Electric vehicle cluster charging and discharging load dynamic prediction method and device

Abstract

The invention provides an electric vehicle cluster charging and discharging load dynamic prediction method and device, and relates to the technical field of data processing, and the method comprises the steps: obtaining an electric vehicle cluster; initial SOC information is collected, and a driving data set is continuously collected; constructing an integrated charging and discharging predictor; multiple pieces of discharge load information are obtained through prediction, multiple pieces of predicted SOC information are obtained through calculation, multiple pieces of charging time information are obtained through prediction, and multiple pieces of charging load information are obtained through analysis; according to the method, the charging power is predicted and optimized, multiple pieces of optimal predicted charging power information are obtained, the charging power is controlled when multiple electric vehicles are charged through charging piles, and multiple pieces of discharging load information and multiple pieces of charging load information are combined to serve as multiple charging and discharging load prediction results. The technical problem of low utilization rate of charging facilities caused by unreasonable distribution of charging power when a plurality of electric vehicles are charged at the same time due to the fact that most electric vehicle data are dispersed in the prior art is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method and device for dynamically predicting charging and discharging loads of an electric vehicle cluster. Background Art

[0002] By optimizing the charging power distribution of electric vehicles, the utilization rate of charging piles can be improved, the excessive load on the power grid during the peak charging period can be avoided, and the safety and efficiency of the charging process can be ensured. In the prior art, the operation data and charging and discharging data of electric vehicles are mostly scattered, lacking effective integration and utilization, and unable to form a global management and optimization strategy; moreover, traditional charging and discharging load prediction methods are mostly simple linear models or models based on single features, lacking in-depth mining and utilization of multidimensional features and historical data, resulting in low prediction accuracy and inability to effectively cope with complex driving and charging behaviors. During the peak charging period, due to the lack of effective optimization strategies, when multiple electric vehicles are charging at the same time, the distribution of charging power is often unreasonable, resulting in low utilization of charging facilities, slow charging speed, and even causing charging overheating problems, affecting battery life and safety. Summary of the invention

[0003] The present application provides a method and device for dynamically predicting the charging and discharging load of an electric vehicle cluster, aiming to solve the technical problem in the prior art that the charging and discharging data of electric vehicles are mostly scattered, lack effective integration and utilization, and cannot form global management, resulting in unreasonable distribution of charging power when multiple electric vehicles are charged at the same time, leading to low utilization of charging facilities.

[0004] In a first aspect disclosed in the present application, a method for dynamically predicting charging and discharging loads of an electric vehicle cluster is provided, the method comprising: acquiring a plurality of electric vehicles of the same type connected by cluster protocol communication to form an electric vehicle cluster; when a preset time node is reached, collecting initial SOC information of a plurality of electric vehicles in the electric vehicle cluster, and continuously collecting driving data sets within a preset time range, wherein each driving behavior data set includes multiple types of driving characteristic information; constructing an integrated charging and discharging predictor based on driving record data and charging and discharging record data of a plurality of electric vehicles within a historical time period, the integrated charging and discharging predictor including a plurality of discharge prediction paths and a plurality of charging prediction paths corresponding to the multiple types of driving characteristic information, each of the discharge prediction path and the charging prediction path including a plurality of discharge prediction paths The invention provides a power prediction branch and multiple charging prediction branches; according to multiple driving data sets, multiple discharge prediction paths in the integrated charge and discharge predictor are used to predict and obtain multiple discharge load information, and multiple initial SOC information is combined to calculate and obtain multiple predicted SOC information, and the multiple charging prediction paths are used to predict and obtain multiple charging time information, and multiple predicted SOC information is combined to analyze and obtain multiple charging load information; according to the multiple charging load information, the charging power of the multiple electric vehicles is predicted and optimized to obtain multiple optimal predicted charging power information, and when the multiple electric vehicles are charged with charging piles, the charging power is controlled, and the multiple discharge load information and the multiple charging load information are combined as multiple charging and discharging load prediction results.

[0005] In a second aspect disclosed in the present application, a device for dynamically predicting charging and discharging loads of an electric vehicle cluster is provided, and the device is used for the above-mentioned method for dynamically predicting charging and discharging loads of an electric vehicle cluster, and the device includes: a module for acquiring electric vehicles of the same type, and the module for acquiring electric vehicles of the same type is used to acquire multiple electric vehicles of the same type that are connected through cluster protocol communication to form an electric vehicle cluster; a driving data acquisition module, and the driving data acquisition module is used to collect initial SOC information of multiple electric vehicles in the electric vehicle cluster when a preset time node is reached, and continuously collect driving data sets within a preset time range, wherein each driving behavior data set includes multiple types of driving characteristic information; a predictor construction module, and the predictor construction module is used to construct an integrated charging and discharging predictor based on driving record data and charging and discharging record data of multiple electric vehicles in a historical time period, and the integrated charging and discharging predictor includes multiple discharge prediction paths and multiple charging and discharging prediction paths corresponding to multiple types of driving characteristic information. An electric prediction path, each discharge prediction path and charging prediction path includes multiple discharge prediction branches and multiple charging prediction branches; a charging load acquisition module, the charging load acquisition module is used to predict and obtain multiple discharge load information based on multiple driving data sets, using multiple discharge prediction paths in the integrated charge and discharge predictor, and combining multiple initial SOC information to calculate and obtain multiple predicted SOC information, using the multiple charging prediction paths, predicting and obtaining multiple charging time information, combining the multiple predicted SOC information, and analyzing and obtaining multiple charging load information; a prediction result acquisition module, the prediction result acquisition module is used to predict and optimize the charging power of the multiple electric vehicles based on the multiple charging load information, obtain multiple optimal predicted charging power information, control the charging power when the multiple electric vehicles are charged using charging piles, and combine the multiple discharge load information and multiple charging load information as multiple charging and discharging load prediction results.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] By connecting similar electric vehicles through cluster protocol communication to form an electric vehicle cluster, multiple electric vehicles in the cluster can be uniformly managed and dispatched, which not only improves the comprehensiveness of the data, but also enables collaborative optimization, improves the utilization rate of charging facilities and the intelligence level of the charging process; collects initial SOC information at a preset time node, and continuously collects multiple types of driving characteristic information within a preset time range, which can comprehensively and real-time grasp the operating status of electric vehicles, provide an accurate data basis for subsequent predictions and optimizations, and improve the accuracy of predictions; based on historical driving record data and charging and discharging record data, an integrated charging and discharging predictor is constructed, which can make full use of the characteristics of historical data, improve the accuracy of the prediction model, and The synergistic effect of paths and branches further improves the prediction accuracy; by using multiple discharge and charging prediction paths within the integrated charge and discharge predictor, the prediction results of different paths and branches can be combined to obtain more accurate discharge load and charging time information. Combining the initial SOC information and the predicted SOC information, the charging load can be accurately calculated to improve the overall prediction effect; charging power prediction optimization based on the charging load information can achieve global optimal allocation of charging power for multiple electric vehicles, which not only improves the charging efficiency, but also effectively controls the charging overheating temperature and protects the battery life. In the actual charging process, by dynamically controlling the charging power, it can adapt to actual demand changes and ensure the efficiency and safety of the charging process.

