Electric vehicle charging control method for new energy charging pile

Abstract

The present invention belongs to the technical field of management and control of new energy charging systems, and in particular relates to a method for controlling electric vehicle charging for new energy charging piles. The method comprises the following steps: S1. Acquiring regional electric vehicle charging load data and drawing a time-varying curve of electric vehicle charging load within a charging cycle; S2. Establishing an objective function and constraints based on a time-of-use electricity price strategy; S3. Acquiring initial load status information and planned time-of-use electricity price data, and periodically retrieving access status; S4. Determining the optimal time-of-use electricity price, and charging electric vehicles in an orderly charging sequence, updating the charging status and load information of the new energy charging pile, and returning to step three; S5. Recording data as initialization data for the next power supply cycle of the time period. The present invention is conducive to reducing the charging load of electric vehicles within a power supply cycle, reducing the load impact of the operation of the new energy charging pile on the power grid, and improving the charging control effect of the new energy charging pile.

Description

Technical Field

[0001] The present invention belongs to the technical field of new energy charging system management and control, and in particular relates to an electric vehicle charging control method for a new energy charging pile. Background Art

[0002] With the vigorous promotion of new energy vehicles, the number of necessary charging facilities has also increased rapidly. The large number of newly added charging piles has had a significant impact on the operating load of the power grid. The power output of a single new energy charging pile can reach over 100 megawatts, and multiple charging piles are usually concentrated in a specific area. During peak charging times, the charging load generated by multiple new energy charging piles running simultaneously can significantly affect the operation of the regional power grid. To ensure the stability of power distribution and power supply, it is hoped that new energy vehicles can be connected to charging piles in a sequential manner. At the same time, when the electric vehicle's power level can meet the next trip plan, it can be actively withdrawn from the charging sequence. This can minimize the peak charging load while ensuring that as many electric vehicles as possible have access to charging opportunities, thereby ensuring continuous and stable power transmission. However, in reality, electric vehicle owners often demand immediate charging upon arrival at the charging station and maximize the power within a limited time. To encourage electric vehicle owners to cooperate with orderly charging plans and reduce the high load and recycling problems caused by excessive occupancy and centralized charging, active or passive guidance methods such as automatic charging control and time-of-use electricity price regulation have been proposed to improve this situation. However, overall, due to the complex operating trajectories of a large number of electric vehicles, the effectiveness of current guidance and control methods remains poor. Summary of the Invention

[0003] The purpose of the present invention is to propose an electric vehicle charging control method for new energy charging piles based on the idea of ​​time-of-use electricity prices, which guides and promotes the orderly and uniform charging of electric vehicle groups, reduces the average charging load of electric vehicles in each regulation cycle, and ultimately reduces the charging load in the power supply cycle.

[0004] To achieve the above objectives, the present invention adopts the following technical solutions.

[0005] A method for controlling charging of an electric vehicle using a new energy charging pile comprises the following steps:

[0006] Step 1: Obtain regional electric vehicle charging load data and draw a time variation curve of electric vehicle charging load during the charging cycle; divide the charging cycle equally to obtain n control cycles. In each control cycle, the electric vehicle charging load demand ΔQ i and electricity price change Δp i The change law is expressed as ΔQ i =ε i ·Δp i , where εi is the elasticity coefficient of charging load demand; defines the correlation coefficient of electricity prices in adjacent regulation cycles where ΔQ i is the change in charging load demand during the ith regulation cycle, Q′ is the original charging load demand during the ith regulation cycle;

[0007] get

[0008] where p i ′ refers to the initial electricity price of the i-th regulation cycle; the electricity demand of electric vehicles after the i-th regulation cycle is expressed as Q i ″=Q i ′(1+ε ij Δp j / p′ j ); the charging load demand after the i-th regulation cycle is expressed as Q i ″=Q i ′(2+ε ii Δp i / p i ′+ε ij Δp j / p j ′);

[0009] Step 2: Establish the electric vehicle charging control objective function of the new energy charging pile:

[0010]

