention

[0005] The purpose of the present invention is to provide a low-carbon optimization method and system for multi-energy virtual power plants taking into account the carbon trading mechanism in order to improve the optimization effect of multi-energy virtual power plants.

[0006] The purpose of the present invention can be achieved by the following technical solutions:

[0007] A low-carbon optimization method for a multi-energy virtual power plant considering a carbon trading mechanism comprises the following steps:

[0008] Considering the characteristics and mutual constraints of each distributed energy conversion device, a multi-energy virtual power plant model is built;

[0009] Based on the multi-energy virtual power plant model, taking into account the carbon trading mechanism, a day-ahead planning and scheduling model and an intraday rolling optimization scheduling model are established respectively;

[0010] A multi-objective particle swarm algorithm is used to solve the day-ahead planning and scheduling model to obtain a day-ahead planning solution at a first time scale;

[0011] Based on the day-ahead planning scheme at the first time scale, the multi-objective particle swarm algorithm is used to solve the intraday rolling optimization scheduling model to obtain the intraday planning scheme at the second time scale, and rolling optimization is performed to complete the low-carbon optimization process.

[0012] Furthermore, the multi-energy virtual power plant model includes three energy networks of electricity, heat and gas and multiple distributed devices, and energy interaction is carried out between the energy networks.

[0013] Furthermore, the day-ahead planning and dispatching model takes the minimum sum of the electricity purchase cost, the equipment operation cost and the carbon trading cost as the optimization goal. The day-ahead planning and dispatching model includes an objective function and corresponding constraints. The objective function is:

[0014] F=C buy +C ope +C carbon

[0015] In the formula, F is the target value, C buy represents the cost of purchasing electricity from the upper power grid, C ope represents the operating cost of each distributed device, C carbon Indicates carbon emission penalty;

[0016] The constraints include:

[0017] (1) Network constraints

[0018] Electric network constraints:

[0019] Grid net +P WT +P PV +e GT P GT -P HP -P E =Load e

[0020] Thermal network constraints:

[0021] h GB P GB +h GT P GT +h HP P HP -P H =Load h

[0022] In the formula, Grid net Indicates the purchase of electricity from the upper power grid, P WT is the power of the fan, P PV is the photovoltaic power, e GT represents the electricity conversion efficiency of the gas turbine, P GT is the power of the gas turbine, P HP is the power of the heat pump, P E and P H Represents electrical energy storage and thermal energy storage, Load e and Load h Represent electrical load and thermal load respectively, h GB is the heat conversion efficiency of the gas boiler, P GB is the power of the gas boiler, hGT is the heat conversion efficiency of the gas turbine, h HP is the heat conversion efficiency of the heat pump;

[0023] (2) Equipment constraints

[0024] Device power constraints:

[0025] P i,min ≤P i ≤P i,max

[0026] Electric energy storage capacity constraints:

[0027] E m ×SOC min ≤∑P E +E0≤E m ×SOC max

[0028] -0.05E m ≤∑P E ≤0.05E m

[0029] Thermal energy storage capacity constraints:

[0030] H m ×EH min ≤∑P H +H0≤H m ×EH max

[0031] -0.05H m ≤∑P H ≤0.05H m

[0032] Where i is equipment i, including batteries, heat storage tanks, gas turbines, gas boilers and heat pumps, P i,min , P i,max is the upper and lower limits of the power of device i, E m Indicates the maximum capacity of the battery, SOC min , SOC max are the upper and lower limits of the state of charge, E0 represents the initial charge, P E Indicates the battery charging and discharging power, H m Indicates the maximum capacity of the heat storage tank, EH min , EH max are the upper and lower limits of the heat storage state, H0 represents the initial heat storage, P H Indicates the charging and discharging power of the heat storage tank.

