Multi-level collaborative optimization control method for virtual power plants based on digital twins

Abstract

The present invention relates to the field of smart grid technology, and specifically to a multi-level collaborative optimization control method for virtual power plants based on digital twins. The present invention establishes a digital twin model through real-time data collection and processing of the power generation end, the power storage end, the power transmission end, and the load end, and can accurately model and monitor the various equipment of the virtual power plant in real time. Through real-time optimization scheduling and collaborative control, it is possible not only to minimize energy loss, but also to balance supply and demand, maximize energy efficiency, and reduce costs; combining genetic algorithms and particle swarm optimization (PSO) algorithms, scheduling optimization is performed based on a multi-level collaborative optimization model, so that the virtual power plant can respond quickly according to changes in the real-time environment and load demand. In the process of power generation equipment scheduling, priority is given to reducing fossil energy and increasing clean energy, which helps to reduce greenhouse gas emissions.

Description

Technical Field

[0001] The present invention relates to the field of smart grid technology, and specifically to a multi-level collaborative optimization control method for a virtual power plant based on digital twins. Background Art

[0002] Smart grid technology is an electric power network system that uses information technology, communication technology, control technology, and automation technology to achieve dynamic monitoring, optimized scheduling, and intelligent management of the power system. Smart grids are not only an upgraded version of traditional power grids, but also enable real-time monitoring and control of energy flows, optimize the allocation of power resources, and improve the reliability, flexibility, sustainability, and economic efficiency of the power grid. Although current power system and virtual power plant management technologies have made certain progress, there are still many problems and shortcomings, especially in terms of optimized scheduling and resource allocation, including the following:

[0003] First, energy efficiency is low, hindering the full utilization of renewable energy. Traditional power system scheduling relies heavily on static models, making it difficult to dynamically adjust the output of various energy sources, especially when it comes to renewable energy sources like wind and photovoltaics. Due to weather changes and forecast errors, clean energy fluctuates significantly, and traditional systems are unable to accurately control the real-time balance between power generation and load, resulting in underutilization of renewable energy.

[0004] Secondly, traditional power grid dispatching systems rely heavily on static dispatching algorithms and often fail to account for seasonal fluctuations in power demand, emergencies, or equipment failures. When faced with load fluctuations or equipment failures, the grid can experience supply-demand imbalances, overloads, or power outages, compromising grid security and stability.

[0005] Finally, traditional grid dispatching only considers optimization at the supply and load ends, failing to account for the significant energy losses associated with long-distance transmission. These losses are often difficult to accurately predict and optimize, resulting in lower overall power system efficiency and increased operating costs.

[0006] In response to the above problems, it is necessary to propose a multi-level collaborative optimization control method for virtual power plants based on digital twins. Summary of the Invention

[0007] The purpose of the present invention is to solve the problems existing in the background technology and propose a multi-level collaborative optimization control method for virtual power plants based on digital twins.

[0008] The purpose of the present invention can be achieved through the following technical solutions:

[0009] The multi-level collaborative optimization control method of a virtual power plant based on digital twins includes the following steps:

[0010] Step 1: Data collection and digital twin model construction;

[0011] The operating data of each device in the virtual power plant from the power generation end, transmission end, storage end to the load end are collected in real time through the sensor group and information collection equipment; environmental data are collected in real time from the environmental data sensors.

[0012] The operating data are specifically:

[0013] Operation data of power generation end The data includes the power generation equipment ID, output power, fossil energy production, real-time power generation, and historical power generation. The power generation equipment includes traditional fossil energy power generation equipment, photovoltaic power generation equipment, and wind power generation equipment. For photovoltaic power generation equipment and wind power generation equipment, the corresponding fossil energy production is set to 0.

[0014] Transmission side operating data , including the ID number of the transmission equipment, the ID number of the upstream equipment connected to the transmission equipment, the ID number of the downstream equipment, the voltage, current, power loss and load rate; among which, the upstream equipment and downstream equipment include a combination of any two of the power generation equipment, energy storage equipment and load-end equipment.

[0015] Operation data of the power storage terminal , including the energy storage device ID, battery remaining capacity, input power, output power and charging efficiency;

[0016] Operating data of load-end equipment , including the ID of the load-end equipment, load demand, weather forecast data, grid frequency and voltage.

[0017] The environmental data are specifically:

[0018] Environmental parameters , including real-time wind speed, temperature and light intensity in the areas where power generation equipment, energy storage equipment and load-end equipment are located.

[0019] As a preferred embodiment of the present invention, a real-time simulation model of a virtual power plant is established using digital twin technology, and the collected operational and environmental data are input into the real-time simulation model. The model includes the physical entity of the virtual power plant, its operational status, operational data, environmental data, energy flows, and the collaborative relationships between devices.

[0020] Step 2: Establish a multi-level collaborative optimization model;

[0021] In the real-time simulation model of the virtual power plant, a multi-level classification is performed, including:

[0022] Level 1: Reality layer, including power generation equipment, transmission equipment, energy storage equipment and load-end equipment in each power generation end, transmission end, storage end and load end, as well as the collected operating data and environmental data of each device. The input data of level 1 is the real data set ;

[0023] Level 2: Virtual layer, including the power generation prediction model at the power generation end, the topology model at the power transmission end, the scheduling and allocation model at the power storage end, and the load prediction model at the load end;

[0024] Level 3: Scheduling optimization layer, which includes several scheduling optimization signals generated by the superposition and comparison of the real layer and the virtual layer;

[0025] In the second level (the virtual layer), the power generation prediction model at the power generation end, the topology structure and transmission loss model at the power transmission end, the scheduling and allocation model at the power storage end, and the load prediction model at the load end are specifically as follows:

[0026] Power generation prediction model: includes fossil energy power generation prediction sub-model, photovoltaic power generation prediction sub-model and wind power generation prediction sub-model. The mathematical model formula is:

[0027] The formula is: ;in, is the predicted total output power of traditional fossil energy power generation equipment at time t; is the predicted total output power of the photovoltaic power generation equipment at the future time t; is the predicted total output power of the wind power generation equipment at the future time t; is the predicted value of total power generation at future time t; and are support vector regression functions trained for traditional fossil energy power generation equipment, photovoltaic power generation equipment, and wind power generation equipment, respectively, and are used to predict the power generation of traditional fossil energy, photovoltaic, and wind power at time t in the future; is the planned output of fossil energy at time t, is the total output power of traditional fossil energy power generation equipment at the current moment, is the predicted average temperature of the power generation equipment at the future time t; is the predicted value of light intensity at the future time t, is the total output power of the photovoltaic power generation equipment at the current moment; is the predicted wind speed at the future time t, is the total output power of the wind power generation equipment at the current moment; and are the prediction errors of the fossil energy power generation prediction sub-model, photovoltaic power generation prediction sub-model and wind power generation prediction sub-model, which obey Gaussian distribution; It is the predicted value of the total output power of all devices at the power generation end at the future time t.

