Tramcar and method and device for predicting energy consumption of tramcar and storage medium

Abstract

The present application provides a tramcar and a tramcar kilometer energy consumption prediction method and device and a storage medium, comprising: determining historical tramcar kilometer energy consumption, outdoor environment temperature, driving mileage, average load and travel speed; determining tramcar kilometer energy consumption model under corresponding environment temperature interval according to the above data, obtaining predicted week sequence number and predicted outdoor environment temperature corresponding to the predicted week sequence number, determining the average value of driving mileage, the average value of average load and the average value of travel speed corresponding to the predicted week sequence number in the prediction data set according to the predicted week sequence number; determining the corresponding target tramcar kilometer energy consumption model according to the predicted outdoor environment temperature; and predicting the tramcar kilometer energy consumption value according to the above data. The present application does not need to analyze the working condition, road environment and driving behavior of the tramcar, greatly simplifies the data processing process, improves the efficiency of data processing, and still has good prediction effect for tramcar operation data accumulated for a short time, and has high feasibility.

Description

Technical Field

[0001] This invention relates to the field of tram technology, and in particular to a tram and a method, device and storage medium for predicting energy consumption per kilometer of the tram. Background Technology

[0002] In related technologies, there are many methods for predicting vehicle energy consumption. For example, methods for predicting vehicle energy consumption analysis include identifying the characteristic information of vehicle operating conditions, road environment, and driving behavior, classifying and refining the characteristic information of vehicle operating conditions and driving behavior, and extracting each basic characteristic information; obtaining energy consumption characteristic samples through preprocessing and Pearson correlation coefficient calculation; constructing an energy consumption random forest model using the random forest algorithm, and performing regression analysis on the energy consumption characteristic samples to obtain the vehicle energy consumption prediction value of the energy consumption random forest model; obtaining the energy consumption weight value corresponding to the energy consumption characteristic sample based on the energy consumption random forest model, and obtaining the corresponding relationship between the vehicle energy consumption prediction value and the change of the energy consumption characteristic sample through single-factor analysis.

[0003] For example, a time series-based method is used to establish a prediction model for the total energy consumption of trains. This method considers the periodicity of train energy consumption and establishes a prediction model according to the length of the energy consumption prediction time. It adopts a parameter estimation method that combines the long autoregressive model method and the nonlinear least squares method. The long autoregressive model method is used for the initial estimation of parameters, and the nonlinear least squares method is used for the fine estimation of parameters.

[0004] However, all of the above-mentioned energy consumption prediction methods have shortcomings. One method, when analyzing vehicle energy consumption, requires analysis of vehicle operating conditions, road environment, and driving behavior, making the establishment of the energy consumption prediction model quite complex. Another method, based on time series data to predict the total energy consumption of trains, requires obtaining operational data over a relatively long period, such as recent years. Since trams have been in operation for a relatively short time, this time series-based prediction model cannot be used to predict the energy consumption of trains. Summary of the Invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art.

[0006] Therefore, the first objective of this invention is to propose a method for predicting the energy consumption per kilometer of trams. This method determines the energy consumption per kilometer under a corresponding ambient temperature range. When determining the week number to be predicted and the corresponding predicted outdoor ambient temperature, it determines the average values ​​of the mileage, average load, and average travel speed corresponding to the week number to be predicted. Based on the predicted outdoor ambient temperature, it determines the corresponding target energy consumption per kilometer model. Thus, based on the target energy consumption per kilometer model and the above average values, it determines the predicted energy consumption per kilometer. This method eliminates the need to analyze the tram's operating conditions, road environment, and driving behavior, simplifying the model structure. The predicted energy consumption per kilometer is stored in the database in real time. When retrieving the energy consumption per kilometer, the method only needs to retrieve the predicted energy consumption from the corresponding table in the database, without recalculating based on the data source. This greatly simplifies the data processing process, improves data processing efficiency, and still has good predictive effects for tram operation data with insufficient operating time, making it highly feasible.

[0007] Therefore, a second objective of the present invention is to provide a tram vehicle-kilometer energy consumption prediction device.

[0008] Therefore, a third objective of this invention is to provide a tram.

[0009] Therefore, a fourth object of the present invention is to provide a computer-readable storage medium.

