An intelligent scheduling system and method for Internet of Things charging

Abstract

The present invention relates to the technical field of charging station scheduling, and discloses an intelligent scheduling system and method for Internet of Things charging. An intelligent scheduling system for Internet of Things charging includes: a first data acquisition module, which collects parameters of electric vehicles being charged and parameters of corresponding charging piles; a second data acquisition module, which constructs a load sequence and an electricity price sequence; a power grid load prediction module, which inputs the load sequence into a power grid load prediction model and outputs the power grid load in the next N time periods; a power grid electricity price prediction module, which inputs the electricity price sequence into the power grid electricity price prediction model and outputs the power grid electricity price in the next N time periods; a charging pile scheduling module, which generates a charging pile scheduling plan through an optimization algorithm; the present invention realizes intelligent scheduling of the output power of the charging piles in multiple time periods in the future through an optimization algorithm based on the predicted power grid load and power grid price, thereby minimizing the electricity price while ensuring maximization of charging efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of charging pile scheduling, and more specifically, to an intelligent scheduling system and method for Internet of Things charging. Background Art

[0002] With the rapid development of electric vehicles, the demand for charging infrastructure has also increased dramatically. Currently, charging stations generally adopt a static scheduling mode, that is, electric vehicles arbitrarily select idle charging piles at the charging station for charging. Due to the lack of adjustment of the output power of the charging piles, the output power of the charging piles is usually fixed. For example, the common 7kW AC charging piles or 60kW and 120kW DC fast charging piles cannot intelligently adjust the output power of the charging piles when facing the charging needs, battery status, grid load and electricity price changes of different electric vehicles, resulting in increased operating costs of the charging station and affecting the charging efficiency of electric vehicles. Summary of the Invention

[0003] The present invention provides an intelligent scheduling system and method for Internet of Things charging, which solves the technical problems in the above-mentioned background technology.

[0004] The present invention provides an intelligent scheduling system for Internet of Things charging, comprising:

[0005] A first data acquisition module, which is used to collect parameters of the electric vehicle being charged and parameters of the corresponding charging pile;

[0006] Electric vehicle parameters include: total battery capacity, remaining battery capacity, maximum allowable charging power, minimum allowable charging power, and estimated charging time;

[0007] The parameters of the charging pile include: rated power, maximum output power, minimum output power and charging efficiency;

[0008] The second data acquisition module is used to collect power grid load data, power grid electricity price data and meteorological data for a historical period, and construct load series and electricity price series respectively;

[0009] Meteorological data include: maximum temperature, maximum rainfall and maximum wind speed;

[0010] A power grid load forecasting module, which is used to input the load sequence into the power grid load forecasting model and output the power grid load in the next N time periods;

[0011] The grid electricity price prediction module is used to input the electricity price sequence into the grid electricity price prediction model and output the grid electricity price in the next N time periods;

[0012] The charging pile scheduling module is used to generate a charging pile scheduling plan through an optimization algorithm based on the parameters of the electric vehicles being charged, the parameters of the corresponding charging piles, and the grid load and grid electricity price in the next N time periods;

[0013] The charging pile scheduling plan is represented by the actual output power of the charging pile in the next N time periods, where N is a custom parameter and the time period is in units of 10 minutes.

[0014] Furthermore, the estimated charging time The calculation formula is as follows:

[0015] ;

[0016] in Indicates the target battery capacity, set to 100%. Indicates the remaining battery power, Q indicates the total battery capacity, It represents the rated power of the charging pile, and effic represents the charging efficiency, which is assigned a value of 0.92.

[0017] Furthermore, each sequence unit of the load sequence is represented by normalized grid load data and meteorological data.

[0018] Furthermore, each sequence unit of the electricity price sequence is represented by normalized grid electricity price data and meteorological data.

[0019] Furthermore, the power grid load forecasting model is constructed based on the GRU model. The number of hidden units of the power grid load forecasting model is consistent with the number of sequence units of the load sequence. The vector output by the last hidden unit is input into N first classifiers respectively, and the classification space of the N first classifiers corresponds to the power grid load in the next N time periods respectively.

