Vehicle charging time prediction method, device, equipment and medium

Abstract

The application belongs to the field of charging, and provides a vehicle charging time prediction method, device, equipment and medium, the method comprising: obtaining a vehicle charging dataset, the vehicle charging dataset comprising vehicle charging data, charging pile equipment data and vehicle charging duration forming a label variable; performing normalization processing on the vehicle charging dataset to obtain a normalized dataset; performing factor correlation analysis on the variables in the normalized dataset, and then constructing a logistic regression classifier for target variable classification; training the logistic regression classifier according to the normalized dataset to obtain a multi-classification normalized exponential function as a prediction model; inputting the current charging parameters of a target vehicle to be tested into the prediction model to obtain the remaining charging duration of the target vehicle, and the application embodiment can accurately predict the remaining charging time, can provide accurate reference for user power consumption, and improves user experience.

Description

Technical Field

[0001] This invention relates to the field of new energy vehicle charging, specifically to a method, device, equipment, and medium for predicting vehicle charging time. Background Technology

[0002] The battery management system (BMS) is the brain of a new energy vehicle, crucial for its safe and efficient operation. While the technical specifications of BMS are becoming increasingly mature, its charging time prediction function remains very rudimentary. Commonly used methods for predicting remaining charging time rely on a simple estimation of the remaining time based on SOC (state-of-charge) and charging current. However, this method is overly simplistic and prone to significant errors.

[0003] It is worth noting that during battery charging, in addition to SOC and charging current, many factors such as battery aging, battery temperature, ambient temperature, charging station equipment data, and the correlation between these data can affect charging time. However, it is difficult to obtain experimental results on how and to what extent these factors affect the charging time, thus making it impossible to establish an effective function prediction model. For example, Chinese patent CN113335125A discloses a vehicle charging method that calculates vehicle power consumption based on vehicle operating data and a pre-trained power consumption model; and calculates vehicle charging time based on vehicle charging data, vehicle power consumption, and the pre-trained charging model; however, it does not consider charging station equipment data or the correlation between these data. Chinese patent CN113968160A discloses a method for estimating vehicle charging time. It calculates a first predicted charging time using a peak temperature and a preset temperature threshold; calculates the charged capacity based on the charging current during the remaining charging phase and the first predicted charging time; calculates a second predicted charging time based on the current remaining battery power, the charged capacity, the total charging capacity of the vehicle battery, and a preset cooling charging current; and adds the first and second predicted charging times to obtain the total predicted charging time for the vehicle. However, it also fails to consider charging pile equipment data and the correlation between data points. Summary of the Invention

[0004] In view of the shortcomings of the prior art described above, this application provides a method, apparatus, device and medium for predicting vehicle charging time, so as to solve the technical problem of inaccurate prediction of vehicle charging time in the prior art.

[0005] In a first aspect, this application provides a method for predicting vehicle charging time, comprising:

[0006] Obtain a vehicle charging dataset, which includes vehicle charging data, charging pile equipment data, and vehicle charging duration forming a label variable.

[0007] The vehicle charging dataset is normalized to obtain a normalized dataset;

[0008] Factor correlation analysis is performed on the variables in the normalized dataset to construct a logistic regression classifier for classifying the target variable.

[0009] The logistic regression classifier is trained based on the normalized dataset to obtain a multi-class normalized exponential function as the prediction model.

[0010] The current charging parameters of the target vehicle are input into the prediction model to obtain the remaining charging time of the target vehicle.

[0011] In one embodiment of the invention, before normalizing the vehicle charging dataset, the method further includes:

[0012] Determine the number of missing variables for each of the vehicle charging datasets;

[0013] Variables with a number of missing variables greater than or equal to a preset missing threshold are deleted. Meanwhile, variables with a number of missing variables less than the preset missing threshold are filled with the mean value to complete the data preprocessing of the vehicle charging dataset.