[0011] Where λ1 and λ2 are weight coefficients, f max and f min are the maximum charging load and the minimum charging load respectively, T is the duration of the regulation cycle, E m is the charging power of the mth electric vehicle connected to the controller, p t is the charging electricity price at time t, Refers to the charge state index;

[0012] Identify the constraints:

[0013]

[0014] Among them, K1 means that the power of the electric vehicle after charging is not less than the power expected by the electric vehicle owner, K2 means that the power of the electric vehicle after charging is not more than the rated power of the vehicle, K3 means that the charging time required for the electric vehicle in each stage does not exceed its actual parking time, K4 means that the actual charging load of the electric vehicle before and after regulation should be consistent, and K5 means that the total charging cost of the electric vehicle after regulation is lower than the charging cost required for the electric vehicle before regulation; T in Indicates the total time required for charging, Tin Indicates the total charging time, T in,m It represents the total time required to charge the mth electric vehicle, Soc m,t It refers to the state of charge of the mth electric vehicle at time t, Soc m,last Refers to the expected state of charge in the mth electric vehicle cycle, C m ′ refers to the rated capacity of the mth electric vehicle battery; T leave,m and T arrive,m They refer to the departure time and arrival time of the mth electric vehicle respectively; represents the total charging load of electric vehicles before regulation, Indicates the total charging load of electric vehicles after regulation

[0015] Step 3: Continuously obtain the initial load status information of the regional electric vehicle charging piles and the planned time-of-use electricity price data, periodically retrieve the electric vehicle access status, and when a new electric vehicle accesses, read the electric vehicle's current power, electric vehicle battery capacity, electric vehicle battery rated capacity, and electric vehicle arrival time; establish an electric vehicle charging expectation interaction mechanism to obtain the expected charging power target given by the user and the electric vehicle's planned departure time;

[0016] Based on the above steps, the required charging time and expected charging cost of the electric vehicle after the electric vehicle is added to the current ordered charging sequence are calculated; the charging time and expected charging cost information of the electric vehicle are output, and the user confirms whether to enter the ordered charging sequence through the electric vehicle charging expectation interaction mechanism; if the user chooses to join the ordered charging, the electric vehicle is added to the ordered charging sequence; if the user chooses not to join the ordered charging, the electric vehicle is charged according to the temporary charging rules;

[0017] Step 4: Based on the changes in the control cycle, determine the optimal time-of-use electricity price within each control cycle based on the current electric vehicle charging load status, the objective function, and the constraints, charge the electric vehicle in an orderly charging sequence, update the charging status and load information of the new energy charging pile, and return to step 3;

[0018] Step 5: loop step 3 and step 4, record the time-of-use electricity price strategy and charging load in different control stages, record the time-of-use electricity price data of electric vehicle charging in the power supply cycle, and use it as the initialization data for the next power supply cycle of the time cycle.

[0019] In a further improvement or preferred embodiment of the aforementioned electric vehicle charging control method for a new energy charging pile, the step 1 further includes a step for realizing electric vehicle charging load prediction, specifically including:

[0020] (1) Collect and analyze historical data on electric vehicle charging load and its associated factors during the power supply cycle, determine the original index data of the associated factors, and classify them according to the relationship between the associated factor index and the electric vehicle charging load, the classification including extremely large factors, extremely small factors, intermediate factors, and interval factors; extremely large factors refer to factors whose larger factor index leads to smaller electric vehicle charging load; extremely small factors refer to factors whose smaller factor index leads to smaller electric vehicle charging load; intermediate factors refer to factors whose factor index approaches the optimal value and whose electric vehicle charging load is smaller; interval factors refer to factors whose factor index is in the optimal interval and whose electric vehicle charging load is the smallest;

[0021] (2) transform any extremely small factor index, intermediate factor index, and interval factor index into an extremely large factor index;

[0022] (3) For each regulation cycle, extract the average charging load of electric vehicles and related factor indicators in each regulation cycle from the historical data of the same period

[0023] Get the electric vehicle charging load sequence Q in the same regulation cycle in the historical data of the same period * ={Q1,Q2...Q i ...Q I};