[0020] Furthermore, the grid electricity price prediction model is constructed based on the GRU model. The number of hidden units in the grid electricity price prediction model is consistent with the number of sequence units in the electricity price sequence. The vector output by the last hidden unit is input into N second classifiers respectively. The classification spaces of the N second classifiers correspond to the grid electricity prices in the next N time periods respectively.

[0021] Furthermore, a charging pile scheduling plan is generated through an optimization algorithm, including the following steps:

[0022] Step S201: randomly generate an initialization population that meets the constraints, and initialize the current number of iterations to 0;

[0023] The number of individuals K in the initialized population is a custom parameter. Each individual is represented by a matrix code. The matrix code Matrix is ​​represented as follows:

[0024] ;

[0025] Where M represents the total number of charging piles in use, and N represents the total number of charging piles in the future time period. Indicates the actual output power of the 1st to Mth charging piles in the first time period in the future. Indicates the actual output power of the first charging pile from the first time period to the Nth time period in the future. Indicates the actual output power of the Mth charging pile in the Nth time period in the future;

[0026] Constraints include:

[0027] Constraint 1: ,in Indicates the actual output power of the charging pile at any time in the future. and They represent the minimum and maximum allowable charging power of electric vehicles, and Respectively represent the minimum output power and maximum output power of the charging pile;

[0028] Constraint 2: ,in Indicates the actual output power of the mth charging pile in the nth time period in the future, It represents the grid load in the nth time period in the future;

[0029] Constraint 3: ,in It represents the average value of the actual output power of the charging pile in the next N time periods. Indicates the rated power of the charging pile;

[0030] Step S202, calculating the fitness values ​​of all individuals in the initialized population through the objective function;

[0031] Step S203, traverse all individuals in the initialized population and generate a first random number between 0 and 1 each time. If the first random number is less than the coefficient of variation, the matrix code of the individual is updated using a random search strategy, otherwise proceed to step S204;

[0032] The coefficient of variation is a custom parameter;

[0033] Step S204: Generate a second random number in the range of 0 to 1, and if it is less than the balance coefficient, update the matrix code of the individual using the global search strategy; otherwise, update the matrix code of the individual using the local search strategy;

[0034] Traverse to the balance coefficient of the i-th individual The calculation formula is as follows:

[0035] ;

[0036] in Indicates the maximum value of fitness value, represents the fitness value of the i-th individual, and tanh represents the hyperbolic tangent function;

[0037] Step S205: The current number of iterations is added by 1, and if it is determined that the current number of iterations is greater than or equal to the maximum number of iterations, the matrix code of the individual with the largest fitness value in the initialized population is output as the charging pile scheduling plan, otherwise, steps S202 to S205 are repeated;

[0038] The maximum number of iterations is a custom parameter.

[0039] Furthermore, the objective function is calculated as follows:

[0040] ;

[0041] Where Fit represents the fitness value, Indicates the actual output power of the mth charging pile in the nth time period in the future, represents the grid electricity price in the nth time period in the future, 、 and They represent the target battery capacity of the electric vehicle corresponding to the mth charging pile, set to 100%, the remaining battery capacity and the total battery capacity, respectively. It represents the average value of the actual output power of the mth charging pile in the next N time periods. effic represents the charging efficiency and is assigned a value of 0.92. and They represent the customized first weight item and second weight item respectively, and the sum of the two is 1.

[0042] Furthermore, the calculation formula of the random search strategy is as follows:

[0043] ;

[0044] Where 1≤i≤K, 1≤t≤Max, where Max represents the maximum number of iterations, represents the matrix encoding of the i-th individual at the t+1-th iteration, Represents the matrix encoding of the individual with the largest fitness value at the t-th iteration, It represents the matrix encoding of the individual with the smallest fitness value at the t-th iteration, and rand represents a random number ranging from 0 to 1;

[0045] The calculation formula of the global search strategy includes:

[0046] ;

[0047] ;

[0048] Where 1≤j≤K, 1≤k≤K, i≠j≠k, offset represents the offset coefficient, and the value range of the offset coefficient is , arctanh represents the inverse hyperbolic tangent function, and denote the matrix codes of the jth and kth individuals at the tth iteration, represents the fitness value of the jth individual at the tth iteration, Indicates the minimum value of fitness value, represents the weight coefficient of the jth individual at the tth iteration, Represents the median fitness value of the initialized population at the tth iteration;

[0049] The calculation formula of the local search strategy includes:

[0050] ;

[0051] ;

[0052] in represents the matrix encoding of the ith individual at the tth iteration, convert represents the conversion coefficient, Express obedience A third random number from a uniform distribution, Express obedience A fourth random number from a uniform distribution, Indicates the fifth random number between 0 and 1.