[0014] In one embodiment of the invention, the vehicle charging dataset is normalized, including:

[0015] The numerical variables in the vehicle charging dataset are normalized to obtain a first normalized dataset, where the vehicle charging data are numerical variables.

[0016] The categorical variables in the vehicle charging dataset are normalized using a dummy variable generation algorithm to obtain a second normalized dataset. Some fields and label variables of the charging pile equipment data are categorical variables. The first normalized dataset and the second normalized dataset constitute the normalized dataset of the same dimension.

[0017] In one embodiment of the invention, factor correlation analysis is performed on the variables within the normalized dataset to construct a logistic regression classifier for classifying the target variable, including:

[0018] Factor correlation analysis was performed on the variables in the normalized dataset to obtain the independent and dependent variables.

[0019] A logistic regression classifier is constructed using the independent and dependent variables, and the variable coefficients Beta and R are calculated. 2 Wherein, the dependent variable is a categorical variable, Beta is used to characterize the influence of each variable on charging time, and R... 2 Used to characterize the reliability of each variable's prediction of charging time;

[0020]

[0021]

[0022] In the formula, the variable coefficients Beta and y are determined by minimizing L(θ). t It's a real label. It is a predicted label. It represents the average of the labels; α is the intercept, y is the label variable, and θ represents...

[0023] A vector combination of intercept and variable coefficients, where X represents a categorical variable and θ T Let θ be the transpose of θ.

[0024] In one embodiment of the invention, training the logistic regression classifier based on the normalized dataset to obtain a multi-class normalized exponential function as a prediction model includes:

[0025] The data carrying labeled variables in the normalized dataset are used as the training set, and the logistic regression classifier is trained using the training set; wherein, the objective function of the logistic regression classifier is:

[0026] P(y=1|x:θ)=g(θ T *[1,x])

[0027] P(y=0|x:θ)=1-g(θ T *[1,x])

[0028] Use the following multi-class normalized exponential function to identify the category of sample X;

[0029]

[0030] In the formula, w c Let X be the weight vector for class c, where the class label y∈{1,2,...,C} has C values, and X is a sample. It is a multi-category predicted label;

[0031] Cross-entropy is used as the loss function, and the weight vector is iteratively calculated using the stochastic gradient descent algorithm to obtain the trained prediction model.

[0032] In one embodiment of the invention, the risk function of the logistic regression classifier is:

[0033]

[0034] Among them, N training samples We use a C-dimensional one-hot vector to represent the category label. For category c, the vector is represented as: y = [I(1 = c), ..., I(C = c)]. T .

[0035] In one embodiment of the invention, the vehicle charging data includes the average temperature, average current, average voltage, remaining battery capacity at the start of charging, and remaining battery capacity at the end of charging for the target vehicle; the charging pile equipment data includes the charging pile operator, charging pile geographical location, charging pile rated current, charging pile rated voltage, charging pile cumulative power, charging pile connected vehicle list information, charging interface cumulative power, and three-phase current and voltage.

[0036] In a second aspect, this application provides a vehicle charging time prediction device, comprising:

[0037] The data acquisition module is used to acquire vehicle charging dataset, which includes vehicle charging data, charging pile equipment data, and vehicle charging time forming a label variable.

[0038] The normalization processing module is used to normalize the vehicle charging dataset to obtain a normalized dataset.

[0039] The correlation analysis module is used to perform factor correlation analysis on the variables in the normalized dataset, and then construct a logistic regression classifier for classifying the target variable.

[0040] The model training module is used to train the logistic regression classifier based on the normalized dataset to obtain a multi-class normalized exponential function as the prediction model.

[0041] The time prediction module is used to input the current charging parameters of the target vehicle to be tested into the prediction model to obtain the remaining charging time of the target vehicle.

[0042] In a third aspect, this application provides an electronic device comprising:

[0043] One or more processors;

[0044] A storage device for storing one or more programs that, when executed by one or more processors, cause the electronic device to implement the vehicle charging time prediction device described above.