[0024] Obtain the index sequence X of the related factors in each regulation cycle in the historical data of the same period * ={X 1,V ,X 2,V ...X i, V ...X I,V};

[0025] Where I is the total number of historical data; Q i Refers to the standardized value of the i-th historical data of the average charging load of electric vehicles during the regulation period; X i,V ={x′ i1 ,x′ i2 ...x′ iv ...x′ iV}, V refers to the total number of categories of associated elements, x′ iv It refers to the standardized value of the vth correlation factor index value in the i-th historical control cycle;

[0026] (4) Determine the correlation between the average charging load of electric vehicles and related factor indicators during each regulation cycle Where β is the resolution coefficient;

[0027] (5) According to the size of the correlation index, several main correlation factor indicators with the greatest correlation with the average charging load of electric vehicles in each regulation cycle are determined respectively, and a long short-term memory neural network model prediction model is established with the main correlation factor indicators as input and the average charging load of electric vehicles as output; the historical data of the average charging load of electric vehicles and the correlation factor indicators in each regulation cycle are used as the original data, and a training data set and a verification data set are established to train the long short-term memory neural network model prediction model until the model output error meets the requirements, and the trained prediction model is obtained.

[0028] (6) The related factor indicators within the control period to be predicted are output to the prediction model to obtain the prediction results of the average charging load of electric vehicles within the control period.

[0029] A further improvement or preferred embodiment of the aforementioned electric vehicle charging control method for a new energy charging pile, wherein the step (2) specifically refers to: for any extremely small element index sequence {x a}, a=1,2....A, A is the total number of very small element indicators, the indicator value x a Replace with x a,max -x a , converting it into a very large index sequence {x a,max -x a}, where x a,max It refers to the element index sequence {x a}'s maximum indicator value;

[0030] For any intermediate element index sequence {x b}, b=1,2....B, B is the total number of intermediate factor indicators, and the indicator value x b Replace with where x b,best It refers to the element index sequence {x b}'s best indicator value;

[0031] For any interval-type element index sequence {x c}, c=1,2....C, C is the total number of interval factor indicators, the indicator value x c Replace with Among them, c1 and c2 refer to the end values ​​of the optimal interval [c1, c2] of the interval factor index.

[0032] A further improvement or preferred embodiment of the aforementioned electric vehicle charging control method for a new energy charging pile, in step (3), Q i ′ refers to the i-th historical data of the average charging load of electric vehicles during the regulation period; x ivIt refers to the vth correlation factor index value in the i-th historical control cycle.

[0033] A further improvement or preferred implementation of the aforementioned electric vehicle charging control method for a new energy charging pile, wherein the electric vehicle charging expected interaction mechanism refers to establishing an interactive terminal for user data input through an in-vehicle APP, a mobile phone APP, or a new energy charging pile. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a flow chart of an electric vehicle charging control method for a new energy charging pile;

[0035] Figure 2 It is a schematic diagram of the relationship between electricity prices and changes in demand when time-of-use electricity prices are adjusted. DETAILED DESCRIPTION

[0036] The present invention is described in detail below with reference to specific embodiments.

[0037] The electric vehicle charging control method for new energy charging piles of the present invention is mainly used to provide a charging control method that can be used to guide electric vehicle owners to use new energy charging piles for orderly and scientific charging, effectively reduce the charging load of electric vehicles during the power supply cycle, reduce the load impact of the operation process of the new energy charging pile on the power grid, and improve the stability of the power grid and the new energy charging process. The charging method is based on the time-of-use electricity price guidance idea, adjusts the time-of-use electricity price by analyzing the expected size of the charging load in different regulation cycles, and establishes a control system with lower charging load size and lower charging cost as the goal, so as to encourage more electric vehicle owners to actively participate in orderly automatic charging and improve the charging control effect of the new energy electric pile.