[0053] The present invention provides an intelligent scheduling method for Internet of Things charging, comprising the following steps:

[0054] Step S301, collecting parameters of the electric vehicle being charged and parameters of the corresponding charging pile;

[0055] Step S302: collecting grid load data, grid electricity price data, and meteorological data for a historical period, and constructing a load series and an electricity price series respectively;

[0056] Step S303: input the load sequence into the power grid load forecasting model, and output the power grid load in the next N time periods;

[0057] Step S304: input the electricity price sequence into the grid electricity price prediction model, and output the grid electricity price in the next N time periods;

[0058] Step S305 , generating a charging pile scheduling plan through an optimization algorithm based on the parameters of the electric vehicle being charged, the parameters of the corresponding charging pile, and the grid load and grid electricity price in the next N time periods.

[0059] The beneficial effect of the present invention is that the present invention realizes intelligent scheduling of the output power of charging piles in multiple time periods in the future according to the predicted grid load and grid electricity price through an optimization algorithm, thereby minimizing the electricity price while ensuring maximum charging efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 It is a schematic diagram of an intelligent scheduling system for Internet of Things charging of the present invention;

[0061] Figure 2 This is a flow chart of the present invention for generating a charging pile scheduling plan through an optimization algorithm;

[0062] Figure 3 This is a flow chart of an intelligent scheduling method for Internet of Things charging of the present invention.

[0063] In the figure: a first data acquisition module 101, a second data acquisition module 102, a power grid load prediction module 103, a power grid electricity price prediction module 104, and a charging pile scheduling module 105. DETAILED DESCRIPTION

[0064] The subject matter described herein will now be discussed with reference to exemplary embodiments. It should be understood that these embodiments are discussed solely to enable those skilled in the art to better understand and implement the subject matter described herein, and that the functions and arrangements of the elements discussed may be varied without departing from the scope of this specification. Various examples may omit, substitute, or add various processes or components as needed. In addition, features described with respect to some examples may also be combined in other examples.

[0065] It should be noted that, unless otherwise defined, the technical or scientific terms used in one or more embodiments of the present invention should have the usual meanings understood by people with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar words used in one or more embodiments of the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprising" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, but do not exclude other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the object being described changes, the relative positional relationship may also change accordingly.

[0066] like Figures 1 to 3 As shown, an intelligent scheduling system for IoT charging includes:

[0067] A first data acquisition module 101, which is used to collect parameters of the electric vehicle being charged and parameters of the corresponding charging pile;

[0068] Electric vehicle parameters include: total battery capacity, remaining battery capacity, maximum allowable charging power, minimum allowable charging power, and estimated charging time;

[0069] The parameters of the charging pile include: rated power, maximum output power, minimum output power and charging efficiency;

[0070] The second data collection module 102 is used to collect the grid load data, grid electricity price data and meteorological data of the historical time period, and respectively construct the load series and electricity price series;

[0071] Meteorological data include: maximum temperature, maximum rainfall and maximum wind speed;

[0072] A power grid load forecasting module 103 is configured to input a load sequence into a power grid load forecasting model and output power grid loads in the next N time periods;

[0073] A grid electricity price prediction module 104 is used to input the electricity price sequence into a grid electricity price prediction model and output the grid electricity price in the next N time periods;

[0074] The charging pile scheduling module 105 is used to generate a charging pile scheduling plan through an optimization algorithm based on the parameters of the electric vehicles being charged, the parameters of the corresponding charging piles, and the grid load and grid electricity price in the next N time periods;

[0075] The charging pile scheduling plan is represented by the actual output power of the charging pile in the next N time periods, where N is a custom parameter and the time period is in units of 10 minutes. Preferably, N is set to 6.

