Service area electric vehicle charging resource allocation guiding method in Internet of Things environment

Abstract

The invention discloses a service area electric vehicle charging resource distribution guiding method in an Internet of Things environment, and belongs to the field of intelligent service area systems and electric vehicle charging distribution guiding. According to the method, the problem that the optimization effect of an existing method on service area electric vehicle charging resource allocation guidance is poor is solved. The information of the vehicles and the service area is collected in real time, the number of the charged vehicles in the future time period is predicted in combination with the collected information, intelligent distribution of the charging resources is achieved through the gray wolf algorithm based on multi-target optimization, and guiding information can be provided for the vehicles and the service area according to the distribution result of the charging resources and the prediction result of the number of the charged vehicles. According to the method, global overall planning of the electric vehicle charging resources on the whole trunk line is achieved, the method can adapt to dynamic changes of vehicle distribution along the line and charging requirements, and the optimization effect of charging resource distribution guidance can be improved under fixed charging resource configuration. The method can be applied to charging distribution guidance of service areas and electric automobiles.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of intelligent service area systems and electric vehicle charging distribution guidance, and particularly relates to a service area electric vehicle charging resource distribution guidance method under an Internet of Things environment. BACKGROUND

[0002] With the popularization of electric vehicles, the charging demand of service areas along national and provincial trunk lines is increasing. At present, although charging stations have been established in service areas and the types and quantities of charging piles have been determined, there are still many problems in charging resource distribution. On the one hand, information exchange between service areas and vehicles is not smooth, and service areas cannot comprehensively master key information such as the positions, residual power, and estimated arrival time of vehicles on trunk lines in real time, and vehicles also cannot accurately obtain the real-time use of charging resources of all service areas, including the residual number of charging piles and the queuing waiting time. This directly leads to the imbalance of charging resource distribution, and some charging piles in service areas are idle for a long time, while vehicles frequently gather to charge and the queuing waiting time is too long in adjacent service areas, causing serious waste of resources and decline of user experience. On the other hand, there is a lack of dynamic guidance of vehicle charging behavior, and vehicles often choose service areas based on their own experience or limited information, which easily leads to a sharp increase in charging pressure in local service areas, further exacerbating the contradiction between supply and demand of charging resources.

[0003] Existing research generally only optimizes the electric vehicle charging resource scheduling in a single service area, ignores the coordination between adjacent service areas, and fails to plan electric vehicle charging resources from the global perspective of the entire trunk line. At the same time, existing research fails to fully utilize the advantages of information interconnection under the Internet of Things environment, and often focuses on a single index in the optimization target, either only pursuing the improvement of electric vehicle charging resource utilization or only focusing on reducing user queuing waiting time, making it difficult to maximize the coordination of both. In addition, existing optimization algorithms are not timely and flexible enough in response to dynamic traffic flow and real-time charging demand, and cannot quickly adapt to the dynamic changes in vehicle distribution and charging demand along the trunk line, resulting in unsatisfactory optimization effect. Therefore, it is of great practical significance to develop a guidance method for optimizing the electric vehicle charging resource distribution of service areas along national and provincial trunk lines under the Internet of Things environment. SUMMARY

[0004] The purpose of the present application is to solve the problem of poor optimization effect of service area electric vehicle charging resource distribution guidance due to the fact that existing methods fail to plan electric vehicle charging resources from the global perspective of the entire trunk line, cannot achieve the coordinated optimization of charging resource utilization and user queuing waiting time, and cannot adapt to the dynamic changes in vehicle distribution and charging demand along the trunk line, and a service area electric vehicle charging resource distribution guidance method under an Internet of Things environment is proposed.

[0005] The technical scheme adopted by the present application to solve the above technical problems is: a service area electric vehicle charging resource allocation guiding method under an Internet of Things environment, which specifically comprises the following steps:

[0006] Step one, the edge end acquires data of all service areas and all electric vehicles on the main line in real time, and uploads the acquired data to a cloud data center;

[0007] The data of the service area includes the number of vehicles in the queue in real time, the estimated waiting time, and the operating state of the charging pile;

[0008] The data of the electric vehicle includes the position, driving speed, remaining power, driving direction, and destination information of the electric vehicle;

[0009] Step two, the cloud calculates the probability of each vehicle selecting each service area according to the received data, generates a single vehicle-service area association matrix according to the calculated probability, and obtains a charging vehicle number prediction result by using a short-term prediction model and a long-term prediction model;

[0010] According to the single vehicle-service area association matrix and the charging vehicle number prediction result, the final prediction result of the total number of charging vehicles in the service area in the future first time period is obtained;

[0011] Step three, a multi-objective grey wolf algorithm is used for multi-objective collaborative optimization to realize dynamic optimal configuration of charging resources;

[0012] Step four, combined with the prediction result of step two and the optimal configuration obtained in step three, personalized charging guidance suggestions are generated for vehicles and service areas, and the generated charging guidance suggestions are sent to vehicle terminals or driver mobile phone APPs to realize two-way collaborative guidance and scheduling of vehicles and service areas.

[0013] Further, the specific process of step two is as follows:

[0014] Step two one, the cloud calculates the probability of each vehicle selecting each service area according to the received data:

[0015]

[0016] wherein, represents the probability of the vehicle selecting the service area ; is a sigmoid normalization function, is a safety weight coefficient; is the distance from the vehicle to the service area , is the current remaining power of the vehicle ; Optimize weighting coefficients for efficiency; For service area During the period The estimated waiting time; To form consistent weighting coefficients; For path matching degree;

[0017] Then the bicycle-service area association matrix for:

[0018]

[0019] in, This represents the probability that the first vehicle chooses the first service area. This indicates that the first vehicle selected the first... The probability of each service area Indicates the first The probability of a vehicle choosing the first service area. Indicates the first The vehicle selection is number one. The probability of each service area This represents the total number of vehicles along the route. The number of service areas along the route;

[0020] Step 22: Predict the number of charging vehicles using short-term and long-term prediction models;

[0021] Step 23: Based on the vehicle-service area correlation matrix obtained in Step 21 and the prediction results obtained in Step 22, obtain the service area... In the future Final prediction of the total number of charging vehicles within a given time period .

[0022] Furthermore, the specific process of step two is as follows:

[0023] Step 221: Constructing External Features :

[0024]

[0025] in, and As a time feature, Represents hours, , Represents the week. , Represents the month. , For holiday signage, =1 indicates that it is currently a holiday. =0 means that it is not a holiday.

