A Method and System for Predicting Air Conditioning Energy Consumption of Electric Vehicles Based on Driving Behavior Analysis

Abstract

This invention provides a method and system for predicting air conditioning energy consumption in electric vehicles based on driving behavior analysis. The method includes: extracting driving segment data from the electric vehicle, cleaning it to obtain effective data, obtaining vehicle driving segments based on driving status, acquiring operational data, and constructing an input feature set; obtaining air conditioning operation segments based on air conditioning status, acquiring energy consumption data to calculate the average air conditioning energy consumption, and obtaining an extended feature set; extracting the air conditioning set temperature and ambient temperature, constructing an air conditioning power prediction model, predicting air conditioning power based on the temperature parameters of the target vehicle, and obtaining a target feature set; constructing a regression model based on the least squares regression algorithm, and training and testing it using the target feature set to obtain an air conditioning energy consumption prediction model; acquiring the target parameters of the target vehicle, and obtaining the predicted air conditioning energy consumption through the air conditioning energy consumption prediction model. This invention improves the accuracy of air conditioning energy consumption prediction, facilitating rational route planning and alleviating range anxiety.

Description

Technical Field

[0001] This invention relates to the field of electric vehicle air conditioning energy consumption prediction technology, and in particular to an electric vehicle air conditioning energy consumption prediction method and system based on driving behavior analysis. Background Technology

[0002] With the increasing popularity of electric vehicles, their performance is receiving more and more attention. Looking at the current state of electric vehicle development worldwide, the main technical challenge lies in battery range. Limited by battery performance and inconvenient charging, electric vehicles cannot meet users' needs for flexible travel, requiring route planning before departure and assessing whether the vehicle's remaining battery power is sufficient. Furthermore, existing electric vehicles typically include a high-powered air conditioning system, which consumes a significant amount of energy, thus increasing the overall vehicle energy consumption. Over time, this can lead to excessive energy consumption, affecting the vehicle's range.

[0003] Therefore, there is an urgent need for a method that can predict the energy consumption of electric vehicle air conditioning, so as to facilitate reasonable travel route planning and alleviate range anxiety. Summary of the Invention

[0004] Therefore, it is necessary to provide a method and system for predicting the energy consumption of electric vehicle air conditioning based on driving behavior analysis to address the aforementioned technical problems.

[0005] A method for predicting air conditioning energy consumption of electric vehicles based on driving behavior analysis includes the following steps: extracting driving segment data of electric vehicles within a preset time period, and cleaning the driving segment data to obtain effective data; segmenting the effective data according to the driving state of the electric vehicle to obtain vehicle driving segments, obtaining the operating data in the vehicle driving segments, and constructing an input feature set based on the operating data; segmenting the air conditioning status in the vehicle driving segments to obtain air conditioning operating segments, calculating the average air conditioning energy consumption based on the energy consumption data in the air conditioning operating segments, and adding it to the input feature set to obtain an extended feature set; extracting the air conditioning set temperature and ambient temperature in the air conditioning operating segments, constructing an air conditioning power prediction model, obtaining the predicted air conditioning power based on the temperature parameters of the target vehicle, and adding it to the extended feature set to obtain a target feature set; establishing a regression model based on the least squares regression method of the machine learning algorithm, training and testing the regression model based on the target feature set to obtain an air conditioning energy consumption prediction model; obtaining the target parameters of the target vehicle and inputting them into the air conditioning energy consumption prediction model to obtain the predicted air conditioning energy consumption.

[0006] In one embodiment, the driving segment data includes a unique vehicle identifier, vehicle mass, vehicle length, vehicle width, vehicle height, ambient temperature, time, driving data timestamp, charging status, driving speed, segment duration, segment start charging state, segment end charging state, segment start mileage, segment end mileage, compressor energy consumption, and air conditioning energy consumption.

[0007] In one embodiment, the step of segmenting the valid data according to the driving state of the electric vehicle to obtain vehicle driving segments, acquiring operational data from the vehicle driving segments, and constructing an input feature set based on the operational data includes: detecting segments in which the electric vehicle is in a driving state based on the valid data; segmenting the valid data to obtain vehicle driving segments; extracting the start time and end time of the segments in a preset format; acquiring operational data from the vehicle driving segments, the operational data including the segment start state of charge, segment end state of charge, segment start mileage, segment end mileage, driving speed, maximum speed, average speed, percentage of high-speed driving time in the segment, ambient temperature, and air conditioning operating status; calculating the unit mileage power consumption of the electric vehicle under multiple operating conditions based on the operational data; and constructing an input feature set by combining the operational data.