[0045] In a fourth aspect, this application provides a vehicle device that includes the aforementioned electronic equipment.

[0046] In a fifth aspect, this application provides a computer-readable storage medium storing computer-readable instructions that, when executed by a computer's processor, cause the computer to execute the aforementioned vehicle charging time prediction device.

[0047] The beneficial effects of this application are as follows: This application obtains a vehicle charging dataset, which includes vehicle charging data, charging pile equipment data, and vehicle charging time as a label variable; it normalizes the vehicle charging dataset to obtain a normalized dataset; it performs factor correlation analysis on the variables in the normalized dataset, and then constructs a logistic regression classifier for classifying the target variable; because it considers the charging pile equipment data and the correlation between data, it not only provides comprehensive training data, but also considers the correlation between variables in the dataset, so that the logistic regression classifier is trained based on the normalized dataset to obtain a prediction model. Thus, the embodiments of this application can accurately predict the remaining charging time, provide accurate reference for users' electricity consumption, and improve user experience.

[0048] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0049] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application. It is obvious that the drawings described below are merely some embodiments of this application, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort. In the drawings:

[0050] Figure 1 This is a flowchart illustrating a vehicle charging time prediction method as an exemplary embodiment of this application;

[0051] Figure 2 This is a complete flowchart illustrating a vehicle charging time prediction method in an exemplary embodiment of this application;

[0052] Figure 3 This is a schematic diagram illustrating a vehicle charging time prediction method as shown in an exemplary embodiment of this application;

[0053] Figure 4 This is a structural block diagram of a vehicle charging time prediction device illustrated in an exemplary embodiment of this application;

[0054] Figure 5 This is a complete structural diagram of a vehicle charging time prediction device shown in an exemplary embodiment of this application;

[0055] Figure 6A schematic diagram of the structure of a computer system suitable for implementing the embodiments of this application is shown; Detailed Implementation

[0056] The embodiments of this application will be described below with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. This application can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this application. It should be understood that the preferred embodiments are only for illustrating this application and are not intended to limit the scope of protection of this application.

[0057] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of this application. Therefore, the drawings only show the components related to this application and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0058] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the present application. However, it will be apparent to those skilled in the art that embodiments of the present application may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the present application.

[0059] Please see Figure 1 An exemplary embodiment of this application illustrates a flowchart of a vehicle charging time prediction method, which is described in detail below:

[0060] Step S110: Obtain vehicle charging dataset, which includes vehicle charging data, charging pile equipment data, and vehicle charging time forming a label variable.

[0061] Specifically, the vehicle is typically a land-based vehicle with three or more wheels, and in this application, it specifically refers to new energy vehicles, including but not limited to pure electric vehicles, plug-in hybrid electric vehicles, range-extended hybrid electric vehicles, and electric vehicles.

[0062] It should be noted that the vehicle charging data includes the target vehicle's average temperature, average current, average voltage, remaining battery capacity at the start of charging, and remaining battery capacity at the end of charging; the charging pile equipment data includes the charging pile operator, charging pile geographical location, charging pile rated current, charging pile rated voltage, charging pile cumulative power, charging pile connected vehicle list information, charging interface cumulative power, and three-phase current and voltage.

[0063] For example, assuming the vehicle has been charging for t hours during the driving process, the details of obtaining the vehicle charging dataset are as follows:

[0064] Vehicle charging data: 1) Average battery temperature T during vehicle charging; 2) Initial SOC and final SOC1 during vehicle charging; 3) Average current I during vehicle charging; 4) Average voltage U during vehicle charging.

[0065] Charging pile equipment data: 1) Charging pile operator; large operators are denoted as 1, medium and small operators as 2, and the variable is denoted as M; 2) Charging pile geographical location: residential areas are denoted as 1, airports as 2, hotels or gas stations as 3, other areas as 4, and the variable is denoted as D; 3) Charging pile rated current (A); 4) Rated voltage upper limit (V) max Rated voltage lower limit V min ;5) Accumulated power of the charging station Q;6) Equipment list information during charging: number of charging devices X;7) Accumulated power of the charging device interface q;8) A, B, and C phase current AI and voltage UI.

