Electric vehicle charging power distribution optimization method, device and equipment and storage medium

Abstract

The invention relates to the technical field of electric vehicles, and discloses an electric vehicle charging power distribution optimization method and device, equipment and a storage medium, and the method comprises the steps: converting a charging power optimization problem of an electric vehicle cluster into a Markov decision process; and a reward function of a Markov decision process is constructed by taking profit maximization of a charging station operator as an optimization target, a reference power scheduling strategy of the charging station is obtained by solving the Markov decision process, and the charging station operator performs power scheduling on the charging station according to the reference power scheduling strategy of the charging station. The method comprises the following steps: establishing an optimal power flow model of a power distribution network operator by taking the operation cost minimization of the power distribution network operator as an optimization target, performing optimization solution to obtain a power dispatching instruction of a charging station, and sequentially distributing charging power to each electric vehicle in an electric vehicle cluster according to the priority through the power dispatching instruction of the charging station. Therefore, the electric vehicle charging power distribution accuracy is improved, and power grid dispatching stability is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of electric vehicles, and in particular to a method, device, equipment and storage medium for optimizing charging power distribution of electric vehicles. Background Art

[0002] As a large number of new energy sources are connected to the power system, the cost of providing flexibility only on the power generation side is high. Therefore, it is necessary to fully tap the flexibility resources on the demand side. There are a large number of electric vehicles in the distribution network, and their parking time is generally much longer than the time required for charging. Therefore, they have the potential to participate in grid dispatching and provide high-quality flexibility.

[0003] However, the characteristics of electric vehicles have brought difficulties to the establishment of aggregation models. On the one hand, the charging and discharging behaviors of electric vehicles have great uncertainty. The traditional offline aggregation method relies on prediction information, resulting in large deviations between the obtained results and the actual results; some model-driven online aggregation methods have large computational complexity, and the results obtained are often too conservative or optimistic. On the other hand, the existing electric vehicle power decomposition methods have large computational complexity and are not suitable for real-time applications. In addition, some algorithms do not consider discharge behavior, which makes the accuracy of electric vehicle charging power allocation low, which easily leads to unstable grid scheduling. Summary of the invention

[0004] In view of this, the present invention provides a method, device, equipment and storage medium for optimizing the charging power distribution of electric vehicles, which solves the technical problems that the existing electric vehicle power decomposition method has a large amount of calculation and is not suitable for real-time application, and some algorithms do not take discharge behavior into consideration, resulting in low accuracy of electric vehicle charging power distribution and easily leading to unstable grid scheduling.

[0005] A first aspect of the present invention provides a method for optimizing charging power distribution of an electric vehicle, comprising:

[0006] Converting the charging power optimization problem of an electric vehicle cluster into a Markov decision process, and determining a state vector and an action vector corresponding to the Markov decision process, wherein the electric vehicle cluster is obtained by aggregating multiple electric vehicles;

[0007] Taking the maximization of the profit of the charging station operator as the optimization goal, the reward function of the Markov decision process is constructed;

[0008] The Markov decision process is optimized and solved to obtain a benchmark power dispatch strategy for the charging station;

[0009] According to the benchmark power dispatch strategy of the charging station, the optimal power flow model of the distribution network operator is established with the minimization of the operating cost of the distribution network operator as the optimization goal and the network power flow and power limit as the constraints;

[0010] Optimizing and solving the optimal power flow model of the distribution network operator to obtain power dispatch instructions for the charging station;

[0011] The power dispatching instruction of the charging station is used to distribute the charging power to each electric vehicle in the electric vehicle cluster in sequence according to the priority, so as to obtain the real-time power dispatching strategy of each electric vehicle in the current period.

[0012] Preferably, the state vector is:

[0013]

[0014] In the formula, is the state vector of period t, is a 0-1 variable indicating whether the current period is a holiday or a non-holiday. For electricity price, ,in, For real-time electricity prices, To increase the flexibility incentive price, To lower the flexibility incentive price, is the power state of electric vehicle i, ,in, The remaining charging demand for electric vehicles i, is the total charging demand, and The remaining time to the charging station. is the number of electric vehicles, , N is the total number of charging piles.

[0015] Preferably, the action vector is:

[0016]

[0017] In the formula, is the action vector in period t, is the charging power of the electric vehicle during period t, and are the flexibility reserve percentages relative to the upper and lower boundaries of the maximum power, respectively;

[0018] The constraints of the action vector are:

[0019]

[0020] ,

[0021]

[0022]

[0023] In the formula, , are the lower and upper limits of charging power respectively. , They are the lower limit and upper limit of the charging power flexibility range respectively.