[0010] To achieve the above objectives, a first aspect of the present invention provides a method for predicting the vehicle-kilometer energy consumption of a tram. The method includes: determining the historical vehicle-kilometer energy consumption, outdoor ambient temperature, mileage, average load, and travel speed of the tram; and determining a vehicle-kilometer energy consumption model for a corresponding ambient temperature range based on the vehicle-kilometer energy consumption, outdoor ambient temperature, mileage, average load, and travel speed. The vehicle-kilometer energy consumption model includes the correspondence between the vehicle-kilometer energy consumption and the outdoor ambient temperature, the average load, the mileage, and the travel speed within the ambient temperature range. The function retrieves the week number to be predicted and the predicted outdoor temperature corresponding to the week number; determines the mean of the mileage, the mean of the average load, and the mean of the travel speed in the prediction dataset corresponding to the week number to be predicted; determines the corresponding target vehicle-kilometer energy consumption model based on the predicted outdoor temperature; and predicts the vehicle-kilometer energy consumption value based on the target vehicle-kilometer energy consumption model and the mean of the predicted outdoor temperature, the mean of the mileage, the mean of the average load, and the mean of the travel speed corresponding to the week number to be predicted.

[0011] The tram energy consumption prediction method according to embodiments of the present invention determines the tram energy consumption model under a corresponding ambient temperature range. When determining the week number to be predicted and the corresponding predicted outdoor ambient temperature, the method determines the average value of the mileage, average load, and average travel speed corresponding to the week number to be predicted, and determines the corresponding target tram energy consumption model based on the predicted outdoor ambient temperature. Thus, the tram energy consumption value is predicted based on the target tram energy consumption model and the aforementioned average values. This method simplifies the model structure by eliminating the need to analyze the tram's operating conditions, road environment, and driving behavior. The tram energy consumption prediction results are stored in the database in real time. When retrieving the tram energy consumption value, the method only needs to retrieve the prediction result from the corresponding table in the database, without recalculating based on the data source. This greatly simplifies the data processing process, improves data processing efficiency, and still provides good prediction results for tram operation data with insufficient operating time, demonstrating high feasibility.

[0012] In some embodiments, determining a vehicle-kilometer energy consumption model under a corresponding ambient temperature range based on the vehicle-kilometer energy consumption, outdoor ambient temperature, mileage, average load, and travel speed includes: determining a first ambient temperature range, a second ambient temperature range, and a third ambient temperature range; determining a corresponding ambient temperature range based on the outdoor ambient temperature; and establishing a vehicle-kilometer energy consumption model of the vehicle-kilometer energy consumption and the outdoor ambient temperature, mileage, average load, and travel speed under the corresponding ambient temperature range.

[0013] In some embodiments, before determining the mean of the mileage, the mean of the average load, and the mean of the travel speed corresponding to the week number to be predicted in the prediction dataset, the method further includes: obtaining all the mileage, average load, and travel speed of the tram within a predetermined time period from a historical database; calculating the mean of the mileage, the mean of the average load, and the mean of the travel speed corresponding to each week number based on all the mileage, average load, and travel speed; and constructing the prediction dataset based on the mean of the mileage, the mean of the average load, and the mean of the travel speed.

[0014] In some embodiments, determining the corresponding target vehicle-kilometer energy consumption model based on the predicted outdoor ambient temperature includes: determining the corresponding ambient temperature range in which the predicted outdoor ambient temperature is located; and using the vehicle-kilometer energy consumption model corresponding to the corresponding ambient temperature range as the target vehicle-kilometer energy consumption model.

[0015] In some embodiments, determining the vehicle-kilometer energy consumption value based on the target vehicle-kilometer energy consumption model and the predicted outdoor ambient temperature, the average mileage, the average load, and the average travel speed corresponding to the week number to be predicted includes: inputting the predicted outdoor ambient temperature and the average mileage, the average load, and the average travel speed in the predicted dataset corresponding to the week number to be predicted into the target vehicle-kilometer energy consumption model to obtain the vehicle-kilometer energy consumption value.

[0016] In some embodiments, after obtaining the vehicle-kilometer energy consumption value, the method further includes: updating the target vehicle-kilometer energy consumption model based on the vehicle-kilometer energy consumption value.

