An energy storage optimization scheduling method for charging station operators

Abstract

The present invention discloses an energy storage optimization scheduling method for charging station operators, which belongs to the field of energy management and smart grid. The present invention has the advantages of optimizing and regulating the energy storage of charging stations through electricity prices in various time periods of the day and the next day, and obtaining an energy storage optimization regulation strategy that meets the best economic benefits of operators by calculating the minimum value of formula H. The method stores electric energy when the electricity price is the lowest and releases electric energy when the electricity price is the highest on two adjacent days of the day and the next day, and optimizes and schedules the energy storage used by the charging station on each of the two adjacent days, thereby realizing optimized and scheduled energy storage for the entire life cycle of the charging station, and achieving the goal of minimizing the total electricity price of the charging station throughout the entire life cycle, thereby maximizing the economic benefits obtained by the operator.

Description

Technical Field

[0001] The present invention relates to the field of energy management and smart grid, and in particular to an energy storage optimization scheduling method for charging station operators. Background Art

[0002] Energy storage optimization and scheduling of charging stations refers to the optimal configuration and management of energy storage equipment in charging stations through intelligent control systems. Currently, energy storage optimization and scheduling of charging stations involves multiple aspects, including but not limited to: static optimization and dynamic optimization, intelligent scheduling, capacity planning and optimization, economic analysis and energy optimization. The above optimization methods can be used to improve the overall performance and user experience of charging stations.

[0003] For charging station operators, when optimizing energy storage scheduling, the main consideration is the economic benefits of the charging station. In the past, the optimization and scheduling of energy storage at charging stations was mainly carried out by means of peak shaving and valley filling, that is, by storing electric energy at the charging station when the electricity price is low, and releasing electric energy to charge the electric vehicle when the electricity price is high. However, the previous methods were all regulated by setting a charging threshold, that is, when the electricity price is lower than the charging threshold, the charging station device is charged, and when the electricity price is higher than the charging threshold, the charging station device releases electric energy to provide electric energy to the electric vehicle. However, this method charges the charging station device when it is lower than the threshold, and often when the electricity price is lower, the charging station device is already fully charged, that is, it is impossible to store electric energy when the electricity price is the lowest, and release electric energy when the electricity price is the highest, resulting in the failure to maximize the economic benefits of the charging station.

[0004] Therefore, it is worth studying how to optimize the energy storage scheduling of charging stations to maximize the economic benefits of charging stations. Summary of the invention

[0005] The purpose of the present invention is to overcome the above problems existing in the prior art and to greatly improve its technical effect on the basis of the original technology; to this end, the present invention provides an energy storage optimization scheduling method for charging station operators, the method comprising:

[0006] First, obtain the charging power curve of the charging station: obtain the historical data information of the daily traffic flow and the charging station load since the installation of the charging station; use the same split point every day to split the time, divide each day into multiple time periods, and extract the traffic flow information and charging station load data information in the corresponding time period;

[0007] The various time periods of the previous day and the traffic flow information corresponding to each time period in the historical data information are used as the input of the neural network, and the power consumption corresponding to each time period of the day is used as the output of the neural network. The neural network is trained to obtain the trained neural network model A; the various time periods of the current date and the traffic flow information corresponding to each time period are extracted, and the extracted information is input into the trained neural network model A to predict the load of the charging station in each time period of the next day;

[0008] The average power value corresponding to each time period is calculated by the load of the charging station in each time period of the second day; the average power value is regarded as the power value corresponding to the charging station at the middle moment of each time period, and the power value coordinate corresponding to the middle moment is extracted (t i , Q i ), t i is the middle moment corresponding to the i-th time period of the next day, Q i is the power of the charging station corresponding to the middle moment of the i-th time period on the second day; the extracted power coordinates are processed by the Newton interpolation algorithm to obtain the charging power curve f(t) of the charging station on the second day;