[0024] Preferably, the reward function of the Markov decision process is:

[0025]

[0026] In the formula, Profits for charging station operators, is the state vector at time t+1, For the revenue of charging station operators, Penalty for charging station dispatch results that fail to comply with the power allocation instructions issued by the distribution network operator. Fines for charging stations that fail to meet the needs of car owners, The reward for actions with charging slack greater than 0;

[0027] in,

[0028]

[0029]

[0030]

[0031]

[0032]

[0033] In the formula, , , and are weight coefficients, is the power allocation result corresponding to the power allocation instruction issued by the distribution network operator in time period t, is the degradation cost coefficient of electric vehicle batteries, is the actual charging power of the charging pile during period t, is the time difference, , are power surplus and power deficit respectively, is the number of charging piles, , are the actual minimum and maximum charging powers of charging pile i during period t, , are the remaining time for the i-th electric car to leave the station in period t and period t+1 respectively. is the power allocation result corresponding to the power allocation instruction issued by the distribution network operator in the t+1 period, For relaxation.

[0034] Preferably, the distribution network operator's optimal power flow model includes an objective function and constraints;

[0035] The objective function and the constraints are:

[0036]

[0037] In the formula, is the decision variable, ,in, and are the active power injected from the common connection point and the active power of generator k, , are the reactive power injected from the common connection point and the reactive power of generator k, is the cost corresponding to the kth generator, is the price coefficient for purchasing electricity from the upper power grid, is the constraint symbol, , are the lower and upper limits of the injected active power, , are the lower and upper limits of injected reactive power, , are the node voltage and current amplitude vectors respectively, , , , These are system parameters. is the output forecast of renewable energy unit m, is the matrix transpose, , are the lower and upper limits of node voltage respectively, , are the lower and upper limits of the current amplitude vector respectively.

[0038] Preferably, after the step of optimizing and solving the optimal power flow model of the distribution network operator to obtain the power dispatch instruction of the charging station, the step further includes:

[0039] With the goal of minimizing the deviation between the actual output and the predicted output of the renewable energy units and the flexible loads of the distribution network, the decision variables of the optimal power flow model of the distribution network operator are corrected.

[0040] Preferably, the power dispatching instruction of the charging station is used to distribute the charging power to each electric vehicle in the electric vehicle cluster in order according to the priority, so as to obtain the real-time power dispatching strategy of each electric vehicle in the current period, including:

[0041] Determine the slack of each electric vehicle in the current period according to the time required for the electric vehicle to complete the charging task and the remaining time to the charging station;

[0042] Determine the priority of the charging task of each electric vehicle according to the size relationship of the slackness of each electric vehicle in the current period;

[0043] Determine the remaining power of the charging station according to the charging power required to complete the charging task of each electric vehicle within the remaining time and the power allocation result corresponding to the power scheduling instruction of the charging station;

[0044] Determining whether the remaining power of the charging station is negative;

[0045] If it is determined that the remaining power of the charging station is negative, the charging power of each charging task is reduced to the minimum charging power in order from low to high according to the priority of the charging task of each electric vehicle, wherein each time a charging task is reduced to the minimum charging power, the remaining power of the charging station is determined again according to the charging power of the charging task to be completed within the remaining time of each electric vehicle and the power allocation result corresponding to the power scheduling instruction of the charging station, until all charging tasks are reduced to the minimum charging power or the remaining power of the charging station is not negative, and the iteration stops;

[0046] If it is determined that the remaining power of the charging station is not negative, the charging power of each charging task is increased to the maximum charging power in sequence from high to low according to the priority of the charging task of each electric vehicle. After each charging task is increased to the maximum charging power, the remaining power of the charging station is determined based on the charging power required to complete the charging task within the remaining time of each electric vehicle and the power allocation result corresponding to the power scheduling instruction of the charging station. The iteration stops until all charging tasks are increased to the maximum charging power or the remaining power of the charging station is negative.

[0047] In a second aspect, the present invention further provides an electric vehicle charging power distribution optimization device, comprising:

[0048] A process conversion module, used for converting the charging power optimization problem of the electric vehicle cluster into a Markov decision process, and determining the state vector and action vector corresponding to the Markov decision process, wherein the electric vehicle cluster is obtained by aggregating multiple electric vehicles;

[0049] A reward building module, used to build a reward function of the Markov decision process with maximizing the profit of the charging station operator as the optimization goal;

[0050] A power optimization module, used to optimize and solve the Markov decision process to obtain a benchmark power scheduling strategy for the charging station;

[0051] A power flow optimization module, used to establish an optimal power flow model for the distribution network operator based on the benchmark power dispatching strategy of the charging station, with minimization of the operating cost of the distribution network operator as the optimization goal, and with network power flow and power limit as constraints;

[0052] A power dispatching module, used to optimize and solve the optimal power flow model of the distribution network operator to obtain power dispatching instructions for the charging station;

[0053] The power distribution module is used to distribute the charging power to each electric vehicle in the electric vehicle cluster in sequence according to the priority of the power dispatching instruction of the charging station, so as to obtain the real-time power dispatching strategy of each electric vehicle in the current period.