[0008] In one embodiment, the process of segmenting the vehicle driving segment based on the air conditioning status to obtain an air conditioning operation segment, calculating the average air conditioning energy consumption based on the energy consumption data in the air conditioning operation segment, and adding it to the input feature set to obtain an extended feature set includes: detecting time nodes in the vehicle driving segment where the compressor energy consumption is greater than 0 or the air conditioning energy consumption is greater than 0, identifying the air conditioning as being in operation, and marking the air conditioning as being turned on with a timestamp; detecting time nodes in the vehicle driving segment where both the compressor energy consumption and the air conditioning energy consumption are 0, identifying the air conditioning as being turned off, and marking the time nodes as being turned off with a timestamp; obtaining the air conditioning operation segment based on the air conditioning as turned on and off timestamps; acquiring the compressor energy consumption, air conditioning energy consumption, and blower energy consumption in the air conditioning operation segment, and calculating the air conditioning energy consumption per trip based on the sampling period; acquiring the mileage of the air conditioning operation segment per trip, and calculating the average air conditioning energy consumption based on the mileage and the air conditioning energy consumption per trip, and adding it to the input feature set to obtain an extended feature set.

[0009] In one embodiment, the step of extracting the air conditioning set temperature and ambient temperature from the air conditioning operation segment, constructing an air conditioning power prediction model, obtaining the predicted air conditioning power based on the temperature parameters of the target vehicle, and adding it to the extended feature set to obtain the target feature set includes: extracting the air conditioning set temperature and ambient temperature from the air conditioning operation segment, and constructing an air conditioning power prediction model, which is:

[0010]

[0011] In the formula, P is the predicted air conditioning power, in W; T env The ambient temperature; T set Set the air conditioner temperature; a1 and b1 are the coefficients of the exponential function of the mathematical model; obtain the temperature parameters of the target vehicle, input the temperature parameters into the air conditioner power prediction model to obtain the predicted air conditioner power, the temperature parameters include the target air conditioner set temperature and the target ambient temperature; add the predicted air conditioner power to the extended feature set to obtain the target feature set.

[0012] In one embodiment, the step of establishing a regression model based on the least squares regression algorithm of the machine learning algorithm, and training and testing the regression model based on the target feature set to obtain an air conditioner energy consumption prediction model includes: establishing a regression model based on the least squares regression algorithm of the machine learning algorithm; dividing the target feature set into a training set and a test set; training the regression model using the training set; and testing the trained regression model using the test set to obtain an air conditioner energy consumption prediction model.

[0013] In one embodiment, obtaining the target parameters of the target vehicle and inputting them into the air conditioning energy consumption prediction model to obtain the predicted air conditioning energy consumption includes: obtaining the target parameters of the target vehicle, the target parameters including the segment start state of charge, segment start mileage, date, vehicle speed, temperature, weather, road type and predicted air conditioning power; and inputting the target parameters into the air conditioning energy consumption prediction model to obtain the predicted air conditioning energy consumption of the target vehicle.

[0014] An electric vehicle air conditioning energy consumption prediction system based on driving behavior analysis is provided to implement the electric vehicle air conditioning energy consumption prediction method based on driving behavior analysis as described above. The system includes: a segment data extraction module for extracting driving segment data of the electric vehicle within a preset time period and cleaning the driving segment data to obtain valid data; a driving segment segmentation module for segmenting the valid data according to the driving state of the electric vehicle to obtain vehicle driving segments, acquiring operational data within the vehicle driving segments, and constructing an input feature set based on the operational data; and an air conditioning segmentation module for segmenting the air conditioning data according to the air conditioning status within the vehicle driving segments to obtain air conditioning operation segments, and constructing an input feature set based on the air conditioning operation data. The energy consumption data in the segment is used to calculate the average energy consumption of the air conditioner and add it to the input feature set to obtain an extended feature set. The air conditioner power prediction module is used to extract the air conditioner set temperature and ambient temperature in the air conditioner operation segment, construct an air conditioner power prediction model, obtain the predicted air conditioner power based on the temperature parameters of the target vehicle, and add it to the extended feature set to obtain a target feature set. The prediction model construction module is used to build a regression model based on the least squares regression method of the machine learning algorithm, train and test the regression model based on the target feature set to obtain an air conditioner energy consumption prediction model. The air conditioner energy consumption prediction module is used to obtain the target parameters of the target vehicle and input them into the air conditioner energy consumption prediction model to obtain the predicted air conditioner energy consumption.

[0015] In one embodiment, the system further includes: a driving state detection module, used to detect segments of the electric vehicle in a driving state based on the valid data, and to segment the valid data to obtain vehicle driving segments; a segment time extraction module, used to extract the segment start time and segment end time in the vehicle driving segments in a preset format; a running acquisition module, used to acquire running data in the vehicle driving segments, the running data including segment start charging state, segment end charging state, segment start mileage, segment end mileage, driving speed, maximum speed, average speed, segment high-speed driving time percentage, ambient temperature, and air conditioning operating status; and a feature set construction module, used to calculate the unit mileage power consumption of the electric vehicle under multiple operating conditions based on the running data, and to construct an input feature set by combining the running data.