[0009] Secondly, obtain the average electricity price in each time period: extract the population density, weather changes and electricity price change data information of each time period in the history of the power grid coverage area corresponding to the charging station; train the neural network through the extracted data information to obtain the trained neural network model B; predict the electricity price h(t j ), t j is the jth time period, h(t j ) represents the electricity price corresponding to the jth time period;

[0010] Finally, energy storage optimization scheduling is performed based on the total electricity price: first, compare the overall changes in electricity prices in each time period of the day and the electricity prices in each time period of the next day; second, based on the overall changes in electricity prices and the energy storage capacity of the charging station equipment, low-price electricity is stored and used to power the charging station when the price is high. The high-priced electricity in the corresponding time period is converted into low-priced electricity, and h'(t k ) represents the price of low-priced electricity that replaces high-priced electricity, k represents the kth time period is high-priced electricity and is replaced by low-priced electricity, t k represents the kth time period; finally, the total electricity price for the next day is calculated by using the low-priced stored electricity to supply the charging station with electricity when the price is high, so as to achieve the optimal scheduling of energy storage at the charging station; the total electricity price calculation formula for the next day is:

[0011]

[0012] Among them, n is the total number of time periods into which the second day is divided, t∈[0,24), and the unit of t is hours; according to the calculation formula of the total electricity price of the second day, the minimum total electricity price is obtained, that is, the best energy storage optimization scheduling method is obtained.

[0013] Furthermore, the acquisition of historical data information on daily vehicle flow and charging station load since the installation of the charging station includes: acquiring data information on daily vehicle flow and charging station load in the power grid coverage area corresponding to the charging station; the charging station load refers to the amount of electricity provided by the charging station to the electric vehicle, that is, the power consumption of the charging station, including the amount of electricity directly converted through the power grid and the amount of electricity stored in the charging station.

[0014] Further, the above method uses each time period of the previous day and the traffic flow information corresponding to each time period in the historical data information as the input of the neural network, and the power consumption corresponding to each time period of the day as the output of the neural network, including: each time period of the previous day and the traffic flow information corresponding to each time period are used as {(T i j－1 ,q i )}, where j represents the historical data information day is week j, j ranges from 1 to 7, and Sunday is considered as week 7; i represents the i-th time segment of week j-1, q i represents the traffic flow corresponding to the i-th time period; the power consumption corresponding to each time period of the day is represented by {P i j}, j means that the current day is week j, the size is 1 to 7, Sunday is regarded as week 7, i means the i-th time segment of week j, P i j represents the power consumption corresponding to the i-th time period of week j; the neural network refers to a neural network used to process time series prediction and analysis, and the selected neural networks include LSTM, GRU, RNN, Transformer and CNN.

[0015] Furthermore, the calculation of the average power value corresponding to the charging station in each time period through the load of the charging station in each time period of the second day includes: obtaining the average power of the charging station in the corresponding time period by dividing the power consumption of the charging station in each time period of the second day by the total time of the corresponding time period; the Newton interpolation algorithm is a method for constructing polynomial interpolation, which aims to construct a polynomial function passing through these points through known data points. Therefore, the power value coordinates corresponding to the intermediate moments can be processed through the Newton interpolation algorithm to obtain the overall charging power curve of the charging station.

[0016] Furthermore, the training of the neural network through the extracted data information includes: taking the population density and weather changes in each time period as the input of the neural network, taking the electricity price change data information in the corresponding time period as the output of the neural network, training the neural network, and obtaining the trained neural network model B; the neural network refers to a neural network used to process time series prediction and analysis, and the selected neural networks include LSTM, GRU, RNN, Transformer and CNN.

