Electric motor coach remaining mileage prediction method and device, electronic equipment and storage medium

Abstract

The invention discloses a method for predicting the remaining mileage of an electric motor coach, which comprises the following steps of: acquiring real-time battery state evaluation data, real-time driving state data, historical driving state data and real-time environment state data of the current electric motor coach, and preprocessing the acquired data; establishing a prediction machine learning model of the electric quantity endurance mileage of the electric motor coach; and according to the prediction machine learning model of the electric quantity endurance mileage of the electric motor coach, estimating the residual endurance mileage of the electric motor coach in real time. According to the method, the prediction machine learning model can be constructed through comprehensive analysis of multiple factors such as the battery state, the environmental condition, historical data and real-time power consumption, and the prediction model is dynamically adjusted according to the actual driving condition of the vehicle, so that the accuracy and timeliness of continuous driving mileage prediction of the electric motor coach are remarkably improved.

Description

Technical Field

[0001] The invention relates to the technical field of electric bus mileage prediction, and in particular to an electric bus remaining mileage prediction method, device, electronic equipment and storage medium. Background Art

[0002] In existing technologies, the remaining mileage of electric buses is generally calculated by combining the battery capacity, the current battery state of charge (SOC), and historical energy consumption for comprehensive prediction and judgment. Some calculation methods also take weather factors and road conditions into consideration. However, due to the large fluctuations in the number of people in electric buses and the changing driving environment, the prediction results are often too different from the actual remaining mileage. In addition, since the data used is not calculated in real time, the prediction results are not very timely. Summary of the invention

[0003] The object of the present invention is to provide a method, device, electronic device and storage medium for predicting the remaining mileage of an electric bus, so as to solve the problem that the prediction result of the remaining mileage of an electric bus in the prior art has poor timeliness.

[0004] The purpose of the present invention can be achieved through the following technical solutions: A method for predicting remaining mileage of an electric bus comprises the following steps: S10, obtaining real-time battery status evaluation data, real-time driving status data, historical driving status data and real-time environmental status data of the current electric bus; S20, pre-processing the acquired real-time battery status evaluation data, real-time driving status data, historical driving status data and real-time environmental status data, and establishing a machine learning model for predicting the battery life of electric buses; S30: Based on the prediction machine learning model of the electric bus's battery life, the remaining range of the electric bus is estimated in real time, and the remaining range is displayed visually and / or announced by voice.

[0005] As a further solution of the present invention: the method for obtaining the real-time battery status evaluation data, real-time driving status data, historical driving status data and real-time environmental status data of the current electric bus in step S10 is: The real-time battery status evaluation data is collected by the vehicle battery management system BMS. The real-time battery status evaluation data includes the current remaining power SOC and the actual battery capacity data. , Get the initial battery capacity data through factory data and initial battery energy ; The real-time driving status data is acquired by the vehicle-mounted data acquisition module, and the real-time driving status data includes real-time current data I, real-time voltage data U and acquisition time interval data t; The historical driving status data is collected by the GPS module and is the vehicle's daily mileage. Daily driving time , based on the vehicle's daily mileage Daily driving time The vehicle's daily driving power consumption is collected through the vehicle's battery management system BMS ; The real-time environmental status data is the current vehicle driving road condition information collected by the GPS module and weather information .

[0006] As a further solution of the present invention: the method for preprocessing the acquired real-time battery status evaluation data, real-time driving status data, historical driving status data and real-time environmental status data in step S20 is: Data cleaning, removing erroneous data and incomplete data in the real-time battery status evaluation data, the real-time driving status data, the historical driving status data and the real-time environmental status data; Data integration, merging multiple different data obtained in the data cleaning step into a unified data set and storing it uniformly; Data transformation: normalizes, discretizes, and reduces the stored data in the unified data set, and converts it into input feature parameters suitable for predictive machine learning models.