[0054] In a third aspect, the present invention further provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the electric vehicle charging power distribution optimization method as described in the first aspect.

[0055] In a fourth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed, the steps of the electric vehicle charging power distribution optimization method as described in the first aspect are implemented.

[0056] It can be seen from the above technical solutions that the present invention has the following advantages:

[0057] The charging power optimization problem of the electric vehicle cluster is converted into a Markov decision process, and the reward function of the Markov decision process is constructed with the profit maximization of the charging station operator as the optimization goal. By solving the Markov decision process, the benchmark power scheduling strategy of the charging station is obtained. According to the benchmark power scheduling strategy of the charging station, the charging station operator takes the minimization of the operating cost of the distribution network operator as the optimization goal, and establishes the optimal power flow model of the distribution network operator to find the optimal solution, and obtains the power scheduling instruction of the charging station. The power scheduling instruction of the charging station is used to distribute the charging power to each electric vehicle in the electric vehicle cluster in sequence according to the priority, thereby improving the accuracy of the charging power distribution of the electric vehicle and ensuring the stability of the power grid scheduling. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1An application environment of an electric vehicle charging power allocation optimization method provided by an embodiment of the present invention;

[0059] Figure 2 A flowchart of a method for optimizing charging power distribution for electric vehicles provided in an embodiment of the present invention;

[0060] Figure 3 A schematic diagram of the structure of a device for optimizing charging power distribution for electric vehicles provided by an embodiment of the present invention;

[0061] Figure 4 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0062] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0063] The electric vehicle charging power distribution optimization method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown, the charging station operator communicates with the server 102 through the network. The data storage system can store data that the server 102 needs to process. The data storage system can be integrated on the server 102, or placed on the cloud or other network servers. The server 102 can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud computing services.

[0064] like Figure 2 As shown, the embodiment of the present application provides a method for optimizing the charging power distribution of an electric vehicle, and the method is applied to Figure 1 The server 102 in the example is used as an example to illustrate the method, which includes the following steps S1 to S6. Among them:

[0065] Step S1, converting the charging power optimization problem of the electric vehicle cluster into a Markov decision process, and determining the state vector and action vector corresponding to the Markov decision process, wherein the electric vehicle cluster is obtained by aggregating multiple electric vehicles.

[0066] Among them, multiple electric vehicles in the same charging station at the same time period are aggregated into an electric vehicle cluster.

[0067] Among them, the charging power optimization problem of the electric vehicle cluster is converted into a Markov decision process, and the state vector is:

[0068]

[0069] In the formula, is the state vector of period t, is a 0-1 variable indicating whether the current period is a holiday or a non-holiday. For electricity price, ,in, For real-time electricity prices, To increase the flexibility incentive price, To lower the flexibility incentive price, is the power state of electric vehicle i, ,in, The remaining charging demand for electric vehicles i, is the total charging demand, and The remaining time to the charging station. is the number of electric vehicles, , N is the total number of charging piles.

[0070] Among them, not all charging piles are occupied during the t period. For charging piles with electric vehicles connected, the regulations stipulate ,on the contrary .

[0071] The charging station operator uses the state vector The baseline power and flexibility intervals for the t period are obtained, which in turn constitute the action vector The three components of , i.e. the action vector, are:

[0072]

[0073] In the formula, is the action vector in period t, is the charging power of the electric vehicle during period t, and are the flexibility reserve percentages relative to the upper and lower boundaries of the maximum power, respectively;

[0074] The constraints on the action vector are:

[0075]

[0076] ,

[0077]

[0078]

[0079] In the formula, , are the lower and upper limits of charging power respectively. , They are the lower limit and upper limit of the charging power flexibility range respectively.

[0080] Among them, for , It will be reported to the distribution network operator for subsequent power allocation.

[0081] Step S2: Taking the maximization of the profit of the charging station operator as the optimization goal, a reward function of the Markov decision process is constructed.

[0082] Among them, the reward function is optimized with the maximum profit of the charging station operator as the objective function. To ensure that the charging station accurately tracks the dispatch instructions issued by the distribution network operator and ensures the satisfaction of the car owner.

[0083] In the embodiment of the present application, the reward function of the Markov decision process is:

[0084]

[0085] In the formula, Profits for charging station operators, is the state vector at time t+1, For the revenue of charging station operators, Penalty for charging station dispatch results that fail to comply with the power allocation instructions issued by the distribution network operator. Fines for charging stations that fail to meet the needs of car owners, The reward for actions with charging slack greater than 0;

[0086] in,

[0087]

[0088]

[0089]

[0090]

[0091]

[0092] In the formula, , , and are weight coefficients, is the power allocation result corresponding to the power allocation instruction issued by the distribution network operator in time period t, is the degradation cost coefficient of electric vehicle batteries, is the actual charging power of the charging pile during period t, is the time difference, , are power surplus and power deficit respectively, is the number of charging piles, , are the actual minimum and maximum charging powers of charging pile i during period t, , are the remaining time for the i-th electric car to leave the station in period t and period t+1 respectively. is the power allocation result corresponding to the power allocation instruction issued by the distribution network operator in the t+1 period, For relaxation.