[0066] Label variables: 1) Charging time interval S: 0-3 hours is recorded as 1; 4-6 hours as 2; 7-9 hours as 3; 10-12 hours as 4; 13-15 hours as 5.

[0067] It should also be noted that the charging station equipment data generated by the target vehicle at the corresponding charging station is associated with and bound to the vehicle, and is also associated with and bound to the tag variable.

[0068] Step S120: Normalize the vehicle charging dataset to obtain a normalized dataset;

[0069] Please see Figure 2 Before step S120, the method further includes:

[0070] Determine the number of missing variables for each of the vehicle charging datasets;

[0071] Step S200: Variables with a number of missing variables greater than or equal to a preset missing threshold are deleted. At the same time, variables with a number of missing variables less than the preset missing threshold are filled with the mean value, so that the vehicle charging dataset completes data preprocessing.

[0072] It should be understood that the preset missing threshold can be 30%. For example, variables with more than 30% missing values ​​can be deleted, and numeric variables with less than 30% missing values ​​can be filled with the mean missing value, while categorical variables can be filled with the mode.

[0073] By incorporating data preprocessing in the above manner, data quality can be improved, and subsequent data processing, analysis, and visualization processes can be made easier and more effective, leading to a better user experience. Preprocessing can take various forms, including data cleaning, data integration, data reduction, and data transformation, which are not limited to these specific steps here.

[0074] It should be noted that step S120 is described in detail below:

[0075] The numerical variables in the vehicle charging dataset are normalized to obtain a first normalized dataset, where the vehicle charging data are numerical variables.

[0076] For example, a data normalization algorithm is used to normalize numerical variables; assuming the numerical variables before preprocessing are X = [x1, x2, ... x...] t Let Y = [y1, y2, ..., y] be the variable after normalization preprocessing. t ],

[0077] The correlation between them is as follows:

[0078] In this embodiment, the vehicle charging data are all numerical variables. By using a data normalization algorithm for data preprocessing, the correlation between the numerical variables can be easily determined.

[0079] The categorical variables in the vehicle charging dataset are normalized using a dummy variable generation algorithm to obtain a second normalized dataset. Some fields and label variables of the charging pile equipment data are categorical variables. The first normalized dataset and the second normalized dataset constitute the normalized dataset of the same dimension.

[0080] For example, suppose the categorical variable X before preprocessing has t categories, and the value of each category is denoted as X(t). Then t-1 dummy variables can be generated, and the transformation relationship between the new variables and the original variables is as follows:

[0081]

[0082] In this embodiment, some fields in the charging pile equipment data, as well as the label variables, are categorical variables. By using a dummy variable generation algorithm for data preprocessing, the categorical variables can be quickly identified.

[0083] Step S130: Perform factor correlation analysis on the variables in the normalized dataset, and then construct a logistic regression classifier for classifying the target variable;

[0084] Specifically, factor correlation analysis is performed on the variables in the normalized dataset to obtain the independent and dependent variables;

[0085] A logistic regression classifier is constructed using the independent and dependent variables, and the variable coefficients Beta and R are calculated. 2 Wherein, the dependent variable is a categorical variable, Beta is used to characterize the influence of each variable on charging time, and R... 2 Used to characterize the reliability of each variable's prediction of charging time;

[0086]

[0087]

[0088] In the formula, the variable coefficients Beta and y are determined by minimizing L(θ). t It's a real label. It is a predicted label. α is the average of the labels; y is the label variable; θ represents the vector combination of the intercept and the variable coefficients; X represents the categorical variable; θ T Let θ be the transpose of θ.