[0026] and is a weather feature, represents temperature, represents precipitation, represents weather type, if the current weather is , then is 0, if the current weather is rain, then is 1, if the current weather is snow, then is 2;

[0027] and is a traffic feature, represents average speed, represents congestion index;

[0028] respectively construct the real-time state vector of each service area:

[0029]

[0030] wherein, represents the real-time state vector of the service area , represents the remaining available charging pile number of the service area , represents the number of queued vehicles of the service area , represents the vehicle arrival rate of the service area

[0031] Step two, taking the external features and the real-time state vector as the input of the short-term prediction model, outputting the short-term charging vehicle number prediction result through the short-term prediction model;

[0032] The charging vehicle number prediction result output by the short-term prediction model is:

[0033]

[0034] wherein, represents the short-term charging vehicle number prediction result of the service area , represents the number of vehicles that need to be charged in the first time period of the service area , represents the number of vehicles that need to be charged in the second time period of the service area , represents the number of vehicles that need to be charged in the time period of the service area , represents the number of vehicles that need to be charged in the time period of the service area ; ​​​​​

[0035] Step two three, taking the external features and the real-time state vector as inputs of the long-term prediction model, outputting the long-term charging vehicle number prediction result through the long-term prediction model;

[0036] The long-term charging vehicle number prediction result output by the long-term prediction model is:

[0037]

[0038] Wherein, represents the long-term charging vehicle number prediction result of the service area .

[0039]

[0040] Wherein, represents the time slot identifier; represents the expected number of vehicles in the service area in the first time period; represents the confidence interval of the service area in the first time period, .

[0041] Further, the specific process of step two three is:

[0042] Step two three one, according to the single-vehicle-service area association matrix, calculating the expected charging vehicle number of the service area in the time period :

[0043]

[0044] In the formula, is a probability selection threshold; represents the time when the vehicle is expected to arrive at the service area ; is the time window width; is an indicator function, when , the value of the indicator function is 1, otherwise, the value of the indicator function is 0;

[0045]

[0046] Wherein, represents the cross-validation and consistency test index, represents taking the absolute value;

[0047] Step two three two, calculating the adaptive fusion weight :

[0048]

[0049] wherein, is a weight adjustment coefficient; is a service area the individual vehicle prediction confidence of the service area in the future period, is a service area the aggregate prediction confidence of the service area set in the future period;

[0050] Step 233, weighted fusion is performed according to the adaptive fusion weight :

[0051]

[0052] wherein, is the predicted number of charging vehicles in the service area in period ;

[0053] Step 234, the conflict resolution strategy is used to process :

[0054]

[0055] wherein, is the predicted number of charging vehicles in the service area in period ; is a manual review mechanism.

[0056] Further, the specific process of step 3 is as follows:

[0057] Step 3 1, a target function of multi-objective collaborative optimization is established

[0058] (1) maximize the charging pile utilization rate :

[0059]

[0060]

[0061] wherein, represents a function of the utilization rate of all charging piles in all service areas of the trunk line in all optimization periods; represents the charging pile utilization rate of the service area in period ; is the charging pile in the service area in period occupancy state variable of the charging pile in the service area occupancy state variable of the charging pile in the service area when the time period is occupied, occupancy state variable of the charging pile in the service area occupancy state variable of the charging pile in the service area when the time period is not occupied, ; total number of charging piles in the service area ;

[0062] (2) Minimize user waiting time :

[0063]

[0064]

[0065] where, denotes the average waiting time of all vehicles to be served; denotes the waiting time of vehicle ; denotes the time when vehicle arrives at the service area; denotes the time when vehicle is assigned to a charging pile; denotes the total number of vehicles to be served on the main line;

[0066] (3) Minimize service area load balancing :

[0067]

[0068]

[0069]

[0070] where, , denotes the charging pile utilization rate vector; denotes the arithmetic mean of all elements in the utilization rate vector ; denotes the deviation of all elements in the utilization rate vector from the mean ;

[0071] The objective function of the multi-objective collaborative optimization is: :

[0072]

[0073]

[0074] wherein the upper index T denotes the transpose, is the current remaining electric quantity of the vehicle, is the current remaining electric quantity of the vehicle, is the anxiety electric quantity, is the battery electric quantity consumption rate of the vehicle driving unit distance; is the state switching threshold value; is the upper limit of the charging queuing time; is the current remaining electric quantity of the vehicle, is the distance from the vehicle to the service area, is the distance from the vehicle to the service area, is the occupancy state variable of the charging pile in the service area is the occupancy state variable of the charging pile in the service area is the occupancy state variable of the charging pile in the service area is the occupancy state variable of the charging pile in the service area

[0075] Step three two, the multi-objective grey wolf algorithm is used to realize the optimization of the charging resource allocation.

[0076] Further, the specific process of the step three two is as follows:

[0077] Step three two one, the maximum iteration number is set as , and the grey wolf population is initialized: the grey wolf population size is set as , the values of each grey wolf individual are initialized, each grey wolf individual respectively represents a charging resource allocation scheme, and the initialized th grey wolf individual is recorded as:

[0078]

[0079] wherein represents the total number of available charging piles of all service areas, ; represents the decision variable in the individual , the value of the decision variable is 0 or 1, when the decision variable , it indicates that the charging pile of the service area is allocated to the vehicle in the time period , when the decision variable , it indicates that the charging pile of the service area is not allocated to the vehicle in the time period ;

[0080] Step three two two, the iteration number is initialized ;

[0081] Step three two three, the objective function value :

[0082]

[0083] wherein, is a charging pile utilization rate function; is an average waiting time function; is a load balancing degree function;

[0084] Step three two four, determining the leader , and :

[0085]

[0086]

[0087]

[0088] wherein, is a current population of grey wolf individuals;

[0089] and according to the current iteration number to update the control parameter:

[0090]

[0091] wherein, is a control parameter;

[0092] Step three two five, calculating the coefficient vector according to the control parameter , , , , and , updating the position of the grey wolf individual in the population according to the coefficient vector and the leader determined in step three two four; specifically:

[0093] Calculate the intermediate variables , and :

[0094]

[0095]

[0096]

[0097] Recalculate the intermediate variables , and :

[0098]

[0099]

[0100]

[0101] The position of the i-th gray wolf individual in the updated population is:

[0102]

[0103] Compare each element value in the position with 1 / 3 in size respectively, reassign the value of the element greater than 1 / 3 as 1, reassign the value of the element less than or equal to 1 / 3 as 0, and take the reassignment result as the value of each element in the position

[0104] Step three two six, respectively judge whether the position of each gray wolf individual in the updated population satisfies the power constraint:

[0105]

[0106] Wherein, For the vehicle :

[0107] (1) If the vehicle satisfies the power constraint, no processing is required.

[0108] (2) If the vehicle does not satisfy the power constraint, release the charging pile originally allocated for the vehicle , that is, set the value of to 0, and calculate the reachable service area set of the vehicle :

[0109]

[0110] Wherein, is the current remaining power of the vehicle ; is the safety redundancy power; is the dynamic energy consumption coefficient, is the service area in the reachable service area set .