[0033] Furthermore, the intraday rolling optimization scheduling model aims to minimize the sum of power purchase cost, equipment operation cost, carbon emission penalty and difference cost. The intraday rolling optimization scheduling model includes an objective function and corresponding constraints. The objective function is:

[0034] F=C buy +C ope +C carbon +C dis

[0035] In the formula, F is the target value, C buy represents the cost of purchasing electricity from the upper power grid, C ope represents the operating cost of each distributed device, C carbon Represents carbon emission penalty, C dis Represents differential cost.

[0036] Furthermore, the step of obtaining the day-ahead planning scheme at the first time scale includes:

[0037] Based on MPC, we can obtain the forecast data of wind power, photovoltaic power and residential load in the next 24 hours;

[0038] Based on the predicted data, the multi-objective particle swarm algorithm is used to solve the day-ahead planning and scheduling model to obtain a day-ahead planning solution at a first time scale.

[0039] Furthermore, the first time scale is 24 hours, each time period is 1 hour, and day-ahead optimization scheduling is performed every hour to solve the day-ahead planning scheme for the operation of the multi-energy virtual power plant model in the next 24 hours.

[0040] Furthermore, the step of obtaining the intraday planning scheme under the second time scale includes:

[0041] Get real-time data on wind power, photovoltaic power and residential loads;

[0042] Based on the actual data and the day-ahead planning scheme at the first time scale, the multi-objective particle swarm algorithm is used to solve the problem, thereby obtaining the intraday planning scheme at the second time scale.

[0043] Furthermore, the second time scale is 4 hours, each time period is 15 minutes, and intraday optimization scheduling is performed every 15 minutes to solve the intraday planning plan for the operation of the multi-energy virtual power plant model in the next 4 hours.

[0044] Furthermore, the solving steps of the multi-objective particle swarm algorithm include:

[0045] (1) Initialize a particle swarm, where each particle represents the power of each distributed device in each scenario;

[0046] (2) Calculating the fitness value of the particle, where the fitness value is the target value of the day-ahead planning scheduling model and the target value of the intraday rolling optimization scheduling model;

[0047] (3) Based on the fitness value of the particle, calculate the individual extreme value P of the particle best ;

[0048] (4) Based on the individual extreme values ​​P of all particles best , calculate the global optimal solution G best ;

[0049] (5) Based on the individual extreme value P of the particle best and the global optimal solution G best , update the particle's velocity and position;

[0050] (6) Repeat steps (3) to (5) until the stopping condition is met, and output the day-ahead planning solution at the first time scale or the intraday planning solution at the second time scale.

[0051] The present invention also provides a multi-energy virtual power plant low-carbon optimization system considering the carbon trading mechanism, comprising:

[0052] Multi-energy virtual power plant model construction module: used to consider the characteristics and mutual constraints of each distributed energy conversion equipment and build a multi-energy virtual power plant model;

[0053] A day-ahead and intraday scheduling model building module: used to establish a day-ahead planning scheduling model and an intraday rolling optimization scheduling model based on the multi-energy virtual power plant model and taking into account the carbon trading mechanism;

[0054] A day-ahead planning solution solving module: used to solve the day-ahead planning scheduling model by using a multi-objective particle swarm algorithm to obtain a day-ahead planning solution at a first time scale;

[0055] The intraday planning solution solving module is used to solve the intraday rolling optimization scheduling model based on the day-ahead planning solution under the first time scale by using the multi-objective particle swarm algorithm to obtain the intraday planning solution under the second time scale, perform rolling optimization, and complete the low-carbon optimization process.

[0056] Compared with the prior art, the present invention has the following beneficial effects:

[0057] (1) The present invention constructs a multi-energy virtual power plant model that can well simulate the energy flow of an actual industrial park. By considering the impact of the carbon trading mechanism, a day-ahead planning and scheduling model and an intraday rolling optimization scheduling model are established. The multi-objective particle swarm algorithm is used to solve the model, which enables advance fine-tuning of day-ahead planning and intraday rolling optimization, thereby reducing the impact of prediction bias on actual scheduling and improving the low-carbon optimization effect of the multi-energy virtual power plant.