[0016] In one embodiment, the system further includes: an air conditioning start-up time detection module, used to detect time points in the vehicle driving segment where the compressor energy consumption is greater than 0 or the air conditioning energy consumption is greater than 0, determine that the air conditioning is in operation, and mark the air conditioning start-up timestamp; an air conditioning stop-down time detection module, used to detect time points in the vehicle driving segment where both the compressor energy consumption and the air conditioning energy consumption are 0, determine that the air conditioning is in operation, and mark the time points as air conditioning stop-down timestamps; a driving segment acquisition module, used to obtain the air conditioning driving segment based on the air conditioning start-up timestamp and the air conditioning stop-down timestamp; an air conditioning energy consumption calculation module, used to obtain the compressor energy consumption, air conditioning energy consumption, and blower energy consumption in the air conditioning driving segment, and calculate the air conditioning energy consumption per driving trip by combining the sampling period; and a feature set expansion module, used to obtain the single driving mileage of the air conditioning driving segment, calculate the average air conditioning energy consumption by combining the single driving mileage and the single driving air conditioning energy consumption, and add it to the input feature set to obtain an expanded feature set.

[0017] Compared with existing technologies, the advantages and beneficial effects of this invention are as follows: By extracting driving segment data of electric vehicles within a preset time period and cleaning it to obtain effective data, the effective data is segmented based on driving status to obtain vehicle driving segments. The operating data from the vehicle driving data is then used to construct an input feature set. Based on air conditioning status, the vehicle driving segments are segmented to obtain air conditioning operating segments. The energy consumption data of the air conditioning operating segments is obtained, the average air conditioning energy consumption is calculated and added to the input feature set to obtain an extended feature set. The air conditioning set temperature and ambient temperature from the air conditioning operating segments are extracted to construct an air conditioning power prediction model. Based on the temperature parameters of the target vehicle, the predicted air conditioning power is obtained and added to the extended feature set to obtain a target feature set. A regression model is constructed based on the least squares regression algorithm and trained and tested using the target feature set to obtain an air conditioning energy consumption prediction model. The target parameters of the target vehicle are obtained, and the predicted air conditioning energy consumption is obtained through the air conditioning energy consumption prediction model. This improves the accuracy of air conditioning energy consumption prediction for the target vehicle in the future, facilitating real-time and accurate energy consumption prediction, enabling drivers to rationally plan their trips and use electricity, and alleviating range anxiety. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating an electric vehicle air conditioning energy consumption prediction method based on driving behavior analysis in one embodiment.

[0019] Figure 2 This is a schematic diagram of the structure of an electric vehicle air conditioning energy consumption prediction system based on driving behavior analysis in one embodiment. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments and accompanying drawings. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0021] In one embodiment, such as Figure 1 As shown, a method for predicting the air conditioning energy consumption of electric vehicles based on driving behavior analysis is provided, including the following steps:

[0022] Step S101: Extract driving segment data of electric vehicles within a preset time period, and clean the driving segment data to obtain valid data.

[0023] Specifically, this embodiment is based on the actual vehicle operation data of electric vehicles. In order to estimate the air conditioning energy consumption demand in the future from the perspective of electric vehicles, it is necessary to extract the driving segment data of electric vehicles within a preset time period based on the vehicle network data warehouse. For example, the driving segment data of artists' pure electric vehicles purchased more than six months ago can be extracted from the vehicle network data warehouse within two years. Each frame of driving segment data is cleaned according to physical rules, abnormal data frames are filtered out, and real and effective data are retained as the data source for subsequent models, so as to obtain more accurate air conditioning energy consumption prediction results.

[0024] The driving segment data includes the vehicle's unique identifier, vehicle weight, vehicle length, vehicle width, vehicle height, ambient temperature, time, driving data timestamp, charging status, driving speed, segment duration, segment start state of charge, segment end state of charge, segment start mileage, segment end mileage, compressor energy consumption, and air conditioning energy consumption.

[0025] Step S102: The effective data is segmented according to the driving status of the electric vehicle to obtain vehicle driving segments, the operation data in the vehicle driving segments is obtained, and the input feature set is constructed based on the operation data.

[0026] Specifically, after obtaining valid data, the valid data is segmented according to whether the vehicle is in a driving state to obtain vehicle driving segments when the vehicle is in a driving state, and the vehicle operation data in the vehicle driving segments are obtained, such as the charging state at the beginning of the segment, the charging state at the end of the segment, the mileage at the beginning of the segment, the mileage at the end of the segment, the driving speed, the maximum speed, the average speed, and other data that affect vehicle energy consumption. An input feature set is constructed based on the operation data.