[0076] In one embodiment of the present invention, the estimated charging time is The calculation formula is as follows:

[0077] ;

[0078] in Indicates the target battery capacity, set to 100%. Indicates the remaining battery power, Q indicates the total battery capacity, It represents the rated power of the charging pile, and effic represents the charging efficiency, which is assigned a value of 0.92.

[0079] For example, if the remaining battery power is 20%, the total battery capacity is 60kWh, and the rated power of the charging pile is 60kW, then according to the above calculation formula, the estimated charging time is approximately 0.87 hours (52 minutes).

[0080] In one embodiment of the present invention, each sequence unit of the load sequence is represented by normalized grid load data and meteorological data, that is, each sequence unit includes four dimensional values, namely grid load data, maximum temperature, maximum rainfall and maximum wind speed.

[0081] In one embodiment of the present invention, each sequence unit of the electricity price sequence is represented by normalized grid electricity price data and meteorological data, that is, each sequence unit includes four dimensional values, namely grid electricity price data, maximum temperature, maximum rainfall and maximum wind speed.

[0082] It should be noted that the normalization method can be Min-Max normalization or Z-Score normalization, which will not be described here.

[0083] In one embodiment of the present invention, the power grid load forecasting model is constructed based on the GRU model, the number of hidden units of the power grid load forecasting model is consistent with the number of sequence units of the load sequence, and the vector output by the last hidden unit is input into N first classifiers respectively, and the classification spaces of the N first classifiers respectively correspond to the power grid load in the next N time periods.

[0084] In one embodiment of the present invention, a grid electricity price prediction model is constructed based on a GRU model. The number of hidden units of the grid electricity price prediction model is consistent with the number of sequence units of the electricity price sequence. The vectors output by the last hidden unit are respectively input into N second classifiers, and the classification spaces of the N second classifiers respectively correspond to the grid electricity prices in the next N time periods.

[0085] In one embodiment of the present invention, the calculation formula of the power grid load forecasting model and the power grid electricity price forecasting model is the same, specifically including:

[0086] ;

[0087] Where 1≤t≤T, T represents the total number of hidden units, and Represent the vectors output by the t-th hidden unit and the t-1-th hidden unit, respectively. Assign a value of 0;

[0088] ;

[0089] in represents the update gate of the t-th hidden unit, Represents the weight parameter corresponding to the update gate, Represents the bias parameter corresponding to the update gate, Concat represents the concatenation function, and sigmoid represents the sigmoid activation function;

[0090] ;

[0091] in represents the reset gate of the t-th hidden unit, Represents the weight parameter corresponding to the reset gate, Indicates the bias parameter corresponding to the reset gate;

[0092] ;

[0093] in represents the candidate vector output by the tth hidden unit, represents the weight parameter corresponding to the candidate vector, Represents the bias parameter corresponding to the candidate vector, and tanh represents the hyperbolic tangent function.

[0094] It should be noted that the weight parameters and bias parameters in the power grid load forecasting model and the power grid electricity price forecasting model are all learnable hyperparameters. They can be trained by constructing load sequences and electricity price sequences based on historical data, which will not be elaborated here.

[0095] In one embodiment of the present invention, Figure 2 As shown in the figure, the charging pile scheduling plan is generated by the optimization algorithm, which includes the following steps:

[0096] Step S201: randomly generate an initialization population that meets the constraints, and initialize the current number of iterations to 0;

[0097] The number of individuals K in the initialized population is a custom parameter. Preferably, K is set to 10. Each individual is represented by a matrix code. The matrix code Matrix is ​​represented as follows:

[0098] ;

[0099] Where M represents the total number of charging piles in use, and N represents the total number of charging piles in the future time period. Indicates the actual output power of the 1st to Mth charging piles in the first time period in the future. Indicates the actual output power of the first charging pile from the first time period to the Nth time period in the future. Indicates the actual output power of the Mth charging pile in the Nth time period in the future;

[0100] Constraints include:

[0101] Constraint 1: ,in Indicates the actual output power of the charging pile at any time in the future. and They represent the minimum and maximum allowable charging power of electric vehicles, and Respectively represent the minimum output power and maximum output power of the charging pile;