[0038] The specific process is as follows Figure 1 As shown, the following steps are included:

[0039] Step 1: Obtain regional electric vehicle charging load data and draw a time variation curve of electric vehicle charging load during the charging cycle;

[0040] When the number of electric vehicles in operation is stable, the electric vehicle charging load in a fixed area remains basically unchanged during the power supply cycle at the same time. Therefore, the load data can be determined based on the historical data of the regional electric vehicle charging load.

[0041] In this application, in order to achieve peak shifting and leveling, and at the same time facilitate more accurate regulation of periodic load and time-of-use electricity prices, the charging cycle is evenly divided according to the preset time-of-use electricity price adjustment cycle to obtain N regulation cycles. Within each regulation cycle, based on the electricity price regulation strategy, the demand decreases when the electricity price increases. After reaching the limit value, the demand no longer changes significantly with the price change. Its change curve is as follows Figure 2As shown; in the nth regulation cycle, the electric vehicle charging load demand ΔQ n and electricity price change Δp n The change law of can be expressed as ΔQ n =ε n ·Δp n , where ε n is the elasticity coefficient of charging load demand; on this basis, the correlation coefficient of electricity prices in adjacent regulation cycles is defined where ΔQ n is the change in charging load demand during the nth regulation cycle, Q n ′ is the original charging load demand of the nth regulation cycle;

[0042] get

[0043] where p n ′ refers to the initial electricity price of the nth regulation cycle; the electricity demand of electric vehicles after the nth regulation cycle can be expressed as Q n ″=Q′ n (1+ε nn Δp n / p′ n ); the charging load demand after the nth regulation cycle can be expressed as Q n ″=Q n ′(2+ε nn Δp n / p n ′+ε nn Δp n / p n ′);

[0044] In particular, the above-mentioned electric vehicle charging load period division scheme is based on historical charging load data, and is therefore suitable for areas with relatively stable regional vehicle flow and operating conditions. However, in some cases, when the vehicle flow varies greatly and the vehicle operating conditions are relatively complex, the spatiotemporal distribution of the electric vehicle charging load in each power supply and distribution cycle may not be consistent, and each may have different load sections. To facilitate the determination of regional electric vehicle charging load data in such cases, this application provides a method for realizing electric vehicle charging load prediction, which specifically includes the following steps:

[0045] (1) Original data collection

[0046] Collect and analyze historical data on electric vehicle charging load and its associated factors during the power supply cycle, determine original indicator data of the associated factors, and classify them according to the relationship between the associated factor indicators and the electric vehicle charging load. The classification includes extremely large factors, extremely small factors, intermediate factors, and interval factors. An extremely large factor refers to a factor whose larger factor indicator results in a smaller electric vehicle charging load; an extremely small factor refers to a factor whose smaller factor indicator results in a smaller electric vehicle charging load; an intermediate factor refers to a factor whose factor indicator is closer to the optimal value and whose electric vehicle charging load is smaller; and an interval factor refers to a factor whose factor indicator is within the optimal interval and whose electric vehicle charging load is the smallest.

[0047] (2) Data preprocessing

[0048] For any extremely small element index sequence {x a}, a=1,2....A, A is the total number of very small element indicators, the indicator value x a Replace with x a,max -x a , converting it into a very large index sequence {x a,max -x a}, where x a,max It refers to the element index sequence {x a}'s maximum indicator value;

[0049] For any intermediate factor index sequence {x b}, b=1,2....B, B is the total number of intermediate factor indicators, and the indicator value x b Replace with where x b,best It refers to the element index sequence {x b}'s best indicator value;

[0050] For any interval-type element index sequence {x c}, c=1,2....C, C is the total number of interval factor indicators, the indicator value x c Replace with Among them, c1 and c2 refer to the end values ​​of the optimal interval [c1, c2] of the interval factor index;

[0051] (3) For each regulation cycle, extract the average charging load of electric vehicles and related factor indicators in each regulation cycle from the historical data of the same period

[0052] Get the electric vehicle charging load sequence Q in the same regulation cycle in the historical data of the same period * ={Q1,Q2...Q i ...Q I};