[0027] Step S102 includes: detecting segments of the electric vehicle in a driving state based on valid data, segmenting the valid data to obtain vehicle driving segments; extracting the start time and end time of the segments in a preset format; acquiring the operating data of the vehicle driving segments, including the starting state of charge, ending state of charge, starting mileage, ending mileage, driving speed, maximum speed, average speed, percentage of high-speed driving time, ambient temperature, and air conditioning operating status; calculating the unit mileage power consumption of the electric vehicle under multiple operating conditions based on the operating data, and constructing an input feature set by combining the operating data.

[0028] Specifically, the system detects segments of valid data where the electric vehicle is in motion, such as time points where the electric vehicle's power gear is not zero, and marks these segments as indicating that the electric vehicle is in motion at those time points. Based on these markings, the valid data is segmented into several vehicle motion segments. The time of each vehicle motion segment is then extracted according to a preset format, such as in hours, minutes, and seconds, to obtain the start and end times of each segment.

[0029] By acquiring operational data from vehicle driving segments, including basic data and processed data, such as vehicle mileage and segment duration, the average speed of the segment can be calculated. Vehicle mileage and segment duration are basic data, while average speed is processed data. Similarly, by acquiring the basic data of vehicle speed, the proportion of high-speed driving time in the processed data segment can be obtained based on a preset speed threshold of 80 km / h, thus obtaining more characteristics affecting energy consumption, which can facilitate accurate prediction of air conditioning energy consumption.

[0030] Based on the combination of fields such as the start and end of the charging state, the start and end mileage of the segment, driving speed, and temperature in the operating data, the power consumption per unit mileage under different operating conditions is calculated. All fields in the operating data and the power consumption per unit mileage are used as input features to construct an input feature set. All input features in the input feature set will affect the air conditioning energy consumption. Therefore, it is necessary to further expand the input feature set to improve the accuracy of air conditioning energy consumption prediction.

[0031] Step S103: Segment the air conditioning status in the vehicle driving segment to obtain the air conditioning operation segment. Calculate the average energy consumption of the air conditioning based on the energy consumption data in the air conditioning operation segment and add it to the input feature set to obtain the extended feature set.

[0032] Specifically, within a vehicle's driving segment, it is necessary to extract segments where the air conditioning is in operation. The driving segment is divided according to the air conditioning switch status to obtain air conditioning operation segments. Energy consumption data is extracted from the air conditioning operation segments, and the average energy consumption of the air conditioning is calculated based on the energy consumption data and the driving mileage corresponding to the air conditioning operation segments. This average energy consumption is then added to the input feature set to expand the feature set, resulting in an expanded feature set, which facilitates better prediction of air conditioning energy consumption.

[0033] Step S103 includes: detecting time points in the vehicle driving segment where the compressor energy consumption is greater than 0 or the air conditioning energy consumption is greater than 0, determining that the air conditioning is in operation, and marking the air conditioning start timestamp; detecting time points in the vehicle driving segment where both the compressor energy consumption and the air conditioning energy consumption are 0, determining that the air conditioning is in off state, and marking the air conditioning off timestamp; obtaining the air conditioning operating segment based on the air conditioning start timestamp and the air conditioning off timestamp; obtaining the compressor energy consumption, air conditioning energy consumption, and blower energy consumption in the air conditioning operating segment, and calculating the air conditioning energy consumption per trip by combining the sampling period; obtaining the single trip mileage of the air conditioning operating segment, and calculating the average air conditioning energy consumption by combining the single trip mileage and the single trip air conditioning energy consumption, and adding it to the input feature set to obtain the extended feature set.

[0034] Specifically, when detecting whether the air conditioner is running, the system determines whether it is running by checking if the compressor energy consumption or air conditioner energy consumption during a vehicle driving segment is greater than 0. If either the compressor energy consumption or the air conditioner energy consumption is greater than 0, the air conditioner is considered to be running, and a timestamp indicating the start of air conditioner operation is added to that time point. Conversely, if both the compressor energy consumption and the air conditioner energy consumption are 0, the air conditioner is considered to be off, and a timestamp indicating the off of air conditioner operation is added to that time point. Based on the air conditioner start and off timestamps in the vehicle driving segment, air conditioner operating segments are obtained. Simultaneously, the compressor energy consumption, air conditioner energy consumption, and blower energy consumption are obtained from the air conditioner operating segments. Combined with the sampling period, the air conditioner energy consumption per trip is calculated. The mileage of each trip in the air conditioner operating segment is obtained, and combined with the corresponding air conditioner energy consumption per trip, the average air conditioner energy consumption is calculated and added to the input feature set to obtain an extended feature set. This expansion of the feature set helps improve the accuracy of air conditioner energy consumption prediction.