[0102] Constraint 2: ,in Indicates the actual output power of the mth charging pile in the nth time period in the future, It represents the grid load in the nth time period in the future;

[0103] Constraint 3: ,in It represents the average value of the actual output power of the charging pile in the next N time periods. Indicates the rated power of the charging pile;

[0104] Step S202, calculating the fitness values ​​of all individuals in the initialized population through the objective function;

[0105] Step S203, traverse all individuals in the initialized population and generate a first random number between 0 and 1 each time. If the first random number is less than the coefficient of variation, the matrix code of the individual is updated using a random search strategy, otherwise proceed to step S204;

[0106] The coefficient of variation is a custom parameter. Preferably, the coefficient of variation is set to 0.03;

[0107] Step S204: Generate a second random number in the range of 0 to 1, and if it is less than the balance coefficient, update the matrix code of the individual using the global search strategy; otherwise, update the matrix code of the individual using the local search strategy;

[0108] Traverse to the balance coefficient of the i-th individual The calculation formula is as follows:

[0109] ;

[0110] in Indicates the maximum value of fitness value, represents the fitness value of the i-th individual, and tanh represents the hyperbolic tangent function;

[0111] Step S205: The current number of iterations is added by 1, and if it is determined that the current number of iterations is greater than or equal to the maximum number of iterations, the matrix code of the individual with the largest fitness value in the initialized population is output as the charging pile scheduling plan, otherwise, steps S202 to S205 are repeated;

[0112] The maximum number of iterations is a custom parameter. Preferably, the maximum number of iterations is set to 20.

[0113] In one embodiment of the present invention, the calculation formula of the objective function is as follows:

[0114] ;

[0115] Where Fit represents the fitness value, Indicates the actual output power of the mth charging pile in the nth time period in the future, represents the grid electricity price in the nth time period in the future, 、 and They represent the target battery capacity of the electric vehicle corresponding to the mth charging pile, set to 100%, the remaining battery capacity and the total battery capacity, respectively. It represents the average value of the actual output power of the mth charging pile in the next N time periods. effic represents the charging efficiency and is assigned a value of 0.92. and They represent the customized first weight item and second weight item respectively, and the sum of the two is 1.

[0116] In one embodiment of the present invention, the calculation formula of the random search strategy is as follows:

[0117] ;

[0118] Where 1≤i≤K, 1≤t≤Max, where Max represents the maximum number of iterations, represents the matrix encoding of the i-th individual at the t+1-th iteration, Represents the matrix encoding of the individual with the largest fitness value at the t-th iteration, It represents the matrix encoding of the individual with the smallest fitness value at the t-th iteration, and rand represents a random number ranging from 0 to 1;

[0119] The calculation formula of the global search strategy includes:

[0120] ;

[0121] ;

[0122] Where 1≤j≤K, 1≤k≤K, i≠j≠k, offset represents the offset coefficient, and the value range of the offset coefficient is , arctanh represents the inverse hyperbolic tangent function, and denote the matrix codes of the jth and kth individuals at the tth iteration, represents the fitness value of the jth individual at the tth iteration, Indicates the minimum value of fitness value, represents the weight coefficient of the jth individual at the tth iteration, Represents the median fitness value of the initialized population at the tth iteration;

[0123] The calculation formula of the local search strategy includes:

[0124] ;

[0125] ;

[0126] in represents the matrix encoding of the ith individual at the tth iteration, convert represents the conversion coefficient, Express obedience A third random number from a uniform distribution, Express obedience A fourth random number from a uniform distribution, Indicates the fifth random number between 0 and 1.

[0127] In one embodiment of the present invention, Figure 3 As shown, an intelligent scheduling method for IoT charging includes the following steps:

[0128] Step S301, collecting parameters of the electric vehicle being charged and parameters of the corresponding charging pile;

[0129] Step S302: collecting grid load data, grid electricity price data, and meteorological data for a historical period, and constructing a load series and an electricity price series respectively;

[0130] Step S303: input the load sequence into the power grid load forecasting model, and output the power grid load in the next N time periods;

[0131] Step S304: input the electricity price sequence into the grid electricity price prediction model, and output the grid electricity price in the next N time periods;

[0132] Step S305 , generating a charging pile scheduling plan through an optimization algorithm based on the parameters of the electric vehicle being charged, the parameters of the corresponding charging pile, and the grid load and grid electricity price in the next N time periods.