[0111]

[0112] Wherein, is the reference energy consumption of the vehicle driving at a constant speed on a flat road, is the energy consumption correction value based on the slope, ​​​​a correction value for the temperature effect on battery efficiency, a wind resistance correction value based on real-time vehicle speed;

[0113] If the set is empty, trigger the emergency alarm mechanism, push the "power crisis warning" to the vehicle terminal and system administrator, and the vehicle stops and contacts the rescue;

[0114] If the set is not empty, determine a service area, and allocate the charging piles in this service area to the vehicle , and update the corresponding decision variable value to 1;

[0115] Step three two seven, judge whether the iteration stop condition is met, the iteration stop condition is to reach the maximum iteration number or reach the upper limit of the calculation time of the iteration process;

[0116] If the iteration stop condition is met, the optimal individual in the individual obtained in the last iteration is taken as the best charging resource allocation scheme;

[0117] If the iteration stop condition is not met, let , return to step three two three.

[0118] Further, when the set is not empty, the service area is determined according to the priority, and the specific process is:

[0119] Priority 1: the remaining power is greater than and the service area closest to ;

[0120] Priority 2: the service area with the shortest predicted waiting time ;

[0121] Priority 3: the service area that optimizes the load balancing degree of the updated individual;

[0122] By comparing each priority, the service area with the highest priority is selected.

[0123] Further, the specific process of step four is:

[0124] Step four one, vehicle guidance strategy

[0125] Step four one one, calculate the reachable service area set of the vehicle based on the current state:

[0126]

[0127] In the formula, represents the optimal result of step three, which is the service area set a set of reachable service areas in all allocated service areas;

[0128] Step Four-ii, calculating the vehicle arrival set the guidance utility value of each service area , and selecting the service area with the minimum guidance utility value as the optimal service area :

[0129]

[0130]

[0131] wherein: is the predicted waiting time of the service area in the time period ; is the probability of the vehicle selecting the service area ; is the predicted charging pile utilization rate of the service area in the time period ; is the system optimal utilization rate, , , and is the weight coefficient;

[0132] Step Four-iii, sending guidance information to the vehicle :

[0133]

[0134] wherein, is the recommended service area name; is the time at which the vehicle is expected to arrive at the optimal service area ; is the expected waiting time of the vehicle in the optimal service area ; is the recommended charging time; is the real-time electricity price; is the alternative service area list;

[0135] Step Four-iv, the vehicle returns satisfaction after confirming receipt of the guidance information :

[0136]

[0137] wherein, ​satisfaction, representing the actual waiting time of the vehicle ;

[0138] Step four two, service area scheduling strategy

[0139] Step four two one, calculate the predicted load rate of all service areas in the time period ;

[0140]

[0141]

[0142] In the formula: is the predicted number of charging vehicles in the time period ; is the average charging time of a single vehicle; is the predicted number of fast charging vehicles in the time period ; is the predicted number of slow charging vehicles in the time period ; is the average fast charging time in the time period ; is the average slow charging time in the time period ; is the total time in the time period ; is the total number of charging piles of all service areas;

[0143] Step four two two, adjust the charging pile resources of the service area according to the predicted load rate ;

[0144] Step four two one, first adjustment of the charging pile resources of the service area:

[0145] When , the standby charging pile needs to be activated and the dynamic pricing needs to be started;

[0146] When , 30% of the charging piles need to be closed and the maintenance mode needs to be started;

[0147] When , no processing is needed;

[0148] Step four two two, second adjustment of the charging pile resources of the service area:

[0149] When the ratio of fast charging demand to slow charging demand of a service area is greater than 2, dynamic conversion of charging pile type is needed;

[0150] Step four two three, when multiple vehicles compete for the same charging pile resource, the vehicles need to be allocated according to their priority;​

[0151] Step four two three, service area state publishing:

[0152] If there is a communication interruption of a service area, it is necessary to return to step one to recalculate the guiding strategy and push the update, and in the recalculation process, do not allocate vehicles to the service area with communication interruption;

[0153] If the communication states of all service areas are normal, each service area publishes real-time state information to the cloud, and the service area The real-time state information published is:

[0154]

[0155] In the formula, is the real-time utilization rate of the service area in the time period ; is the predicted waiting time of the service area in the time period ; is the number of available charging piles of the service area in the time period ; is the real-time electricity price of the service area in the time period ; is the state of the service area in the time period , ;

[0156] The normal state of the service area is that the number of available charging piles ;

[0157] The busy state of the service area is that the number of available charging piles ;

[0158] The saturation state of the service area is that the number of available charging piles .

[0159] Further, in step four two three, the priority calculation method of the vehicle is:

[0160]

[0161] In the formula, represents the priority, is a binary variable representing the degree of power crisis, if , it means that the power is in crisis, and the value of is 1; if , indicates a non-electric quantity crisis, is 0; is a vehicle type coefficient, ; is a reservation state, .

[0162] Further, the method monitors the resource usage of each service area in real time:

[0163]

[0164] and every time, the following update conditions are detected:

[0165]

[0166] wherein, is the predicted waiting time of the service area in the time period; is the actual waiting time of the service area in the time period; is the waiting time deviation threshold value; is the real-time distance of the vehicle to the service area ; is the maximum acceptable detour distance; when or

[0167] the value is 1, then it is necessary to return to step one to recalculate the guidance strategy and push the update; Otherwise, no processing is required.

[0168]

[0169] The beneficial effects of the present application are:

[0170] The present application collects information of vehicles and service areas in real time, and combines the collected information of vehicles and service areas to predict the number of charging vehicles in the future period, uses the grey wolf algorithm based on multi-objective optimization to realize intelligent allocation of charging resources, and according to the allocation result of charging resources and the prediction result of the number of charging vehicles, guidance information can be provided for vehicles and service areas, realizing global overall planning of electric vehicle charging resources on the whole trunk line. Moreover, the method of the present application can effectively adapt to the dynamic changes of vehicle distribution and charging demand along the line through re-planning, can improve charging pile utilization rate and user satisfaction at the same time under fixed charging resource configuration, and improves the optimization effect of charging resource allocation guidance. The method of the present application is especially suitable for electric vehicle charging management of long-distance roads, and can effectively solve the problem of uneven allocation of charging resources. BRIEF DESCRIPTION OF DRAWINGS

[0171] ​​Figure 1 A flowchart of a service area electric vehicle charging resource allocation guidance method in an Internet of Things environment;

[0172] Figure 2 A flowchart of a charging demand prediction; DETAILED DESCRIPTION

[0173] To make the method of the present application clearer and easier to understand, the method of the present application will be further described in detail below in combination with the accompanying drawings of the present application.