[0035] Step S104: Extract the air conditioning set temperature and ambient temperature from the air conditioning operation segment, construct an air conditioning power prediction model, obtain the predicted air conditioning power based on the temperature parameters of the target vehicle, and add it to the extended feature set to obtain the target feature set.

[0036] Specifically, since the air conditioning set temperature and the ambient temperature are correlated, the air conditioning set temperature and the ambient temperature can be extracted during the air conditioning operation segment and used to build an air conditioning power prediction model. When predicting the air conditioning energy consumption of a target vehicle, the predicted air conditioning power can be obtained based on the temperature parameters of the target vehicle, namely the air conditioning set temperature and the ambient temperature. The predicted air conditioning power is then added to the extended feature set to obtain the target feature set, thereby further expanding the feature set and adding features related to air conditioning energy consumption to facilitate accurate prediction of air conditioning energy consumption.

[0037] Step S104 includes: extracting the air conditioner set temperature and ambient temperature from the air conditioner operation segment, and constructing an air conditioner power prediction model, which is:

[0038]

[0039] In the formula, P is the predicted air conditioning power, in W; T env The ambient temperature; T set Set the air conditioner temperature; a1 and b1 are the coefficients of the exponential function of the mathematical model; obtain the temperature parameters of the target vehicle, input the temperature parameters into the air conditioner power prediction model to obtain the predicted air conditioner power, the temperature parameters include the target air conditioner set temperature and the target ambient temperature; add the predicted air conditioner power to the extended feature set to obtain the target feature set.

[0040] Specifically, based on the air conditioning set temperature and ambient temperature extracted from the air conditioning operation segment, an air conditioning power prediction model is constructed. When it is necessary to predict the air conditioning energy consumption of a target vehicle, the temperature parameters of the target vehicle are obtained and input into the air conditioning power prediction model. The temperature parameters are the target air conditioning set temperature and the target ambient temperature, so as to predict the air conditioning power of the target vehicle and obtain the predicted air conditioning power. This predicted air conditioning power is then added to the extended feature set to obtain the target feature set. The air conditioning energy consumption prediction model can be trained through the target feature set.

[0041] Step S105: Establish a regression model based on the least squares regression method of the machine learning algorithm, train and test the regression model based on the target feature set, and obtain the air conditioning energy consumption prediction model.

[0042] Specifically, a regression model is established using the least squares regression method of machine learning algorithms. The acquired target feature set is used to train and test the constructed regression model to obtain an air conditioning energy consumption prediction model. This model can accurately predict the air conditioning energy consumption of a target vehicle over a future period, facilitating reasonable electricity planning. The target feature set includes fields such as vehicle weight, vehicle length, vehicle width, vehicle height, ambient temperature, weather, date, whether the air conditioning is running, time, morning peak hours, evening peak hours, drive position, and dashboard.

[0043] Step S105 includes: establishing a regression model based on the least squares regression algorithm of machine learning algorithm; dividing the target feature set into training set and test set, training the regression model with the training set, and testing the trained regression model with the test set to obtain the air conditioning energy consumption prediction model.

[0044] Specifically, after constructing a regression model using the least squares regression algorithm of machine learning, the target feature set is divided into a training set and a test set. For example, the training set:test set = 7:3 is randomly divided. The regression model is trained using the obtained training set, and the trained regression model is tested using the test set. After passing the test, an air conditioning energy consumption prediction model is obtained, which can accurately predict air conditioning energy consumption.

[0045] Step S106: Obtain the target parameters of the target vehicle and input them into the air conditioning energy consumption prediction model to obtain the predicted air conditioning energy consumption.

[0046] Specifically, when it is necessary to predict the air conditioning energy consumption of a target vehicle, the target parameters of the target vehicle, such as the state of charge at the start of the segment and the mileage at the start of the segment, are obtained and input into the air conditioning energy consumption prediction model. The model predicts the air conditioning energy consumption of the target vehicle in the future for a period of time. Based on the air conditioning energy consumption, the vehicle range can be predicted in real time, so that the driver can make reasonable plans for subsequent routes and battery power.

[0047] Step S106 includes: obtaining target parameters of the target vehicle, including the segment start state of charge, segment start mileage, date, vehicle speed, temperature, weather, road type and predicted air conditioning power; inputting the target parameters into the air conditioning energy consumption prediction model to obtain the predicted air conditioning energy consumption of the target vehicle.

[0048] Specifically, the target vehicle is identified, its target parameters are obtained, and these parameters are input into the air conditioning energy consumption prediction model. The predicted air conditioning energy consumption of the target vehicle is then obtained through the model. Based on the vehicle's external environmental conditions and its own characteristics, the prediction can be accurate down to the individual target vehicle. The prediction results are timely and improve the accuracy of air conditioning energy consumption prediction, which helps drivers to rationally plan their travel routes and alleviate range anxiety.