[0007] As a further solution of the present invention: the method for establishing a prediction machine learning model for the electric bus's battery life in step S20 is: According to the real-time battery status evaluation data, the current remaining power SOC input characteristic parameters and the initial battery power are determined Enter the characteristic parameters and calculate the formula Obtain battery health SOH input characteristic parameters; According to the real-time driving status data, and by calculating the formula Get the real-time power consumption of the vehicle Input feature parameters, where , and P is a certain moment Real-time power; According to the historical driving status data, and by calculating the formula Get historical energy consumption per mile parameter; According to the real-time environmental status data, a vehicle unit mileage energy consumption model is established, and the historical unit mileage energy consumption is converted into Adjust to current energy consumption per mileage Input feature parameters; The current remaining power SOC, the initial battery power Input characteristic parameters, the battery health SOH input characteristic parameters, the real-time power consumption Input characteristic parameters and the current unit mileage energy consumption After the input feature parameters are determined, the training and establishment of the predictive machine learning model is completed.

[0008] As a further solution of the present invention: in the process of establishing the vehicle unit mileage energy consumption model, the current unit mileage energy consumption pass =F( , , ) Function training is established.

[0009] As a further solution of the present invention: the method for estimating the remaining cruising range of the electric bus in real time in step S30 is: The prediction machine learning model is calculated by formula Predict the remaining range of the vehicle at the current power level .

[0010] As a further solution of the present invention: a method for visually displaying and / or voice broadcasting the remaining cruising range is: The vehicle control unit ECU displays the Visual display on the vehicle instrument panel and / or voice module Make voice announcement.

[0011] A device for predicting remaining mileage of an electric bus, comprising: An acquisition unit, used to acquire real-time battery status evaluation data, real-time driving status data, historical driving status data and real-time environmental status data of the current electric bus; The calculation unit is configured to extract the input feature parameters from the acquisition unit and input them into the prediction machine learning model for calculation, so as to obtain a real-time estimate of the remaining cruising range of the electric bus according to the input feature parameters; a control unit for estimating the remaining range of the electric bus in real time using the predictive machine learning model output; The warning unit is used to output the control unit to estimate the remaining cruising range of the electric bus in real time for visual display and / or voice broadcast.

[0012] An electronic device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor is used to obtain real-time battery status evaluation data, real-time driving status data, historical driving status data and real-time environmental status data of a current electric bus through an interface circuit, and the processor implements the above method when executing the computer program.

[0013] A computer-readable storage medium stores a computer program, which implements the above method when executed by a processor.

[0014] The present invention has at least the following beneficial effects: the present application can construct a predictive machine learning model through a comprehensive analysis of multiple factors such as battery status, environmental conditions, historical data and real-time power consumption, and dynamically adjust the predictive model according to the actual driving conditions of the vehicle, thereby significantly improving the accuracy and timeliness of the prediction of the continued mileage of the electric bus; and can remind the driver of the mileage that the electric bus can still travel under the influence of various factors at present through voice broadcast, screen display, etc. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The present invention will be further described below in conjunction with the accompanying drawings.

[0016] Figure 1 is a method flow chart of the remaining mileage prediction method of an electric bus of the present invention; Figure 2 is a flow chart of preprocessing the acquired data in step S20 of the method of the present invention; Figure 3 is an example diagram of a predictive machine learning model used in step S30 of the method of the present invention; Figure 4 It is a schematic diagram of a remaining mileage prediction device for an electric bus according to the present invention; Figure 5 It is a schematic diagram of another remaining mileage prediction device for an electric bus according to the present invention. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] In the description of the present invention, it should be understood that the terms indicating orientation or positional relationship are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention; in the description of the present invention, the meaning of "multiple" and "several" is at least two, for example two, three, etc., unless otherwise clearly and specifically defined.

[0019] See also Figure 1 As shown, the present invention is a method for predicting the remaining mileage of an electric bus, comprising the following steps: S10, obtaining real-time battery status evaluation data, real-time driving status data, historical driving status data and real-time environmental status data of the current electric bus.