[0174] Specific implementation method one: in combination with Figure 1 This embodiment describes a service area electric vehicle charging resource allocation guidance method in an Internet of Things environment. The method specifically includes the following steps:

[0175] Step one, the edge end acquires data of all service areas and all electric vehicles on the main line in real time, and uploads the acquired data to the cloud data center;

[0176] The data of the service area includes the number of vehicles in the queue in real time, the estimated waiting time based on the charging rate, and the operating state of the charging pile;

[0177] The data of the electric vehicle includes the position, speed, remaining power, direction of travel, and destination information (through latitude and longitude coordinates) of the electric vehicle;

[0178] Step two, the cloud calculates the probability of each vehicle selecting each service area according to the received data, generates a single vehicle-service area association matrix according to the calculated probability, and obtains a charging vehicle number prediction result by using a short-term prediction model and a long-term prediction model;

[0179] According to the single vehicle-service area association matrix and the charging vehicle number prediction result, the service area In the future period, the final prediction result of the total number of charging vehicles;

[0180] The specific process of step two will be described in detail as follows: Figure 2 Step two one, the cloud calculates the probability of each vehicle selecting each service area according to the received data:

[0181]

[0182]

[0183] Among them, represents the probability of the vehicle selecting the service area ; is a sigmoid normalization function, ​A safety weighting factor is used to ensure that vehicles do not run out of power and become stranded. For vehicles to service area distance, For vehicles Current remaining battery power; Optimize weighting coefficients to reduce user waiting time and improve satisfaction; For service area During the period The estimated waiting time; To ensure consistent weighting coefficients, the user's original travel plan must be respected; Path fit is used to quantify service areas. Geographical location and vehicles This is an indicator of the degree of overlap between the original driving route (the planned route without considering charging needs), with a value ranging from 0 to 1. Its core function is to measure the distance to the service area. Whether charging will significantly deviate from the user's original travel route is calculated based on the obtained information on the electric vehicle's location, direction of travel, and destination.

[0184] Then the bicycle-service area association matrix for:

[0185]

[0186] in, This represents the probability that the first vehicle chooses the first service area. This indicates that the first vehicle selected the first... The probability of each service area Indicates the first The probability of a vehicle choosing the first service area. Indicates the first The vehicle selection is number one. The probability of each service area This represents the total number of vehicles along the route. The number of service areas along the route;

[0187] Step 22: Predict the number of charging vehicles using short-term and long-term prediction models;

[0188] Step 221: Constructing External Features :

[0189]

[0190] in, and As a time feature, Represents hours, , Represents the week. , Represents the month. , For holiday signage, =1 indicates that it is currently a holiday. =0 means that it is not a holiday.

[0191] and As a meteorological feature, Represents temperature. Represents precipitation. This represents the weather type. If the current weather is... ,but The value is 0. If the current weather is raining, then... The value is 1. If the current weather is snowing, then... The value of is 2;

[0192] and As a traffic feature, Represents average vehicle speed. Represents the congestion index;

[0193] Construct the real-time state vector for each service area:

[0194]

[0195] in, Indicates service area The real-time state vector, Indicates service area The remaining number of available charging stations (obtained based on the operating status of the charging stations). Indicates service area The number of vehicles in the queue Indicates service area Vehicle arrival rate, in units of vehicles per hour;

[0196] It should be noted that before constructing the real-time state vector, the number of remaining available charging piles, the number of vehicles in queue, and the vehicle arrival rate in each service area need to be standardized.

[0197] Step 222: Select the ARIMA-LSTM hybrid model as the short-time prediction model, use the external features and real-time state vector as inputs to the short-time prediction model, and output the short-time charging vehicle number prediction result through the short-time prediction model.

[0198] The short-term accurate prediction targets charging demand for the next 0-2 hours, with the output results using a 15-minute time window. The short-term prediction emphasizes high accuracy and detailed segmentation, dividing the time window into eight segments: 0-15 minutes (segment 1), 15-30 minutes (segment 2), and so on, for a total of eight segments from 0 to 2 hours. The predicted number of charging vehicles output by the short-term prediction model is as follows:

[0199]

[0200] in, Indicates service area The predicted number of vehicles charging for short periods of time. Indicates the service area within the predicted first time period. The number of vehicles that need charging. Indicates the service area during the predicted second time period. The number of vehicles that need charging. Indicates the predicted first Service area within a certain time period The number of vehicles that need charging;

[0201] Step 223: Select the Prophet model as the long-term prediction model, use external features and real-time state vectors as inputs to the long-term prediction model, and output the prediction result of the number of vehicles charging for a long time through the long-term prediction model.

[0202] The long-term accurate forecast targets charging demand from 2 hours to 24 hours in the future. The output results are in 1-hour time windows. The long-term forecast focuses on the overall trend and macro distribution. 2 hours to 3 hours is the first period of the long-term forecast, 3 hours to 4 hours is the second period, and so on, with a total of 22 periods from 2 hours to 24 hours.

[0203] The predicted number of vehicles undergoing long-term charging output by the long-term prediction model is as follows:

[0204]

[0205] in, Indicates service area The predicted number of vehicles charging for extended periods;

[0206]

[0207] in, This indicates a time slot identifier (specifically, a time period, such as 8:00~8:15). Representing the Service area within a certain time period The expected number of vehicles, representing the number of vehicles in the first place. Expected arrival time at the service area within the specified time period the total number of charging vehicles; represent the confidence interval of the service area in the first time period, ;

[0208] Step 3, according to the single-vehicle-service area association matrix obtained in Step 2 and the prediction results obtained in Step 2, obtain the total number of charging vehicles in the future first time period, the final prediction result, ;

[0209] The specific process of Step 3 is as follows:

[0210] Construct vectors and ;

[0211]

[0212]

[0213] wherein, represents the time when the vehicle is expected to arrive at the service area , is the predicted number of charging vehicles in the service area in the time period (which can be obtained according to the prediction results of different time periods by the ARIMA-LSTM model and the Prophet model, i.e., for the service area , is equal to the predicted );

[0214] Step 3.1, according to the single-vehicle-service area association matrix, calculate the expected number of charging vehicles in the service area in the time period :

[0215]

[0216] In the formula, is a probability selection threshold (usually 0.3-0.5); represents the time when the vehicle is expected to arrive at the service area ; is the width of the time window (for the time period corresponding to the short-time model, is 15 min; for the time period corresponding to the long-time model, is 1 h); is an indicator function, when ​When the value is 1, the indicator function evaluates to 1; otherwise, the indicator function evaluates to 0.

[0217]

[0218] in, This represents the cross-validation and consistency test index. Indicates taking the absolute value;

[0219] Step 2.3. Calculate the adaptive fusion weights :

[0220]

[0221] Among them, adaptive fusion weights The value range is [0,1]; This is the weighting adjustment coefficient, typically ranging from 2.0 to 5.0; For service area In the future Individual vehicle prediction confidence levels for each time period For service area In the future Service area aggregated prediction confidence level for each time period;

[0222] The calculation method is as follows:

[0223]

[0224] in, This indicates that under historical scenarios with identical temporal and meteorological characteristics, the service area During the period Actual number of charging vehicles Standard deviation; This indicates that under historical scenarios with identical temporal and meteorological characteristics, the service area During the period Actual number of charging vehicles The mean;

[0225] The calculation method is as follows:

[0226]

[0227] in, This indicates the service area predicted by the model under historical scenarios with the same temporal and meteorological characteristics. During the period Number of charging vehicles Standard deviation; This indicates the service area predicted by the model under historical scenarios with the same temporal and meteorological characteristics. During the period Number of charging vehicles The mean.