[0089] In this embodiment, considering that the dependent variable is a categorical variable and that the correlation between independent variables needs to be removed, logistic regressions are performed on each independent variable against the dependent variable to calculate the variable coefficients Beta and R. 2 (Because there is only one independent variable, R) 2 (This can directly measure the reliability of the variable)

[0090] The formula for logistic regression is shown below (α is the intercept, Beta is the coefficient of the variable, y is the label variable, and θ represents the vector combination of the intercept and the coefficient of the variable, i.e., θ = [α, beta]):

[0091] Prediction function:

[0092]

[0093] Based on this prediction function, the categorical expression for the target variable can be obtained:

[0094] P(y=1|x:θ)=g(θ T *[1,x])

[0095] P(y=0|x:θ)=1-g(θ T *[1,x])

[0096] Combining these two formulas, we obtain the log-likelihood function:

[0097]

[0098] Finding its minimum value yields the value of Beta.

[0099] After obtaining the value of Beta, R 2 The formula for calculating the value of is as follows:

[0100]

[0101] Among them, y t It's a real label. It is a predicted label. It is the average value of the labels.

[0102] Calculate Beta and R 2 Then, thresholds can be set according to the actual situation to filter variables; Beta represents the influence of each variable on charging time. 2 This refers to the reliability of the prediction results of each variable for the charging time interval.

[0103] Step S140: Train the logistic regression classifier based on the normalized dataset to obtain a multi-class normalized exponential function as the prediction model;

[0104] Here, the dataset can be divided into a test set and a training set according to a preset ratio. The logistic regression classifier is trained using the training set to obtain a multi-class normalized exponential function as the prediction model. The accuracy of the prediction model is tested using the test set. For example, when the accuracy of the prediction model, i.e., the confidence level, reaches the preset accuracy rate, the prediction model is considered to have been successfully trained and can be used for subsequent testing of charging vehicles.

[0105] Step S150: Input the current charging parameters of the target vehicle to be tested into the prediction model to obtain the remaining charging time of the target vehicle.

[0106] The data carrying labeled variables in the normalized dataset are used as the training set, and the logistic regression classifier is trained using the training set; wherein, the objective function of the logistic regression classifier is:

[0107] P(y=1|x:θ)=g(θ T *[1,x])

[0108] P(y=0|x:θ)=1-g(θ T *[1,x])

[0109] Use the following multi-class normalized exponential function to identify the category of sample X;

[0110]

[0111] In the formula, w c Let X be the weight vector for the c-th class, where the class label y∈{1,2,…,C} has C values, and X is the sample. It is a multi-category predicted label;

[0112] Cross-entropy is used as the loss function, and the weight vector is iteratively calculated using the stochastic gradient descent algorithm to obtain the trained prediction model.

[0113] Specifically, the risk function of the logistic regression classifier is:

[0114]

[0115] Among them, N training samples We use a C-dimensional one-hot vector to represent the category label. For category c, the vector is represented as: y = [I(1 = c), ..., I(C = c)]. T .

[0116] In this embodiment, softmax regression solves multi-class classification problems (as opposed to binary classification problems solved by logistic regression), where the class label y can take multiple different values ​​(instead of two). The softmax function maps the outputs of multiple neurons (x values) to the (0, 1) interval, which can be understood as probabilities, thus enabling multi-class classification.

[0117] For a multi-class problem, the class label y∈{1,2,…,C} can have C values. Given a sample X, the conditional probability of belonging to class C predicted by Softmax regression is:

[0118] Among them, w c Let be the weight vector for class c. Therefore, the decision function of Softmax regression can be expressed as:

[0119]

[0120] The vector representation is as follows:

[0121]

[0122] Generally, Softmax regression uses cross-entropy as the loss function and optimizes the model parameters using the gradient descent algorithm, as follows:

[0123] Given N training samples We use a C-dimensional one-hot vector to represent the category label. For category c, the vector is represented as: y = [I(1 = c), ..., I(C = c)]. T .