[0020] The method for obtaining the real-time battery status evaluation data of the current electric bus in the above step S10 is as follows: the real-time battery status evaluation data is collected and obtained by the vehicle battery management system BMS, and the real-time battery status evaluation data includes the current remaining power SOC, the actual battery capacity data , Get the initial battery capacity data through factory data and initial battery energy For example, during the acquisition process of the vehicle battery management system (BMS), the data acquisition card in the vehicle battery management system (BMS) has the characteristics of high sampling rate and high resolution, which can completely record the change trajectory of battery parameters. Even when the battery status changes rapidly, no important transient data will be missed. Based on the battery status acquisition data, the vehicle battery management system (BMS) can evaluate the actual available capacity of the battery. , thereby dynamically adjusting the battery capacity parameters in the predictive machine learning model to improve the prediction accuracy.

[0021] The method for obtaining real-time driving status data in the above step S10 is: the real-time driving status data is obtained by the on-board data acquisition module, and the real-time driving status data includes real-time current data I, real-time voltage data U and acquisition time interval data t; for example, the on-board data acquisition module includes a voltage sensor and a current sensor, which are used to collect key operating data of the vehicle in real time. The voltage sensor and the current sensor can be installed at the output end of the battery pack to measure the output voltage and current of the battery in real time. The measurement data is transmitted to the on-board control unit ECU through the CAN bus. The ECU is responsible for summarizing these data and transmitting them to the central computing module for further analysis.

[0022] The method for obtaining the historical driving status data in the above step S10 is: the historical driving status data is the vehicle daily mileage collected and obtained by the GPS module Daily driving time , based on the vehicle's daily mileage Daily driving time The vehicle's daily driving power consumption is collected through the vehicle's battery management system BMS For example, the current vehicle can collect driving data within seven days, including daily driving mileage , daily vehicle power consumption ,in , respectively represent the Data from days ago.

[0023] The method for obtaining the real-time environmental status data in the above step S10 is: the real-time environmental status data is the current vehicle driving road condition information collected and obtained by the GPS module and weather information ; During vehicle driving, different weather and road conditions will affect the endurance performance of electric vehicles.

[0024] S20, pre-processing the acquired real-time battery status assessment data, real-time driving status data, historical driving status data and real-time environmental status data, and establishing a machine learning model for predicting the battery life of electric buses.

[0025] like Figure 2 As shown, the preprocessing method for the acquired real-time battery status evaluation data, real-time driving status data, historical driving status data and real-time environmental status data in the above step S20 is: data cleaning is to remove erroneous data and incomplete data in the real-time battery status evaluation data, the real-time driving status data, the historical driving status data and the real-time environmental status data; data integration is to merge multiple different data acquired in the data cleaning step into a unified data set, and store them uniformly; data transformation is to normalize, discretize and reduce the stored data in the unified data set, and convert it into input feature parameters suitable for predicting machine learning models; preprocessing of the acquired data refers to a series of processing processes performed on the original data before data analysis and modeling, aiming to reduce errors and deviations in subsequent processing and improve the quality and reliability of the data.

[0026] In this specific implementation, the process of building a predictive machine learning model can be divided into several processes, including data feature selection, label calculation, data set division, model building and training, and model prediction. Figure 3 As shown in the figure, the features are the input of the model, and the label calculation result is the output of the model, the remaining range. In the following, the implementation process of the predictive machine learning model will be explained in detail.

[0027] The method for establishing a prediction machine learning model for the electric bus's battery life in step S20 is: Determine the current remaining capacity SOC input characteristic parameters and initial battery energy based on real-time battery status evaluation data Enter the characteristic parameters and calculate the formula Obtain the battery health SOH input characteristic parameters.