[0028] The topological structure and transmission loss model of the transmission end are as follows:

[0029] Where i1 and i2 are the ID numbers of the upstream device and the downstream device connected to each transmission device, respectively. I is a set of ID numbers of all power generation equipment, energy storage equipment, and load-end equipment. C(i1, i2) is the transmission dispatch symbol. The value of C(i1, i2) is 1, indicating that power transmission is being performed between the upstream device i1 and the downstream device i2. The value of C(i1, i2) is 0, indicating that power transmission is not being performed between the upstream device i1 and the downstream device i2. represents the transmission power between the upstream device i1 and the downstream device i2, where V(i1, i2) is the transmission line voltage between the upstream device i1 and the downstream device i2.

[0030] The specific dispatching and allocation model of the power storage terminal is:

[0031] ;in is the predicted storage capacity of all energy storage devices at time t, where is the actual storage capacity of the energy storage device at the previous historical moment t; is the total input power and total output power of all energy storage devices collected at time x; are charge and discharge efficiency respectively; Forecast management costs for charging and discharging behavior, are the management costs generated by the charge and discharge scheduling of unit power; and They are the maximum thresholds for total input power, total output power, and storage capacity of all energy storage devices respectively.

[0032] Load forecasting model at the load end: A trained LSTM long short-term memory network, including a forget gate, an input gate, a candidate memory unit, an update memory unit, and an output gate. The total input of the input gate is the historical load data, the real-time wind speed, temperature, and light intensity of the area where the load-end equipment is located, and the dates of special events including holidays and weekdays. The output is the total power forecast value of the load end at the future time t. .

[0033] Among them, the core calculation formula of LSTM long short-term memory network is:

[0034] Forget Gate: ;in is the output value of the forget gate, is the Sigmoid activation function; is the hidden state at the previous moment; The input at the current moment is a data vector consisting of historical load data, real-time wind speed, temperature and light intensity of the area where the load-end equipment is located, and special event dates including holidays and weekdays; are the weight term and bias term of the forget gate respectively.

[0035] Input Gate: ;in is the output value of the input gate, where are the weight term and bias term of the input gate respectively.

[0036] Candidate memory gate: ;in is the output value of the candidate memory gate; tanh is the hyperbolic tangent activation function, are the weight term and bias term of the candidate memory gate respectively.

[0037] Update memory unit: ;in is the output value of the updated memory unit; Update the output value of the memory unit for the previous moment;

[0038] Output Gate: ;in is the output value of the output gate, is the hidden state at the current moment, after all The total output of the LSTM long short-term memory network is obtained by the full link ;in are the weight term and bias term of the output gate respectively.

[0039] As a preferred embodiment of the present invention, the multi-objective optimization function of the second level is set, including:

[0040] Minimize energy loss objective optimization function: , that is, to minimize the total power loss at the transmission end;

[0041] Optimization function for balancing supply and demand objectives: , that is, the total output power of the power generation end and the power storage end is equal to the total output power of the load end;

[0042] Minimize the power generation cost objective optimization function: , that is, to minimize the cost of fossil energy production and storage scheduling management, among which The cost required to put a unit of fossil energy into production;

[0043] Minimum carbon emission objective optimization function: , that is, the planned output of fossil energy at the power generation end is minimized.

[0044] Step 3: Data processing and collaborative optimization;

[0045] The weight coefficients λ1, λ2, λ3, and λ4 are assigned to the optimization functions for minimizing energy loss, balancing supply and demand, minimizing power generation costs, and minimizing carbon emissions, respectively, representing the importance of each multi-objective optimization function. The sum of λ1, λ2, λ3, and λ4 is 1, and their specific values ​​are dynamically adjusted based on scheduling requirements. The final optimization objective function F is obtained by taking a weighted sum of the weights assigned to the optimization functions for minimizing energy loss, balancing supply and demand, minimizing power generation costs, and minimizing carbon emissions.

[0046] The optimization objective function F is specifically: .

[0047] Furthermore, whenever an optimization operation instruction is received from the administrator, a joint optimization model based on GA genetic algorithm and PSO particle swarm optimization is built to perform multi-objective optimization operations, and the final optimization objective function F is solved. Through selection, crossover and mutation operations, the virtual power plant optimization control scheme that satisfies the condition: the final optimization objective function F is minimized is obtained. The specific process is as follows:

[0048] Initialize the population particles of GA and PSO, initialize the GA population particles, generate a set of GA population particles representing the initial candidate solutions, each candidate solution represents a control scheme for the virtual power plant; each particle represents a control scheme for the virtual power plant; select, crossover and mutate the GA population particles, and evaluate the fitness of each GA population particle according to the specific value of the objective function F; the smaller the objective function F of a specific particle in the GA population particles, the higher the fitness of the particle, and the more likely it is to be selected into the next generation. By calculating the objective function F of each particle in the GA population particles, the probability P of each particle being selected into the next generation is obtained; and a crossover operation is performed on all particles in the GA population particles, and a preset number of particles are extracted with the probability P of each particle being selected into the next generation. The specific values ​​of each dimension in the multidimensional vector corresponding to the extracted particles are randomly exchanged through the crossover operation, simulating partial gene exchange to obtain a new solution after the crossover operation;

[0049] The new solution obtained after the crossover operation is converted into PSO population particles, initialized to generate a set of initial PSO population particles, where each particle represents a control scheme of the virtual power plant;

[0050] All PSO swarm particles obtained by initializing the new solution after the crossover operation are input into the PSO velocity update and position update formulas to perform PSO swarm optimization.