[0049] In this embodiment, driving segment data of electric vehicles within a preset time period is extracted and cleaned to obtain effective data. Based on the driving status, the effective data is segmented to obtain vehicle driving segments. Operational data from the vehicle driving data is obtained and used to construct an input feature set. Based on the air conditioning status, the vehicle driving segments are segmented to obtain air conditioning operation segments. Energy consumption data for the air conditioning operation segments is obtained, the average air conditioning energy consumption is calculated and added to the input feature set to obtain an extended feature set. The air conditioning set temperature and ambient temperature from the air conditioning operation segments are extracted to construct an air conditioning power prediction model. Based on the temperature parameters of the target vehicle, the predicted air conditioning power is obtained and added to the extended feature set to obtain a target feature set. A regression model is constructed based on the least squares regression algorithm and trained and tested using the target feature set to obtain an air conditioning energy consumption prediction model. The target parameters of the target vehicle are obtained, and the predicted air conditioning energy consumption is obtained through the air conditioning energy consumption prediction model. This improves the accuracy of air conditioning energy consumption prediction for the target vehicle in the future, facilitating real-time and accurate energy consumption prediction. This allows drivers to rationally plan their trips and use electricity, alleviating range anxiety.

[0050] like Figure 2 As shown, an electric vehicle air conditioning energy consumption prediction system 20 based on driving behavior analysis is provided to implement the electric vehicle air conditioning energy consumption prediction method based on driving behavior analysis as described above. It includes: a segment data extraction module 21, a driving segment segmentation module 22, an air conditioning segment segmentation module 23, an air conditioning power prediction module 24, a prediction model construction module 25, and an air conditioning energy consumption prediction module 26, wherein:

[0051] The segment data extraction module 21 is used to extract driving segment data of electric vehicles within a preset time period, and to clean the driving segment data to obtain valid data;

[0052] The driving segment segmentation module 22 is used to segment the effective data according to the driving status of the electric vehicle to obtain vehicle driving segments, acquire the running data in the vehicle driving segments, and construct an input feature set based on the running data;

[0053] The air conditioning segment segmentation module 23 is used to segment the air conditioning status in the vehicle driving segment to obtain the air conditioning operation segment. Based on the energy consumption data in the air conditioning operation segment, the average energy consumption of the air conditioning is calculated and added to the input feature set to obtain the extended feature set.

[0054] The air conditioning power prediction module 24 is used to extract the air conditioning set temperature and ambient temperature in the air conditioning operation segment, construct the air conditioning power prediction model, obtain the predicted air conditioning power based on the temperature parameters of the target vehicle, and add it to the extended feature set to obtain the target feature set.

[0055] The prediction model building module 25 is used to build a regression model based on the least squares regression method of the machine learning algorithm, train and test the regression model based on the target feature set, and obtain the air conditioning energy consumption prediction model.

[0056] The air conditioning energy consumption prediction module 26 is used to obtain the target parameters of the target vehicle and input them into the air conditioning energy consumption prediction model to obtain the predicted air conditioning energy consumption.

[0057] In one embodiment, the system further includes: a driving state detection module, used to detect segments of the electric vehicle in a driving state based on valid data, and segment the valid data to obtain vehicle driving segments; a segment time extraction module, used to extract the segment start time and segment end time in the vehicle driving segments in a preset format; a running acquisition module, used to acquire running data in the vehicle driving segments, including segment start state of charge, segment end state of charge, segment start mileage, segment end mileage, driving speed, maximum speed, average speed, segment high-speed driving time percentage, ambient temperature, and air conditioning operating status; and a feature set construction module, used to calculate the unit mileage power consumption of the electric vehicle under multiple operating conditions based on the running data, and construct an input feature set by combining the running data.

[0058] In one embodiment, the system further includes: an air conditioning start-up time detection module, used to detect time points in a vehicle driving segment where the compressor energy consumption is greater than 0 or the air conditioning energy consumption is greater than 0, to determine that the air conditioning is in operation, and to mark the air conditioning start-up timestamp; an air conditioning stop-down time detection module, used to detect time points in a vehicle driving segment where both the compressor energy consumption and the air conditioning energy consumption are 0, to determine that the air conditioning is in operation, and to mark the time points as air conditioning stop-down timestamps; a driving segment acquisition module, used to obtain the air conditioning driving segment based on the air conditioning start-up timestamp and the air conditioning stop-down timestamp; an air conditioning energy consumption calculation module, used to obtain the compressor energy consumption, air conditioning energy consumption, and blower energy consumption in the air conditioning driving segment, and to calculate the air conditioning energy consumption per driving trip by combining the sampling period; and a feature set expansion module, used to obtain the single driving mileage of the air conditioning driving segment, and to calculate the average air conditioning energy consumption by combining the single driving mileage and the single driving air conditioning energy consumption, and to add it to the input feature set to obtain an expanded feature set.