[0028] According to the real-time driving status data, and through the calculation formula Get the real-time power consumption of the vehicle Input feature parameters, where , and P is a certain moment In this input characteristic parameter determination process, after the central computing module receives the real-time voltage and current data from the sensor, the real-time current is I, the real-time voltage is U, and t is the interval of data collection; the computing unit receives the I and U data and , get the real-time power P of the vehicle; it needs to be understood that I and U are respectively The instantaneous current and instantaneous voltage, P is the instantaneous current and voltage at a certain moment. The real-time power of the vehicle is calculated by integrating the instantaneous power over time. t is the interval for data collection, which is set to 1 second by default. Real-time power consumption .

[0029] According to the historical driving status data, and through the calculation formula Get historical energy consumption per mile Parameters, in the process of determining the input characteristic parameters, the driving data within seven days was collected, among which, , respectively represent the The data from days ago, j is the number of days within seven days when the driving data is not 0, so that the unit mileage energy consumption of the current vehicle under historical driving conditions can be accurately obtained. .

[0030] According to the real-time environmental status data, a vehicle unit mileage energy consumption model is established, and the historical unit mileage energy consumption is converted into Adjust to current energy consumption per mileage Input characteristic parameters; in the process of establishing the vehicle unit mileage energy consumption model, the current unit mileage energy consumption pass =F( , , ) function training is established; in the process of determining the input feature parameters, the weather and road condition parameters are used to build an energy consumption model. Through machine learning algorithms, the energy consumption of vehicles in different weather and road conditions is analyzed to build a prediction model. In the real-time prediction process, the system incorporates these factors into the energy consumption calculation and adjusts the energy consumption per unit mileage in real time. , so as to obtain the current unit mileage energy consumption after adjustment ,in: , Parameters are compared with normal power consumption By comparison, we can obtain the weather parameter comparison table shown in Table 1 and the road condition parameter comparison table shown in Table 2.

[0031] Table 1 Weather parameter comparison table

[0032] Table 2 Road condition parameter comparison table

[0033] Combining Table 1 and Table 2, we can see that taking a rainy day as an example: the energy consumption of driving in rainy days is recorded in the historical driving data. ,at this time ,at this time ; Other weather and road conditions and driving in rainy days to obtain the current unit mileage energy consumption Similar, no further explanation is given; if there are multiple weather conditions in Table 1, then All parameter selections need to be considered. The negative slope value in Table 2 represents the downward direction. In this implementation process, after the comparison relationship between different weather and different road conditions is calibrated, the comparison relationship can be stored in the vehicle control unit ECU in the form of a table, and the table output can be performed when used.

[0034] The current remaining power SOC, initial battery power Input characteristic parameters, battery health SOH Input characteristic parameters, real-time power consumption Input characteristic parameters and current energy consumption per unit mileage After the input feature parameters are determined, the training and establishment of the predictive machine learning model is completed.

[0035] S30: Based on the prediction machine learning model of the electric bus's battery life, the remaining range of the electric bus is estimated in real time, and the remaining range is displayed visually and / or announced by voice.

[0036] The method for estimating the remaining cruising range of the electric bus in real time in step S30 is as follows: Predictive machine learning models are calculated by formula Predict the remaining range of the vehicle at the current power level .

[0037] Based on the vehicle's unit mileage energy consumption model, the on-board system can adjust the parameters in the remaining mileage prediction formula in real time to make the prediction more accurate and timely. For example, when the weather is detected to be getting colder, the system will increase the energy consumption estimate because the battery efficiency will decrease at low temperatures. Similarly, when going uphill, the system will adjust the energy consumption value in real time to reflect the increased energy consumption demand, thereby more accurately predicting the remaining mileage. ; For example, when the weather changes from normal to rainy, the computing unit will record the moment Energy consumption data, when calculating real-time unit mileage, first calculate the current time arrive Mileage and power consumption , and then calculate the real-time energy consumption per unit mileage , the calculation formula is as follows: ; ; .

[0038] This application takes into account the historical driving conditions and external environmental factors, and the system predicts the remaining mileage. More accurate and more timely.