[0093] Among them, the penalty for charging station dispatch results that fail to comply with the power allocation instructions issued by the distribution network operator can prevent the charging station operators from over-reporting in the early stage. , This leads to falls outside the feasible region. For example: When all EVs in the charging station are charged at the maximum power, the actual power required is , which is consistent with the actual dispatch power of DSO The difference is .

[0094] Step S3: optimizing and solving the Markov decision process to obtain a benchmark power dispatching strategy for the charging station.

[0095] In the embodiment of the present application, the electric vehicle aggregation stage is modeled as a Markov decision process (MDP). As the basis for the next step of neural network training, the goal of MDP is defined as: finding a strategy π that maximizes the expected discounted return J, that is,

[0096]

[0097] In the formula, For the discounted return expectation, is the discount factor, the strategy π can be based on the current state Produce Action ; It is the development trajectory of the entire MDP.

[0098] In some embodiments, this patent uses the SAC algorithm to train neural networks and solve the MDP process. Other open source reinforcement learning algorithms are also applicable. Any modifications or partial replacements that do not depart from the spirit and scope of the invention should be covered by the scope of the claims of the invention.

[0099] Among them, the state transition of the Markov decision process: the transition probability Affected by many factors, such as the regulation signal of the distribution network operator, the state of electric vehicles, etc. To be more realistic, it is considered here Unknown. After receiving the regulation signal from the distribution network operator, the charging station operator configures the charging / discharging power of each electric vehicle according to the ILLLP algorithm, that is, , and then the charging pile status is updated as follows:

[0100]

[0101]

[0102] in is the electric power corresponding to the charging / discharging of the i-th electric vehicle on the grid side, They are charge and discharge efficiency respectively; is the charging demand of the i-th EV in period t. is a function whose output value is 0 or 1; is a set threshold, and has the same dimension of time. That is, the remaining time for the electric vehicle to leave the station is less than hour, The greater the probability of outputting 0. That is, the charging station can only allow new EVs to connect after the existing EV leaves.

[0103] Step S4: According to the benchmark power dispatching strategy of the charging station, with minimizing the operating cost of the distribution network operator as the optimization goal, and with network power flow and power limit as constraints, an optimal power flow model of the distribution network operator is established.

[0104] Among them, the distribution network operator receives the power benchmark reported by the charging station and flexibility range Finally, it is necessary to consider factors such as network trends and power output costs to make decisions on actual dispatch instructions. And inform the charging station operator.

[0105] The optimal power flow model for distribution network operators includes objective functions and constraints;

[0106] The objective function and constraints are:

[0107]

[0108] In the formula, is the decision variable, ,in, and are the active power injected from the common connection point and the active power of generator k, , are the reactive power injected from the common connection point and the reactive power of generator k, is the cost corresponding to the kth generator, is the price coefficient for purchasing electricity from the upper power grid, is the constraint symbol, , are the lower and upper limits of the injected active power, , are the lower and upper limits of injected reactive power, , are the node voltage and current amplitude vectors respectively, , , , These are system parameters. is the output forecast of renewable energy unit m, is the matrix transpose, , are the lower and upper limits of node voltage respectively, , are the lower and upper limits of the current amplitude vector respectively.

[0109] Step S5: optimizing and solving the optimal power flow model of the distribution network operator to obtain the power dispatch instruction of the charging station.

[0110] Among them, genetic algorithms can be used to optimize and solve the optimal power flow model of the distribution network operator to obtain the power dispatch instructions of the charging station.

[0111] In some embodiments, since the actual output of the renewable energy unit may deviate from the predicted value, the embodiments of the present application also add a correction step to correct the deviation by matching the output of the non-renewable energy unit and the flexible load.

[0112] Specifically, with the goal of minimizing the deviation between the actual output and the predicted output of renewable energy units and flexible loads in the distribution network, the decision variables of the optimal power flow model of the distribution network operator are corrected.

[0113] The correction process can be described as:

[0114]

[0115] The goal is to minimize the deviation between the actual output and the predicted output of renewable energy units and flexible loads in the distribution network, and the objective function is to minimize the correction cost at the generator and the public connection point. ,in and They are the output of generator k and the injected power at the common connection point, which are used to smooth the output fluctuations of renewable energy units.

[0116] By seeking to solve the objective function, the deviation between the actual output and the predicted output of the renewable energy units and flexible loads in the distribution network is minimized, and the decision variables including the power dispatch instructions of the charging station are obtained.