[0133] The above describes the embodiments of this embodiment, but this embodiment is not limited to the above specific implementation methods. The above specific implementation methods are merely illustrative and not restrictive. Ordinary technicians in this field can also make many forms based on the inspiration of this embodiment, all of which are protected by this embodiment.

Claims

1. An intelligent scheduling system for Internet of Things charging, characterized in that: include: A first data acquisition module, which is used to collect parameters of the electric vehicle being charged and parameters of the corresponding charging pile; Electric vehicle parameters include: total battery capacity, remaining battery capacity, maximum allowable charging power, minimum allowable charging power, and estimated charging time; The parameters of the charging pile include: rated power, maximum output power, minimum output power and charging efficiency; The second data acquisition module is used to collect power grid load data, power grid electricity price data and meteorological data for a historical period, and construct load series and electricity price series respectively; Meteorological data include: maximum temperature, maximum rainfall and maximum wind speed; A power grid load forecasting module, which is used to input the load sequence into the power grid load forecasting model and output the power grid load in the next N time periods; The grid electricity price prediction module is used to input the electricity price sequence into the grid electricity price prediction model and output the grid electricity price in the next N time periods; The charging pile scheduling module is used to generate a charging pile scheduling plan through an optimization algorithm based on the parameters of the electric vehicles being charged, the parameters of the corresponding charging piles, and the grid load and grid electricity price in the next N time periods; The charging pile scheduling plan is represented by the actual output power of the charging pile in the next N time periods, where N is a custom parameter and the time period is in 10-minute units. Generating a charging pile scheduling plan through an optimization algorithm includes the following steps: Step S201: randomly generate an initialization population that meets the constraints, and initialize the current number of iterations to 0; The number of individuals K in the initialized population is a custom parameter. Each individual is represented by a matrix code. The matrix code Matrix is ​​represented as follows: ； Where M represents the total number of charging piles in use, and N represents the total number of charging piles in the future time period. Indicates the actual output power of the 1st to Mth charging piles in the first time period in the future. Indicates the actual output power of the first charging pile from the first time period to the Nth time period in the future. Indicates the actual output power of the Mth charging pile in the Nth time period in the future; Constraints include: Constraint 1: ,in Indicates the actual output power of the charging pile at any time in the future. and They represent the minimum and maximum allowable charging power of electric vehicles, and Respectively represent the minimum output power and maximum output power of the charging pile; Constraint 2: ,in Indicates the actual output power of the mth charging pile in the nth time period in the future, It represents the grid load in the nth time period in the future; Constraint 3: ,in It represents the average value of the actual output power of the charging pile in the next N time periods. Indicates the rated power of the charging pile; Step S202, calculating the fitness values ​​of all individuals in the initialized population through the objective function; Step S203, traverse all individuals in the initialized population and generate a first random number between 0 and 1 each time. If the first random number is less than the coefficient of variation, the matrix code of the individual is updated using a random search strategy. Otherwise, proceed to step S204; The coefficient of variation is a custom parameter; Step S204: Generate a second random number in the range of 0 to 1, and if it is less than the balance coefficient, update the matrix code of the individual using the global search strategy; otherwise, update the matrix code of the individual using the local search strategy; Traverse to the balance coefficient of the i-th individual The calculation formula is as follows: ； in Indicates the maximum value of fitness value, represents the fitness value of the i-th individual, and tanh represents the hyperbolic tangent function; Step S205: The current number of iterations is added by 1, and if it is determined that the current number of iterations is greater than or equal to the maximum number of iterations, the matrix code of the individual with the largest fitness value in the initialized population is output as the charging pile scheduling plan, otherwise, steps S202 to S205 are repeated; The maximum number of iterations is a custom parameter.