[0008] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 A schematic flow chart of a method for dynamically predicting charging and discharging loads of an electric vehicle cluster provided in an embodiment of the present application;

[0010] Figure 2 A schematic diagram of the structure of a dynamic prediction device for charging and discharging loads of an electric vehicle cluster provided in an embodiment of the present application.

[0011] Description of the reference numerals: similar electric vehicle acquisition module 10, driving data acquisition module 20, predictor construction module 30, charging load acquisition module 40, prediction result acquisition module 50. DETAILED DESCRIPTION

[0012] The embodiment of the present application provides a method for dynamically predicting the charging and discharging load of an electric vehicle cluster, thereby solving the technical problem in the prior art that the charging and discharging data of electric vehicles are mostly scattered, lack effective integration and utilization, and cannot form global management, resulting in unreasonable distribution of charging power when multiple electric vehicles are charged at the same time, leading to low utilization of charging facilities.

[0013] After introducing the basic principles of the present application, various non-limiting implementation methods of the present application will be specifically introduced below in conjunction with the accompanying drawings of the specification. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0014] like Figure 1 As shown, an embodiment of the present application provides a method for dynamically predicting charging and discharging loads of an electric vehicle cluster, the method comprising:

[0015] A plurality of electric vehicles of the same type connected by cluster protocol communication are obtained to form an electric vehicle cluster.

[0016] First, a standardized cluster protocol is developed to unify the communication between similar electric vehicles. This method can be implemented by charging service providers. The owner submits an application through the vehicle's control system, such as an onboard computer or mobile phone application, and fills in the vehicle information, including vehicle model, region, etc. The cluster management system verifies the vehicle information and confirms that the vehicle meets the cluster requirements, including the same region and the same model. After passing the authentication, the vehicle obtains a unique cluster ID. When the vehicle is authenticated as a cluster member, the network parameters are configured on the vehicle to ensure that the vehicle can communicate in the cluster network. The cluster protocol is used to discover nodes between vehicles, identify and connect to other cluster members that accept the protocol. These cluster members are similar electric vehicles, including electric vehicles of the same region and model.

[0017] When a preset time node is reached, initial SOC information of multiple electric vehicles in the electric vehicle cluster is collected, and driving data sets within a preset time range are continuously collected, wherein each driving behavior data set includes multiple types of driving characteristic information.

[0018] The preset time node may be the time point when the electric vehicle is started, and this time point can be automatically detected by the vehicle system. When the vehicle is started, the vehicle system records the current time and sets it as the preset time node. At the preset time node, the initial SOC information of each electric vehicle in the electric vehicle cluster is collected. The SOC information represents the current state of charge of the battery, expressed as a percentage. For example, an SOC of 80% means that the battery currently has 80% of its charge.

[0019] Within a preset time range, such as one hour or half a day, the driving data sets of electric vehicles are continuously collected. These data sets include multiple types of driving characteristic information such as driving speed records, driving time records, and driving location records. The collected driving data is uploaded to the cluster management system through the cluster network for subsequent discharge load prediction.

[0020] Furthermore, when a preset time node is reached, initial SOC information of multiple electric vehicles in the electric vehicle cluster is collected, and driving data sets within a preset time range are continuously collected, including:

[0021] When a preset time node is reached, the initial SOC information of multiple electric vehicles in the electric vehicle cluster is collected; within a preset time range after the preset time node, the driving speed records, driving time records and driving location records of the multiple electric vehicles are collected as multiple types of driving characteristic information to obtain multiple driving data sets.

[0022] The preset time node may be the time point when the electric vehicle is started, and this time point can be automatically detected by the vehicle system. When the vehicle is started, the vehicle system records the current time and sets it as the preset time node. At the preset time node, the initial SOC information of each electric vehicle in the electric vehicle cluster is collected. The SOC information represents the current state of charge of the battery, expressed as a percentage. For example, an SOC of 80% means that the battery currently has 80% of its charge.

[0023] The preset time range can be one hour, half a day or a whole day, and is set according to specific needs. Within the preset time range after the preset time node, multiple types of driving characteristic information are continuously collected, including recording the vehicle's speed information during driving, recording the vehicle's total driving time and specific time period, using GPS to record the vehicle's driving trajectory and position, and aggregating the data to obtain multiple driving data sets, providing accurate data support for subsequent predictions and optimization.

[0024] Based on the driving record data and charging and discharging record data of multiple electric vehicles in historical time, an integrated charging and discharging predictor is constructed, wherein the integrated charging and discharging predictor includes multiple discharge prediction paths and multiple charging prediction paths corresponding to multiple types of driving characteristic information, and each discharge prediction path and charging prediction path includes multiple discharge prediction branches and multiple charging prediction branches.

[0025] Acquire the driving record data and charging and discharging record data of multiple electric vehicles in the historical period. The driving record data includes driving speed record, driving time record and driving location record, etc. The charging and discharging record data includes the time of each charging and discharging, SOC change, charging time length, etc.