[0058] (2) The day-ahead planning scheduling model and the intraday rolling optimization scheduling model established by the present invention both take the minimum cost as the optimization goal, which reduces the optimization cost to a certain extent. In addition, in the day-ahead planning and intraday rolling optimization, different event scales are used for scheduling optimization. In the intraday rolling optimization, the errors of the day-ahead planning are continuously rolled forward by real-time data, thereby improving the real-time nature of scheduling and the accuracy of the planning scheme.

[0059] (3) Compared with the prior art that uses CPLEX solver to solve the model, it needs to rely on strict and accurate linear mathematical models. It is difficult to find a mathematical model with high matching degree in practical applications, which makes the solution efficiency low and the accuracy low. The present invention uses multi-objective particle swarm algorithm to solve the planning and scheduling problems of distributed energy that integrates renewable energy and traditional energy generation by traversing various situations, avoiding the low efficiency and low precision problems caused by model nonlinearity.

[0060] (4) Through the carbon trading mechanism, gas turbines and gas boilers will be subject to additional "carbon taxes" for power generation methods that cause carbon emissions, thereby increasing the power generation costs of both. Therefore, in order to reduce operating costs, the power generation capacity of gas turbines and gas boilers will be reduced, and other cleaner and more economical power generation methods will be used instead, thereby reducing carbon emissions. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 It is a schematic diagram of the method flow of the present invention;

[0062] Figure 2 It is a multi-objective particle swarm algorithm solution flow chart of the present invention;

[0063] Figure 3 This is the grid structure of the multi-energy virtual power plant. DETAILED DESCRIPTION

[0064] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0065] Example 1

[0066] This embodiment provides a low-carbon optimization method for a multi-energy virtual power plant under a carbon trading mechanism. Figure 1 As shown, the method comprises the following steps:

[0067] S1. Considering the characteristics and mutual constraints of each distributed energy conversion device, a multi-energy virtual power plant model is built, which includes three energy networks of electricity, heat and gas and a large number of distributed equipment.

[0068] The multi-energy virtual power plant model is as follows Figure 3 As shown in the figure, Heat system is the heat system, Power system is the electric system, Gas system is the gas system, Electricity Flow is the electric energy flow, Heat Flow is the heat energy flow, Gas Flow is the gas energy flow, Load is the load, TST is the heat storage tank, GT is the gas turbine, GB is the gas boiler, Battery is the battery, HP is the heat pump, WT is the fan, PV is the photovoltaic, and the number represents the node number. This model specifically involves the energy interaction within and between networks, and the goal is to optimize the economic benefits of the virtual power plant while meeting the user's load requirements and environmental considerations.

[0069] S2. Considering the carbon trading mechanism, a day-ahead planning and scheduling model and an intraday rolling optimization scheduling model are established respectively.

[0070] The objective function of day-ahead planning is: F = C buy +C ope +C carbon , where C buy represents the cost of purchasing electricity from the upper power grid, C ope represents the operating cost of each distributed device, C carbon Represents the carbon emission penalty. The optimization goal is to minimize the objective function F, and the particle swarm algorithm is used for optimization.

[0071] Electricity purchase cost C buy It includes both the cost of purchasing electricity from the upper-level power grid and the profit from selling electricity to the upper-level power grid. Where P buy and P sell Represent the unit price of electricity purchase and electricity sales, P E,i represents the amount of electricity purchased at time i. E,i When it is negative, it indicates electricity sales.

[0072] Equipment operating cost C ope Including the sum of the operating costs of each device at each time period, where c i represents the unit operating cost of equipment i, P i,t represents the power of device i at time t.

[0073] Carbon trading costs C carbon =k×(E ac -E p ), where k represents the carbon trading market price, E p represents the carbon quota, E ac represents the actual carbon emissions, Where P GT and P GB Represent the power of gas turbine and gas boiler respectively, e GT represents the electricity conversion efficiency of the gas turbine, h i represents the thermal conversion efficiency of device i, represents the conversion rate of electrical energy to thermal energy, κ represents the carbon quota coefficient, GT and κ GB Represent the carbon emission coefficients of gas turbines and gas boilers respectively.