[0228] Steps 2 and 3: Based on adaptive fusion weights Perform weighted fusion:

[0229]

[0230] In the formula, For the service area obtained after the integration During the period The estimated number of vehicles to be charged;

[0231] Steps two, three, and four: Adopting conflict resolution decision-making methods Processing:

[0232]

[0233] In the formula, For the final predicted service area During the period The number of vehicles charging; The system employs a manual review mechanism. When prediction conflicts become severe, the system administrator will manually intervene, i.e., manually assign values.

[0234] Step 3: Employ the Multi-Objective Gray Wolf Algorithm (MOGWO) for multi-objective collaborative optimization to achieve dynamic optimal allocation of charging resources;

[0235] The specific process of step three is as follows:

[0236] Step 31: Establish the objective function for multi-objective collaborative optimization.

[0237] (1) Maximize the utilization rate of charging piles :

[0238]

[0239]

[0240] In the formula, A function representing the utilization rate of all charging piles within all service areas along the main line during all optimization periods; Indicates service area During the period The utilization rate of charging piles; For service area Charging stations inside During the period The occupancy status variable, when the service area Charging stations inside During the period When occupied, When the service area Charging stations inside During the period When not occupied, ; For service area The total number of charging stations;

[0241] (2) Minimize user waiting time :

[0242]

[0243]

[0244] In the formula, This represents the average waiting time for all vehicles awaiting service. Indicates vehicle Waiting time; For vehicles The time of arrival at the service area; For vehicles The time allocated to the charging station; The total number of vehicles waiting to be served on the main line;

[0245] (3) Minimize service area load balancing :

[0246]

[0247]

[0248]

[0249] In the formula, , This represents the charging pile utilization rate vector. Represents the utilization vector The arithmetic mean of all elements, i.e., the average utilization of all service areas over all optimization periods; Represents the utilization vector All elements and mean The degree of deviation;

[0250] The objective function of multi-objective collaborative optimization for:

[0251]

[0252]

[0253] In the formula, the superscript T indicates transpose. For vehicles The current remaining battery power, For anxiety about battery life, This is the battery power consumption rate per unit distance traveled by the vehicle, representing the energy efficiency of the vehicle during its journey from its current location to the target service area. This is the threshold for state transition; The maximum charging queue time is set to protect user experience. For vehicles to service area The distance; For service area Charging stations inside During the period Occupied state variables;

[0254] Step 32: Optimize charging resource allocation using the Multi-Objective Gray Wolf (MOGWO) algorithm;

[0255] The specific process of step three-two is as follows:

[0256] Step 321: Set the maximum number of iterations to... and initialize the gray wolf population: set the gray wolf population size to... Initialize the values ​​of each individual gray wolf, with each gray wolf representing a charging resource allocation scheme. The initialized values ​​of the first individual gray wolf are then set. A gray wolf Recorded as:

[0257]

[0258] In the formula, This indicates the total number of available charging stations across all service areas. ; Represents an individual Decision variables in The value of can be 0 or 1, when the decision variable "Time" indicates the period of time. Service area charging stations Assigned to vehicles When decision variables "Time" indicates the period of time. Service areas charging stations Assigned to vehicles When the number of vehicles that need to be charged at the same time exceeds the number of available charging stations, the same charging station can be assigned to different vehicles, and the charging order of the vehicles is determined according to priority.

[0259] Step 3.2.2 Initialize the number of iterations ;

[0260] Step 3: Calculate the objective function value for each individual gray wolf in the current population. :

[0261]

[0262] In the formula, This is a function for the utilization rate of charging piles; This is a function of the average waiting time; This is the load balancing function;

[0263] With the first gray wolf individual For example, (This corresponds to an overall utilization rate of 85% for charging piles; the negative sign is introduced because the concept of "maximizing" has been changed to "minimizing".) (The average waiting time for all vehicles is 12 minutes). (Corresponding load balancing ratio of 0.18). Then, the individual gray wolf... The corresponding target vector is:

[0264]

[0265] The target vector contains three specific values, clearly quantifying the individual. The performance across the three dimensions of "utilization, latency, and load balancing" provides a comprehensive picture of multi-objective performance.

[0266] Steps 3-4: Identifying the Leader Based on Pareto Dominance , and :

[0267]

[0268]

[0269]

[0270] In the formula, This refers to the set of individual gray wolves in the current population.

[0271] And based on the current iteration number To update the control parameters:

[0272]

[0273] In the formula, For control parameters;

[0274] Step 325: Calculate the coefficient vector based on the control parameters. , , , , and The positions of individual gray wolves in the population are updated based on the coefficient vector and the leader determined in steps three, two, and four; specifically:

[0275] Calculate intermediate variables , and :

[0276]

[0277]

[0278]

[0279] Then calculate the intermediate variables , and :

[0280]

[0281]

[0282]

[0283] Then the updated population of the first The location of an individual gray wolf for:

[0284]

[0285] Compare positions separately For each element value in the array, compare it to 1 / 3 of the original value. Reassign values ​​to elements greater than 1 / 3, and reassign values ​​to elements less than or equal to 1 / 3. Use the reassignment result as the position. The values ​​of each element in the array;

[0286] Step 326: Determine whether the position of each gray wolf in the updated population satisfies the power constraint (i.e., determine whether there is a service area that the current vehicle can reach with its current power level within the assigned service area):

[0287]

[0288] in, For vehicles :

[0289] (1) If the vehicle If the power constraint is met, no further action is required;

[0290] (2) If the vehicle If the power constraint is not met, the vehicle will be released. The originally allocated charging stations (for other vehicles) will soon be... The value is 0, and the vehicle is calculated. Accessible service area collection That is, to select service areas that meet the power constraints:

[0291]

[0292] in, For vehicles Current remaining battery power; For safety redundancy (usually set at 5%); The dynamic energy consumption coefficient, Collection of reachable service areas Service areas within;

[0293]

[0294] in, This represents the baseline energy consumption of a vehicle traveling at a constant speed on a straight road. This is the energy consumption correction value for slope based on high-precision map data. This is a correction value for the effect of temperature on battery efficiency. This is a drag correction value based on real-time vehicle speed;

[0295] If set If the number is empty (the vehicle has no service area available), an emergency alarm mechanism will be triggered, and a "battery crisis warning" will be pushed to the vehicle terminal and system administrator. The vehicle will pull over and contact roadside assistance. In subsequent iterations and optimizations, no charging station will be assigned to this vehicle.

[0296] If set If the condition is not empty, a service area is determined, and the charging stations within this service area are allocated to the vehicles. Based on the vehicle's estimated arrival time at the service area, the corresponding decision variable value is updated to 1, meaning that at the estimated arrival time, all charging stations in this service area can be allocated to the vehicle. For vehicles choose;

[0297] The service areas are determined based on priority, and the specific process is as follows:

[0298] Priority 1: Remaining battery power greater than And closest Service areas (i.e., service areas that are "just within reach" to avoid excessive detours);

[0299] Priority 2: Predict waiting time The shortest service area;

[0300] Priority 3: The service area that provides the best load balancing for the updated individual units;

[0301] By comparing each priority level, the service area with the highest priority is selected.