[0039] The method for visually displaying and / or voice broadcasting the remaining cruising range is as follows: the vehicle control unit ECU displays the remaining cruising range through the display module Visual display on the vehicle instrument panel and / or voice module Voice broadcast; for example, the calculation results are updated to the vehicle display system in real time, with an update interval of 1 second; at the same time, the system will make voice broadcasts according to the set time, for example, at the current speed, it is expected to travel 100 km.

[0040] The method of the present application constructs a predictive machine learning model through a comprehensive analysis of multiple factors such as battery status, environmental conditions, historical data and real-time power consumption, and dynamically adjusts the predictive model according to the actual driving conditions of the vehicle, thereby significantly improving the accuracy and timeliness of the prediction of the continued mileage of the electric bus; and can remind the driver of the mileage that the electric bus can still travel under the influence of various factors at present through voice broadcasts, screen displays, etc.

[0041] like Figure 4As shown, the present invention also provides an electric bus remaining mileage prediction device, including an acquisition unit 101, a calculation unit 102, a control unit 100 and an alarm unit 103; the acquisition unit 101 is used to obtain real-time battery status evaluation data, real-time driving status data, historical driving status data and real-time environmental status data of the current electric bus; the calculation unit 102 is configured to extract input feature parameters from the acquisition unit 101 and input them into the predictive machine learning model for calculation, so as to obtain a real-time estimate of the remaining cruising range of the electric bus according to the input feature parameters; the control unit 100 is used to use the predictive machine learning model to output a real-time estimate of the remaining cruising range of the electric bus; the alarm unit 103 is used to visualize and / or voice broadcast the real-time estimate of the remaining cruising range of the electric bus output by the control unit 100.

[0042] It should be noted that for other corresponding descriptions of the functional units involved in the device of the present application, reference can be made to the corresponding descriptions in the above method and will not be repeated here.

[0043] like Figure 5 As shown, the present invention also provides an electronic device, including a memory 201, a processor 200, and a computer program stored in the memory 201 and executable on the processor 200, the processor 200 is used to obtain real-time battery status evaluation data, real-time driving status data, historical driving status data, and real-time environmental status data of the current electric bus through the interface circuit 202, and the processor 200 implements the above method when executing the computer program.

[0044] A computer-readable storage medium stores a computer program, which implements the above method when executed by a processor.

[0045] It should be understood that all or part of the steps of implementing the above method embodiment can be completed by hardware related to program instructions, and the above program can be stored in a computer-readable storage medium. When the program code is called by the processor 200, the processor 200 is used to execute the method executed by the UE in the above method embodiment. The embodiment of the present invention does not limit the form and quantity of the memory 201 and the processor 200. For example, the memory 201 can be a CPU or other processor 7200 that can call a program, and the memory 201 can be a read-only memory ROM, a random access memory RAM, a disk or an optical disk, and other media that can store program codes.

[0046] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A method for predicting the remaining mileage of an electric bus, characterized in that: The following steps are involved: S10, obtaining real-time battery status evaluation data, real-time driving status data, historical driving status data and real-time environmental status data of the current electric bus; S20, pre-processing the acquired real-time battery status evaluation data, real-time driving status data, historical driving status data and real-time environmental status data, and establishing a machine learning model for predicting the battery life of electric buses; S30: Based on the prediction machine learning model of the electric bus's battery life, the remaining range of the electric bus is estimated in real time, and the remaining range is displayed visually and / or announced by voice.