[0117] Step S6: Allocate the charging power to each electric vehicle in the electric vehicle cluster in turn according to the priority level using the power dispatching instruction of the charging station, and obtain a real-time power dispatching strategy for each electric vehicle in the current period.

[0118] In the embodiment of the present application, the charging station operator follows the dispatching instructions issued by the distribution network operator. Perform power decomposition and allocate the charging power of each charging pile in time period t.

[0119] To this end, the embodiment of the present application uses an improved LLLP (The Laxity and Longer Remaining Processing Time) algorithm to Assign them to each electric vehicle in turn according to the priority, and obtain the real-time dispatch strategy of electric vehicles in the current period .

[0120] Specifically, in step S6, the power dispatching instructions of the charging station are used to distribute the charging power to each electric vehicle in the electric vehicle cluster in order according to the priority, so as to obtain the real-time power dispatching strategy of each electric vehicle in the current period, including:

[0121] Step S601: Determine the slackness of each electric vehicle in the current time period according to the time required for the electric vehicle to complete the charging task and the remaining time to the charging station.

[0122] The calculation process of relaxation is:

[0123]

[0124] In the formula, is the relaxation, The time required to complete the charging task for the i-th electric vehicle, where From the calculation process of relaxation, we can get: The larger the value, the more flexibility the electric vehicle can provide. In addition, for two different electric vehicles i and j, if Or both and , then j is said to have a higher priority.

[0125] Step S602: Determine the priority of the charging task of each electric vehicle according to the magnitude relationship of the slackness of each electric vehicle in the current time period.

[0126] Among them, the data released by the distribution network operator during period t can be entered , the task set of the charging station during period t ,Will All charging tasks in the system are arranged in descending order of priority and are recorded as .

[0127] Step S603: Determine the remaining power of the charging station according to the charging power required to complete the charging task within the remaining time of each electric vehicle and the power allocation result corresponding to the power dispatching instruction of the charging station.

[0128] Among them, the actual charging power of all charging tasks The initial value of is assigned to ensure that the charging power of the electric vehicle i completes the charging task within the remaining time , and calculate the remaining power: .

[0129] In this embodiment, the power feasible region of electric vehicle i in period t is defined as:

[0130]

[0131]

[0132]

[0133] in and Respectively represent the rated charging and discharging power of the charging pile; and They represent the actual maximum and minimum charging power of charging pile i during period t respectively; It is used to ensure that the electric vehicle i completes the charging task within the remaining time.

[0134] Step S604: determine whether the remaining power of the charging station is negative.

[0135] Step S605: If it is determined that the remaining power of the charging station is negative, the charging power of each charging task is reduced to the minimum charging power in order from low to high according to the priority of the charging task of each electric vehicle. After each charging task is reduced to the minimum charging power, the remaining power of the charging station is determined again according to the charging power of the charging task to be completed within the remaining time of each electric vehicle and the power allocation result corresponding to the power scheduling instruction of the charging station. The iteration stops until all charging tasks are reduced to the minimum charging power or the remaining power of the charging station is not negative.

[0136] Step S606: If it is determined that the remaining power of the charging station is not negative, the charging power of each charging task is increased to the maximum charging power in order from high to low according to the priority of the charging task of each electric vehicle. After each charging task is increased to the maximum charging power, the remaining power of the charging station is determined again according to the charging power of the charging task to be completed within the remaining time of each electric vehicle and the power allocation result corresponding to the power scheduling instruction of the charging station, until all charging tasks are increased to the maximum charging power or the remaining power of the charging station is negative, and the iteration stops.

[0137] It can be understood that the real-time power scheduling strategy of the electric vehicle charging station in the current period is obtained through the ILLLP algorithm. At this point, the t period step is completely completed, let t=t+1 to enter the next period and repeat the steps until a scheduling cycle ends.

[0138] It should be noted that the embodiment of the present application converts the charging power optimization problem of the electric vehicle cluster into a Markov decision process, and constructs the reward function of the Markov decision process with maximizing the profit of the charging station operator as the optimization goal. By solving the Markov decision process, the benchmark power scheduling strategy of the charging station is obtained. The charging station operator establishes an optimal power flow model of the distribution network operator based on the benchmark power scheduling strategy of the charging station and takes minimizing the operating cost of the distribution network operator as the optimization goal to find the optimal solution and obtain the power scheduling instruction of the charging station. The power scheduling instruction of the charging station is used to distribute the charging power to each electric vehicle in the electric vehicle cluster in sequence according to the priority, thereby improving the accuracy of the charging power distribution of the electric vehicles and ensuring the stability of the grid scheduling.

[0139] Based on the same inventive concept, an embodiment of the present application also provides an electric vehicle charging power distribution optimization device for implementing the above-mentioned electric vehicle charging power distribution optimization method.

[0140] The solution to the problem provided by the device is similar to the solution described in the above method. Therefore, the specific limitations in one or more embodiments of the electric vehicle charging power distribution optimization device provided below can be referred to the limitations on the electric vehicle charging power distribution optimization method above, and will not be repeated here.