[0074] The constraints of the current planning include network constraints and equipment constraints. Network constraints include grid constraints net +P WT +P PV +e GT P GT -P HP -P E =Load e With thermal network constraints h GB P GB +h GT P GT +h HP P HP -P H =Load h WT stands for wind turbine, PV stands for photovoltaic, HP stands for heat pump, GT stands for gas turbine, and GB stands for gas boiler. i represents the power of device i, P E and P H Represents electrical energy storage and thermal energy storage, Load e and Load h Represents electrical load and thermal load respectively, Grid net Indicates purchasing electricity from the upper power grid, e GT represents the electricity conversion efficiency of the gas turbine, h i represents the thermal conversion efficiency of device i.

[0075] Equipment constraints include equipment power constraints and energy storage capacity constraints. Equipment power constraints: P i,min ≤P i ≤P i,max, where i includes batteries, thermal storage tanks, gas turbines, gas boilers and heat pumps. Electric energy storage capacity constraint: E m ×SOC min ≤∑P E +E0≤E m ×SOC max and -0.05E m ≤∑P E ≤0.05E m , where E m Indicates the maximum capacity of the battery, SOC (state of charge) indicates the state of charge, E0 indicates the initial charge, P E Represents the battery charging and discharging power. Thermal energy storage capacity constraint: H m ×EH min ≤ΣP H +H0≤H m ×EH max and -0.05H m ≤ΣP H ≤0.05H m , where H m represents the maximum capacity of the heat storage tank, EH represents the heat storage state, H0 represents the initial heat storage, P H Indicates the charging and discharging power of the heat storage tank.

[0076] Intraday rolling optimization objective function F = C buy +C ope +C carbon +C dis , where C buy represents the cost of purchasing electricity from the upper power grid, C ope represents the operating cost of each distributed device, C carbon Represents carbon emission penalty, C dis Represents the difference cost. The optimization goal is to minimize the objective function F, and the particle swarm algorithm is still used for optimization.

[0077] Differential Cost C dis It refers to the operating cost caused by the power difference between the intraday rolling optimization and the day-ahead planning. Its function is: where ω i represents the differential cost coefficient of equipment i, Indicates the power of intraday rolling optimization, Indicates the power planned for the day before.

[0078] The constraints of intraday rolling optimization include network constraints and device constraints, which are the same as those of day-ahead planning.

[0079] S3. Use multi-objective particle swarm algorithm for day-ahead planning and intra-day planning.

[0080] This step mainly combines day-ahead planning with intraday rolling optimization to reduce the interference of new energy and load forecast errors on scheduling planning, and uses an improved multi-objective particle swarm algorithm to solve and obtain the best planning scheme. Multi-objective refers to the optimal solution for the three optimization goals of power purchase cost, equipment operation cost and carbon trading cost. The algorithm flow chart is as follows: Figure 2 shown.

[0081] First, the particle swarm algorithm is used for day-ahead planning. The forecast data of wind power, photovoltaic power and residential load for the next day based on MPC (Model Predictive Control) is input. According to the objective function and constraints of the day-ahead planning mentioned above, the best planning scheme is obtained. Then, during the intra-day rolling optimization, fine-tuning is performed on the basis of the day-ahead planning according to the actual data of wind power, photovoltaic power and residential load in real time, so as to obtain the real-time equipment power.

[0082] The time scale of the day-ahead planning is 24 hours, with each time period being 1 hour. Each optimization will make an overall plan for the operation of each device in the next 24 hours. The time scale of the intraday rolling optimization is 4 hours, with each time period being 15 minutes. Each operation will solve the optimal result for the next 4 hours, and retain the operation result of the first time period (15 minutes). Then, the second step of optimization will be performed again after 15 minutes, with the time scale still being 4 hours, and so on, rolling forward continuously until the 24-hour operation plan is optimized.