[0017] Furthermore, the comparison of the overall changes in electricity prices in various time periods of the day and in various time periods of the next day includes: comparing the overall average electricity prices of the day and the next day, and on which day the three time periods with the lowest electricity prices in two days are mainly concentrated; if the average electricity price of the day is lower than the average electricity price of the next day, and the three time periods with the lowest electricity prices in two days are mainly concentrated on the day, then the charging station stores electricity in the three lowest time periods of the day, and releases the stored electricity in the several time periods with the highest electricity prices in the next day; if the average electricity price of the day is lower than the average electricity price of the next day, but the three time periods with the lowest electricity prices in two days are mainly concentrated on the second day, it is necessary to consider the energy storage capacity of the charging station equipment; if the energy storage capacity of the charging station is low, the charging station performs appropriate energy storage regulation by storing electricity at its own low price on the second day and releasing electricity at a high price; if the energy storage capacity of the charging station is high, the charging station performs appropriate energy storage regulation by storing electricity at low electricity prices on the day and the next day, and releasing electricity at high electricity prices on the second day; if the average electricity price of the day is higher than the average electricity price of the second day, and the three time periods with the lowest electricity prices in two days are mainly concentrated on the second day, then By storing electricity at a low price on the second day and releasing electricity at a high price, appropriate energy storage regulation is carried out; if the average electricity price of the day is higher than the average electricity price of the second day, but the three time periods with the lowest electricity prices in two days are mainly concentrated on the day, if the energy storage capacity of the charging station is low, and the storage capacity is lower than the electricity stored in the three time periods with the lowest electricity prices in two days, then the charging station stores electricity in the three time periods with the lowest electricity prices on the day, and releases electricity in the corresponding time periods with the highest electricity prices on the second day, so as to achieve appropriate energy storage regulation; if the energy storage capacity of the charging station is high, that is, the energy storage capacity is higher than the electricity stored in the three time periods with the lowest electricity prices in two days, then in addition to storing electricity in the three time periods with the lowest electricity prices on the day, the charging station also needs to store electricity in the corresponding time periods with the lowest electricity prices on the second day, and release electricity in the corresponding time periods with the highest electricity prices on the second day; the energy storage optimization scheduling of the charging station is carried out according to the overall size changes of electricity prices in each time period of the day and the second day and the energy storage capacity of the charging station equipment, and the most suitable energy storage optimization scheduling strategy is determined by the minimum value of the total electricity price calculation formula H on the second day, so that the operator can obtain the maximum economic benefit. The beneficial effects of the present invention are:

[0018] The present invention proposes an energy storage optimization scheduling method for charging station operators; the present invention has the advantages of providing an energy storage optimization scheduling method for charging station operators by obtaining the charging power curve of the charging station, obtaining the average electricity price in each time period and performing energy storage optimization scheduling according to the total electricity price; the present invention optimizes and controls the energy storage of the charging station by the electricity prices in each time period of the day and the next day, and obtains an energy storage optimization regulation strategy that meets the best economic benefits for the operator by calculating the minimum value of the formula H; the method can store electricity when the electricity price is the lowest and release electricity when the electricity price is the highest on the two adjacent days of the day and the next day, and optimizes and schedules the energy storage of the charging station for the electricity used in each adjacent two days, thereby realizing energy storage optimization scheduling for the entire life cycle of the charging station, and achieving the goal of minimizing the total electricity price of the charging station throughout the entire life cycle, thereby maximizing the economic benefits obtained by the operator. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 : A flow chart of an energy storage optimization scheduling method for charging station operators of the present invention. DETAILED DESCRIPTION

[0020] The specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings; it should be understood that the specific embodiments given here are only used to illustrate and explain the present invention and cannot be used to limit the present invention.