[0051] Among them, the GA population particles and PSO population particles are both multidimensional vectors, each dimension corresponds to the real-time output power of each power generation equipment at each time t, the planned output of fossil energy at each time t, the real-time output power of the power generation equipment at each time t, the transmission scheduling symbol C (i1, i2) of all upstream devices i1 and downstream devices i2 at each time t, and the transmission power of any upstream device i1 and downstream device i2 at each time t. , the total input power and total output power of the energy storage device at each time t, and the load end power consumption limit at each time t.

[0052] The PSO velocity update and position update formulas are: ;in is the update formula of particle velocity, where is the updated speed of particle k in the q+1th operation; k is the particle number index, q is the operation iteration number index; w is the preset inertia weight, which controls the influence of the particle's historical speed on the updated speed in the next operation. The speed of particle k in the qth operation; c1 and c2 are preset learning factors, which control the degree of dependence of the particle on the personal best position and the global best position respectively; r1 and r2 are random numbers between [0, 1]; is the position of particle k in the qth operation, that is, the multidimensional vector representing the virtual power plant control scheme; is the personal optimal solution of particle k, that is, the particle position of particle k that minimizes the objective function F in the iterative operation; is the global optimal solution, that is, the particle position where all particles optimize the minimum objective function F in the iterative operation; Update the formula for particle position.

[0053] The conditions for stopping the iteration operation of PSO population optimization are: Condition 1: the number of iterations q is greater than the preset upper limit of the number of optimization operations qmax; Condition 2: the updated speed of all particles k All converge;

[0054] When it is recognized that condition 1 or condition 2 is met, the optimization is determined to be successful, the iterative operation of the PSO velocity update and position update formulas is stopped, and the global optimal solution is output as the final optimization result;

[0055] Obtain the multidimensional vector corresponding to the global optimal solution, and match the parameter interpretation corresponding to each dimension to obtain the virtual power plant optimization control solution.

[0056] As a preferred method of the present invention, whenever an optimization operation instruction is received from the administrator, a joint optimization model based on the GA genetic algorithm and the PSO particle swarm optimization is built to perform multi-objective optimization operations, and the final optimization objective function is solved. Through selection, crossover and mutation operations, a virtual power plant optimization control scheme that meets the condition: the final optimization objective function F is minimized is obtained.

[0057] The virtual power plant optimization control scheme includes:

[0058] The dispatching strategy of power generation equipment, i.e., the real-time output power of power generation equipment at each time t and the planned output of fossil energy at each time t;

[0059] The transmission network scheduling strategy, i.e. the transmission scheduling symbol at each time t;

[0060] The scheduling strategy of the energy storage device, that is, the total input power and total output power of the energy storage device at each time t.

[0061] The scheduling strategy of the load end, that is, the power limit of the load end at each time t.

[0062] Step 4: Collaboratively optimize signal matching;

[0063] As a preferred embodiment of the present invention, based on the virtual power plant optimization control solution, the corresponding control signals are matched and saved to the third level: the scheduling optimization layer. The control signals include:

[0064] Power regulation signal at the power generation end: the fossil energy production increase / decrease signal at each time t, as well as the target quality of the increase / decrease;

[0065] Storage terminal control signal: the charge and discharge instructions of the storage terminal at each time t, as well as the target power of charging / discharging of the storage terminal;

[0066] Transmission end control signal: the on / off instruction of the transmission end equipment at each time t;

[0067] Load regulation signal at the load end: the power limit instruction at the load end dispatched by the power grid at each time t, and the restricted load end target power.

[0068] The control signals at each time t saved in the third level: scheduling optimization layer are sent to the management platform for manual review. After the manual review is confirmed, they are output to the first level: reality layer for control signal execution.

[0069] Step 5: Optimize control strategy execution and supervision;

[0070] During the execution of the control signal, the execution effect is jointly supervised by the PID supervisor and the objective function. The specific process is as follows:

[0071] Obtain the target quality of the increase / decrease in fossil energy production contained in the power regulation signal at the power generation end; calculate the difference between the target quality of the increase / decrease and the actual increase / decrease at every preset time interval t to obtain a first execution deviation.

[0072] Obtaining the target charging / discharging power included in the storage terminal control signal; calculating the difference between the target charging / discharging power and the actual charging / discharging power at every preset time interval t to obtain a second execution deviation.

[0073] The load end target power is included in the load end load regulation signal; and every preset time interval t, the difference between the load end target power and the actual load end power is calculated to obtain a third execution deviation.

[0074] The average of the first, second, and third execution deviations is input into the PID supervisor as the total execution deviation e(t): Calculate the PID deviation characteristic value u(t); where They are respectively the preset weight coefficients of the preset proportion, integration and differentiation; among them, t1 is the time when the control signal is generated, that is, the time when the control instruction is output after the optimization operation instruction sent by the administrator is received for the last time.

[0075] When it is identified that the PID deviation characteristic value u(t) is greater than the first preset threshold, or when it is identified that the increase rate of the final optimization objective function F over time t is greater than the second preset threshold, it is determined that there is a huge deviation between the control signal and the actual control effect, and the isolation between the execution level 1: the reality layer and the level 3: the scheduling optimization layer is stopped, and the generation of all new control signals and the execution of existing control signals are stopped.

[0076] Compared with the prior art, the present invention has the following beneficial effects:

[0077] 1. This invention establishes a digital twin model by collecting and processing data from the power generation, storage, transmission, and load ends in real time. This allows for accurate modeling and real-time monitoring of each device in the virtual power plant. Through real-time optimized scheduling and coordinated control, it can not only minimize energy loss but also balance supply and demand, maximize energy efficiency, and reduce costs. For example, by precisely adjusting the output of power generation equipment and the charging and discharging strategies of energy storage equipment, the overall efficiency and economy of the power system can be optimized.