[0059] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0060] It is obvious to those skilled in the art that the modules or steps of the present invention described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Optionally, they can be implemented using computer-executable program code, thereby storing them in a computer storage medium (ROM / RAM, magnetic disk, optical disk) for execution by the computing device. In some cases, the steps shown or described can be performed in a different order than those described herein, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Therefore, the present invention is not limited to any particular hardware and software combination.

[0061] The above description, in conjunction with specific embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various simple deductions or substitutions can be made without departing from the concept of the present invention, and all such deductions or substitutions should be considered within the scope of protection of the present invention.

Claims

1. A method for predicting the air conditioning energy consumption of electric vehicles based on driving behavior analysis, characterized in that, Includes the following steps: Extract driving segment data of electric vehicles within a preset time period, and clean the driving segment data to obtain valid data; The effective data is segmented according to the driving status of the electric vehicle to obtain vehicle driving segments, and the operation data in the vehicle driving segments is obtained. An input feature set is constructed based on the operation data. The air conditioning operation segment is segmented based on the air conditioning status during the vehicle driving segment. An air conditioning operation segment is obtained. The average air conditioning energy consumption is calculated based on the energy consumption data within the air conditioning operation segment and added to the input feature set to obtain an extended feature set. This includes: detecting time points in the vehicle driving segment where either the compressor energy consumption or the air conditioning energy consumption is greater than 0, identifying the air conditioning as running, and marking the time point as air conditioning on / off timestamp; detecting time points in the vehicle driving segment where both compressor energy consumption and air conditioning energy consumption are 0, identifying the air conditioning as off, and marking the time point as air conditioning off timestamp; obtaining the air conditioning operation segment based on the air conditioning on / off timestamp; acquiring compressor energy consumption, air conditioning energy consumption, and blower energy consumption within the air conditioning operation segment, and calculating the air conditioning energy consumption per trip based on the sampling period; acquiring the single-trip mileage of the air conditioning operation segment, and calculating the average air conditioning energy consumption based on the single-trip mileage and single-trip air conditioning energy consumption, and adding this to the input feature set to obtain an extended feature set. Extract the air conditioning set temperature and ambient temperature from the air conditioning operation segment, construct an air conditioning power prediction model, obtain the predicted air conditioning power based on the temperature parameters of the target vehicle, and add it to the extended feature set to obtain the target feature set; A regression model is established based on the least squares regression method of the machine learning algorithm. The regression model is trained and tested based on the target feature set to obtain the air conditioning energy consumption prediction model. The target parameters of the target vehicle are obtained and input into the air conditioning energy consumption prediction model to obtain the predicted air conditioning energy consumption.

2. The method for predicting the energy consumption of electric vehicle air conditioning based on driving behavior analysis according to claim 1, characterized in that, The driving segment data includes the vehicle's unique identifier, vehicle weight, vehicle length, vehicle width, vehicle height, ambient temperature, time, driving data timestamp, charging status, driving speed, segment duration, segment start charging status, segment end charging status, segment start mileage, segment end mileage, compressor energy consumption, and air conditioning energy consumption.

3. The method for predicting the energy consumption of electric vehicle air conditioning based on driving behavior analysis according to claim 2, characterized in that, The effective data is segmented according to the driving status of the electric vehicle to obtain vehicle driving segments, and the operation data in the vehicle driving segments is obtained. An input feature set is constructed based on the operation data, including: Based on the valid data, segments of the electric vehicle in motion are detected, and the valid data are segmented to obtain vehicle motion segments; Extract the start and end times of the vehicle driving segments using a preset format; The vehicle's driving segment is used to obtain the operating data, which includes the segment's starting state of charge, the segment's ending state of charge, the segment's starting mileage, the segment's ending mileage, the driving speed, the maximum speed, the average speed, the percentage of high-speed driving time in the segment, the ambient temperature, and the air conditioning's operating status. Based on the operational data, the power consumption per unit mileage of the electric vehicle under multiple operating conditions is calculated, and an input feature set is constructed by combining the operational data.

4. The method for predicting the energy consumption of electric vehicle air conditioning based on driving behavior analysis according to claim 1, characterized in that, The process involves extracting the air conditioning set temperature and ambient temperature from the air conditioning operation segment, constructing an air conditioning power prediction model, obtaining the predicted air conditioning power based on the temperature parameters of the target vehicle, and adding it to the extended feature set to obtain the target feature set, which includes: Extracting the air conditioner set temperature and ambient temperature from the air conditioner operation segment, an air conditioner power prediction model is constructed as follows: In the formula, P is the predicted air conditioning power, in W; T env The ambient temperature; T set Set the temperature for the air conditioner; a1 and b1 are the coefficients of the exponential function in the mathematical model; The temperature parameters of the target vehicle are obtained and input into the air conditioning power prediction model to obtain the predicted air conditioning power. The temperature parameters include the target air conditioning set temperature and the target ambient temperature. The predicted air conditioning power is added to the extended feature set to obtain the target feature set.