2. The method for predicting the remaining mileage of an electric bus according to claim 1, characterized in that: The method for obtaining the real-time battery status evaluation data, real-time driving status data, historical driving status data and real-time environmental status data of the current electric bus in step S10 is: The real-time battery status evaluation data is collected by the vehicle battery management system BMS. The real-time battery status evaluation data includes the current remaining power SOC and the actual battery capacity data. , Get the initial battery capacity data through factory data and initial battery energy ; The real-time driving status data is acquired by the vehicle-mounted data acquisition module, and the real-time driving status data includes real-time current data I, real-time voltage data U and acquisition time interval data t; The historical driving status data is collected by the GPS module and is the vehicle's daily mileage. Daily driving time , based on the vehicle's daily mileage Daily driving time The vehicle's daily driving power consumption is collected through the vehicle's battery management system BMS ; The real-time environmental status data is the current vehicle driving road condition information collected by the GPS module and weather information .

3. The method for predicting the remaining mileage of an electric bus according to claim 2, characterized in that: In step S20, the method for preprocessing the acquired real-time battery status evaluation data, real-time driving status data, historical driving status data and real-time environmental status data is as follows: Data cleaning, removing erroneous data and incomplete data from the real-time battery status evaluation data, the real-time driving status data, the historical driving status data, and the real-time environmental status data; Data integration, merging multiple different data obtained in the data cleaning step into a unified data set and storing it uniformly; Data transformation: normalizes, discretizes, and reduces the stored data in the unified data set, and converts it into input feature parameters suitable for predictive machine learning models.

4. The method for predicting the remaining mileage of an electric bus according to claim 3, characterized in that: The method for establishing a prediction machine learning model for the electric bus's battery life in step S20 is: According to the real-time battery status evaluation data, the current remaining power SOC input characteristic parameters and the initial battery power are determined Enter the characteristic parameters and calculate the formula Obtain battery health SOH input characteristic parameters; According to the real-time driving status data, and by calculating the formula Get the real-time power consumption of the vehicle Input feature parameters, where , and P is a certain moment Real-time power; According to the historical driving status data, and by calculating the formula Get historical energy consumption per mile parameter; According to the real-time environmental status data, a vehicle unit mileage energy consumption model is established, and the historical unit mileage energy consumption is converted into Adjust to current energy consumption per mileage Input feature parameters; The current remaining power SOC, the initial battery power Input characteristic parameters, the battery health SOH input characteristic parameters, the real-time power consumption Input characteristic parameters and the current unit mileage energy consumption After the input feature parameters are determined, the prediction machine learning model training is completed.

5. The method for predicting the remaining mileage of an electric bus according to claim 4, characterized in that: In the process of establishing the vehicle unit mileage energy consumption model, the current unit mileage energy consumption pass =F( , , ) Function training is established.

6. The method for predicting the remaining mileage of an electric bus according to claim 5, characterized in that: The method for estimating the remaining cruising range of the electric bus in real time in step S30 is: The prediction machine learning model is calculated by formula Predict the remaining range of the vehicle at the current power level .

7. The method for predicting the remaining mileage of an electric bus according to claim 6, characterized in that: The remaining cruising range can be displayed visually and / or announced by voice as follows: The vehicle control unit ECU displays the Visual display on the vehicle instrument panel and / or voice module Make voice announcement.

8. A device for predicting remaining mileage of an electric bus, characterized in that: include: An acquisition unit (101) is used to acquire real-time battery status evaluation data, real-time driving status data, historical driving status data and real-time environmental status data of the current electric bus; A calculation unit (102) is configured to extract input feature parameters from the acquisition unit (101) and input them into a prediction machine learning model for calculation, so as to obtain a real-time estimate of the remaining cruising range of the electric bus based on the input feature parameters; A control unit (100) for estimating the remaining cruising range of the electric bus in real time using the output of the predictive machine learning model; The warning unit (103) is used to visually display and / or voice broadcast the real-time estimated remaining cruising range of the electric bus output by the control unit (100).

9. An electronic device comprising a memory (201), a processor (200), and a computer program stored in the memory (201) and executable on the processor (200), characterized in that: The processor (200) is used to obtain real-time battery status evaluation data, real-time driving status data, historical driving status data and real-time environmental status data of the current electric bus through the interface circuit (202), and the processor (200) implements the method described in any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images