[0026] Based on historical data, an integrated learning method is used to build and train multiple discharge prediction paths and charging prediction paths. One path corresponds to one type of data, and each path contains multiple prediction branches. The discharge load is predicted according to the driving conditions within the preset time range, and the time when charging can be predicted. For example, if the driving characteristic information reflects that the vehicle is busy, the charging time is only one hour, or the charging time is relatively long, the accuracy of each prediction path is evaluated using the test set, and weighted according to the accuracy. Multiple discharge prediction paths and multiple charging prediction paths are integrated to form an integrated charge and discharge predictor, which provides accurate prediction support for the charging and discharging management of electric vehicle clusters.

[0027] Furthermore, based on the driving record data and charging and discharging record data of multiple electric vehicles in historical time, an integrated charging and discharging predictor is constructed, including:

[0028] According to the driving record data and charging and discharging record data of multiple electric vehicles in historical time, a sample driving speed record set, a sample driving time record set, a sample driving location record set, a sample discharge load information set and a sample charging time information set are collected; the sample driving speed record set, the sample driving time record set and the sample driving location record set are respectively used as input, and combined with the sample discharge load information set and the sample charging time information set as output to obtain a discharge speed training data set, a charging speed training data set, a discharge time training data set, a charging time training data set, a discharge location training data set and a charging location training data set; based on ensemble learning, multiple discharge prediction paths and multiple charging prediction paths are trained to obtain an integrated charge and discharge predictor.

[0029] According to the driving record data of multiple electric vehicles in historical time, a sample driving speed record set, a sample driving time record set, and a sample driving location record set are extracted, wherein the sample driving speed record set records the driving speed data of the electric vehicle at different time points in the historical time, the sample driving time record set records the driving time data of the electric vehicle at different time points in the historical time, and the sample driving location record set records the driving location data of the electric vehicle at different time points in the historical time.

[0030] Based on the charging and discharging record data of multiple electric vehicles in historical time, a sample discharge load information set and a sample charging time information set are extracted, wherein the sample discharge load information set records the discharge load of the electric vehicle at different time points in historical time, that is, the power consumption data, and the sample charging time information set records the charging time data of the electric vehicle at different time points in historical time.

[0031] A training data set is constructed using a sample data set. Specifically, a sample driving speed record set is taken as input and a sample discharge load information set is taken as output to obtain a discharge speed training data set; a sample driving speed record set is taken as input and a sample charging time information set is taken as output to obtain a charging speed training data set; a sample driving time record set is taken as input and a sample discharge load information set is taken as output to obtain a discharge time training data set; a sample driving time record set is taken as input and a sample charging time information set is taken as output to obtain a charging time training data set; a sample driving location record set is taken as input and a sample discharge load information set is taken as output to obtain a discharge location training data set; a sample driving location record set is taken as input and a sample charging time information set is taken as output to obtain a charging location training data set.

[0032] An ensemble learning method is used to train multiple discharge prediction paths and multiple charging prediction paths. Specifically, based on a training data set, an ensemble learning algorithm, such as an ensemble tree model, an ensemble neural network, etc., is used to train multiple discharge prediction paths and multiple charging prediction paths. Each discharge prediction path predicts a discharge load according to different input features; each charging prediction path predicts a charging time according to different input features.

[0033] Furthermore, based on ensemble learning, multiple discharge prediction paths and multiple charge prediction paths are trained to obtain an integrated charge and discharge predictor, including:

[0034] A preset proportion of training data is randomly extracted from the discharge speed training data set to train the first speed discharge prediction branch in the speed discharge prediction path; a preset proportion of training data is randomly extracted from the discharge speed training data set again to train the second speed discharge prediction branch; training is continued to obtain multiple speed discharge prediction branches, which are integrated to obtain the first discharge prediction path; the charging speed training data set, the discharge time training data set, the charging time training data set, the discharge location training data set and the charging location training data set are continued to be used to train multiple discharge prediction paths and multiple charging prediction paths to obtain an integrated charge and discharge predictor.

[0035] Through the random sampling method, a certain proportion of data is randomly selected from the discharge speed training data set as training data to ensure the diversity of the training data. The training data uses a sample driving speed record set as input and a sample discharge load information set as output to perform model training. For example, a machine learning model such as a random forest, a neural network, etc. is used to train and obtain a first speed discharge prediction branch so that it can accurately predict the discharge load of electric vehicles at different speeds.

[0036] Another part of data is randomly extracted from the discharge speed training data set as training data, and the sample driving speed record set is also used as input and the sample discharge load information set is used as output to train the second speed discharge prediction branch.

[0037] Repeat the above steps to train more speed discharge prediction branches, and use ensemble learning techniques such as voting and weighted average to combine multiple trained discharge prediction branches to form the first discharge prediction path. By integrating multiple prediction branches, the accuracy and stability of discharge load prediction can be improved to cope with different driving conditions and data changes.

[0038] Repeat the training process of the first discharge prediction path, use the discharge time training data set and the discharge location training data set respectively, train to obtain the second discharge prediction path and the third discharge prediction path, and integrate to obtain multiple discharge prediction paths; use the charging speed training data set, the charging time training data set, and the charging location training data set respectively, train to obtain the first charging prediction path, the second charging prediction path, and the third charging prediction path, and integrate to obtain multiple charging prediction paths. Integrate multiple discharge prediction paths and multiple charging prediction paths to obtain an integrated charge and discharge predictor.

[0039] According to multiple driving data sets, multiple discharge prediction paths in the integrated charge and discharge predictor are used to predict and obtain multiple discharge load information, and multiple initial SOC information is combined to calculate and obtain multiple predicted SOC information. The multiple charging prediction paths are used to predict and obtain multiple charging time information, and the multiple predicted SOC information is combined to analyze and obtain multiple charging load information.