5. The method for predicting the energy consumption of electric vehicle air conditioning based on driving behavior analysis according to claim 1, characterized in that, The regression model is established based on the least squares regression method of the machine learning algorithm. The regression model is trained and tested based on the target feature set to obtain the air conditioning energy consumption prediction model, including: A regression model is established based on the least squares regression algorithm using machine learning algorithms; The target feature set is divided into a training set and a test set. The regression model is trained using the training set and tested using the test set to obtain the air conditioning energy consumption prediction model.

6. The method for predicting the energy consumption of electric vehicle air conditioning based on driving behavior analysis according to claim 1, characterized in that, The process of acquiring the target parameters of the target vehicle and inputting them into the air conditioning energy consumption prediction model to obtain the predicted air conditioning energy consumption includes: Obtain target parameters for the target vehicle, including segment start state of charge, segment start mileage, date, vehicle speed, temperature, weather, road type, and predicted air conditioning power. The target parameters are input into the air conditioning energy consumption prediction model to obtain the predicted air conditioning energy consumption of the target vehicle.

7. An electric vehicle air conditioning energy consumption prediction system based on driving behavior analysis, characterized in that, A method for predicting the air conditioning energy consumption of an electric vehicle based on driving behavior analysis as described in any one of claims 1-6 includes: The segment data extraction module is used to extract driving segment data of electric vehicles within a preset time period, and to clean the driving segment data to obtain valid data; The driving segment segmentation module is used to segment the effective data into segments according to the driving status of the electric vehicle to obtain vehicle driving segments, acquire the running data in the vehicle driving segments, and construct an input feature set based on the running data. The air conditioning segment segmentation module is used to segment the vehicle driving segment according to the air conditioning status to obtain air conditioning operation segments. Based on the energy consumption data in the air conditioning operation segments, the average energy consumption of the air conditioning is calculated and added to the input feature set to obtain an extended feature set, including: an air conditioning start-up time detection module, used to detect time points in the vehicle driving segment where either the compressor energy consumption or the air conditioning energy consumption is greater than 0, identifying the air conditioning as running and marking it with an air conditioning start-up timestamp; and an air conditioning stop-down time detection module, used to detect time points in the vehicle driving segment where both compressor energy consumption and air conditioning energy consumption are 0. The system identifies when the air conditioner is off and marks the time point as an air conditioner off timestamp; the operation segment acquisition module obtains the air conditioner operation segment based on the air conditioner on and off timestamps; the air conditioner energy consumption calculation module obtains the compressor energy consumption, air conditioner energy consumption, and blower energy consumption in the air conditioner operation segment, and calculates the air conditioner energy consumption per trip by combining the sampling period; the feature set expansion module obtains the single trip mileage of the air conditioner operation segment, calculates the average air conditioner energy consumption by combining the single trip mileage and the single trip air conditioner energy consumption, and adds it to the input feature set to obtain the expanded feature set. An air conditioning power prediction module is used to extract the air conditioning set temperature and ambient temperature in the air conditioning operation segment, construct an air conditioning power prediction model, obtain the predicted air conditioning power based on the temperature parameters of the target vehicle, and add it to the extended feature set to obtain the target feature set; The prediction model building module is used to build a regression model based on the least squares regression method of the machine learning algorithm, and to train and test the regression model based on the target feature set to obtain the air conditioner energy consumption prediction model. The air conditioning energy consumption prediction module is used to obtain the target parameters of the target vehicle and input them into the air conditioning energy consumption prediction model to obtain the predicted air conditioning energy consumption.

8. The electric vehicle air conditioning energy consumption prediction system based on driving behavior analysis according to claim 7, characterized in that, The system also includes: The driving status detection module is used to detect segments of the electric vehicle in a driving state based on the valid data, and to segment the valid data to obtain vehicle driving segments. The segment time extraction module is used to extract the start time and end time of the vehicle driving segment in a preset format. The operation acquisition module is used to acquire the operation data in the vehicle driving segment. The operation data includes the charging state at the beginning of the segment, the charging state at the end of the segment, the mileage at the beginning of the segment, the mileage at the end of the segment, the driving speed, the maximum speed, the average speed, the proportion of high-speed driving time in the segment, the ambient temperature, and the air conditioning operation status. The feature set construction module is used to calculate the power consumption per unit mileage of the electric vehicle under multiple operating conditions based on the operating data, and to construct the input feature set by combining the operating data.

Images