[0017] In some embodiments, obtaining the tram's energy consumption per kilometer, outdoor ambient temperature, mileage, average load, and travel speed for the previous day includes: obtaining the tram's historical battery pack's current total voltage, current total current, first pressure value, second pressure value, current station location, terminal station location, and tram speed; determining the energy consumption per kilometer based on the battery pack's current total voltage, current total current, and mileage; determining the average load based on the first pressure value and second pressure value; and determining the travel speed based on the current station location, terminal station location, and tram speed.

[0018] To achieve the above objectives, a second aspect of the present invention provides a tram-kilometer energy consumption prediction device, the device comprising: a first determining module for determining the tram's historical kilometer energy consumption, outdoor ambient temperature, mileage, average load, and travel speed; and a second determining module for determining a kilometer energy consumption model under a corresponding ambient temperature range based on the kilometer energy consumption, outdoor ambient temperature, mileage, average load, and travel speed, wherein the kilometer energy consumption model includes a correspondence function between the kilometer energy consumption and the outdoor ambient temperature, the average load, the mileage, and the travel speed under the ambient temperature range; The third determining module is used to obtain the week number to be predicted and the predicted outdoor ambient temperature corresponding to the week number to be predicted, and to determine the mean value of the driving mileage, the mean value of the average load, and the mean value of the travel speed in the prediction dataset corresponding to the week number to be predicted; the fourth determining module is used to determine the corresponding target vehicle-kilometer energy consumption model based on the predicted outdoor ambient temperature; the prediction module is used to predict the vehicle-kilometer energy consumption value based on the target vehicle-kilometer energy consumption model and the mean values ​​of the predicted outdoor ambient temperature, the mean value of the driving mileage, the mean value of the average load, and the mean value of the travel speed corresponding to the week number to be predicted.

[0019] The tram vehicle-kilometer energy consumption prediction device according to an embodiment of the present invention determines a vehicle-kilometer energy consumption model under a corresponding ambient temperature range. When determining the week number to be predicted and the corresponding predicted outdoor ambient temperature, it determines the average values ​​of the mileage, average load, and average travel speed corresponding to the week number, and determines the corresponding target vehicle-kilometer energy consumption model based on the predicted outdoor ambient temperature. Thus, it predicts the vehicle-kilometer energy consumption value based on the target vehicle-kilometer energy consumption model and the aforementioned average values. This eliminates the need to analyze the tram's operating conditions, road environment, and driving behavior, simplifying the model structure. The vehicle-kilometer energy consumption prediction results are stored in a database in real time. When retrieving the vehicle-kilometer energy consumption value, it only needs to retrieve the energy consumption prediction result from the corresponding table in the database, without recalculating based on the data source. This greatly simplifies the data processing process, improves data processing efficiency, and still provides good prediction results for tram operation data with insufficient operating time, demonstrating high feasibility.

[0020] To achieve the above objectives, a third aspect of the present invention provides a tram that includes a tram-kilometer energy consumption prediction device as described in the above embodiments.

[0021] According to embodiments of the present invention, the tram system determines a vehicle-kilometer energy consumption model within a corresponding ambient temperature range. When determining the week number to be predicted and the corresponding predicted outdoor ambient temperature, it calculates the average mileage, average load, and average travel speed corresponding to the week number. Based on the predicted outdoor ambient temperature, it determines the corresponding target vehicle-kilometer energy consumption model. Thus, it predicts vehicle-kilometer energy consumption values ​​based on the target model and the aforementioned averages. This eliminates the need to analyze the tram's operating conditions, road environment, and driving behavior, simplifying the model structure. The predicted vehicle-kilometer energy consumption results are stored in a database in real time. When retrieving vehicle-kilometer energy consumption values, the prediction results can be retrieved from the corresponding table in the database without recalculating based on the data source, greatly simplifying the data processing process and improving efficiency. Furthermore, it maintains good predictive performance even for tram operation data with insufficient running time, demonstrating high feasibility.

[0022] To achieve the above objectives, an embodiment of the fourth aspect of the present invention provides a non-transitory computer-readable storage medium storing a tram vehicle-kilometer energy consumption prediction program, which, when executed by a processor, implements the tram vehicle-kilometer energy consumption prediction method as described in the above embodiments.