[0040] The discharge load information of each electric vehicle is predicted using multiple discharge prediction paths in the integrated charge and discharge predictor. Specifically, the driving data set of each electric vehicle is input into the discharge prediction path. Each discharge prediction path performs discharge load prediction based on the driving data to obtain multiple discharge load prediction results. The prediction results of multiple branches of each prediction path are averaged and weighted calculation is performed to obtain the final discharge load information.

[0041] Combined with the initial SOC information of each electric vehicle, the predicted SOC information is calculated. The predicted SOC is the predicted power consumption after the driving is completed. Specifically, the initial SOC is the power consumption of each electric vehicle at the time of departure, and the discharge load information is the predicted power consumption during driving. The predicted SOC is calculated by subtracting the discharge load information from the initial SOC, and the predicted SOC information is obtained through calculation.

[0042] The charging time information of each electric vehicle is predicted using multiple charging prediction paths in the integrated charging and discharging predictor. Specifically, the driving data set and predicted SOC information of each electric vehicle are input into the charging prediction path. Each charging prediction path predicts the charging time according to the driving data and the predicted SOC to obtain multiple charging time information. The prediction results of multiple branches of each prediction path are averaged, and weighted calculation is performed to obtain the final charging time information.

[0043] The charging load information is obtained by combining multiple predicted SOC information and charging time information. Specifically, the predicted SOC is the predicted power after the driving is completed, the charging time information is the predicted charging time required for a full charge, and the charging load information is calculated as the predicted SOC divided by the charging time, which represents the charging power requirement for a full charge, that is, the predicted charging load per time unit for a full charge.

[0044] Further, according to multiple driving data sets, multiple discharge prediction paths in the integrated charge and discharge predictor are used to predict and obtain multiple discharge load information, and multiple initial SOC information is combined to calculate and obtain multiple predicted SOC information, and multiple charging prediction paths are used to predict and obtain multiple charging time information, and multiple charging load information is analyzed and obtained in combination with the multiple predicted SOC information, including:

[0045] The multiple driving data sets are respectively input into the multiple discharge prediction paths in the integrated charge and discharge predictor for prediction, and the mean of the prediction results of the multiple discharge prediction branches is calculated to obtain multiple discharge load information sets; the multiple discharge prediction paths are tested to obtain multiple discharge prediction accuracy rates, and the multiple discharge load information sets are weighted to obtain multiple discharge load information; based on the multiple discharge load information, in combination with the multiple initial SOC information, multiple predicted SOC information is calculated; the multiple driving data sets are respectively input into the multiple charging prediction paths in the integrated charge and discharge predictor for prediction, and the mean of the prediction results of the multiple charging prediction branches is calculated to obtain multiple charging time information sets; the multiple charging prediction paths are tested to obtain multiple charging prediction accuracy rates, and the multiple charging time information sets are weighted to obtain multiple charging time information; based on the multiple predicted SOC information, multiple charging SOC information is calculated, and in combination with the multiple charging time information, multiple charging load information is analyzed and calculated.

[0046] Each driving data set is input into multiple discharge prediction paths in the integrated charge and discharge predictor for prediction. Each discharge prediction path predicts the future discharge load information of the electric vehicle based on historical data and model training results. For any discharge prediction path, multiple discharge load prediction results of multiple discharge prediction branches of the path are obtained. These prediction results are averaged to obtain the discharge load information of the discharge prediction path. Multiple discharge prediction paths are traversed in turn to obtain multiple discharge load information sets.

[0047] For each discharge prediction path, a test data set is used, which can be a separately reserved data set, to test and evaluate its prediction accuracy and generalization ability. The prediction accuracy of each discharge prediction path is calculated, which measures the performance of the model by comparing the difference between the prediction result and the actual observation value. According to the accuracy of each discharge prediction path, the predicted discharge load information is weighted to obtain the discharge load information of each path, among which the prediction path with higher accuracy will be given a higher weight to improve the reliability and accuracy of the overall prediction result. Through weighted calculation, the prediction results of multiple discharge prediction paths can be comprehensively utilized to obtain more accurate electric vehicle discharge load information.

[0048] The predicted SOC is the predicted power after the driving is completed. Specifically, the initial SOC is the power of each electric vehicle at the time of departure, and the discharge load information is the predicted power consumption during driving. The predicted SOC is calculated by subtracting the discharge load information from the initial SOC, and multiple predicted SOC information is obtained through subtraction calculation.

[0049] Each driving data set is input into multiple charging prediction paths in the integrated charging and discharging predictor for prediction. Each charging prediction path predicts the charging time information of the electric vehicle based on historical data and model training results. For each charging prediction path, multiple charging time prediction results of multiple charging prediction branches are obtained. The average of these prediction results is taken to obtain a charging time information set for each charging prediction path.

[0050] For each charging prediction path, a test data set is used to test and evaluate its prediction accuracy and generalization ability, and the prediction accuracy of each charging prediction path is calculated. According to the accuracy, the predicted charging time information is weighted and calculated to obtain multiple charging time information.

[0051] The predicted SOC information reflects the predicted SOC of each electric vehicle at the end of charging, and the charging time information represents the estimated charging time period for each electric vehicle. According to each predicted SOC information and the corresponding charging time information, the change in the SOC value of the electric vehicle at each time point during the charging process is calculated. This can be estimated by a simple integration method, that is, the charging rate at each time point in the charging process, that is, the charging power, is multiplied by the time period, and the change in the charging SOC is gradually accumulated. After calculation, multiple charging SOC information is obtained. The charging SOC information reflects the actual SOC change of each electric vehicle during the charging process. According to the charging SOC information and the charging time information, the charging power demand and charging efficiency of the electric vehicle during the charging process are analyzed. The average charging power demand of each electric vehicle in different time periods can be calculated, and then the energy consumption and charging efficiency in the charging process are evaluated. After calculation, multiple charging load information is obtained, which provides data support and decision-making basis for optimizing charging strategies and improving charging efficiency.

[0052] Based on the multiple charging load information, the charging power of the multiple electric vehicles is predicted and optimized to obtain multiple optimal predicted charging power information. When the multiple electric vehicles are charged using charging piles, the charging power is controlled and combined with the multiple discharge load information and the multiple charging load information as multiple charging and discharging load prediction results.