[0021] like Figure 1 As shown, a flow chart of an energy storage optimization scheduling method for charging station operators according to an embodiment of the present invention includes: step S100, obtaining a charging power curve of the charging station; step S200, obtaining an average electricity price in each time period; step S300, performing energy storage optimization scheduling according to the total electricity price; wherein step S100 includes: step S101, obtaining historical data information of daily vehicle flow and charging station load since the installation of the charging station; using the same segmentation point every day to segment the time, dividing each day into multiple time periods on average, and extracting the vehicle flow information and charging station load data information in the corresponding time period; step S102, extracting the historical data information of the previous day The time periods of the current date and the traffic flow information corresponding to each time period are used as the input of the neural network, and the power consumption corresponding to each time period of the day is used as the output of the neural network. The neural network is trained to obtain the trained neural network model A; the time periods of the current date and the traffic flow information corresponding to each time period are extracted, and the extracted information is input into the trained neural network model A to predict the load of the charging station in each time period of the next day; step S103, calculate the average power value corresponding to the charging station in each time period according to the load of the charging station in each time period of the next day; regard the average power value as the power value corresponding to the charging station at the middle moment of each time period, and extract the power value coordinate (t i, Q i ), t i is the middle time corresponding to the i-th time period of the next day, Q i is the power of the charging station corresponding to the middle moment of the i-th time period of the second day; the extracted power coordinates are processed by the Newton interpolation algorithm to obtain the charging power curve f(t) of the charging station on the second day; step S200 includes: step S201, extracting the population density, weather changes and electricity price change data information of each time period in the history of the power grid coverage area corresponding to the charging station; training the neural network through the extracted data information to obtain the trained neural network model B; predicting the electricity price h(t j ), t j is the jth time period, h(t j ) represents the electricity price corresponding to the jth time period; step S300 comprises: step S301, first, comparing the overall size changes of the electricity prices in each time period of the day and the electricity prices in each time period of the next day; secondly, according to the overall size changes of the electricity prices and the energy storage capacity of the charging station equipment, storing electric energy at a low price, and using the low-priced stored electric energy to power the charging station when the price is high, and the high-priced electric energy in the corresponding time period is converted into low-priced electric energy, and h'(t k ) represents the price of low-priced electricity that replaces high-priced electricity, k represents the kth time period is high-priced electricity and is replaced by low-priced electricity, t k represents the kth time period; finally, by using the low-priced stored electric energy to power the charging station when the price is high, the total electricity price for the next day is calculated to achieve optimal energy storage scheduling of the charging station.

[0022] Specifically, the present invention first divides each day into multiple time periods on average, extracts the traffic flow information and charging station load data information in the corresponding time period; uses the historical data information as the input and output of the neural network, trains the neural network, and obtains the trained neural network model A; and predicts the load of the charging station in each time period of the next day through the trained neural network model A; calculates the average power value corresponding to the charging station in each time period according to the load of the charging station in each time period of the next day; regards the average power value as the power value corresponding to the charging station at the middle moment of each time period, and extracts the power value coordinate (t i , Q i ), the extracted power coordinates are processed by Newton interpolation algorithm to obtain the charging power curve f(t) of the charging station on the next day; secondly, the historical data information of the grid coverage area corresponding to the charging station is extracted, and the neural network is trained by the extracted historical data information to obtain the trained neural network model B; the trained neural network model B is used to predict the electricity price h(t j); Finally, the energy storage is optimally dispatched according to the total electricity price, and the total electricity price for the next day is calculated to achieve optimal energy storage dispatch of the charging station.

[0023] Step S100, obtaining the charging power curve of the charging station:

[0024] Step S101, obtaining historical data information of daily vehicle flow and charging station load since the charging station was installed; using the same segmentation point every day to segment time, each day is evenly divided into multiple time periods, and the vehicle flow information and charging station load data information in the corresponding time period are extracted.

[0025] In the above embodiment, specifically, the daily vehicle flow information of the power grid coverage area corresponding to the charging station and the data information of the charging station load are obtained; the charging station load refers to the amount of electricity provided by the charging station to the electric vehicle, that is, the power consumption of the charging station, including the amount of electricity directly converted through the power grid and the amount of electricity stored in the charging station.