[0124] The hazard function of the Softmax regression model is expressed in the following form:

[0125]

[0126] The gradient of the risk function with respect to W is:

[0127]

[0128] The training process for Softmax regression using gradient descent is as follows:

[0129] 1. Initialize W0←0;

[0130] 2. Update parameters according to the following rules:

[0131] See details Figure 3 The schematic diagram of a vehicle charging time prediction method shown in an exemplary embodiment of this application is described in detail below:

[0132] Data collection involves acquiring a vehicle charging dataset, which includes vehicle charging data, charging pile equipment data, and vehicle charging duration as a label variable.

[0133] Data preprocessing improves the data quality of vehicle charging datasets.

[0134] Normalization processing: The vehicle charging dataset is normalized to obtain a normalized dataset.

[0135] Factor correlation analysis is performed on the variables in the normalized dataset to construct a logistic regression classifier for classifying the target variable. Since the charging pile equipment data and the correlation between data are considered, the training data is not only comprehensive, but also the correlation between variables in the dataset is taken into account.

[0136] The training and testing process involves dividing the normalized dataset into a training set and a test set. Training is performed first, followed by testing. Once the test results meet the requirements, a prediction model is obtained. For example, the logistic regression classifier can be trained using the normalized dataset to obtain the prediction model. Thus, this embodiment of the application can accurately predict the remaining charging time, providing users with an accurate reference for electricity consumption and improving user experience.

[0137] For example, if the accuracy of the prediction model reaches 70%, then the prediction model training is complete. The relevant data of vehicle charging is input into the prediction model, and the prediction model makes predictions. The probability that the label variable corresponding to any charging time interval is equal to 1 is calculated. If the probability is greater than 0.5, the remaining charging time interval of the vehicle is informed to the car owner.

[0138] like Figure 4 As shown, the exemplary vehicle charging time prediction device 400 includes:

[0139] Data acquisition module 401 is used to acquire vehicle charging dataset, which includes vehicle charging data, charging pile equipment data and vehicle charging time forming a label variable.

[0140] The normalization processing module 402 is used to normalize the vehicle charging dataset to obtain a normalized dataset.

[0141] The correlation analysis module 403 is used to perform factor correlation analysis on the variables in the normalized dataset, and then construct a logistic regression classifier for classifying the target variable.

[0142] The model training module 404 is used to train the logistic regression classifier based on the normalized dataset to obtain a multi-class normalized exponential function as a prediction model.

[0143] The time prediction module 405 is used to input the current charging parameters of the target vehicle to be tested into the prediction model to obtain the remaining charging time of the target vehicle.

[0144] Based on the above embodiments, see details. Figure 5 ,and Figure 4 The difference lies in the fact that, prior to the normalization processing module 402, the vehicle charging time prediction device 400 also includes:

[0145] The preprocessing module 50 is used to determine the number of missing variables corresponding to each of the vehicle charging datasets; delete variables whose number of missing variables is greater than or equal to a preset missing threshold, and fill variables whose number of missing variables is less than the preset missing threshold with the mean, so that the vehicle charging datasets complete data preprocessing.

[0146] This exemplary vehicle charging time prediction device acquires a vehicle charging dataset, which includes vehicle charging data, charging pile equipment data, and vehicle charging duration as a label variable. The vehicle charging dataset is then normalized to obtain a normalized dataset. Factor correlation analysis is performed on the variables within the normalized dataset to construct a logistic regression classifier for classifying the target variable. Because the charging pile equipment data and the correlations between data points are considered, the training data is comprehensive and also considers the correlations between variables within the dataset. This allows the logistic regression classifier to be trained based on the normalized dataset, resulting in a prediction model. Thus, this embodiment can accurately predict the remaining charging time, providing users with an accurate reference for electricity usage and improving user experience.

[0147] It should be noted that the vehicle charging time prediction device and the vehicle charging time prediction method provided in the above embodiments belong to the same concept. The specific operation methods of each module and unit have been described in detail in the method embodiments and will not be repeated here. In practical applications, the vehicle charging time prediction device provided in the above embodiments can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. This is not a limitation here.