[0053] A charging power optimization model is established to increase the charging capacity and reduce the charging overheating temperature. Specifically, the possible range of electric vehicle charging power is defined to meet the safety threshold, that is, it cannot exceed the maximum allowable power. The optimization goal is to increase the charging capacity and reduce the charging overheating temperature. A cluster charging function is constructed to predict and optimize the actual charging power of multiple electric vehicles in the charging power space to obtain multiple optimal predicted charging power information.

[0054] In the actual charging process, the optimized predicted charging power information is used to control the charging power of each electric vehicle, and the charging status and temperature of each electric vehicle are monitored in real time, and the charging power is adjusted to ensure that it is within a safe range. According to real-time data, the charging power is dynamically adjusted to ensure charging efficiency and safety. Combined with the multiple discharge load information and multiple charging load information, multiple charging and discharging load prediction results are used.

[0055] Through the above steps, the charging power of electric vehicles is optimized, the charging efficiency is improved, and the risk of overheating during charging is reduced, providing a safe and effective charging management solution for electric vehicle clusters.

[0056] Furthermore, according to the plurality of charging load information, the charging power of the plurality of electric vehicles is predicted and optimized to obtain a plurality of optimal predicted charging power information, including:

[0057] Obtain the charging power space of the multiple electric vehicles; for the purpose of increasing the charging amount of the multiple electric vehicles and reducing the charging overheating temperature, construct a cluster charging function for predicting and optimizing the charging power of the multiple electric vehicles, as shown in the following formula:

[0058]

[0059] Among them, CCF is the cluster fitness, w1 and w2 are weights, M is the number of electric vehicles in the electric vehicle cluster, is the predicted charging power information of the i-th electric vehicle in the optimization, is the charging load information of the i-th electric vehicle, T y is the standard temperature of electric vehicle batteries, The charging overheat temperature of the i-th electric vehicle being optimized when charging according to the predicted charging power information; within the electric vehicle charging power space, the charging powers of the multiple electric vehicles are predicted and optimized according to the cluster charging function to obtain multiple optimal predicted charging power information.

[0060] The electric vehicle charging power space is the possible charging power range of each electric vehicle during the charging process. This range can be affected by factors such as charging pile capacity, electric vehicle battery technology and safety requirements.

[0061] Construct the cluster charging function as follows:

[0062]

[0063] and Represent the predicted charging power information and charging load information respectively. The first part of the function reflects the relationship between the predicted charging power and the actual charging demand; T y It is the standard temperature of the battery, which affects the thermal management and efficiency of the battery during charging. It is the charging overheat temperature predicted based on the predicted charging power information, which guides the temperature control during the charging process. According to the actual needs and optimization goals, the values ​​of weights w1 and w2 are set to adjust the balance between charging power optimization and thermal management. The sum of the two parts is calculated separately to obtain the final cluster fitness. Through this cluster charging function, the charging strategy of the electric vehicle cluster can be evaluated, so that not only the charging amount is maximized during the charging process, but also the temperature during the charging process can be effectively controlled to extend the battery life and improve the charging efficiency.

[0064] In the electric vehicle charging power space, the charging power of each electric vehicle is predicted and optimized according to the cluster charging function. The optimization goals include increasing the charging amount of the electric vehicle while reducing the charging overheating temperature to ensure charging safety and battery life. The specific optimization process is elaborated in subsequent steps and will not be repeated here.

[0065] Furthermore, within the electric vehicle charging power space, the charging powers of the plurality of electric vehicles are predicted and optimized according to the cluster charging function, including:

[0066] In the electric vehicle charging power space, a plurality of first predicted charging power information of a plurality of electric vehicles is randomly generated; based on the plurality of first predicted charging power information, a plurality of first charging overheat temperatures are predicted, and in combination with the plurality of charging load information, a first cluster fitness is calculated based on the cluster charging function, wherein a charging temperature predictor is trained by collecting a sample charging power set and a sample charging overheat temperature set, and a plurality of first charging overheat temperatures are predicted; according to the deviation amplitudes between the plurality of first charging overheat temperatures and the first average charging overheat temperature, a plurality of adjustment amplitudes are set, and in the electric vehicle charging power space, the plurality of first predicted charging power information are adjusted to obtain a plurality of second predicted charging power information, and a second cluster fitness is calculated; and the prediction optimization of the charging power of a plurality of electric vehicles is continued in the charging power space until convergence, and a plurality of predicted charging power information with the largest cluster fitness is output, and a plurality of optimal predicted charging power information is obtained.

[0067] In the defined charging power space, a predicted charging power value is randomly generated for each electric vehicle to obtain multiple first predicted charging power information. For each electric vehicle, the corresponding charging overheat temperature is predicted based on its predicted charging power information. This is accomplished by a pre-trained charging temperature predictor, which is trained with a sample charging power set as input data and a sample charging overheat temperature set as output data, and can map the charging power to the expected charging overheat temperature. After prediction by the predictor, multiple first charging overheat temperatures are obtained. Multiple first predicted charging power information, multiple first charging overheat temperatures, and multiple charging load information are input into the cluster charging function to calculate the first cluster fitness.

[0068] An average value of multiple first charging overheat temperatures is calculated as the first average charging overheat temperature. For each electric vehicle, a deviation between its predicted charging overheat temperature and the first average charging overheat temperature is calculated to obtain multiple deviation amplitudes. Multiple adjustment amplitudes are set according to the multiple deviation amplitudes, for example, the deviation amplitude is directly normalized and used as the adjustment amplitude.

[0069] Apply these adjustment ranges to the first predicted charging power information of each electric vehicle to generate multiple second predicted charging power information, and use the cluster charging function to calculate the second cluster fitness, similar to the above calculation method. In this way, the predicted charging power information can be adjusted based on the actual situation, and the effect of the adjusted charging strategy can be evaluated, thereby optimizing the power distribution and temperature control during the charging process and improving the charging efficiency.