[0026] Step S102, taking each time period of the previous day in the historical data information and the traffic flow information corresponding to each time period as the input of the neural network, and the power consumption corresponding to each time period of the day as the output of the neural network, the neural network is trained to obtain the trained neural network model A; extracting each time period of the current date and the traffic flow information corresponding to each time period, inputting the extracted information into the trained neural network model A, and predicting the load of the charging station in each time period of the next day.

[0027] In the above embodiment, specifically, each time period of the previous day and the traffic flow information corresponding to each time period are represented by {(T i j－1 ,q i )}, where j represents the historical data information day is week j, j ranges from 1 to 7, and Sunday is considered as week 7; i represents the i-th time segment of week j-1, q i represents the traffic flow corresponding to the i-th time period; the power consumption corresponding to each time period of the day is represented by {P i j}, j means that the current day is week j, the size is 1 to 7, Sunday is regarded as week 7, i means the i-th time segment of week j, P i j represents the power consumption corresponding to the ith time period of week j; the neural network refers to a neural network used for processing time series prediction and analysis, and the selected neural network includes one of LSTM, GRU, RNN, Transformer and CNN. The selected neural network input {(T i j－1 ,q i )}, the corresponding output {P i j}, obtain the trained neural network model A, and use the trained neural network model A to predict the load of the charging station in each time period of the next day.

[0028] In the above embodiment, preferably, if it is found that the difference between the predicted result and the actual charging station load in a certain time period exceeds a preset threshold, usually the difference is more than 5%; the actual data is input into the trained neural network model A to optimize the neural network. In this way, the neural network is continuously optimized to make the prediction result of model A more accurate.

[0029] Step S103, calculate the average power value corresponding to each time period of the charging station according to the load of the charging station in each time period of the second day; regard the average power value as the power value corresponding to the charging station at the middle moment of each time period, and extract the power value coordinate corresponding to the middle moment (t i , Q i ), t i is the middle moment corresponding to the i-th time period of the next day, Q i is the charging station power corresponding to the middle moment of the i-th time period on the second day; the extracted power coordinates are processed by the Newton interpolation algorithm to obtain the charging power curve f(t) of the charging station on the second day.

[0030] In the above embodiment, specifically, the average power of the charging station in the corresponding time period is obtained by dividing the power consumption of the charging station in each time period of the second day by the total time of the corresponding time period; the Newton interpolation algorithm is a method for constructing polynomial interpolation, which aims to construct a polynomial function passing through these points through known data points. Therefore, the power value coordinates corresponding to the intermediate moments can be processed by the Newton interpolation algorithm to obtain the overall charging power curve of the charging station.

[0031] Step S200, obtaining the average electricity price in each time period:

[0032] Step S201, extracting data information on population density, weather changes, and electricity price changes in the power grid coverage area corresponding to the charging station in each time period in history; training a neural network using the extracted data information to obtain a trained neural network model B; and predicting the electricity price h(t j ), t j is the jth time period, h(t j ) represents the electricity price corresponding to the jth time period.

[0033] In the above embodiment, specifically, the population density and weather changes in each time period are used as the input of the neural network, and the electricity price change data information in the corresponding time period is used as the output of the neural network, the neural network is trained, and the trained neural network model B is obtained; the neural network refers to a neural network used to process time series prediction and analysis, and the selected neural network includes any one of LSTM, GRU, RNN, Transformer and CNN neural networks. The extracted data information is analyzed by the selected neural network algorithm to obtain the trained neural network model B, and the trained neural network model B is used to predict the electricity prices in each time period of the next day.

[0034] In the above embodiment, preferably, if it is found that the difference between the predicted result and the actual electricity price in a certain time period exceeds a preset threshold, usually the difference is more than 5%, the actual electricity price data is input into the trained neural network model B to optimize the neural network. In this way, the neural network is continuously optimized to make the prediction result of model B more accurate.