[0078] 2. This invention adopts a scheduling control scheme based on multi-objective optimization, with minimizing carbon emissions as a key optimization objective. This can effectively reduce the use of fossil energy in virtual power plants and reduce the emission of pollutants such as carbon dioxide. During the power generation equipment scheduling process, the system will prioritize the use of clean energy and reduce carbon emissions by controlling the increase or decrease of fossil energy power generation. This approach not only helps reduce greenhouse gas emissions, but also supports the development of virtual power plants towards a green and sustainable energy structure.

[0079] 3. This invention combines genetic algorithms and particle swarm optimization algorithms to optimize scheduling based on a multi-level collaborative optimization model, enabling the virtual power plant to rapidly respond to real-time environmental and load demand changes. Through an intelligent decision-making system, the system dynamically adjusts to external conditions and internal equipment status. This flexible adaptive capability effectively responds to emergencies, ensures the stable operation of the virtual power plant, and enables optimal scheduling decisions in the face of complex power demands. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings:

[0081] Figure 1 is a flow chart of the method of the present invention;

[0082] Figure 2 This is a topological diagram of the LSTM long short-term memory network proposed in the embodiment of the present invention; DETAILED DESCRIPTION

[0083] The technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0084] See also Figure 1 As shown in FIG, the multi-level collaborative optimization control method of a virtual power plant based on digital twin includes the following steps:

[0085] Step 1: Data collection and digital twin model construction;

[0086] The operating data of each device in the virtual power plant from the power generation end, transmission end, storage end to the load end are collected in real time through the sensor group and information collection equipment; environmental data are collected in real time from the environmental data sensors.

[0087] The operating data are specifically:

[0088] Operation data of power generation end The data includes the power generation equipment ID, output power, fossil energy production, real-time power generation, and historical power generation. The power generation equipment includes traditional fossil energy power generation equipment, photovoltaic power generation equipment, and wind power generation equipment. For photovoltaic power generation equipment and wind power generation equipment, the corresponding fossil energy production is set to 0.

[0089] Transmission side operating data , including the ID number of the transmission equipment, the ID number of the upstream equipment connected to the transmission equipment, the ID number of the downstream equipment, the voltage, current, power loss and load rate; among which, the upstream equipment and downstream equipment include a combination of any two of the power generation equipment, energy storage equipment and load-end equipment.

[0090] Operation data of the power storage terminal , including the energy storage device ID, battery remaining capacity, input power, output power and charging efficiency;

[0091] Operating data of load-end equipment , including the ID of the load-end equipment, load demand, weather forecast data, grid frequency and voltage.

[0092] The environmental data are specifically:

[0093] Environmental parameters , including real-time wind speed, temperature and light intensity in the areas where power generation equipment, energy storage equipment and load-end equipment are located.

[0094] Furthermore, a real-time simulation model of the virtual power plant is established using digital twin technology, and the collected operating data and environmental data are input into the real-time simulation model. The real-time simulation model of the virtual power plant includes the physical entity of the virtual power plant, operating status, operating data, environmental data, energy flow, and collaborative relationships between devices.

[0095] Step 2: Establish a multi-level collaborative optimization model;

[0096] In the real-time simulation model of the virtual power plant, a multi-level classification is performed, including:

[0097] Level 1: Reality layer, including power generation equipment, transmission equipment, energy storage equipment and load-end equipment in each power generation end, transmission end, storage end and load end, as well as the collected operating data and environmental data of each device. The input data of level 1 is the real data set ;

[0098] Level 2: Virtual layer, including the power generation prediction model at the power generation end, the topology model at the power transmission end, the scheduling and allocation model at the power storage end, and the load prediction model at the load end;

[0099] Level 3: Scheduling optimization layer, which includes several scheduling optimization signals generated by the superposition and comparison of the real layer and the virtual layer;

[0100] In the second level (the virtual layer), the power generation prediction model at the power generation end, the topology structure and transmission loss model at the power transmission end, the scheduling and allocation model at the power storage end, and the load prediction model at the load end are specifically as follows:

[0101] Power generation prediction model: includes fossil energy power generation prediction sub-model, photovoltaic power generation prediction sub-model and wind power generation prediction sub-model. The mathematical model formula is:

[0102] The formula is: ;in, is the predicted total output power of traditional fossil energy power generation equipment at time t; is the predicted total output power of the photovoltaic power generation equipment at the future time t; is the predicted total output power of the wind power generation equipment at the future time t; is the predicted value of total power generation at future time t; and are support vector regression functions trained for traditional fossil energy power generation equipment, photovoltaic power generation equipment, and wind power generation equipment, respectively, and are used to predict the power generation of traditional fossil energy, photovoltaic, and wind power at time t in the future; is the planned output of fossil energy at time t, is the total output power of traditional fossil energy power generation equipment at the current moment, is the predicted average temperature of the power generation equipment at the future time t; is the predicted value of light intensity at the future time t, is the total output power of the photovoltaic power generation equipment at the current moment; is the predicted wind speed at the future time t, is the total output power of the wind power generation equipment at the current moment; and are the prediction errors of the fossil energy power generation prediction sub-model, photovoltaic power generation prediction sub-model and wind power generation prediction sub-model, which obey Gaussian distribution; It is the predicted value of the total output power of all devices at the power generation end at the future time t.

[0103] The topological structure and transmission loss model of the transmission end are as follows:

[0104] Where i1 and i2 are the ID numbers of the upstream device and the downstream device connected to each transmission device, respectively. I is a set of ID numbers of all power generation equipment, energy storage equipment, and load-end equipment. C(i1, i2) is the transmission dispatch symbol. The value of C(i1, i2) is 1, indicating that power transmission is being performed between the upstream device i1 and the downstream device i2. The value of C(i1, i2) is 0, indicating that power transmission is not being performed between the upstream device i1 and the downstream device i2. represents the transmission power between the upstream device i1 and the downstream device i2, where V(i1, i2) is the transmission line voltage between the upstream device i1 and the downstream device i2.

[0105] The specific dispatching and allocation model of the power storage terminal is:

[0106] ;in is the predicted storage capacity of all energy storage devices at time t, where is the actual storage capacity of the energy storage device at the previous historical moment t; is the total input power and total output power of all energy storage devices collected at time x; are charge and discharge efficiency respectively; Forecast management costs for charging and discharging behavior, are the management costs generated by the charge and discharge scheduling of unit power; and They are the maximum thresholds for total input power, total output power, and storage capacity of all energy storage devices respectively.