[0070] Repeat the above steps, continuously adjust the predicted charging power information, and obtain the corresponding cluster fitness, compare all the obtained cluster fitnesses, and finally obtain multiple predicted charging power information with the largest cluster fitness as multiple optimal predicted charging power information.

[0071] In summary, the method for dynamic prediction of charging and discharging load of electric vehicle cluster provided in the embodiment of the present application has the following technical effects:

[0072] By connecting similar electric vehicles through cluster protocol communication to form an electric vehicle cluster, multiple electric vehicles in the cluster can be uniformly managed and dispatched, which not only improves the comprehensiveness of the data, but also enables collaborative optimization, improves the utilization rate of charging facilities and the intelligence level of the charging process; collects initial SOC information at a preset time node, and continuously collects multiple types of driving characteristic information within a preset time range, which can comprehensively and real-time grasp the operating status of electric vehicles, provide an accurate data basis for subsequent predictions and optimizations, and improve the accuracy of predictions; based on historical driving record data and charging and discharging record data, an integrated charging and discharging predictor is constructed, which can make full use of the characteristics of historical data, improve the accuracy of the prediction model, and The synergistic effect of paths and branches further improves the prediction accuracy; by using multiple discharge and charging prediction paths within the integrated charge and discharge predictor, the prediction results of different paths and branches can be combined to obtain more accurate discharge load and charging time information. Combining the initial SOC information and the predicted SOC information, the charging load can be accurately calculated to improve the overall prediction effect; charging power prediction optimization based on the charging load information can achieve global optimal allocation of charging power for multiple electric vehicles, which not only improves the charging efficiency, but also effectively controls the charging overheating temperature and protects the battery life. In the actual charging process, by dynamically controlling the charging power, it can adapt to actual demand changes and ensure the efficiency and safety of the charging process.

[0073] Based on the same inventive concept as the method for dynamic prediction of charging and discharging load of an electric vehicle cluster in the aforementioned embodiment, Figure 2 As shown, an embodiment of the present application provides a device for dynamically predicting charging and discharging loads of an electric vehicle cluster, the device comprising:

[0074] A similar electric vehicle acquisition module 10, wherein the similar electric vehicle acquisition module 10 is used to acquire a plurality of similar electric vehicles connected via cluster protocol communication to form an electric vehicle cluster;

[0075] A driving data collection module 20, wherein the driving data collection module 20 is used to collect initial SOC information of multiple electric vehicles in the electric vehicle cluster when a preset time node is reached, and continuously collect driving data sets within a preset time range, wherein each driving behavior data set includes multiple types of driving characteristic information;

[0076] A predictor construction module 30, the predictor construction module 30 is used to construct an integrated charge and discharge predictor based on the driving record data and the charging and discharging record data of multiple electric vehicles in the historical time, the integrated charge and discharge predictor includes multiple discharge prediction paths and multiple charging prediction paths corresponding to multiple types of driving characteristic information, each discharge prediction path and charging prediction path includes multiple discharge prediction branches and multiple charging prediction branches;

[0077] A charging load acquisition module 40, the charging load acquisition module 40 is used to predict and obtain multiple discharge load information based on multiple driving data sets, using multiple discharge prediction paths in the integrated charge and discharge predictor, and calculating and obtaining multiple predicted SOC information in combination with multiple initial SOC information, and using the multiple charging prediction paths to predict and obtain multiple charging time information, and analyzing and obtaining multiple charging load information in combination with the multiple predicted SOC information;

[0078] The prediction result acquisition module 50 is used to predict and optimize the charging power of the multiple electric vehicles based on the multiple charging load information, obtain multiple optimal predicted charging power information, control the charging power when the multiple electric vehicles are charged using charging piles, and combine the multiple discharge load information and multiple charging load information as multiple charging and discharging load prediction results.

[0079] Furthermore, the device further includes a driving data set acquisition module to perform the following operation steps:

[0080] When a preset time node is reached, the initial SOC information of multiple electric vehicles in the electric vehicle cluster is collected; within a preset time range after the preset time node, the driving speed records, driving time records and driving location records of the multiple electric vehicles are collected as multiple types of driving characteristic information to obtain multiple driving data sets.

[0081] Furthermore, the device further includes a charge and discharge predictor acquisition module to perform the following operation steps:

[0082] According to the driving record data and charging and discharging record data of multiple electric vehicles in historical time, a sample driving speed record set, a sample driving time record set, a sample driving location record set, a sample discharge load information set and a sample charging time information set are collected; the sample driving speed record set, the sample driving time record set and the sample driving location record set are respectively used as input, and combined with the sample discharge load information set and the sample charging time information set as output to obtain a discharge speed training data set, a charging speed training data set, a discharge time training data set, a charging time training data set, a discharge location training data set and a charging location training data set; based on ensemble learning, multiple discharge prediction paths and multiple charging prediction paths are trained to obtain an integrated charge and discharge predictor.

[0083] Furthermore, the device further includes an integrated charge and discharge predictor acquisition module to perform the following operation steps:

[0084] A preset proportion of training data is randomly extracted from the discharge speed training data set to train the first speed discharge prediction branch in the speed discharge prediction path; a preset proportion of training data is randomly extracted from the discharge speed training data set again to train the second speed discharge prediction branch; training is continued to obtain multiple speed discharge prediction branches, which are integrated to obtain the first discharge prediction path; the charging speed training data set, the discharge time training data set, the charging time training data set, the discharge location training data set and the charging location training data set are continued to be used to train multiple discharge prediction paths and multiple charging prediction paths to obtain an integrated charge and discharge predictor.

[0085] Furthermore, the device further includes a charging load information acquisition module to perform the following operation steps:

[0086] The multiple driving data sets are respectively input into the multiple discharge prediction paths in the integrated charge and discharge predictor for prediction, and the mean of the prediction results of the multiple discharge prediction branches is calculated to obtain multiple discharge load information sets; the multiple discharge prediction paths are tested to obtain multiple discharge prediction accuracy rates, and the multiple discharge load information sets are weighted to obtain multiple discharge load information; based on the multiple discharge load information, in combination with the multiple initial SOC information, multiple predicted SOC information is calculated; the multiple driving data sets are respectively input into the multiple charging prediction paths in the integrated charge and discharge predictor for prediction, and the mean of the prediction results of the multiple charging prediction branches is calculated to obtain multiple charging time information sets; the multiple charging prediction paths are tested to obtain multiple charging prediction accuracy rates, and the multiple charging time information sets are weighted to obtain multiple charging time information; based on the multiple predicted SOC information, multiple charging SOC information is calculated, and in combination with the multiple charging time information, multiple charging load information is analyzed and calculated.