[0023] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0024] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which:

[0025] Figure 1 This is a block diagram of a module for predicting the vehicle-kilometer energy consumption of a tram according to an embodiment of the present invention.

[0026] Figure 2 This is a flowchart of a method for predicting the vehicle-kilometer energy consumption of a tram according to an embodiment of the present invention;

[0027] Figure 3 This is a scatter plot of vehicle-kilometer energy consumption versus outdoor ambient temperature according to an embodiment of the present invention;

[0028] Figure 4 This is a scatter plot of vehicle-kilometer energy consumption and average load according to an embodiment of the present invention;

[0029] Figure 5 This is a scatter plot of vehicle-kilometer energy consumption and driving mileage according to an embodiment of the present invention;

[0030] Figure 6 This is a scatter plot of vehicle-kilometer energy consumption versus travel speed according to an embodiment of the present invention;

[0031] Figure 7 This is a schematic diagram comparing the actual energy consumption per vehicle kilometer with the energy consumption per vehicle kilometer according to an embodiment of the present invention.

[0032] Figure 8 This is a flowchart of a method for predicting the vehicle-kilometer energy consumption of a tram according to a specific embodiment of the present invention;

[0033] Figure 9 This is a block diagram of a tram vehicle-kilometer energy consumption prediction device according to an embodiment of the present invention;

[0034] Figure 10 This is a block diagram of a tram according to an embodiment of the present invention. Detailed Implementation

[0035] The embodiments of the present invention are described in detail below. The embodiments described with reference to the accompanying drawings are exemplary. The embodiments of the present invention are described in detail below.

[0036] When analyzing the energy consumption per kilometer of tram lines, since the tram routes are fixed, the road environment is stable, the trams involved in operation are fixed, the operating conditions of the trams are stable, and the drivers on the lines are also basically fixed with stable driving behavior, it is not necessary to analyze tram operating conditions, road environment, and driving behavior when predicting the energy consumption per kilometer of tram lines. Only the influence of factors such as outdoor ambient temperature, average load, mileage, and travel speed on the energy consumption per kilometer needs to be analyzed. It is understood that the energy consumption per kilometer prediction method of this invention includes factors such as tram operating conditions, road environment, and driving behavior for the line to be predicted. Furthermore, when predicting the energy consumption per kilometer of trams, it is necessary to train the energy consumption per kilometer model. The larger the number of training samples, the higher the accuracy of the energy consumption per kilometer model. In the tram energy consumption prediction method of this embodiment, relevant data from the previous day are added to retrain the model each time the energy consumption per kilometer is predicted, so as to predict the energy consumption per kilometer for the next few days and gradually improve the prediction accuracy.

[0037] like Figure 1 The diagram shown is a block diagram of the tram's vehicle-kilometer energy consumption and prediction module according to an embodiment of the present invention. The calculation module calculates and stores model-related variables based on historical data of the tram; the modeling module establishes a corresponding vehicle-kilometer energy consumption model based on the tram's historical operating parameters and the corresponding ambient temperature range; the prediction module constructs a prediction dataset based on the predicted outdoor ambient temperature and operating data from the line within the most recent predetermined time period, for example, three months, to predict the vehicle-kilometer energy consumption for the next few days.

[0038] Therefore, the tram kilometer energy consumption prediction method of the present invention is easy to implement and simple to use to predict the tram kilometer energy consumption for the next few days.

[0039] The following is an example illustrating the method for predicting the vehicle-kilometer energy consumption of a tram according to an embodiment of the present invention.

[0040] The following is combined with Figures 2-8 The method for predicting the vehicle-kilometer energy consumption of trams according to embodiments of the present invention is described, such as... Figure 2 As shown, the tram vehicle-kilometer energy consumption prediction method of this embodiment of the invention includes at least steps S1-S4.

[0041] Step S1: Determine the historical energy consumption per kilometer, outdoor ambient temperature, mileage, average load, and travel speed of the tram. It can be understood that the energy consumption per kilometer is the energy consumption per kilometer per carriage.

[0042] Among them, the energy consumption per vehicle kilometer is related to the current total voltage of the battery pack, the current total current of the battery pack, and the mileage of the tram. The energy consumption per vehicle kilometer of the tram can be determined by determining the current total voltage of the battery pack, the current total current of the battery pack, and the mileage.