[0107] Load forecasting model at the load end: A trained LSTM long short-term memory network, including a forget gate, an input gate, a candidate memory unit, an update memory unit, and an output gate. The total input of the input gate is the historical load data, the real-time wind speed, temperature, and light intensity of the area where the load-end equipment is located, and the dates of special events including holidays and weekdays. The output is the total power forecast value of the load end at the future time t. .

[0108] See also Figure 2 As shown in the figure, the topological structure diagram of the LSTM long short-term memory network, where the core operation formula of the LSTM long short-term memory network is:

[0109] Forget Gate: ;in is the output value of the forget gate, is the Sigmoid activation function; is the hidden state at the previous moment; The input at the current moment is a data vector consisting of historical load data, real-time wind speed, temperature and light intensity of the area where the load-end equipment is located, and special event dates including holidays and weekdays; are the weight term and bias term of the forget gate respectively.

[0110] Input Gate: ;in is the output value of the input gate, where are the weight term and bias term of the input gate respectively.

[0111] Candidate memory gate: ;in is the output value of the candidate memory gate; tanh is the hyperbolic tangent activation function, are the weight term and bias term of the candidate memory gate respectively.

[0112] Update memory unit: ;in is the output value of the updated memory unit; Update the output value of the memory unit for the previous moment;

[0113] Output Gate: ;in is the output value of the output gate, is the hidden state at the current moment, after all The total output of the LSTM long short-term memory network is obtained by the full link ;in are the weight term and bias term of the output gate respectively.

[0114] Furthermore, the multi-objective optimization function of the second level is set, including:

[0115] Minimize energy loss objective optimization function: , that is, to minimize the total power loss at the transmission end;

[0116] Optimization function for balancing supply and demand objectives: , that is, the total output power of the power generation end and the power storage end is equal to the total output power of the load end;

[0117] Minimize the power generation cost objective optimization function: , that is, to minimize the cost of fossil energy production and storage scheduling management, among which The cost required to put a unit of fossil energy into production;

[0118] Minimum carbon emission objective optimization function: , that is, the planned output of fossil energy at the power generation end is minimized.

[0119] Step 3: Data processing and collaborative optimization;

[0120] The weight coefficients λ1, λ2, λ3, and λ4 are assigned to the optimization function f1 (minimizing energy loss), the optimization function f2 (balancing supply and demand), the optimization function f3 (minimizing power generation costs), and the optimization function f4 (minimizing carbon emissions), respectively. These weight coefficients represent the importance of each multi-objective optimization function. The sum of λ1, λ2, λ3, and λ4 is 1, and their specific values ​​are dynamically adjusted based on scheduling requirements. The weighted sum of the optimization functions f1 (minimizing energy loss), f2 (balancing supply and demand), f3 (minimizing power generation costs), and f4 (minimizing carbon emissions) is calculated based on their assigned weight coefficients to obtain the final optimization objective function F.

[0121] The optimization objective function F is specifically:

[0122] .

[0123] Furthermore, whenever an optimization operation instruction is received from the administrator, a joint optimization model based on GA genetic algorithm and PSO particle swarm optimization is built to perform multi-objective optimization operations, and the final optimization objective function F is solved. Through selection, crossover and mutation operations, the virtual power plant optimization control scheme that satisfies the condition: the final optimization objective function F is minimized is obtained. The specific process is as follows:

[0124] Initialize the population particles of GA and PSO, initialize the GA population particles, generate a set of GA population particles representing the initial candidate solutions, each candidate solution represents a control scheme for the virtual power plant; each particle represents a control scheme for the virtual power plant; select, crossover and mutate the GA population particles, and evaluate the fitness of each GA population particle according to the specific value of the objective function F; the smaller the objective function F of a specific particle in the GA population particles, the higher the fitness of the particle, and the more likely it is to be selected into the next generation. By calculating the objective function F of each particle in the GA population particles, the probability P of each particle being selected into the next generation is obtained; and a crossover operation is performed on all particles in the GA population particles, and a preset number of particles are extracted with the probability P of each particle being selected into the next generation. The specific values ​​of each dimension in the multidimensional vector corresponding to the extracted particles are randomly exchanged through the crossover operation, simulating partial gene exchange to obtain a new solution after the crossover operation;

[0125] The new solution obtained after the crossover operation is converted into PSO population particles, initialized to generate a set of initial PSO population particles, where each particle represents a control scheme of the virtual power plant;

[0126] All PSO swarm particles obtained by initializing the new solution after the crossover operation are input into the PSO velocity update and position update formulas to perform PSO swarm optimization.

[0127] Among them, the GA population particles and PSO population particles are both multidimensional vectors, each dimension corresponds to the real-time output power of each power generation equipment at each time t, the planned output of fossil energy at each time t, the real-time output power of the power generation equipment at each time t, the transmission scheduling symbol C (i1, i2) of all upstream devices i1 and downstream devices i2 at each time t, and the transmission power of any upstream device i1 and downstream device i2 at each time t. , the total input power and total output power of the energy storage device at each time t, and the load end power consumption limit at each time t.

[0128] The PSO velocity update and position update formulas are: ;in is the update formula of particle velocity, where is the updated speed of particle k in the q+1th operation; k is the particle number index, q is the operation iteration number index; w is the preset inertia weight, which controls the influence of the particle's historical speed on the updated speed in the next operation. The speed of particle k in the qth operation; c1 and c2 are preset learning factors, which control the degree of dependence of the particle on the personal best position and the global best position respectively; r1 and r2 are random numbers between [0, 1]; is the position of particle k in the qth operation, that is, the multidimensional vector representing the virtual power plant control scheme; is the personal optimal solution of particle k, that is, the particle position of particle k that minimizes the objective function F in the iterative operation; is the global optimal solution, that is, the particle position where all particles optimize the minimum objective function F in the iterative operation; Update the formula for particle position.