[0083] This project uses the particle swarm algorithm to solve the problem. The particle swarm algorithm (PSO) is a type of swarm intelligence algorithm, which is designed by simulating the predation behavior of bird flocks. The particle swarm algorithm simulates the birds in a flock by designing a massless particle. The particle has only two properties: speed V and position X. The speed represents the speed of movement, and the position represents the direction of movement. Each particle represents a possible scenario. The power of each distributed device in each scenario meets its own device constraints but is random. Therefore, it is necessary to search for the best optimization solution through the continuous movement of particles. The specific implementation process is that each particle searches for the optimal solution in the search space separately, and records it as the current individual extreme value P best , and share the individual extreme value with other particles in the entire particle swarm, and find the optimal individual extreme value as the current global optimal solution G of the entire particle swarm best , all particles in the particle swarm find their own current individual extreme value P best The current global optimal solution G shared by the entire particle swarm best To adjust your speed and position.

[0084] The idea of ​​the particle swarm algorithm is relatively simple, mainly divided into: 1. Initialize the particle swarm; 2. Evaluate the particles, that is, calculate the fitness value, which is the target value of the day-ahead planning scheduling model and the target value of the intraday rolling optimization scheduling model; 3. Based on the fitness value of the particles, find the individual extreme value P best ; 4. Based on the individual extreme values ​​P of all particles best , find the global optimal solution G best ; 5. Based on the individual extreme value P of the particle best and the global optimal solution G best , update the speed and position of the particle; (6) repeat steps (3)-(5) until the stop condition is met, and output the day-ahead planning scheme under the first time scale or the intra-day planning scheme under the second time scale.

[0085] In summary, this embodiment builds a multi-energy virtual power plant including electricity, heat, and gas and a large number of multi-energy conversion equipment, analyzes the interactions and constraints of various distributed devices, and the flow and loss of energy in the network, and establishes algorithm constraints. Comprehensively consider the economy, environment and stability of the power grid, convert the environment and stability into a certain economic cost, and establish an objective function with the lowest total operating cost as the purpose. Scheduling planning is carried out by combining day-ahead planning with intraday rolling optimization to reduce the impact of prediction errors on the actual power grid operation. Day-ahead planning is planned based on the forecast data of wind and solar loads, with a time interval of 1 hour; intraday rolling optimization is based on day-ahead planning, inputs real-time data of wind, photovoltaic and load, and makes small-scale fine-tuning of day-ahead planning, so that it can meet the constraints of various equipment and networks, and be as economical, environmentally friendly and stable as possible, with a time interval of 15 minutes. The above method adopts a multi-energy virtual power plant model, including three energy networks of electricity, heat and gas and a large number of distributed equipment, which can well simulate the energy flow of actual industrial parks; the combination of day-ahead planning and intraday rolling optimization is adopted to reduce the impact of prediction errors on actual scheduling. By adopting the multi-objective particle swarm optimization (MOPSO), various situations are traversed to solve the planning and scheduling problems of distributed energy that integrates renewable energy and traditional energy generation, avoiding the problems of low efficiency and low precision caused by model nonlinearity.

[0086] Example 2

[0087] This embodiment provides a multi-energy virtual power plant low-carbon optimization system considering a carbon trading mechanism, including:

[0088] Multi-energy virtual power plant model construction module: used to consider the characteristics and mutual constraints of each distributed energy conversion equipment and build a multi-energy virtual power plant model;

[0089] A day-ahead and intraday scheduling model building module: used to establish a day-ahead planning scheduling model and an intraday rolling optimization scheduling model based on the multi-energy virtual power plant model and taking into account the carbon trading mechanism;

[0090] A day-ahead planning solution solving module: used to solve the day-ahead planning scheduling model by using a multi-objective particle swarm algorithm to obtain a day-ahead planning solution at a first time scale;

[0091] The intraday planning solution solving module is used to solve the intraday rolling optimization scheduling model based on the day-ahead planning solution under the first time scale by using the multi-objective particle swarm algorithm to obtain the intraday planning solution under the second time scale, perform rolling optimization, and complete the low-carbon optimization process.