[0302] The process begins by evaluating service areas using priority 1. If multiple service areas simultaneously meet priority 1, priority 2 is then applied. If multiple service areas simultaneously meet priority 2, priority 3 is applied. When only one service area meets the criteria for a given priority, that service area is selected.

[0303] Step 327: Determine whether the iteration stopping condition is met. The iteration stopping condition is reaching the maximum number of iterations. Or it may reach the upper limit of the computation time for the iterative process;

[0304] If the iteration stopping condition is met, the best individual among the individuals obtained in the last iteration will be taken as the optimal charging resource allocation scheme.

[0305] If the iteration stopping condition is not met, then let Return to step three.

[0306] It should be further explained that the coefficient vector , , , , and The calculation method is as follows:

[0307] With coefficient vector For example, initialize a random vector with each element value in the range [0,1]. Then the coefficient vector for:

[0308]

[0309] coefficient vector and The calculation process is consistent with same;

[0310] With coefficient vector For example, initialize a random vector with each element value in the range [0,1]. Then the coefficient vector for:

[0311]

[0312] coefficient vector and The calculation process is consistent with same.

[0313] Step 4: Combining the prediction results from Step 2 and the optimal configuration obtained in Step 3, generate personalized charging guidance suggestions for vehicles and service areas, and send the generated charging guidance suggestions to vehicle terminals or driver mobile apps through vehicle-to-everything (V2X) or mobile networks to achieve two-way collaborative guidance and scheduling for individual vehicles and service areas.

[0314] The specific process of step four is as follows:

[0315] Step 41: Vehicle Guidance Strategy

[0316] Step 411: Calculate the vehicle based on the current state. A collection of reachable service areas:

[0317]

[0318] In the formula, This indicates that the optimization results in step three represent vehicles. The set of reachable service areas from all allocated service areas;

[0319] Step 4.12: Calculate the vehicle Arrive at the meeting Each service area The guiding utility value The service area with the lowest guiding utility value is selected as the optimal service area. :

[0320]

[0321]

[0322] In the formula: For service area In time period Predicted waiting time; For vehicles Select service area The probability of; For service area In time period The predicted utilization rate of charging piles (calculated based on the optimization results of step three); To achieve optimal system utilization, for example, it is set to 80% in this invention. , , and These are weighting coefficients, which can be dynamically adjusted.

[0323] Step 413, To the vehicle Send guidance information :

[0324]

[0325] In the formula, Recommended service area name; Predict the optimal service area for the vehicle. The time (can be obtained based on the vehicle speed uploaded from the edge device); For vehicles in the optimal service area The estimated waiting time; Recommended charging time; For real-time electricity prices; The list of alternative service areas can be determined based on step 412. That is, in step 412, in addition to the optimal service area, some service areas with low guidance utility values ​​can be selected to form the alternative service area list. When the optimal service area triggers an alarm, the vehicle... You can then select a service area from the list of alternative service areas.

[0326] Step 414, Vehicle Confirm receipt of guidance information Then, return the satisfaction level. :

[0327]

[0328] in, Indicates satisfaction. , Indicates vehicle The actual waiting time;

[0329] Step 4.2 Service Area Dispatch Strategy

[0330] Step 421: Calculate all service areas during the time period Predicted load factor The predicted load rate Used to measure time period The degree of tension between supply and demand for charging resources;

[0331]

[0332]

[0333] In the formula: For time period The predicted number of charging vehicles (taken from the final fused prediction value in step two) ); Average charging time for a single vehicle; For time period The predicted number of fast-charging vehicles; For time period The predicted number of slow-charging vehicles, and This can be obtained based on the decision results of the optimization process in step three; For time period The average charging time; For time period The average charging time for slow charging, and Obtained from historical data; For time period Total duration; This represents the total number of charging stations across all service areas.

[0334] Step 422: Based on the predicted load factor Adjust the charging station resources in service areas;

[0335] Step 4221: First adjustment of charging pile resources in the service area:

[0336] when At this time, it is necessary to activate the backup charging station and initiate dynamic pricing;

[0337] when When this happens, the charging station needs to be shut down at 30% capacity and maintenance mode activated.

[0338] when No processing is required at this time;

[0339] Step 4222: Make a second adjustment to the charging pile resources in the service area:

[0340] When the ratio of fast charging demand to slow charging demand in a service area is greater than 2, the charging pile type needs to be dynamically switched (when the slow charging demand is large, the charging pile type does not need to be switched, and fast charging piles can be used for slow charging demand by default. After optimizing the allocation results in step three, the fast charging demand and slow charging demand within the service area can be obtained based on the charging preferences of the vehicles allocated in step three).

[0341] Step 4223: Vehicles entering the server can choose a charging station based on their preference for fast charging or slow charging. When multiple vehicles compete for the same charging station resource, it needs to be allocated according to the vehicle's priority.

[0342] The priority calculation method for the vehicle is as follows:

[0343]

[0344] In the formula, Indicates priority. It is a binary variable representing the degree of power urgency. If This indicates a critical power shortage, that is... The value of is 1; if This indicates a non-electrical energy emergency. The value is 0; For vehicle type coefficients, ; It is in reservation status. ;

[0345] Steps 4-2-3: Service Area Status Release

[0346] If communication is interrupted in a service area, it is necessary to return to step one to recalculate the guidance policy and push the update, and during the recalculation process, vehicles should not be assigned to the service area with communication interruption.

[0347] If the communication status of all service areas is normal, each service area will publish real-time status information to the cloud. Published real-time status information for:

[0348]

[0349] In the formula, For service area In time period Real-time utilization rate; For service area In time period Predicted waiting time; For service area In time period The number of available charging stations; For service area In time period Real-time electricity price; For service area In time period state, ;

[0350] service area Under normal conditions, the following is satisfied: Number of available charging stations ;

[0351] service area Under busy conditions, the following must be met: Number of available charging stations ;

[0352] service area Under saturation conditions, the following is satisfied: Number of available charging stations ;

[0353] Meanwhile, the resource usage of each service area is monitored in real time:

[0354]

[0355] And every Time checks the following update conditions:

[0356]

[0357] In the formula, For service area In time period The estimated waiting time; For service area In time period The actual waiting time; The waiting time deviation threshold; For vehicles to service area Real-time distance; The maximum acceptable detour distance; if any condition is triggered, The value is 1;

[0358] when or If the value is 1, then it is necessary to return to step one to recalculate the bootstrapping strategy and push the update;

[0359] Otherwise, no action is required.