[0035] Step S300, optimizing energy storage scheduling according to the total electricity price:

[0036] Step S301, first, compare the overall changes in electricity prices in each time period of the day and the electricity prices in each time period of the next day; secondly, store electric energy at a low price according to the overall changes in electricity prices and the energy storage capacity of the charging station equipment, and use the low-priced stored electric energy to power the charging station when the price is high, so that the high-priced electric energy in the corresponding time period becomes low-priced electric energy, and use h'(t k ) represents the price of low-priced electricity that replaces high-priced electricity, k represents the kth time period is high-priced electricity and is replaced by low-priced electricity, t k represents the kth time period; finally, by using the low-priced stored electric energy to power the charging station when the price is high, the total electricity price for the next day is calculated to achieve optimal energy storage scheduling of the charging station.

[0037] In the above embodiment, specifically, the total electricity price calculation formula for the second day is:

[0038]

[0039] Among them, n is the total number of time periods into which the second day is divided, t∈[0,24), and the unit of t is hours; according to the calculation formula of the total electricity price of the second day, the minimum total electricity price is obtained, that is, the best energy storage optimization scheduling method is obtained.

[0040] In the above embodiment, specifically, the overall average electricity price of the current day and the next day is compared, and on which day the three time periods with the lowest electricity prices in two days are mainly concentrated; if the average electricity price of the current day is lower than the average electricity price of the next day, and the three time periods with the lowest electricity prices in two days are mainly concentrated on the current day, the charging station stores electric energy in the three time periods with the lowest electricity prices on the current day, and releases the stored electric energy in the several time periods with the highest electricity prices on the next day; if the average electricity price of the current day is lower than the average electricity price of the next day, but the three time periods with the lowest electricity prices in two days are mainly concentrated on the next day, the energy storage capacity of the charging station equipment needs to be considered; if the energy storage capacity of the charging station is low, the charging station performs appropriate energy storage regulation by performing its own low electricity price storage on the next day and releasing electric energy at a high electricity price; if the energy storage capacity of the charging station is high, the charging station performs appropriate energy storage regulation by storing electric energy at low electricity prices on the current day and the next day, and releasing electric energy at a high electricity price on the next day; if the average electricity price of the current day is higher than the average electricity price of the next day, and the three time periods with the lowest electricity prices in two days are mainly concentrated on the second day, the charging station performs its own low electricity price storage on the second day, and the charging station performs appropriate energy storage regulation by performing its own low electricity price storage on the second day, and releasing electric energy at a high electricity price on the second day; If the average electricity price of the day is higher than the average electricity price of the next day, but the three time periods with the lowest electricity prices in the two days are mainly concentrated on the day, if the energy storage capacity of the charging station is low, the storage capacity is lower than the electricity stored in the three time periods with the lowest electricity prices in the two days, then the charging station stores electricity in the three time periods with the lowest electricity prices on the day, and releases electricity in the corresponding time periods with the highest electricity prices on the next day, so as to achieve appropriate energy storage regulation; if the energy storage capacity of the charging station is high, that is, the energy storage capacity is higher than the electricity stored in the three time periods with the lowest electricity prices in the two days, then in addition to storing electricity in the three time periods with the lowest electricity prices on the day, the charging station also needs to store electricity in the corresponding time periods with the lowest electricity prices on the next day, and release electricity in the corresponding time periods with the highest electricity prices on the next day; according to the overall changes in electricity prices in each time period of the day and the next day and the energy storage capacity of the charging station equipment, the energy storage optimization scheduling of the charging station is carried out, and the most suitable energy storage optimization scheduling strategy is determined by the minimum value of the total electricity price calculation formula H on the next day, so that the operator can obtain the maximum economic benefits.

[0041] It should be understood that the above-mentioned embodiments are one or more embodiments of the present invention, and there are many other embodiments and variations thereof based on the present invention; the variations and modifications made by ordinary technicians in this industry through the present invention without making pioneering innovations all fall within the scope of protection of the present invention.