[0141] like Figure 3 As shown, the embodiment of the present application also provides an electric vehicle charging power distribution optimization device, comprising:

[0142] The process conversion module 100 is used to convert the charging power optimization problem of the electric vehicle cluster into a Markov decision process, and determine the state vector and action vector corresponding to the Markov decision process, wherein the electric vehicle cluster is obtained by aggregating multiple electric vehicles;

[0143] A reward building module 200 is used to build a reward function of a Markov decision process with maximizing the profit of the charging station operator as the optimization goal;

[0144] The power optimization module 300 is used to optimize and solve the Markov decision process to obtain a benchmark power scheduling strategy for the charging station;

[0145] The power flow optimization module 400 is used to establish an optimal power flow model for the distribution network operator based on the benchmark power dispatching strategy of the charging station, with minimization of the operation cost of the distribution network operator as the optimization goal, and with network power flow and power limit as constraints;

[0146] The power dispatching module 500 is used to optimize and solve the optimal power flow model of the distribution network operator to obtain the power dispatching instructions of the charging station;

[0147] The power distribution module 600 is used to distribute the charging power to each electric vehicle in the electric vehicle cluster according to the priority according to the power dispatching instruction of the charging station, and obtain the real-time power dispatching strategy of each electric vehicle in the current period.

[0148] In some embodiments, the state vector is:

[0149]

[0150] In the formula, is the state vector of period t, is a 0-1 variable indicating whether the current period is a holiday or a non-holiday. For electricity price, ,in, For real-time electricity prices, To increase the flexibility incentive price, To lower the flexibility incentive price, is the power state of electric vehicle i, ,in, The remaining charging demand for electric vehicles i, is the total charging demand, and The remaining time to the charging station. is the number of electric vehicles, , N is the total number of charging piles.

[0151] In some embodiments, the action vector is:

[0152]

[0153] In the formula, is the action vector in period t, is the charging power of the electric vehicle during period t, and are the flexibility reserve percentages relative to the upper and lower boundaries of the maximum power, respectively;

[0154] The constraints of the action vector are:

[0155]

[0156] ,

[0157]

[0158]

[0159] In the formula, , are the lower and upper limits of charging power respectively. , They are the lower limit and upper limit of the charging power flexibility range respectively.

[0160] In some embodiments, the reward function of the Markov decision process is:

[0161]

[0162] In the formula, Profits for charging station operators, is the state vector at time t+1, For the revenue of charging station operators, Penalty for charging station dispatch results that fail to comply with the power allocation instructions issued by the distribution network operator. Fines for charging stations that fail to meet the needs of car owners, The reward for actions with charging slack greater than 0;

[0163] in,

[0164]

[0165]

[0166]

[0167]

[0168]

[0169] In the formula, , , and are weight coefficients, is the power allocation result corresponding to the power allocation instruction issued by the distribution network operator in time period t, is the degradation cost coefficient of electric vehicle batteries, is the actual charging power of the charging pile during period t, is the time difference, , are power surplus and power deficit respectively, is the number of charging piles, , are the actual minimum and maximum charging powers of charging pile i during period t, , are the remaining time for the i-th electric car to leave the station in period t and period t+1 respectively. is the power allocation result corresponding to the power allocation instruction issued by the distribution network operator in the t+1 period, For relaxation.

[0170] In some embodiments, the distribution network operator optimal power flow model includes an objective function and constraints;

[0171] The objective function and the constraints are:

[0172]

[0173] In the formula, is the decision variable, ,in, and are the active power injected from the common connection point and the active power of generator k, , are the reactive power injected from the common connection point and the reactive power of generator k, is the cost corresponding to the kth generator, is the price coefficient for purchasing electricity from the upper power grid, is the constraint symbol, , are the lower and upper limits of the injected active power, , are the lower and upper limits of injected reactive power, , are the node voltage and current amplitude vectors respectively, , , , These are system parameters. is the output forecast of renewable energy unit m, is the matrix transpose, , are the lower and upper limits of node voltage respectively, , are the lower and upper limits of the current amplitude vector respectively.

[0174] In some embodiments, the device further comprises:

[0175] The correction module is used to correct the decision variables of the optimal power flow model of the distribution network operator with the goal of minimizing the deviation between the actual output and the predicted output of the renewable energy units and flexible loads of the distribution network.