[0360] The above examples of the present invention are merely illustrative of the computational model and process of the present invention, and are not intended to limit the implementation of the present invention. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is impossible to exhaustively list all possible implementations here. Any obvious variations or modifications derived from the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. A method for guiding the allocation of electric vehicle charging resources in a service area under an Internet of Things (IoT) environment, characterized in that, The method specifically includes the following steps: Step 1: The edge device acquires real-time data on all service areas and all electric vehicles on the main line and uploads the acquired data to the cloud data center; The data for the service area includes the real-time number of vehicles in queue, the estimated waiting time, and the operating status of the charging piles; The data for the electric vehicle includes its location, speed, remaining battery power, direction of travel, and destination information. Step 2: The cloud calculates the probability of each vehicle selecting each service area based on the received data, and then generates a vehicle-service area correlation matrix based on the calculated probabilities; and uses short-term prediction model and long-term prediction model to obtain the prediction result of the number of charging vehicles. Based on the vehicle-service area correlation matrix and the predicted number of charging vehicles, the service area is obtained. In the future Final prediction of the total number of charging vehicles in each time period; Step 3: Employ the multi-objective gray wolf algorithm for multi-objective collaborative optimization to achieve dynamic optimal allocation of charging resources; Step 4: Combining the prediction results from Step 2 and the optimal configuration obtained in Step 3, generate personalized charging guidance suggestions for vehicles and service areas, and send the generated charging guidance suggestions to the vehicle terminal or driver's mobile APP to achieve two-way collaborative guidance and scheduling for individual vehicles and service areas.

2. The method for guiding the allocation of electric vehicle charging resources in a service area under the Internet of Things environment according to claim 1, characterized in that, The specific process of step two is as follows: Step 21: The cloud calculates the probability of each vehicle selecting each service area based on the received data. in, Indicates vehicle Select service area The probability of; The sigmoid normalization function. This is a safety weighting coefficient; For vehicles to service area distance, For vehicles Current remaining battery power; Optimize weighting coefficients for efficiency; For service area During the period The estimated waiting time; To form consistent weighting coefficients; For path matching degree; Then the bicycle-service area association matrix for: in, This represents the probability that the first vehicle chooses the first service area. This indicates that the first vehicle selected the first... The probability of each service area Indicates the first The probability of a vehicle choosing the first service area. Indicates the first The vehicle selection is number one. The probability of each service area This represents the total number of vehicles along the route. The number of service areas along the route; Step 22: Predict the number of charging vehicles using short-term and long-term prediction models; Step 23: Based on the vehicle-service area correlation matrix obtained in Step 21 and the prediction results obtained in Step 22, obtain the service area... In the future Final prediction of the total number of charging vehicles within a given time period .

3. The method for guiding the allocation of electric vehicle charging resources in a service area under the Internet of Things environment according to claim 2, characterized in that, The specific process of step two is as follows: Step 221: Constructing External Features : in, and As a time feature, Represents hours, , Represents the week. , Represents the month. , For holiday signage, =1 indicates that it is currently a holiday. =0 means that it is not a holiday. and As a meteorological feature, Represents temperature. Represents precipitation. This represents the weather type. If the current weather is sunny, then... The value is 0. If the current weather is raining, then... The value is 1. If the current weather is snowing, then... The value of is 2; and As a traffic feature, Represents average vehicle speed. Represents the congestion index; Construct the real-time state vector for each service area: in, Indicates service area The real-time state vector, Indicates service area The number of remaining available charging stations, Indicates service area The number of vehicles in the queue Indicates service area Vehicle arrival rate; Step 222: Use the external features and real-time state vector as input to the short-time prediction model, and output the short-time charging vehicle number prediction result through the short-time prediction model; The predicted number of charging vehicles output by the short-time prediction model is as follows: in, Indicates service area The predicted number of vehicles charging for short periods of time. Indicates the service area within the predicted first time period. The number of vehicles that need charging. Indicates the service area during the predicted second time period. The number of vehicles that need charging. Indicates the predicted first Service area within a certain time period The number of vehicles that need charging; Step 223: Use the external features and real-time state vector as input to the long-term prediction model, and output the prediction result of the number of vehicles charging for a long time through the long-term prediction model; The predicted number of vehicles undergoing long-term charging output by the long-term prediction model is as follows: in, Indicates service area The predicted number of vehicles charging for extended periods; in, Indicates the time slot identifier; Representing the Service area within a certain time period The expected number of vehicles; Indicates the first Service area within a certain time period The confidence interval, .

4. The method for guiding the allocation of electric vehicle charging resources in a service area under the Internet of Things environment according to claim 3, characterized in that, The specific process of steps two and three is as follows: Step 231: Calculate the service area association matrix based on the bicycle-service area association matrix. During the period The expected number of charging vehicles : In the formula, Choose a threshold for the probability; Indicates vehicle Expected arrival at service area Time; The width of the time window; For indicator functions, when When the value is 1, the indicator function evaluates to 1; otherwise, the indicator function evaluates to 0. in, This represents the cross-validation and consistency test index. Indicates taking the absolute value; Step 2.

3. Calculate the adaptive fusion weights. : in, This is the weighting adjustment coefficient; For service area In the future Individual vehicle prediction confidence levels for each time period For service area In the future Service area aggregated prediction confidence level for each time period; Steps 2 and 3: Based on adaptive fusion weights Perform weighted fusion: In the formula, The service area obtained after the integration During the period The estimated number of vehicles to be charged; Steps two, three, and four: Adopting conflict resolution decision-making methods Processing: In the formula, For the final predicted service area During the period The number of vehicles charging; It uses a manual review mechanism.

5. The method for guiding the allocation of electric vehicle charging resources in a service area under the Internet of Things environment according to claim 4, characterized in that, The specific process of step three is as follows: Step 31: Establish the objective function for multi-objective collaborative optimization. (1) Maximize the utilization rate of charging piles : In the formula, A function representing the utilization rate of all charging piles within all service areas along the main line during all optimization periods; Indicates service area During the period The utilization rate of charging piles; For service area Charging stations inside During the period The occupancy status variable, when the service area Charging stations inside During the period When occupied, When the service area Charging stations inside During the period When not occupied, ; For service area The total number of charging stations; (2) Minimize user waiting time : In the formula, This represents the average waiting time for all vehicles awaiting service. Indicates vehicle Waiting time; For vehicles The time of arrival at the service area; For vehicles The time allocated to the charging station; The total number of vehicles waiting to be served on the main line; (3) Minimize service area load balancing : In the formula, , This represents the utilization rate vector of charging piles. Represents the utilization vector The arithmetic mean of all elements in the set; Represents the utilization vector All elements and mean The degree of deviation; The objective function of multi-objective collaborative optimization for: In the formula, the superscript T indicates transpose. For vehicles The current remaining battery power, For anxiety about battery life, Battery power consumption rate per unit distance traveled by the vehicle; This is the threshold for state transition; This sets the maximum charging queue time. For vehicles to service area The distance; For service area Charging stations inside During the period Occupied state variables; Step 3.2: Optimize the allocation of charging resources using the multi-objective gray wolf algorithm.