Claims

1. A method for optimizing energy storage scheduling for charging station operators, characterized in that: The method comprises the following steps: (1) Obtaining the charging power curve of the charging station: Obtain the historical data information of the daily vehicle flow and the charging station load since the installation of the charging station; Use the same segmentation point every day to divide the time, divide each day into multiple time periods, and extract the vehicle flow information and charging station load data information in the corresponding time period; The various time periods of the previous day and the traffic flow information corresponding to each time period in the historical data information are used as the input of the neural network, and the power consumption corresponding to each time period of the day is used as the output of the neural network. The neural network is trained to obtain the trained neural network model A; the various time periods of the current date and the traffic flow information corresponding to each time period are extracted, and the extracted information is input into the trained neural network model A to predict the load of the charging station in each time period of the next day; The average power value corresponding to each time period is calculated by the load of the charging station in each time period of the second day; the average power value is regarded as the power value corresponding to the charging station at the middle moment of each time period, and the power value coordinate corresponding to the middle moment is extracted (t i , Q i ), t i is the middle time corresponding to the i-th time period of the next day, Q i is the power of the charging station corresponding to the middle moment of the i-th time period on the second day; the extracted power coordinates are processed by the Newton interpolation algorithm to obtain the charging power curve f(t) of the charging station on the second day; (ii) Obtaining the average electricity price for each time period: extracting the population density, weather changes, and electricity price change data information for each time period in the history of the power grid coverage area corresponding to the charging station; training the neural network with the extracted data information to obtain the trained neural network model B; predicting the electricity price h(t j ), t j is the jth time period, h(t j ) represents the electricity price corresponding to the jth time period; (III) Optimize energy storage scheduling based on total electricity price: First, compare the overall changes in electricity prices in each time period of the day and the electricity prices in each time period of the next day; second, store electricity at a low price based on the overall changes in electricity prices and the energy storage capacity of the charging station equipment, and use the low-priced stored electricity to power the charging station when the price is high. The high-priced electricity in the corresponding time period is converted into low-priced electricity, and h'(t k ) represents the price of low-priced electricity that replaces high-priced electricity, k represents the kth time period is high-priced electricity and is replaced by low-priced electricity, t k represents the kth time period; finally, the total electricity price for the next day is calculated by using the low-priced stored electric energy to supply the charging station with electricity when the price is high, so as to realize the optimal scheduling of energy storage at the charging station; The total electricity price for the next day is calculated as follows: Among them, n is the total number of time periods into which the second day is divided, t∈[0,24), and the unit of t is hours; according to the calculation formula of the total electricity price of the second day, the minimum total electricity price is obtained, that is, the best energy storage optimization scheduling method is obtained.

2. The energy storage optimization scheduling method for charging station operators according to claim 1, characterized in that: The acquisition of historical data information on daily vehicle flow and charging station load since the installation of the charging station includes: acquiring daily vehicle flow information and charging station load data information in the power grid coverage area corresponding to the charging station; the charging station load refers to the amount of electricity provided by the charging station to the electric vehicle, that is, the power consumption of the charging station, including the amount of electricity directly converted through the power grid and the amount of electricity stored in the charging station.

3. The energy storage optimization scheduling method for charging station operators according to claim 1 is characterized in that: The above method uses each time period of the previous day and the traffic flow information corresponding to each time period in the historical data information as the input of the neural network, and the power consumption corresponding to each time period of the day as the output of the neural network, including: each time period of the previous day and the traffic flow information corresponding to each time period are represented by {(T i j－1 ,q i )}, where j represents the historical data information day is week j, j ranges from 1 to 7, and Sunday is considered as week 7; i represents the i-th time segment of week j-1, q i represents the traffic flow corresponding to the i-th time period; the power consumption corresponding to each time period of the day is represented by {P i j }, j means that the current day is week j, the size is 1 to 7, Sunday is regarded as week 7, i means the i-th time segment of week j, P i j represents the power consumption corresponding to the i-th time period of week j; the neural network refers to a neural network used to process time series prediction and analysis, and the selected neural networks include LSTM, GRU, RNN, Transformer and CNN.