[0043] The average load is related to the pressure detected by the pressure sensor. By determining the pressure values ​​detected by different sensors, the average load of the tram can be determined.

[0044] The travel speed is related to the tram's travel time and speed. The travel time between any two stations is determined by the current station location and the destination location, and the tram's travel speed is determined by combining this with the tram's speed.

[0045] In this embodiment, the historical data of the tram, such as the energy consumption per kilometer, outdoor temperature, mileage, average load, and travel speed of the previous day, are determined and stored in a historical database, such as a database energy consumption data table. The database energy consumption data table is used to store the historical daily energy consumption per kilometer, outdoor temperature, mileage, average load, and travel speed of each tram on the route.

[0046] Step S2: Determine the vehicle-kilometer energy consumption model for the corresponding ambient temperature range based on vehicle-kilometer energy consumption, outdoor ambient temperature, mileage, average load, and travel speed. The vehicle-kilometer energy consumption model includes the correspondence function between vehicle-kilometer energy consumption and outdoor ambient temperature, average load, mileage, and travel speed for the ambient temperature range.

[0047] Among these factors, vehicle-kilometer energy consumption is correlated with outdoor ambient temperature, mileage, average load, and travel speed. For example... Figure 3 The diagram shown is a scatter plot illustrating the vehicle-kilometer energy consumption versus outdoor ambient temperature according to an embodiment of the present invention. Figure 3 It can be seen that when the outdoor ambient temperature is higher than 22℃, the energy consumption per vehicle kilometer increases with the increase of outdoor ambient temperature; when the outdoor ambient temperature is lower than 18℃, the energy consumption per vehicle kilometer increases with the decrease of outdoor ambient temperature. Therefore, the relationship between outdoor ambient temperature and energy consumption per vehicle kilometer can be fitted using a piecewise function.

[0048] like Figure 4 The diagram shown is a scatter plot illustrating vehicle-kilometer energy consumption versus average load according to an embodiment of the present invention. Figure 4 It can be seen that when the average load is less than 20, there is a certain correlation between the vehicle-kilometer energy consumption and the average load.

[0049] like Figure 5 The diagram shown is a scatter plot illustrating vehicle-kilometer energy consumption versus driving mileage according to an embodiment of the present invention. Figure 5It can be seen that there is a certain correlation between driving mileage and vehicle energy consumption per kilometer.

[0050] like Figure 6 The diagram shown is a scatter plot illustrating vehicle-kilometer energy consumption versus travel speed according to an embodiment of the present invention. Figure 6 It can be seen that there is a certain correlation between travel speed and vehicle energy consumption per kilometer.

[0051] In this embodiment, the energy consumption per kilometer, outdoor ambient temperature, mileage, average load, and travel speed are stored in a historical database, such as a database energy consumption data table. Then, the daily energy consumption per kilometer, outdoor ambient temperature, mileage, average load, and travel speed of the tram after it officially starts operating are obtained from the database energy consumption data table.

[0052] After obtaining the above data, a vehicle-kilometer energy consumption model is determined based on the ambient temperature range. It can be understood that, within the corresponding ambient temperature range, vehicle-kilometer energy consumption has a functional relationship with the ambient temperature, average load, distance traveled, and travel speed.

[0053] The vehicle-kilometer energy consumption model for the corresponding ambient temperature range is trained based on the operating data of the previous day and previous dates. Therefore, the vehicle-kilometer energy consumption model obtained daily for the corresponding ambient temperature range may differ. By determining the vehicle-kilometer energy consumption model for the corresponding ambient temperature range, it is easier to predict the vehicle-kilometer energy consumption for the next few days based on the vehicle-kilometer energy consumption model.

[0054] Step S3: Obtain the week number to be predicted and the predicted outdoor temperature corresponding to the week number to be predicted. Based on the week number to be predicted, determine the mean of the driving mileage, the mean of the average load, and the mean of the travel speed in the prediction dataset corresponding to the week number to be predicted.

[0055] In this embodiment, the prediction dataset includes predicted outdoor temperatures for the next few days obtained from a weather network, as well as operational data for a specific route over a recent period, such as daily mileage, average load, and travel speed for the tram route over the past three months.