6. The method for guiding the allocation of electric vehicle charging resources in a service area under the Internet of Things environment according to claim 5, characterized in that, The specific process of step 32 is as follows: Step 321: Set the maximum number of iterations to... and initialize the gray wolf population: set the gray wolf population size to... Initialize the values ​​of each individual gray wolf, with each gray wolf representing a charging resource allocation scheme. The initialized values ​​of the first individual gray wolf are then set. A gray wolf Recorded as: In the formula, This indicates the total number of available charging stations across all service areas. ; Represents an individual Decision variables in The value of can be 0 or 1, when the decision variable "Time" indicates the period of time. Service area charging stations Assigned to vehicles When decision variables "Time" indicates the period of time. Service areas charging stations Assigned to vehicles ; Step 3.2.2: Initialize the number of iterations ; Step 3: Calculate the objective function value for each individual gray wolf in the current population. : In the formula, This is a function representing the utilization rate of charging piles. This is a function of the average waiting time; This is the load balancing function; Steps 3-4: Determine the Leader , and : In the formula, This refers to the set of individual gray wolves in the current population. And based on the current iteration number To update the control parameters: In the formula, For control parameters; Step 325: Calculate the coefficient vector based on the control parameters. , , , , and The positions of individual gray wolves in the population are updated based on the coefficient vector and the leader determined in steps three, two, and four; specifically: Calculate intermediate variables , and : Then calculate the intermediate variables , and : Then the updated population of the first The location of an individual gray wolf for: Compare positions separately For each element value in the array, compare it to 1 / 3 of the original value. Reassign values ​​to elements greater than 1 / 3, and reassign values ​​to elements less than or equal to 1 / 3. Use the reassignment result as the position. The values ​​of each element in the array; Step 326: Determine whether the position of each gray wolf in the updated population satisfies the energy constraint. in, For vehicles : (1) If the vehicle If the power constraint is met, no further action is required; (2) If the vehicle If the power constraint is not met, the vehicle will be released. The originally allocated charging stations will soon be... The value is 0, and the vehicle is calculated. Accessible service area collection : in, For vehicles Current remaining battery power; For safety redundancy; The dynamic energy consumption coefficient, Collection of reachable service areas Service areas within; in, This represents the baseline energy consumption of a vehicle traveling at a constant speed on a straight road. This is an energy consumption correction value based on slope. This is a correction value for the effect of temperature on battery efficiency. This is a drag correction value based on real-time vehicle speed; If set If the value is empty, an emergency alarm mechanism will be triggered, and a "battery crisis warning" will be pushed to the vehicle terminal and system administrator. The vehicle will then pull over and contact roadside assistance. If set If the condition is not empty, a service area is determined, and the charging stations within this service area are allocated to the vehicles. Update the corresponding decision variable value to 1; Step 327: Determine whether the iteration stopping condition is met. The iteration stopping condition is reaching the maximum number of iterations. Or it may reach the upper limit of the computation time for the iterative process; If the iteration stopping condition is met, the best individual among the individuals obtained in the last iteration is taken as the optimal charging resource allocation scheme. If the iteration stopping condition is not met, then let Return to step three.

7. The method for guiding the allocation of electric vehicle charging resources in a service area under an Internet of Things environment according to claim 6, characterized in that, The set When not empty, the service area is determined according to priority, and the specific process is as follows: Priority 1: Remaining battery power greater than And closest Service areas; Priority 2: Predict waiting time The shortest service area; Priority 3: The service area that provides the best load balancing for the updated individuals; By comparing each priority level, the service area with the highest priority is selected.

8. The method for guiding the allocation of electric vehicle charging resources in a service area under the Internet of Things environment according to claim 7, characterized in that, The specific process of step four is as follows: Step 41: Vehicle Guidance Strategy Step 411: Calculate the vehicle based on the current state. A collection of reachable service areas: In the formula, This indicates that the optimization results in step three represent vehicles. The set of reachable service areas from all allocated service areas; Step 4.12: Calculate the vehicle Arrive at the meeting Each service area The guiding utility value The service area with the lowest guiding utility value is selected as the optimal service area. : In the formula: For service area In time period Predicted waiting time; For vehicles Select service area The probability of; For service area In time period Predicted charging pile utilization rate; To achieve optimal system utilization, , , and These are the weighting coefficients; Step 413, To the vehicle Send guidance information : In the formula, Recommended service area name; Predict the optimal service area for the vehicle. Time; For vehicles in the optimal service area The estimated waiting time; Recommended charging time; For real-time electricity prices; List of alternative service areas; Step 414, Vehicle Confirm receipt of guidance information Then, return the satisfaction level. : in, Indicates satisfaction. Indicates vehicle The actual waiting time; Step 4.2 Service Area Dispatch Strategy Step 421: Calculate all service areas during the time period Predicted load factor ; In the formula: For time period The predicted number of charging vehicles; Average charging time for a single vehicle; For time period The predicted number of fast-charging vehicles; For time period The predicted number of vehicles charging slowly; For time period The average fast charging time; For time period The average charging time for slow charging; For time period Total duration; This represents the total number of charging stations across all service areas. Step 422: Based on the predicted load factor Adjust the charging station resources in service areas; Step 4221: First adjustment of charging pile resources in the service area: when At this time, it is necessary to activate the backup charging station and initiate dynamic pricing; when When this happens, the charging station needs to be shut down at 30% capacity and maintenance mode activated. when No processing is required at this time; Step 4222: Make a second adjustment to the charging pile resources in the service area: When the ratio of fast charging demand to slow charging demand in a service area is greater than 2, the type of charging station needs to be dynamically switched. Step 4223: When multiple vehicles compete for the same charging station resource, allocation needs to be based on vehicle priority. Steps 4-2-3: Service Area Status Release If communication is interrupted in a service area, it is necessary to return to step one to recalculate the guidance policy and push the update, and during the recalculation process, vehicles should not be assigned to the service area with communication interruption. If the communication status of all service areas is normal, each service area will publish real-time status information to the cloud. Published real-time status information for: In the formula, For service area In time period Real-time utilization rate; For service area In time period Predicted waiting time; For service area In time period The number of available charging stations; For service area In time period Real-time electricity price; For service area In time period state, ; service area The normal status is: number of available charging stations ; service area The busy status is: number of available charging stations ; service area The saturation state is: the number of available charging stations. .

9. A method for guiding the allocation of electric vehicle charging resources in a service area under an Internet of Things environment, as described in claim 8, is characterized in that... In step four, two, two, three, the vehicle priority is calculated as follows: In the formula, Indicates priority. It is a binary variable representing the degree of power urgency. If This indicates a critical power shortage, that is... The value of is 1; if This indicates a non-electrical energy emergency. The value is 0; For vehicle type coefficients, ; It is in reservation status. .

10. A method for guiding the allocation of electric vehicle charging resources in a service area under an Internet of Things environment, as described in claim 9, is characterized in that... The method describes real-time monitoring of resource usage in each service area: And every Time checks the following update conditions: In the formula, For service area In time period The estimated waiting time; For service area In time period The actual waiting time; The waiting time deviation threshold; For vehicles to service area Real-time distance; The maximum acceptable detour distance; when or If the value is 1, then it is necessary to return to step one to recalculate the bootstrapping strategy and push the update; Otherwise, no action is required.

Images