[0053] Obtain the index sequence X of the related factors in each regulation cycle in the historical data of the same period * ={X 1,V ,X 2,V ...X i, V ...X I,V}

[0054] Where I is the total number of historical data; Refers to the i-th historical data Q of the average charging load of electric vehicles during the regulation period i ′’s normalized value; X i,V ={x′ i1 ,x′ i2 ...x′ iv ...x′ iV}, V refers to the total number of categories of associated elements, It refers to the vth correlation factor index value x in the i-th historical control cycle iv The standardized value of

[0055] (4) Determine the correlation between the average charging load of electric vehicles and related factor indicators during each regulation cycle Where β is the resolution coefficient;

[0056] (5) According to the size of the correlation index, several main correlation factor indicators with the greatest correlation with the average charging load of electric vehicles in each regulation cycle are determined respectively, and a long short-term memory neural network model prediction model is established with the main correlation factor indicators as input and the average charging load of electric vehicles as output; the historical data of the average charging load of electric vehicles and the correlation factor indicators in each regulation cycle are used as the original data, and a training data set and a verification data set are established to train the long short-term memory neural network model prediction model until the model output error meets the requirements, and the trained prediction model is obtained.

[0057] (6) The related factor indicators within the control period to be predicted are output to the prediction model to obtain the prediction results of the average charging load of electric vehicles within the control period.

[0058] Step 2: Based on the above, in order to guide users to charge in an orderly manner in a time-sharing manner, it is necessary to establish a clear control target based on the electricity price control strategy. Generally speaking, the control target model can be determined by several objective functions. There are two goals in this application: one is to average the charging load within the charging cycle, which can be expressed as minimizing the difference between the maximum charging load and the minimum charging load within the charging cycle; the other is to minimize the total charging cost of electric vehicle owners, so as to maximize the total number of electric vehicles that actively participate in the periodic charging load control and improve the control effect;

[0059] The objective function can be expressed as:

[0060]

[0061] Where λ1 and λ2 are weight coefficients, f max and f min are the maximum charging load and the minimum charging load respectively, T is the duration of the regulation cycle, E m is the charging power of the mth electric vehicle connected to the controller, p t is the charging electricity price at time t, It refers to the charging state index of the mth electric vehicle at time t;

[0062] At the same time, in order to ensure the effective charging load demand and orderly charging of electric vehicles within the power supply cycle, the following constraints should also be met:

[0063]

[0064] Among them, K1 indicates that the power of the electric vehicle after charging in each regulation cycle is not less than the power expected by the electric vehicle owner, K2 indicates that the power of the electric vehicle after charging in each regulation cycle does not exceed the rated power of the vehicle, K3 indicates that the charging time required for the electric vehicle in each stage does not exceed its actual parking time, K4 indicates that the actual charging load of the electric vehicle before and after regulation should remain the same, and K5 indicates that the total charging cost of the electric vehicle after regulation is lower than the charging cost required for the electric vehicle before regulation; T in Indicates the total time required for charging, T in Indicates the total charging time, T in,m It represents the total time required to charge the mth electric vehicle, Soc m,t It refers to the state of charge of the mth electric vehicle at time t, Soc m,last Refers to the expected state of charge in the mth electric vehicle cycle, C m ′ refers to the rated capacity of the mth electric vehicle battery; T leave,m and T arrive,m They refer to the departure time and arrival time of the mth electric vehicle respectively; represents the total charging load of electric vehicles before regulation, represents the total charging load of electric vehicles after regulation; p n ′ refers to the electricity price before the regulation in the nth regulation cycle, Q n ′ refers to the electric vehicle charging load before the regulation of the nth regulation cycle, p n It refers to the electricity price in the nth regulation cycle;

[0065] Step 3: Continuously obtain the initial load status information of the regional electric vehicle charging piles and the planned time-of-use electricity price data, periodically search the electric vehicle access status, and when a new electric vehicle accesses, read the electric vehicle's current power, electric vehicle battery capacity, electric vehicle battery rated capacity, and electric vehicle arrival time;

[0066] Establish an interactive mechanism for electric vehicle charging expectations to obtain the expected charging power target and the planned departure time of the electric vehicle given by the user;