2. The intelligent scheduling system for Internet of Things charging according to claim 1, characterized in that: Estimated charging time The calculation formula is as follows: ； in Indicates the target battery capacity, set to 100%. Indicates the remaining battery power, Q indicates the total battery capacity, It represents the rated power of the charging pile, and effic represents the charging efficiency, which is assigned a value of 0.

92.

3. The intelligent scheduling system for Internet of Things charging according to claim 1, characterized in that: Each sequence unit of the load sequence is represented by normalized grid load data and meteorological data.

4. The intelligent scheduling system for Internet of Things charging according to claim 1, characterized in that: Each sequence unit of the electricity price series is represented by normalized grid electricity price data and meteorological data.

5. The intelligent scheduling system for Internet of Things charging according to claim 1, characterized in that: The power grid load forecasting model is constructed based on the GRU model. The number of hidden units in the power grid load forecasting model is consistent with the number of sequence units in the load sequence. The vector output by the last hidden unit is input into N first classifiers respectively. The classification space of the N first classifiers corresponds to the power grid load in the next N time periods respectively.

6. The intelligent scheduling system for Internet of Things charging according to claim 1, characterized in that: The grid electricity price prediction model is built based on the GRU model. The number of hidden units in the grid electricity price prediction model is consistent with the number of sequence units in the electricity price sequence. The vector output by the last hidden unit is input into N second classifiers respectively. The classification space of the N second classifiers corresponds to the grid electricity prices in the next N time periods respectively.

7. The intelligent scheduling system for Internet of Things charging according to claim 1, characterized in that: The objective function is calculated as follows: ； Where Fit represents the fitness value, Indicates the actual output power of the mth charging pile in the nth time period in the future, represents the grid electricity price in the nth time period in the future, 、 and They represent the target battery capacity of the electric vehicle corresponding to the mth charging pile, set to 100%, the remaining battery capacity and the total battery capacity, respectively. It represents the average value of the actual output power of the mth charging pile in the next N time periods. effic represents the charging efficiency and is assigned a value of 0.

92. and They represent the customized first weight item and second weight item respectively, and the sum of the two is 1.

8. The intelligent scheduling system for Internet of Things charging according to claim 1, characterized in that: The calculation formula of the random search strategy is as follows: ； Where 1≤i≤K, 1≤t≤Max, where Max represents the maximum number of iterations, represents the matrix encoding of the i-th individual at the t+1-th iteration, Represents the matrix encoding of the individual with the largest fitness value at the t-th iteration, It represents the matrix encoding of the individual with the smallest fitness value at the t-th iteration, and rand represents a random number ranging from 0 to 1; The calculation formula of the global search strategy includes: ； ； Where 1≤j≤K, 1≤k≤K, i≠j≠k, offset represents the offset coefficient, and the value range of the offset coefficient is , arctanh represents the inverse hyperbolic tangent function, and denote the matrix codes of the jth and kth individuals at the tth iteration, represents the fitness value of the jth individual at the tth iteration, Indicates the minimum value of fitness value, represents the weight coefficient of the jth individual at the tth iteration, Represents the median fitness value of the initialized population at the tth iteration; The calculation formula of the local search strategy includes: ； ； in represents the matrix encoding of the ith individual at the tth iteration, convert represents the conversion coefficient, Express obedience A third random number from a uniform distribution, Express obedience A fourth random number from a uniform distribution, Indicates the fifth random number between 0 and 1.

9. An intelligent scheduling method for Internet of Things charging, executing an intelligent scheduling system for Internet of Things charging according to any one of claims 1 to 8, characterized in that: The following steps are involved: Step S301, collecting parameters of the electric vehicle being charged and parameters of the corresponding charging pile; Step S302: collecting grid load data, grid electricity price data, and meteorological data for a historical period, and constructing a load series and an electricity price series respectively; Step S303: input the load sequence into the power grid load forecasting model, and output the power grid load in the next N time periods; Step S304: input the electricity price sequence into the grid electricity price prediction model, and output the grid electricity price in the next N time periods; Step S305 , generating a charging pile scheduling plan through an optimization algorithm based on the parameters of the electric vehicle being charged, the parameters of the corresponding charging pile, and the grid load and grid electricity price in the next N time periods.

Images