[0148] Embodiments of this application also provide an electronic device, including: one or more processors; and a storage device for storing one or more programs, which, when executed by the one or more processors, cause the electronic device to implement the vehicle charging time prediction device provided in the above embodiments.

[0149] Figure 6 A schematic diagram of a computer system suitable for implementing the embodiments of this application is shown. It should be noted that... Figure 6 The computer system 600 of the electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0150] like Figure 6 As shown, the computer system 600 includes a Central Processing Unit (CPU) 601, which can perform various appropriate actions and processes based on programs stored in Read-Only Memory (ROM) 602 or programs loaded from storage portion 608 into Random Access Memory (RAM) 603, such as performing the methods described in the above embodiments. The RAM 603 also stores various programs and data required for system operation. The CPU 601, ROM 602, and RAM 603 are interconnected via a bus 604. An Input / Output (I / O) interface 605 is also connected to the bus 604.

[0151] The following components are connected to I / O interface 605: an input section 606 including a keyboard, mouse, etc.; an output section 607 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN (Local Area Network) card, modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to I / O interface 605 as needed. A removable medium 611, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on drive 610 as needed so that computer programs read from it can be installed into storage section 608 as needed.

[0152] Specifically, according to embodiments of this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program including a computer program for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 609, and / or installed from removable medium 6811. When the computer program is executed by central processing unit (CPU) 601, it performs various functions defined in the system of this application.

[0153] It should be noted that the computer-readable medium shown in the embodiments of this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying a computer-readable computer program. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media can also be any computer-readable medium other than computer-readable storage media, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The computer program contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to wireless, wired, etc., or any suitable combination thereof.

[0154] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. Each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0155] The units described in the embodiments of this application can be implemented in software or hardware, and the described units can also be located in a processor. The names of these units do not necessarily limit the specific unit itself.

[0156] This application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the vehicle charging time prediction device as described above. This computer-readable storage medium may be included in the electronic device described in the above embodiments, or it may exist independently and not assembled into the electronic device.

[0157] The above embodiments are merely illustrative of the principles and effects of this application and are not intended to limit this application. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of this application. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in this application should still be covered by the claims of this application.

Claims

1. A method for predicting vehicle charging time, characterized in that, include: Obtain a vehicle charging dataset, which includes vehicle charging data, charging pile equipment data, and vehicle charging duration forming a label variable. The vehicle charging dataset is normalized to obtain a normalized dataset; Factor correlation analysis is performed on the variables in the normalized dataset to construct a logistic regression classifier for classifying the target variable. Specifically, factor correlation analysis is performed on the variables in the normalized dataset to obtain independent and dependent variables; a logistic regression classifier is constructed using the independent and dependent variables, and the variable coefficients Beta and R are calculated. 2 Wherein, the dependent variable is a categorical variable, Beta is used to characterize the influence of each variable on charging time, and R... 2 Used to characterize the reliability of each variable's prediction of charging time; The logistic regression classifier is trained based on the normalized dataset to obtain a multi-class normalized exponential function as the prediction model. Input the current charging parameters of the target vehicle to be tested into the prediction model to obtain the remaining charging time of the target vehicle; The vehicle charging data includes the target vehicle's average temperature, average current, average voltage, remaining battery capacity at the start of charging, and remaining battery capacity at the end of charging. The charging pile equipment data includes the charging pile operator, charging pile geographical location, charging pile rated current, charging pile rated voltage, charging pile cumulative power, charging pile connected vehicle list information, charging interface cumulative power, and three-phase current and voltage.

2. The vehicle charging time prediction method according to claim 1, characterized in that, Before normalizing the vehicle charging dataset, the following steps are also included: Determine the number of missing variables for each of the vehicle charging datasets; Variables with a number of missing variables greater than or equal to a preset missing threshold are deleted. Meanwhile, variables with a number of missing variables less than the preset missing threshold are filled with the mean value to complete the data preprocessing of the vehicle charging dataset.