[0092] The rest is the same as in Example 1.

[0093] If the above functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program code.

[0094] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes. The schemes in the embodiments of the present invention may be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.

[0095] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0096] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0097] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0098] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0099] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A low-carbon optimization method for a multi-energy virtual power plant under a carbon trading mechanism, characterized in that: The following steps are involved: Considering the characteristics and mutual constraints of each distributed energy conversion device, a multi-energy virtual power plant model is built; Based on the multi-energy virtual power plant model, taking into account the carbon trading mechanism, a day-ahead planning and scheduling model and an intraday rolling optimization scheduling model are established respectively; A multi-objective particle swarm algorithm is used to solve the day-ahead planning and scheduling model to obtain a day-ahead planning solution at a first time scale; Based on the day-ahead planning scheme at the first time scale, the multi-objective particle swarm algorithm is used to solve the intraday rolling optimization scheduling model to obtain the intraday planning scheme at the second time scale, and rolling optimization is performed to complete the low-carbon optimization process.

2. According to the low-carbon optimization method of a multi-energy virtual power plant under the carbon trading mechanism of claim 1, it is characterized in that: The multi-energy virtual power plant model includes three energy networks of electricity, heat and gas and multiple distributed devices, and energy interaction is carried out between the energy networks.

3. According to the low-carbon optimization method of a multi-energy virtual power plant under the carbon trading mechanism of claim 1, it is characterized in that: The day-ahead planning and scheduling model takes the minimum sum of power purchase cost, equipment operation cost and carbon trading cost as the optimization goal. The day-ahead planning and scheduling model includes an objective function and corresponding constraints. The objective function is: F=C buy +C ope +C carbon In the formula, F is the target value, C buy represents the cost of purchasing electricity from the upper power grid, C ope represents the operating cost of each distributed device, C carbon Indicates carbon emission penalty; The constraints include: (1) Network constraints Electric network constraints: Grid net +P WT +P PV +e GT P GT -P HP -P E =Load e Thermal network constraints: h GB P GB +h GT P GT +h HP P HP -P H =Load h In the formula, Grid net Indicates the purchase of electricity from the upper power grid, P WT is the power of the fan, P PV is the photovoltaic power, e GT represents the electricity conversion efficiency of the gas turbine, P GT is the power of the gas turbine, P HP is the power of the heat pump, P E and P H Represents electrical energy storage and thermal energy storage, Load e and Load h Represent electrical load and thermal load respectively, h GB is the heat conversion efficiency of the gas boiler, P GB is the power of the gas boiler, h GT is the heat conversion efficiency of the gas turbine, h HP is the heat conversion efficiency of the heat pump; (2) Equipment constraints Device power constraints: P i,min ≤P i ≤P i,max Electric energy storage capacity constraints: AND m ×SOC min ≤∑P E +E0≤E m ×SOC max -0.05E m ≤∑P E ≤0.05E m Thermal energy storage capacity constraints: H m ×EH min ≤∑P H +H0≤H m ×EH max -0.05H m ≤∑P H ≤0.05H m Where i is equipment i, including batteries, heat storage tanks, gas turbines, gas boilers and heat pumps, P i,min , P i,max is the upper and lower limits of the power of device i, E m Indicates the maximum capacity of the battery, SOC min , SOC max are the upper and lower limits of the state of charge, E0 represents the initial charge, P E Indicates the battery charging and discharging power, H m Indicates the maximum capacity of the heat storage tank, EH min , EH max are the upper and lower limits of the heat storage state, H0 represents the initial heat storage, P H Indicates the charging and discharging power of the heat storage tank.