[0087] Furthermore, the device further includes an optimal predicted charging power information acquisition module to perform the following operation steps:

[0088] Obtain the charging power space of the multiple electric vehicles; for the purpose of increasing the charging amount of the multiple electric vehicles and reducing the charging overheating temperature, construct a cluster charging function for predicting and optimizing the charging power of the multiple electric vehicles, as shown in the following formula:

[0089]

[0090] Among them, CCF is the cluster fitness, w1 and w2 are weights, M is the number of electric vehicles in the electric vehicle cluster, is the predicted charging power information of the i-th electric vehicle in the optimization, is the charging load information of the i-th electric vehicle, T y is the standard temperature of electric vehicle batteries, The charging overheat temperature of the i-th electric vehicle being optimized when charging according to the predicted charging power information; within the electric vehicle charging power space, the charging powers of the multiple electric vehicles are predicted and optimized according to the cluster charging function to obtain multiple optimal predicted charging power information.

[0091] Furthermore, the device further includes an optimal predicted charging power information acquisition module to perform the following operation steps:

[0092] In the electric vehicle charging power space, a plurality of first predicted charging power information of a plurality of electric vehicles is randomly generated; based on the plurality of first predicted charging power information, a plurality of first charging overheat temperatures are predicted, and in combination with the plurality of charging load information, a first cluster fitness is calculated based on the cluster charging function, wherein a charging temperature predictor is trained by collecting a sample charging power set and a sample charging overheat temperature set, and a plurality of first charging overheat temperatures are predicted; according to the deviation amplitudes between the plurality of first charging overheat temperatures and the first average charging overheat temperature, a plurality of adjustment amplitudes are set, and in the electric vehicle charging power space, the plurality of first predicted charging power information are adjusted to obtain a plurality of second predicted charging power information, and a second cluster fitness is calculated; and the prediction optimization of the charging power of a plurality of electric vehicles is continued in the charging power space until convergence, and a plurality of predicted charging power information with the largest cluster fitness is output, and a plurality of optimal predicted charging power information is obtained.

[0093] Through the above-mentioned detailed description of a method for dynamically predicting charging and discharging loads of an electric vehicle cluster, those skilled in the art can clearly understand a device for dynamically predicting charging and discharging loads of an electric vehicle cluster in this embodiment. Since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the description of the method part.

[0094] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for dynamic prediction of charging and discharging load of electric vehicle clusters, characterized in that: The method comprises: Acquire multiple electric vehicles of the same type connected by cluster protocol communication to form an electric vehicle cluster; When a preset time node is reached, initial SOC information of multiple electric vehicles in the electric vehicle cluster is collected, and driving data sets within a preset time range are continuously collected, wherein each driving behavior data set includes multiple types of driving characteristic information; Based on the driving record data and charging and discharging record data of multiple electric vehicles in historical time, an integrated charging and discharging predictor is constructed, wherein the integrated charging and discharging predictor includes multiple discharge prediction paths and multiple charging prediction paths corresponding to multiple types of driving characteristic information, and each discharge prediction path and charging prediction path includes multiple discharge prediction branches and multiple charging prediction branches; According to multiple driving data sets, multiple discharge prediction paths in the integrated charge and discharge predictor are used to predict and obtain multiple discharge load information, and multiple initial SOC information is combined to calculate and obtain multiple predicted SOC information, and multiple charging prediction paths are used to predict and obtain multiple charging time information, and multiple charging load information is analyzed and obtained in combination with the multiple predicted SOC information; Based on the multiple charging load information, the charging power of the multiple electric vehicles is predicted and optimized to obtain multiple optimal predicted charging power information. When the multiple electric vehicles are charged using charging piles, the charging power is controlled and combined with the multiple discharge load information and the multiple charging load information as multiple charging and discharging load prediction results.

2. The method for dynamic prediction of charging and discharging load of electric vehicle cluster according to claim 1, characterized in that: When a preset time node is reached, the initial SOC information of multiple electric vehicles in the electric vehicle cluster is collected, and the driving data set within the preset time range is continuously collected, including: When a preset time node is reached, initial SOC information of multiple electric vehicles in the electric vehicle cluster is collected; Within a preset time range after the preset time node, the driving speed records, driving time records and driving location records of the multiple electric vehicles are collected as multiple types of driving characteristic information to obtain multiple driving data sets.

3. The method for dynamic prediction of charging and discharging load of electric vehicle cluster according to claim 1, characterized in that: Based on the driving record data and charging and discharging record data of multiple electric vehicles in the historical time, an integrated charging and discharging predictor is constructed, including: According to the driving record data and charging and discharging record data of multiple electric vehicles in historical time, a sample driving speed record set, a sample driving time record set, a sample driving location record set, a sample discharge load information set and a sample charging time information set are collected; The sample driving speed record set, the sample driving time record set, and the sample driving location record set are respectively used as inputs, and combined with the sample discharge load information set and the sample charging time information set as outputs to obtain a discharge speed training data set, a charging speed training data set, a discharge time training data set, a charging time training data set, a discharge location training data set, and a charging location training data set; Based on ensemble learning, multiple discharge prediction paths and multiple charge prediction paths are trained to obtain an integrated charge and discharge predictor.