3. The vehicle charging time prediction method according to claim 2, characterized in that, The vehicle charging dataset is normalized, including: The numerical variables in the vehicle charging dataset are normalized to obtain a first normalized dataset, where the vehicle charging data are numerical variables. The categorical variables in the vehicle charging dataset are normalized using a dummy variable generation algorithm to obtain a second normalized dataset. Some fields and label variables of the charging pile equipment data are categorical variables. The first normalized dataset and the second normalized dataset constitute the normalized dataset of the same dimension.

4. The vehicle charging time prediction method according to claim 1, characterized in that, The expression for the logistic regression classifier is: （1）； （2）； In the formula, by... Finding the minimum value determines the coefficient Beta of the variable. It's a real label. It is a predicted label. It is the average value of the labels; This is the intercept, and y is the label variable. This represents a vector combination of the intercept and variable coefficients, where X represents a categorical variable. for The transpose of .

5. The vehicle charging time prediction method according to claim 4, characterized in that, The logistic regression classifier is trained based on the normalized dataset to obtain a multi-class normalized exponential function as the prediction model, including: The data carrying labeled variables in the normalized dataset are used as the training set, and the logistic regression classifier is trained using the training set; wherein, the objective function of the logistic regression classifier is: Use the following multi-class normalized exponential function to identify the category of sample X; In the formula, Let c be the weight vector of class c, and let class label be 'c'. There are C possible values, and X is a sample. It is a multi-category predicted label; Cross-entropy is used as the loss function, and the weight vector is iteratively calculated using the stochastic gradient descent algorithm to obtain the trained prediction model.

6. The vehicle charging time prediction method according to claim 5, characterized in that, The risk function of the logistic regression classifier is: Among them, N training samples We use C-dimensional one-hot vectors to represent category labels. For category c, the vector representation is as follows: .

7. A vehicle charging time prediction device, characterized in that, include: The data acquisition module is used to acquire vehicle charging dataset, which includes vehicle charging data, charging pile equipment data, and vehicle charging time forming a label variable. The normalization processing module is used to normalize the vehicle charging dataset to obtain a normalized dataset. The correlation analysis module is used to perform factor correlation analysis on the variables in the normalized dataset, and then construct a logistic regression classifier for classifying the target variable. Specifically, factor correlation analysis is performed on the variables in the normalized dataset to obtain independent and dependent variables; a logistic regression classifier is constructed using the independent and dependent variables, and the variable coefficients Beta and R are calculated. 2 Wherein, the dependent variable is a categorical variable, Beta is used to characterize the influence of each variable on charging time, and R... 2 Used to characterize the reliability of each variable's prediction of charging time; The model training module is used to train the logistic regression classifier based on the normalized dataset to obtain a multi-class normalized exponential function as the prediction model. The time prediction module is used to input the current charging parameters of the target vehicle to be tested into the prediction model to obtain the remaining charging time of the target vehicle. The vehicle charging data includes the target vehicle's average temperature, average current, average voltage, remaining battery capacity at the start of charging, and remaining battery capacity at the end of charging. The charging pile equipment data includes the charging pile operator, charging pile geographical location, charging pile rated current, charging pile rated voltage, charging pile cumulative power, charging pile connected vehicle list information, charging interface cumulative power, and three-phase current and voltage.

8. An electronic device, characterized in that, include: One or more processors; A storage device for storing one or more programs, which, when executed by the one or more processors, cause the electronic device to perform the vehicle charging time prediction method according to any one of claims 1 to 6.

9. A vehicle device, characterized in that, Includes the electronic device as described in claim 8.

10. A computer-readable storage medium, characterized in that, It stores computer-readable instructions that, when executed by the processor of a computer, cause the computer to perform the vehicle charging time prediction method according to any one of claims 1 to 6.

Images