[0129] The conditions for stopping the iteration operation of PSO population optimization are: Condition 1: the number of iterations q is greater than the preset upper limit of the number of optimization operations qmax; Condition 2: the updated speed of all particles k All converge;

[0130] When it is recognized that condition one or condition two is met, the optimization is determined to be successful, the iterative operation of the PSO velocity update and position update formulas is stopped, and the global optimal solution is output as the final optimization result.

[0131] It's important to note that GA and PSO are two common heuristic optimization algorithms. Genetic algorithm (GA) is an optimization algorithm that simulates the principles of natural selection and genetics. It gradually approaches the optimal solution by simulating the evolution of biological species. The core concepts of GA include selection, crossover, and mutation. Particle swarm optimization (PSO) is a swarm intelligence optimization algorithm that simulates the foraging behavior of bird flocks. Each solution is considered a particle, which updates its position based on its own experience and that of its neighbors to find the optimal solution.

[0132] Obtain the multidimensional vector corresponding to the global optimal solution, and match the parameter interpretation corresponding to each dimension to obtain the virtual power plant optimization control solution.

[0133] The virtual power plant optimization control scheme includes:

[0134] The dispatching strategy of power generation equipment, i.e., the real-time output power of power generation equipment at each time t and the planned output of fossil energy at each time t;

[0135] The transmission network scheduling strategy, i.e. the transmission scheduling symbol at each time t;

[0136] The scheduling strategy of the energy storage device, that is, the total input power and total output power of the energy storage device at each time t.

[0137] The scheduling strategy of the load end, that is, the power limit of the load end at each time t.

[0138] Step 4: Collaboratively optimize signal matching;

[0139] Furthermore, based on the virtual power plant optimization control scheme, the corresponding control signals are matched and saved to the third level: the scheduling optimization layer. The control signals include:

[0140] Power regulation signal at the power generation end: the fossil energy production increase / decrease signal at each time t, as well as the target quality of the increase / decrease;

[0141] Storage terminal control signal: the charge and discharge instructions of the storage terminal at each time t, as well as the target power of charging / discharging of the storage terminal;

[0142] Transmission end control signal: the on / off instruction of the transmission end equipment at each time t;

[0143] Load regulation signal at the load end: the power limit instruction at the load end dispatched by the power grid at each time t, and the restricted load end target power.

[0144] The control signals at each time t saved in the third level: scheduling optimization layer are sent to the management platform for manual review. After the manual review is confirmed, they are output to the first level: reality layer for control signal execution.

[0145] Step 5: Optimize control strategy execution and supervision;

[0146] During the execution of the control signal, the execution effect is jointly supervised by the PID supervisor and the objective function. The specific process is as follows:

[0147] Obtain the target quality of the increase / decrease in fossil energy production contained in the power regulation signal at the power generation end; calculate the difference between the target quality of the increase / decrease and the actual increase / decrease at every preset time interval t to obtain a first execution deviation.

[0148] Obtaining the target charging / discharging power included in the storage terminal control signal; calculating the difference between the target charging / discharging power and the actual charging / discharging power at every preset time interval t to obtain a second execution deviation.

[0149] The load end target power is included in the load end load regulation signal; and every preset time interval t, the difference between the load end target power and the actual load end power is calculated to obtain a third execution deviation.

[0150] The average of the first, second, and third execution deviations is input into the PID supervisor as the total execution deviation e(t): Calculate the PID deviation characteristic value u(t); where They are respectively the preset weight coefficients of the preset proportion, integration and differentiation; among them, t1 is the time when the control signal is generated, that is, the time when the control instruction is output after the optimization operation instruction sent by the administrator is received for the last time.

[0151] When it is identified that the PID deviation characteristic value u(t) is greater than the first preset threshold, or when it is identified that the increase rate of the final optimization objective function F over time t is greater than the second preset threshold, it is determined that there is a huge deviation between the control signal and the actual control effect, and the isolation between the execution level 1: the reality layer and the level 3: the scheduling optimization layer is stopped, and the generation of all new control signals and the execution of existing control signals are stopped.

[0152] It should be understood that the terms “include” and “comprising” used in the specification and claims of the present disclosure indicate the presence of described features, integers, steps, operations, elements and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or collections thereof.

[0153] It should also be understood that the terms used in this disclosure are for the purpose of describing particular embodiments only and are not intended to limit the disclosure. As used in this disclosure and the claims, the singular forms "a," "an," and "the" are intended to include the plural forms unless the context clearly indicates otherwise. It should also be further understood that the term "and / or" used in this disclosure and the claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations;

[0154] The preferred embodiments of the present invention disclosed above are intended only to help illustrate the present invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the present invention to specific embodiments. Obviously, many modifications and variations are possible based on the contents of this specification. These embodiments are selected and described in detail in this specification to better explain the principles and practical applications of the present invention, thereby enabling those skilled in the art to better understand and utilize the present invention. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A multi-level collaborative optimization control method for virtual power plants based on digital twins, characterized by: The following steps are involved: Step 1: Data collection and digital twin model construction; The operation data of each device in the virtual power plant from the power generation end, the power transmission end, the power storage end to the load end are collected in real time through the sensor group and information collection equipment; the environmental data are collected in real time from the environmental data sensor; a real-time simulation model of the virtual power plant is established through digital twin technology, including physical entities, operating status, operating data, environmental data, energy flow and collaborative relationships between devices; the collected operation data and environmental data are input into the real-time simulation model; Step 2: Establish a multi-level collaborative optimization model; A multi-level classification is performed in the real-time simulation model of the virtual power plant, resulting in Level 1: Reality Level; Level 2: Virtual Level; and Level 3: Scheduling Optimization Level. In Level 2: Virtual Level, a multi-objective optimization function is set for scheduling optimization operations. Step 3: Data processing and collaborative optimization; Based on the joint optimization model of weight coefficient allocation, GA genetic algorithm and PSO particle swarm optimization, multi-objective optimization calculation is carried out to obtain the optimal control scheme of the virtual power plant; Step 4: Collaboratively optimize signal matching; Based on the virtual power plant optimization control plan, the corresponding control signal is matched and saved to the third level: the scheduling optimization layer; and then sent to the management platform for manual review. Once the manual review is confirmed, it is output to the first level: the reality layer for control signal execution; Step 5: Optimize control strategy execution and supervision; During the execution of the control signal, the execution effect is jointly supervised by a PID supervisor and an objective function; the target quality of the increase / decrease in fossil energy production contained in the power regulation signal at the power generation end is obtained; and the difference between the target quality of the increase / decrease in production and the actual increase / decrease in production is calculated at every preset time interval t to obtain a first execution deviation; Obtaining a target charging / discharging power included in the storage terminal control signal; calculating a difference between the target charging / discharging power and the actual charging / discharging power at every preset time interval t to obtain a second execution deviation; The load end target power included in the load end load regulation signal; calculating the difference between the load end target power and the actual load end power at every preset time interval t to obtain a third execution deviation; The average of the first, second, and third execution deviations is input into the PID supervisor as the total execution deviation e(t): Calculate the PID deviation characteristic value u(t); where are preset weight coefficients for the preset proportional, integral and differential respectively; Where t1 is the time when the control signal is generated, that is, the time when the control signal is output after the optimization operation instruction is received from the administrator for the most recent time; When it is identified that the PID deviation characteristic value u(t) is greater than the first preset threshold, or when it is identified that the increase rate of the final optimization objective function F over time t is greater than the second preset threshold, it is determined that there is a huge deviation between the control signal and the actual control effect, and the isolation between the execution level 1: the reality layer and the level 3: the scheduling optimization layer is stopped, and the generation of all new control signals and the execution of existing control signals are stopped.