4. The method for dynamic prediction of charging and discharging load of electric vehicle cluster according to claim 3 is characterized in that: Based on ensemble learning, multiple discharge prediction paths and multiple charge prediction paths are trained to obtain an integrated charge and discharge predictor, including: Randomly extracting a preset proportion of training data from the discharge speed training data set to train a first speed discharge prediction branch in a speed discharge prediction path; Re-randomly extracting a preset proportion of training data from the discharge speed training data set to train a second speed discharge prediction branch; Continue training to obtain multiple speed discharge prediction branches, and integrate to obtain the first discharge prediction path; Continue to use the charging speed training data set, the discharging time training data set, the charging time training data set, the discharging location training data set and the charging location training data set to train and obtain multiple discharge prediction paths and multiple charging prediction paths to obtain an integrated charge and discharge predictor.

5. The method for dynamic prediction of charging and discharging load of electric vehicle cluster according to claim 4, characterized in that: According to multiple driving data sets, multiple discharge prediction paths in the integrated charge and discharge predictor are used to predict and obtain multiple discharge load information, and multiple initial SOC information is combined to calculate and obtain multiple predicted SOC information, and multiple charging prediction paths are used to predict and obtain multiple charging time information, and multiple charging load information is analyzed and obtained in combination with the multiple predicted SOC information, including: Inputting the multiple driving data sets into multiple discharge prediction paths in the integrated charge and discharge predictor for prediction, and calculating the average of the prediction results of multiple discharge prediction branches to obtain multiple discharge load information sets; Testing the multiple discharge prediction paths to obtain multiple discharge prediction accuracy rates, and performing weighted calculation on the multiple discharge load information sets to obtain multiple discharge load information; Calculating and obtaining a plurality of predicted SOC information according to the plurality of discharge load information and in combination with the plurality of initial SOC information; Inputting the multiple driving data sets into multiple charging prediction paths in the integrated charging and discharging predictor for prediction, and calculating the average of the prediction results of multiple charging prediction branches to obtain multiple charging time information sets; Testing the multiple charging prediction paths to obtain multiple charging prediction accuracy rates, and performing weighted calculation on the multiple charging time information sets to obtain multiple charging time information; Based on the multiple predicted SOC information, multiple charging SOC information is calculated and obtained, and combined with the multiple charging time information, multiple charging load information is analyzed and calculated to obtain multiple charging load information.

6. The method for dynamic prediction of charging and discharging load of electric vehicle cluster according to claim 1, characterized in that: According to the plurality of charging load information, the charging power of the plurality of electric vehicles is predicted and optimized to obtain a plurality of optimal predicted charging power information, including: Acquire the multiple electric vehicle charging power spaces; In order to increase the charging capacity of multiple electric vehicles and reduce the overheating temperature of charging, a cluster charging function is constructed to predict and optimize the charging power of multiple electric vehicles, as shown in the following formula: Among them, CCF is the cluster fitness, w1 and w2 are weights, M is the number of electric vehicles in the electric vehicle cluster, is the predicted charging power information of the i-th electric vehicle in the optimization, is the charging load information of the i-th electric vehicle, T y is the standard temperature of electric vehicle batteries, is the charging overheat temperature of the i-th electric vehicle in the optimization when charging according to the predicted charging power information; In the electric vehicle charging power space, the charging powers of the plurality of electric vehicles are predicted and optimized according to the cluster charging function to obtain a plurality of optimal predicted charging power information.

7. The method for dynamic prediction of charging and discharging load of electric vehicle cluster according to claim 6, characterized in that: In the electric vehicle charging power space, according to the cluster charging function, the charging powers of the plurality of electric vehicles are predicted and optimized, including: In the electric vehicle charging power space, randomly generating a plurality of first predicted charging power information of a plurality of electric vehicles; Predicting and obtaining a plurality of first charging overheat temperatures according to the plurality of first predicted charging power information, and calculating and obtaining a first cluster fitness based on the cluster charging function in combination with the plurality of charging load information, wherein a charging temperature predictor is trained by collecting a sample charging power set and a sample charging overheat temperature set, and a plurality of first charging overheat temperatures are predicted; According to the deviation ranges of the multiple first charging overheat temperatures and the first average charging overheat temperature, multiple adjustment ranges are set, and within the electric vehicle charging power space, the multiple first predicted charging power information are adjusted to obtain multiple second predicted charging power information, and the second cluster fitness is calculated; Continue to perform prediction optimization of the charging power of multiple electric vehicles in the charging power space until convergence, output multiple predicted charging power information with the maximum cluster fitness, and obtain multiple optimal predicted charging power information.

8. A dynamic prediction device for charging and discharging load of electric vehicle clusters, characterized in that: The device is used to implement a method for dynamically predicting charging and discharging loads of an electric vehicle cluster as described in any one of claims 1 to 7, and comprises: A module for acquiring electric vehicles of the same type, wherein the module is used to acquire a plurality of electric vehicles of the same type connected by cluster protocol communication to form an electric vehicle cluster; A driving data collection module, the driving data collection module is used to collect initial SOC information of multiple electric vehicles in the electric vehicle cluster when a preset time node is reached, and continuously collect driving data sets within a preset time range, wherein each driving behavior data set includes multiple types of driving characteristic information; A predictor construction module, the predictor construction module is used to construct an integrated charge and discharge predictor based on the driving record data and the charging and discharging record data of multiple electric vehicles in the historical time, the integrated charge and discharge predictor includes multiple discharge prediction paths and multiple charging prediction paths corresponding to multiple types of driving characteristic information, each discharge prediction path and charging prediction path includes multiple discharge prediction branches and multiple charging prediction branches; A charging load acquisition module, the charging load acquisition module is used to predict and obtain multiple discharge load information based on multiple driving data sets, using multiple discharge prediction paths in the integrated charge and discharge predictor, and calculating and obtaining multiple predicted SOC information in combination with multiple initial SOC information, and using the multiple charging prediction paths to predict and obtain multiple charging time information, and analyzing and obtaining multiple charging load information in combination with the multiple predicted SOC information; A prediction result acquisition module, the prediction result acquisition module is used to predict and optimize the charging power of the multiple electric vehicles based on the multiple charging load information, obtain multiple optimal predicted charging power information, control the charging power when the multiple electric vehicles are charged using charging piles, and combine the multiple discharge load information and multiple charging load information as multiple charging and discharging load prediction results.

Images