[0067] The electric vehicle charging expectation interaction mechanism refers to the establishment of a data input interaction terminal or tool through an in-vehicle APP or mobile phone APP or a new energy charging pile interactive panel, which is used to obtain the aforementioned expected information collection and upload;

[0068] Calculating the required charging time and expected charging cost of the electric vehicle after the electric vehicle is added to the current ordered charging sequence based on the aforementioned steps; outputting the electric vehicle charging time and expected charging cost information to the user for confirmation via the electric vehicle charging expectation interaction mechanism whether to enter the ordered charging sequence;

[0069] If the user chooses to join the orderly charging, they will be included in the newly added orderly charging sequence. If the user chooses not to join the orderly charging, they will be charged according to the temporary charging rules;

[0070] Step 4: Based on the changes in the control cycle, determine the optimal time-of-use electricity price within each control cycle based on the current electric vehicle charging load status, the objective function, and the constraints, charge the electric vehicle in an orderly charging sequence, update the charging status and load information of the new energy charging pile, and return to step 3;

[0071] Step 5: loop step 3 and step 4, record the time-of-use electricity price strategy and charging load in different control stages, record the time-of-use electricity price data of electric vehicle charging in the power supply cycle, and use it as the initialization data for the next power supply cycle of the time cycle.

[0072] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the scope of protection of the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the essence and scope of the technical solutions of the present invention.

Claims

1. A method for controlling electric vehicle charging using a new energy charging pile, characterized in that: The steps include: Step 1: Obtain the regional electric vehicle charging load data and draw the electric vehicle charging load time variation curve during the charging cycle; divide the charging cycle equally to obtain In each regulation cycle, the electric vehicle charging load demand Changes in electricity prices The changing law of ,in is the elasticity coefficient of charging load demand; defines the correlation coefficient of electricity prices in adjacent regulation cycles ;in It is The change in charging load demand during a regulation cycle is: It is The original charging load demand of a regulation cycle; get ; in It refers to the The initial electricity price of the first regulation cycle; The charging load demand after a regulation cycle is expressed as ; Step 2: Establish the electric vehicle charging control objective function of the new energy charging pile: ； in and is the weight coefficient, are the maximum charging load and the minimum charging load, is the duration of the regulation cycle, It is the first The charging power of electric vehicles, yes The charging price at the time, Refers to the charge state index; Identify the constraints: ； in It means that the electric vehicle's power level after charging is not lower than the electric vehicle owner's expectation. It means that the electric vehicle's power does not exceed the rated power after charging. It means that the charging time required for electric vehicles at each stage does not exceed their actual parking time. This means that the actual charging load of electric vehicles should remain consistent before and after regulation. It means that the total charging cost of electric vehicles after regulation is lower than the charging cost of electric vehicles before regulation; Indicates the total time required for charging. Indicates the The total time required to charge an electric vehicle, It refers to the electric vehicles The current state of charge, It refers to the The expected state of charge during an electric vehicle cycle, It refers to the The rated capacity of an electric vehicle battery; They refer to The departure and arrival times of electric vehicles; represents the total charging load of electric vehicles before regulation, Indicates the total charging load of electric vehicles after regulation; Step 3: Continuously obtain the initial load status information of the regional electric vehicle charging piles and the planned time-of-use electricity price data, periodically retrieve the electric vehicle access status, and when a new electric vehicle accesses, read the electric vehicle's current power, electric vehicle battery capacity, electric vehicle battery rated capacity, and electric vehicle arrival time; establish an electric vehicle charging expectation interaction mechanism to obtain the expected charging power target given by the user and the electric vehicle's planned departure time; Based on the above steps, the required charging time and expected charging cost of the electric vehicle after the electric vehicle is added to the current ordered charging sequence are calculated; the charging time and expected charging cost information of the electric vehicle are output, and the user confirms whether to enter the ordered charging sequence through the electric vehicle charging expectation interaction mechanism; if the user chooses to join the ordered charging, the electric vehicle is added to the ordered charging sequence; if the user chooses not to join the ordered charging, the electric vehicle is charged according to the temporary charging rules; Step 4: Based on the changes in the control cycle, determine the optimal time-of-use electricity price within each control cycle based on the current electric vehicle charging load status, the objective function, and the constraints, charge the electric vehicle in an orderly charging sequence, update the charging status and load information of the new energy charging pile, and return to step 3; Step 5: loop step 3 and step 4, record the time-of-use electricity price strategy and charging load in different control stages, record the time-of-use electricity price data of electric vehicle charging in the power supply cycle, and use it as the initialization data for the next power supply cycle of the time cycle.