[0176] In some embodiments, the power allocation module 600 is used to determine the slack of each electric vehicle in the current time period according to the time required for the electric vehicle to complete the charging task and the remaining time from the charging station; determine the priority of the charging task of each electric vehicle according to the size relationship of the slack of each electric vehicle in the current time period; determine the remaining power of the charging station according to the charging power required to complete the charging task in the remaining time of each electric vehicle and the power allocation result corresponding to the power scheduling instruction of the charging station; determine whether the remaining power of the charging station is negative; if the remaining power of the charging station is determined to be negative, reduce the charging power of each charging task to the minimum charging power in order from low to high according to the priority of the charging task of each electric vehicle, wherein, each time a charging task is reduced to the minimum charging power, it is transferred back to the remaining power of each electric vehicle. The remaining power of the charging station is determined according to the charging power of the charging task to be completed within the remaining time and the power allocation result corresponding to the power scheduling instruction of the charging station, until all charging tasks are reduced to the minimum charging power or the remaining power of the charging station is not negative, and the iteration stops; if it is determined that the remaining power of the charging station is not negative, the charging power of each charging task is increased to the maximum charging power in sequence from high to low according to the priority of the charging task of each electric vehicle, wherein, after each charging task is increased to the maximum charging power, it is turned back to the power allocation result corresponding to the charging power of the charging task to be completed within the remaining time of each electric vehicle and the power scheduling instruction of the charging station to determine the remaining power of the charging station, until all charging tasks are increased to the maximum charging power or the remaining power of the charging station is negative, and the iteration stops.

[0177] like Figure 4As shown, an embodiment of the present application further provides an electronic device, the electronic device 10 includes a memory 20 and a processor 30, the memory 20 stores a computer program, and when the computer program is executed by the processor 30, the processor 30 executes the steps of the electric vehicle charging power distribution optimization method in any of the above embodiments.

[0178] An embodiment of the present application also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed, the steps of the electric vehicle charging power distribution optimization method in any of the above embodiments are implemented.

[0179] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described system, electronic device and computer storage medium can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0180] It should be noted that the terms "including" and "having" and any variations thereof in the specification and claims of the present invention and the above-mentioned drawings are intended to cover non-exclusive inclusions. For example, a process, method, system, product or apparatus comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or apparatus.

[0181] In several embodiments provided by the present invention, it is understood that each box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and a part of a module, a program segment or a code includes one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from that marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved.

[0182] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, electronic devices, computer storage media and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0183] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0184] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0185] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or all or 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 executing all or part of the steps of the method described in each embodiment of the present invention through a computer device (which can be a personal computer, server, or network device, etc.). The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (full name in English: Read-Only Memory, English abbreviation: ROM), random access memory (full name in English: Random Access Memory, English abbreviation: RAM), disk or optical disk and other media that can store program codes.

[0186] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for optimizing charging power distribution of electric vehicles, characterized in that: include: Converting the charging power optimization problem of an electric vehicle cluster into a Markov decision process, and determining a state vector and an action vector corresponding to the Markov decision process, wherein the electric vehicle cluster is obtained by aggregating multiple electric vehicles; Taking the maximization of the profit of the charging station operator as the optimization goal, the reward function of the Markov decision process is constructed; The Markov decision process is optimized and solved to obtain a benchmark power dispatch strategy for the charging station; According to the benchmark power dispatch strategy of the charging station, the optimal power flow model of the distribution network operator is established with the minimization of the operating cost of the distribution network operator as the optimization goal and the network power flow and power limit as the constraints; Optimizing and solving the optimal power flow model of the distribution network operator to obtain power dispatch instructions for the charging station; The power dispatching instruction of the charging station is used to distribute the charging power to each electric vehicle in the electric vehicle cluster in sequence according to the priority, so as to obtain the real-time power dispatching strategy of each electric vehicle in the current period.

2. The method for optimizing the charging power distribution of electric vehicles according to claim 1, characterized in that: The state vector is: In the formula, is the state vector of period t, is a 0-1 variable indicating whether the current period is a holiday or a non-holiday. For electricity price, ,in, For real-time electricity prices, To increase the flexibility incentive price, To lower the flexibility incentive price, is the power state of electric vehicle i, ,in, The remaining charging demand for electric vehicles i, is the total charging demand, and The remaining time to the charging station. is the number of electric vehicles, , N is the total number of charging piles.

3. The method for optimizing the charging power distribution of electric vehicles according to claim 1, characterized in that: The action vector is: In the formula, is the action vector in period t, is the charging power of the electric vehicle during period t, and are the flexibility reserve percentages relative to the upper and lower boundaries of the maximum power, respectively; The constraints of the action vector are: ， In the formula, , are the lower and upper limits of charging power respectively. , They are the lower limit and upper limit of the charging power flexibility range respectively.