2. The multi-level collaborative optimization control method for virtual power plants based on digital twins according to claim 1 is characterized in that: The specific process of multi-level classification in the real-time simulation model of the virtual power plant is as follows: Level 1: Reality layer, including power generation equipment, transmission equipment, energy storage equipment and load-end equipment in each power generation end, transmission end, storage end and load end, as well as the collected operating data and environmental data of each device. The input data of level 1 is the real data set ; Level 2: Virtual layer, including the power generation prediction model at the power generation end, the topology model at the power transmission end, the scheduling and allocation model at the power storage end, and the load prediction model at the load end; Level 3: Scheduling optimization layer, which includes several scheduling optimization signals generated by the superposition and comparison of the real layer and the virtual layer.

3. The multi-level collaborative optimization control method for virtual power plants based on digital twins according to claim 2 is characterized in that: The power generation prediction model at the power generation end is specifically as follows: Power generation prediction model: includes fossil energy power generation prediction sub-model, photovoltaic power generation prediction sub-model and wind power generation prediction sub-model. The mathematical model formula is: ;in, is the predicted total output power of traditional fossil energy power generation equipment at time t; is the predicted total output power of the photovoltaic power generation equipment at the future time t; is the predicted total output power of the wind power generation equipment at the future time t; is the predicted value of total power generation at future time t; and are support vector regression functions trained for traditional fossil energy power generation equipment, photovoltaic power generation equipment, and wind power generation equipment, respectively, and are used to predict the power generation of traditional fossil energy, photovoltaic, and wind power at time t in the future; is the planned output of fossil energy at time t, is the total output power of traditional fossil energy power generation equipment at the current moment, is the predicted average temperature of the power generation equipment at the future time t; is the predicted value of light intensity at the future time t, is the total output power of the photovoltaic power generation equipment at the current moment; is the predicted wind speed at the future time t, is the total output power of the wind power generation equipment at the current moment; are the prediction errors of the fossil energy power generation prediction sub-model, photovoltaic power generation prediction sub-model and wind power generation prediction sub-model, which obey Gaussian distribution; It is the predicted value of the total output power of all devices at the power generation end at the future time t.

4. The multi-level collaborative optimization control method for virtual power plants based on digital twins according to claim 2 is characterized in that: The topological structure and transmission loss model of the transmission end are as follows: Among them, i1 and i2 are the ID numbers of the upstream device and the downstream device connected to each transmission device, respectively. I is a set of the ID numbers of all power generation equipment, energy storage equipment and load-end equipment. C(i1, i2) is the transmission scheduling symbol. The value of C(i1, i2) is 1, which means that power transmission is carried out between the upstream device i1 and the downstream device i2; the value of C(i1, i2) is 0, which means that power transmission is not carried out between the upstream device i1 and the downstream device i2; represents the transmission power between the upstream device i1 and the downstream device i2, and V(i1, i2) is the transmission line voltage between the upstream device i1 and the downstream device i2.

5. The multi-level collaborative optimization control method for virtual power plants based on digital twins according to claim 2 is characterized in that: The specific dispatching and allocation model of the power storage terminal is: ;in is the predicted storage capacity of all energy storage devices at time t, where is the actual storage capacity of the energy storage device at the previous historical moment t; is the total input power and total output power of all energy storage devices collected at time x; are charge and discharge efficiency respectively; Forecast management costs for charging and discharging behavior, are the management costs generated by the charge and discharge scheduling of unit power; They are the maximum thresholds for total input power, total output power, and storage capacity of all energy storage devices respectively.

6. The multi-level collaborative optimization control method for virtual power plants based on digital twins according to claim 1 is characterized in that: The multi-objective optimization function includes: Minimize energy loss objective optimization function: , that is, to minimize the total power loss at the transmission end; Optimization function for balancing supply and demand objectives: , that is, the total output power of the power generation end and the power storage end is equal to the total output power of the load end; Minimize the power generation cost objective optimization function: , that is, to minimize the cost of fossil energy production and storage scheduling management, among which The cost required to put a unit of fossil energy into production; Minimum carbon emission objective optimization function: , that is, the planned output of fossil energy at the power generation end is minimized.