2. The electric vehicle charging control method for a new energy charging pile according to claim 1, characterized in that: The step 1 also includes steps for realizing electric vehicle charging load prediction, specifically including: (1) Collect and analyze historical data of electric vehicle charging load and its related factors during the power supply cycle, determine the original index data of the related factors, and classify them according to the relationship between the related factor index and the electric vehicle charging load. The classification includes extremely large factors, extremely small factors, intermediate factors, and interval factors; extremely large factors refer to factors whose larger factor index leads to smaller electric vehicle charging load; extremely small factors refer to factors whose smaller factor index leads to smaller electric vehicle charging load; intermediate factors refer to factors whose factor index approaches the optimal value and whose electric vehicle charging load is smaller; interval factors refer to factors whose factor index is in the optimal interval and whose electric vehicle charging load is the smallest; (2) Transform any extremely small factor index, intermediate factor index, and interval factor index into an extremely large factor index; (3) For each regulation cycle, extract the average charging load of electric vehicles and related factor indicators in each regulation cycle in the historical data of the same period, and obtain the charging load sequence of electric vehicles in the same regulation cycle in the historical data of the same period. ; Obtain the indicator sequence of related factors in each regulation cycle in the historical data of the same period ;in is the total amount of historical data; Refers to the average charging load of electric vehicles during the regulation period. Normalized value of historical data; , is the total number of categories of associated elements, It refers to the The first regulation cycle in the same historical period The standardized value of the index value of each associated factor; (4) Determine the correlation between the average charging load of electric vehicles and related factor indicators during each regulation period ;in is the resolution coefficient; (5) According to the size of the correlation index, several main correlation factor indicators with the greatest correlation with the average charging load of electric vehicles in each regulation cycle are determined respectively, and a long short-term memory neural network model prediction model is established with the main correlation factor indicators as input and the average charging load of electric vehicles as output; the historical data of the average charging load of electric vehicles and the correlation factor indicators in each regulation cycle are used as the original data, and a training data set and a validation data set are established to train the long short-term memory neural network model prediction model until the model output error meets the requirements, and the trained prediction model is obtained; (6) Output the related factor indicators within the regulation period to be predicted into the prediction model to obtain the prediction results of the average charging load of electric vehicles within the regulation period.

3. The electric vehicle charging control method for a new energy charging pile according to claim 2, characterized in that: The step (2) specifically refers to: for any extremely small element index sequence , , is the total number of very small element indicators, and the indicator value Replace with , converting it into a very large index sequence ,in It refers to the element index sequence The maximum index value of For any intermediate factor index sequence , , is the total number of intermediate factor indicators, and the indicator value Replace with ,in It refers to the element index sequence The best indicator value of For any interval factor index sequence , , is the total number of interval factor indicators, and the indicator value Replace with ,in Refers to the optimal interval of interval-type factor indicators The end value of .

4. The electric vehicle charging control method for a new energy charging pile according to claim 2, characterized in that: In the step (3), , Refers to the average charging load of electric vehicles during the regulation period. Historical data; , It refers to the The first regulation cycle in the same historical period The value of the associated factor indicator.

5. The electric vehicle charging control method for a new energy charging pile according to claim 1, characterized in that: The electric vehicle charging expected interaction mechanism refers to establishing an interactive terminal that enables user data input through an in-vehicle APP, a mobile phone APP, or a new energy charging pile.

Images