4. The energy storage optimization scheduling method for charging station operators according to claim 1, characterized in that: The method of calculating the average power value corresponding to the charging station in each time period through the load of the charging station in each time period of the second day includes: obtaining the average power of the charging station in the corresponding time period by dividing the power consumption of the charging station in each time period of the second day by the total time of the corresponding time period; the Newton interpolation algorithm is a method for constructing polynomial interpolation, which aims to construct a polynomial function passing through these points through known data points. Therefore, the power value coordinates corresponding to the intermediate moments can be processed through the Newton interpolation algorithm to obtain the overall charging power curve of the charging station.

5. The energy storage optimization scheduling method for charging station operators according to claim 1 is characterized in that: The training of the neural network through the extracted data information includes: taking the population density and weather changes in each time period as the input of the neural network, taking the electricity price change data information in the corresponding time period as the output of the neural network, training the neural network, and obtaining the trained neural network model B; the neural network refers to a neural network used to process time series prediction and analysis, and the selected neural networks include LSTM, GRU, RNN, Transformer and CNN.

6. The energy storage optimization scheduling method for charging station operators according to claim 1, characterized in that: The comparison of the overall changes in electricity prices in various time periods of the day and electricity prices in various time periods of the next day includes: comparing the overall average electricity prices of the day and the next day, and on which day the three time periods with the lowest electricity prices in two days are mainly concentrated; if the average electricity price of the day is lower than the average electricity price of the next day, and the three time periods with the lowest electricity prices in two days are mainly concentrated on the day, then the charging station stores electricity in the three time periods with the lowest electricity prices on the day, and releases the stored electricity in the time periods with the highest electricity prices on the next day; if the average electricity price of the day is lower than the average electricity price of the next day, but the three time periods with the lowest electricity prices in two days are mainly concentrated on the second day, it is necessary to consider the energy storage capacity of the charging station equipment; if the energy storage capacity of the charging station is low, the charging station performs appropriate energy storage regulation by storing electricity at its own low price on the next day and releasing electricity at a high price on the next day; if the energy storage capacity of the charging station is high, the charging station performs appropriate energy storage regulation by storing electricity at low electricity prices on the day and the next day, and releasing electricity at high electricity prices on the next day; if the average electricity price of the day is higher than the average electricity price of the next day, and the three time periods with the lowest electricity prices in two days are mainly concentrated on the second day, then On the second day, the charging station stores electricity at a low price and releases electricity at a high price, and performs appropriate energy storage regulation. If the average electricity price of the day is higher than the average electricity price of the next day, but the three time periods with the lowest electricity prices in the two days are mainly concentrated on the day, if the energy storage capacity of the charging station is low, and the storage capacity is lower than the electricity stored in the three time periods with the lowest electricity prices in the two days, the charging station stores electricity in the three time periods with the lowest electricity prices on the day, and releases electricity in the corresponding time periods with the highest electricity prices on the next day, so as to achieve appropriate energy storage regulation. If the energy storage capacity of the charging station is high, that is, the energy storage capacity is higher than the electricity stored in the three time periods with the lowest electricity prices in the two days, in addition to storing electricity in the three time periods with the lowest electricity prices on the day, the charging station also needs to store electricity in the corresponding time periods with the lowest electricity prices on the next day, and release electricity in the corresponding time periods with the highest electricity prices on the next day. The energy storage optimization scheduling of the charging station is performed according to the overall changes in electricity prices in each time period of the day and the next day and the energy storage capacity of the charging station equipment, and the most suitable energy storage optimization scheduling strategy is determined by the minimum value of the second day's total electricity price calculation formula H, so that the operator can obtain the maximum economic benefits.

Images