4. According to a low-carbon optimization method of a multi-energy virtual power plant considering a carbon trading mechanism according to claim 1, it is characterized in that: The intraday rolling optimization scheduling model aims to minimize the sum of power purchase cost, equipment operation cost, carbon emission penalty and difference cost. The intraday rolling optimization scheduling model includes an objective function and corresponding constraints. The objective function is: F=C buy +C ope +C carbon +C dis In the formula, F is the target value, C buy represents the cost of purchasing electricity from the upper power grid, C ope represents the operating cost of each distributed device, C carbon Represents carbon emission penalty, C dis Represents differential cost.

5. According to a low-carbon optimization method of a multi-energy virtual power plant considering a carbon trading mechanism according to claim 1, it is characterized in that: The step of obtaining the day-ahead planning scheme under the first time scale includes: Based on MPC, we can obtain the forecast data of wind power, photovoltaic power and residential load in the next 24 hours; Based on the predicted data, the multi-objective particle swarm algorithm is used to solve the day-ahead planning and scheduling model to obtain a day-ahead planning solution at a first time scale.

6. A low-carbon optimization method for a multi-energy virtual power plant considering a carbon trading mechanism according to claim 5, characterized in that: The first time scale is 24 hours, each time period is 1 hour, and day-ahead optimization scheduling is performed every hour to solve the day-ahead planning scheme for the operation of the multi-energy virtual power plant model in the next 24 hours.

7. The low-carbon optimization method of a multi-energy virtual power plant considering a carbon trading mechanism according to claim 1 is characterized in that: The step of obtaining the intraday planning scheme under the second time scale comprises: Get real-time data on wind power, photovoltaic power and residential loads; Based on the actual data and the day-ahead planning scheme at the first time scale, the multi-objective particle swarm algorithm is used to solve the problem, thereby obtaining the intraday planning scheme at the second time scale.

8. A low-carbon optimization method for a multi-energy virtual power plant considering a carbon trading mechanism according to claim 7, characterized in that: The second time scale is 4 hours, each time period is 15 minutes, and intraday optimization scheduling is performed every 15 minutes to solve the intraday planning plan for the operation of the multi-energy virtual power plant model in the next 4 hours.

9. The low-carbon optimization method of a multi-energy virtual power plant considering a carbon trading mechanism according to claim 1 is characterized in that: The solution steps of the multi-objective particle swarm algorithm include: (1) Initialize a particle swarm, where each particle represents the power of each distributed device in each scenario; (2) Calculating the fitness value of the particle, where the fitness value is the target value of the day-ahead planning scheduling model and the target value of the intraday rolling optimization scheduling model; (3) Based on the fitness value of the particle, calculate the individual extreme value P of the particle best ; (4) Based on the individual extreme values ​​P of all particles best , calculate the global optimal solution G best ; (5) Based on the individual extreme value P of the particle best and the global optimal solution G best , update the particle's velocity and position; (6) Repeat steps (3) to (5) until the stopping condition is met, and output the day-ahead planning solution at the first time scale or the intraday planning solution at the second time scale.

10. A multi-energy virtual power plant low-carbon optimization system considering the carbon trading mechanism, characterized in that: include: Multi-energy virtual power plant model construction module: used to consider the characteristics and mutual constraints of each distributed energy conversion equipment and build a multi-energy virtual power plant model; A day-ahead and intraday scheduling model building module: used to establish a day-ahead planning scheduling model and an intraday rolling optimization scheduling model based on the multi-energy virtual power plant model and taking into account the carbon trading mechanism; A day-ahead planning solution solving module: used to solve the day-ahead planning scheduling model by using a multi-objective particle swarm algorithm to obtain a day-ahead planning solution at a first time scale; The intraday planning solution solving module is used to solve the intraday rolling optimization scheduling model based on the day-ahead planning solution under the first time scale by using the multi-objective particle swarm algorithm to obtain the intraday planning solution under the second time scale, perform rolling optimization, and complete the low-carbon optimization process.

Images