7. The multi-level collaborative optimization control method for virtual power plants based on digital twins according to claim 1 is characterized in that: The specific process of obtaining the optimal control scheme of the virtual power plant through calculation is as follows: Assign weight coefficients λ1, λ2, λ3, and λ4 to the energy loss minimization objective optimization function f1, the supply and demand balancing objective optimization function f2, the power generation cost minimization objective optimization function f3, and the carbon emission minimization objective optimization function f4, respectively, representing the importance of each multi-objective optimization function; the sum of λ1, λ2, λ3, and λ4 is 1, and their specific values ​​are dynamically adjusted according to the scheduling requirements; according to the weight coefficients assigned to each, perform a weighted sum of the energy loss minimization objective optimization function f1, the supply and demand balancing objective optimization function f2, the power generation cost minimization objective optimization function f3, and the carbon emission minimization objective optimization function f4 to obtain the final optimization objective function F; Whenever an optimization operation instruction is received from the administrator, a joint optimization model based on GA genetic algorithm and PSO particle swarm optimization is built to perform multi-objective optimization operations, and the final optimization objective function is solved. Through selection, crossover and mutation operations, the virtual power plant optimization control scheme that satisfies the condition: the final optimization objective function F is minimized is obtained; The virtual power plant optimization control scheme includes: The dispatching strategy of power generation equipment, i.e., the real-time output power of power generation equipment at each time t and the planned output of fossil energy at each time t; The transmission network scheduling strategy, i.e. the transmission scheduling symbol at each time t; The scheduling strategy of the energy storage device, that is, the total input power and total output power of the energy storage device at each time t; The scheduling strategy of the load end, that is, the power limit of the load end at each time t.

8. The multi-level collaborative optimization control method for virtual power plants based on digital twins according to claim 1 is characterized in that: The control signal includes: Power regulation signal at the power generation end: the fossil energy production increase / decrease signal at each time t, as well as the target quality of the increase / decrease; Storage terminal control signal: the charge and discharge instructions of the storage terminal at each time t, as well as the target power of charging / discharging of the storage terminal; Transmission end control signal: the on / off instruction of the transmission end equipment at each time t; Load regulation signal at the load end: the power limit instruction at the load end dispatched by the power grid at each time t, and the restricted load end target power.

9. The multi-level collaborative optimization control method for virtual power plants based on digital twins according to claim 1 is characterized in that: Based on the joint optimization model of weight coefficient allocation, GA genetic algorithm and PSO particle swarm optimization, multi-objective optimization calculation is performed. The specific process of obtaining the optimal control scheme of the virtual power plant through calculation is as follows: Assign weight coefficients λ1, λ2, λ3, and λ4 to the objective optimization function of minimizing energy loss, balancing supply and demand, minimizing power generation cost, and minimizing carbon emissions, respectively, representing the importance of each multi-objective optimization function. The sum of λ1, λ2, λ3, and λ4 is 1, and their specific values ​​are dynamically adjusted according to scheduling requirements. According to the weight coefficients assigned to each, the objective optimization function of minimizing energy loss, balancing supply and demand, minimizing power generation cost, and minimizing carbon emissions are weighted and summed to obtain the final optimization objective function F. The optimization objective function F is specifically: ; Furthermore, whenever an optimization operation instruction is received from the administrator, a joint optimization model based on GA genetic algorithm and PSO particle swarm optimization is built to perform multi-objective optimization operations, and the final optimization objective function F is solved. Through selection, crossover and mutation operations, the virtual power plant optimization control scheme that satisfies the condition: the final optimization objective function F is minimized is obtained. The specific process is as follows: Initialize the population particles of GA and PSO, initialize the GA population particles, generate a set of GA population particles representing the initial candidate solutions, each candidate solution represents a control scheme for the virtual power plant; each particle represents a control scheme for the virtual power plant; select, crossover and mutate the GA population particles, and evaluate the fitness of each GA population particle according to the specific value of the objective function F; the smaller the objective function F of a specific particle in the GA population particles, the higher the fitness of the particle, and the more likely it is to be selected into the next generation. By calculating the objective function F of each particle in the GA population particles, the probability P of each particle being selected into the next generation is obtained; and a crossover operation is performed on all particles in the GA population particles, and a preset number of particles are extracted with the probability P of each particle being selected into the next generation. The specific values ​​of each dimension in the multidimensional vector corresponding to the extracted particles are randomly exchanged through the crossover operation, simulating partial gene exchange to obtain a new solution after the crossover operation; The new solution obtained after the crossover operation is converted into PSO population particles, initialized to generate a set of initial PSO population particles, where each particle represents a control scheme of the virtual power plant; All PSO swarm particles obtained by initializing the new solution after the crossover operation are input into the PSO velocity update and position update formulas to perform PSO swarm optimization; Among them, the GA population particles and PSO population particles are both multidimensional vectors, each dimension corresponds to the real-time output power of each power generation equipment at each time t, the planned output of fossil energy at each time t, the real-time output power of the power generation equipment at each time t, the transmission scheduling symbol C (i1, i2) of all upstream devices i1 and downstream devices i2 at each time t, and the transmission power of any upstream device i1 and downstream device i2 at each time t. , the total input power and total output power of the energy storage device at each time t, and the load end power consumption limit at each time t; The PSO velocity update and position update formulas are: ;in is the update formula of particle velocity, where is the updated speed of particle k in the q+1th operation; k is the particle number index, q is the operation iteration number index; w is the preset inertia weight, which controls the influence of the particle's historical speed on the updated speed in the next operation. The speed of particle k in the qth operation; c1 and c2 are preset learning factors, which control the degree of dependence of the particle on the personal best position and the global best position respectively; r1 and r2 are random numbers between [0, 1]; is the position of particle k in the qth operation, that is, the multidimensional vector representing the virtual power plant control scheme; is the personal optimal solution of particle k, that is, the particle position of particle k that minimizes the objective function F in the iterative operation; is the global optimal solution, that is, the particle position where all particles optimize the minimum objective function F in the iterative operation; Update formula for particle position; The conditions for stopping the iteration operation of PSO population optimization are: Condition 1: the number of iterations q is greater than the preset upper limit of the number of optimization operations qmax; Condition 2: the updated speed of all particles k All converge; When it is recognized that condition 1 or condition 2 is met, the optimization is determined to be successful, the iterative operation of the PSO velocity update and position update formulas is stopped, and the global optimal solution is output as the final optimization result; Obtain the multidimensional vector corresponding to the global optimal solution, and match the parameter interpretation corresponding to each dimension to obtain the virtual power plant optimization control solution.

Images