4. The method for optimizing the charging power distribution of electric vehicles according to claim 1, characterized in that: The reward function of the Markov decision process is: In the formula, Profits for charging station operators, is the state vector at time t+1, For the revenue of charging station operators, Penalty for charging station dispatch results that fail to comply with the power allocation instructions issued by the distribution network operator. Fines for charging stations that fail to meet the needs of car owners, The reward for actions with charging slack greater than 0; in, In the formula, , , and are weight coefficients, is the power allocation result corresponding to the power allocation instruction issued by the distribution network operator in time period t, is the degradation cost coefficient of electric vehicle batteries, is the actual charging power of the charging pile during period t, is the time difference, , are power surplus and power deficit respectively, is the number of charging piles, , are the actual minimum and maximum charging powers of charging pile i during period t, , are the remaining time for the i-th electric car to leave the station in period t and period t+1 respectively. is the power allocation result corresponding to the power allocation instruction issued by the distribution network operator in the t+1 period, For relaxation.

5. The method for optimizing the charging power distribution of electric vehicles according to claim 1, characterized in that: The optimal power flow model for distribution network operators includes an objective function and constraints; The objective function and the constraints are: In the formula, is the decision variable, ,in, and are the active power injected from the common connection point and the active power of generator k, , are the reactive power injected from the common connection point and the reactive power of generator k, is the cost corresponding to the kth generator, is the price coefficient for purchasing electricity from the upper power grid, is the constraint symbol, , are the lower and upper limits of the injected active power, , are the lower and upper limits of injected reactive power, , are the node voltage and current amplitude vectors respectively, , , , These are system parameters. is the output forecast of renewable energy unit m, is the matrix transpose, , are the lower and upper limits of node voltage respectively, , are the lower and upper limits of the current amplitude vector respectively.

6. The method for optimizing the charging power distribution of electric vehicles according to claim 5, characterized in that: After the step of optimizing and solving the optimal power flow model of the distribution network operator to obtain the power dispatch instruction of the charging station, the method further includes: With the goal of minimizing the deviation between the actual output and the predicted output of the renewable energy units and the flexible loads of the distribution network, the decision variables of the optimal power flow model of the distribution network operator are corrected.

7. The method for optimizing the charging power distribution of electric vehicles according to claim 1, characterized in that: The power dispatching instruction of the charging station is used to distribute the charging power to each electric vehicle in the electric vehicle cluster in order according to the priority, so as to obtain the real-time power dispatching strategy of each electric vehicle in the current period, including: Determine the slack of each electric vehicle in the current period according to the time required for the electric vehicle to complete the charging task and the remaining time to the charging station; Determine the priority of the charging task of each electric vehicle according to the size relationship of the slackness of each electric vehicle in the current period; Determine the remaining power of the charging station according to the charging power required to complete the charging task of each electric vehicle within the remaining time and the power allocation result corresponding to the power scheduling instruction of the charging station; Determining whether the remaining power of the charging station is negative; If it is determined that the remaining power of the charging station is negative, the charging power of each charging task is reduced to the minimum charging power in order from low to high according to the priority of the charging task of each electric vehicle, wherein each time a charging task is reduced to the minimum charging power, the remaining power of the charging station is determined again according to the charging power of the charging task to be completed within the remaining time of each electric vehicle and the power allocation result corresponding to the power scheduling instruction of the charging station, until all charging tasks are reduced to the minimum charging power or the remaining power of the charging station is not negative, and the iteration stops; If it is determined that the remaining power of the charging station is not negative, the charging power of each charging task is increased to the maximum charging power in sequence from high to low according to the priority of the charging task of each electric vehicle. After each charging task is increased to the maximum charging power, the remaining power of the charging station is determined based on the charging power required to complete the charging task within the remaining time of each electric vehicle and the power allocation result corresponding to the power scheduling instruction of the charging station. The iteration stops until all charging tasks are increased to the maximum charging power or the remaining power of the charging station is negative.

8. An electric vehicle charging power distribution optimization device, characterized in that: include: A process conversion module, used for converting the charging power optimization problem of the electric vehicle cluster into a Markov decision process, and determining the state vector and action vector corresponding to the Markov decision process, wherein the electric vehicle cluster is obtained by aggregating multiple electric vehicles; A reward building module, used to build a reward function of the Markov decision process with maximizing the profit of the charging station operator as the optimization goal; A power optimization module, used to optimize and solve the Markov decision process to obtain a benchmark power scheduling strategy for the charging station; A power flow optimization module, used to establish an optimal power flow model for the distribution network operator based on the benchmark power dispatching strategy of the charging station, with minimization of the operating cost of the distribution network operator as the optimization goal, and with network power flow and power limit as constraints; A power dispatching module, used to optimize and solve the optimal power flow model of the distribution network operator to obtain power dispatching instructions for the charging station; The power distribution module is used to distribute the charging power to each electric vehicle in the electric vehicle cluster in sequence according to the priority of the power dispatching instruction of the charging station, so as to obtain the real-time power dispatching strategy of each electric vehicle in the current period.

9. An electronic device, characterized in that: The electronic device includes a memory and a processor, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the processor executes the steps of the electric vehicle charging power distribution optimization method as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed, the steps of the electric vehicle charging power distribution optimization method as described in any one of claims 1 to 7 are implemented.

Images