[0056] For example, after obtaining the week number to be predicted and the corresponding predicted outdoor temperature, the mean of the mileage, the mean of the load, and the mean of the travel speed corresponding to the week number are determined based on the week number. For instance, when the week number to be predicted is Tuesday, the mean of the mileage, the mean of the load, and the mean of the travel speed corresponding to Tuesday in the prediction dataset are determined.

[0057] The outdoor ambient temperature in the predicted dataset is obtained by crawling a weather network; the average mileage in the predicted dataset is determined based on the mileage in the historical database within a predetermined time period, such as three months of operating data; the average load in the predicted dataset is determined based on the average load in the three months of operating data; and the average travel speed in the predicted dataset is determined based on the travel speed in the three months of operating data. It is understood that the determination of the above average mileage, average load, and average travel speed needs to be calculated by grouping according to the week number.

[0058] Step S4: Determine the corresponding target vehicle-kilometer energy consumption model based on the predicted outdoor ambient temperature and the vehicle-kilometer energy consumption model under the corresponding ambient temperature range.

[0059] The determination of the target vehicle-kilometer energy consumption model is related to the predicted outdoor ambient temperature. After determining the predicted outdoor ambient temperature, the corresponding ambient temperature range is determined based on the predicted outdoor ambient temperature, and the target vehicle-kilometer energy consumption model is determined.

[0060] For example, the corresponding ambient temperature will be determined differently depending on the predicted outdoor ambient temperature. For instance, when the predicted outdoor ambient temperature is 17℃, the corresponding target vehicle-kilometer energy consumption model is t18; when the predicted outdoor ambient temperature is 28℃, the corresponding target vehicle-kilometer energy consumption model is considered to be t22; and when the predicted outdoor ambient temperature is 20℃, the corresponding target vehicle-kilometer energy consumption model is considered to be t1822. Therefore, the target vehicle-kilometer energy consumption model can be determined based on the corresponding ambient temperature range of the predicted outdoor ambient temperature.

[0061] Step S5: Predict the vehicle-kilometer energy consumption value based on the target vehicle-kilometer energy consumption model and the predicted outdoor ambient temperature, average mileage, average load, and average travel speed corresponding to the week number to be predicted.

[0062] The predicted vehicle-kilometer energy consumption value is determined based on the target vehicle-kilometer energy consumption model and the predicted outdoor ambient temperature, average mileage, average load, and average trip corresponding to the week number to be predicted.

[0063] In this embodiment, if the predicted outdoor ambient temperature is greater than 22°C, the target vehicle-kilometer energy consumption model corresponding to the predicted outdoor ambient temperature is determined to be the t22 model. At this time, if the predicted vehicle-kilometer energy consumption value for next Monday is predicted, for example, if the predicted outdoor ambient temperature for next Monday is 27°C, the predicted outdoor ambient temperature, i.e., 27°C, and the average values ​​of the average load, average mileage, and average travel speed for all Mondays in the last three months are substituted into the t22 model to obtain the vehicle-kilometer energy consumption value for next Monday.

[0064] The tram's vehicle-kilometer energy consumption prediction method according to embodiments of the present invention determines a vehicle-kilometer energy consumption model within a corresponding ambient temperature range. When determining the week number to be predicted and the corresponding predicted outdoor ambient temperature, it calculates the average mileage, average load, and average travel speed corresponding to the week number. Based on the predicted outdoor ambient temperature, it determines the corresponding target vehicle-kilometer energy consumption model. Thus, it predicts the vehicle-kilometer energy consumption value based on the target model and the aforementioned averages. This method simplifies the model structure by eliminating the need to analyze the tram's operating conditions, road environment, and driving behavior. The vehicle-kilometer energy consumption prediction results are stored in a database in real time. When retrieving the vehicle-kilometer energy consumption value, the prediction result can be retrieved from the corresponding table in the database without recalculating based on the data source, greatly simplifying the data processing process and improving efficiency. Furthermore, it maintains good predictive performance even for tram operation data with insufficient running time, demonstrating high feasibility.

[0065] In some embodiments, determining a vehicle-kilometer energy consumption model under a corresponding ambient temperature range based on vehicle-kilometer energy consumption, outdoor ambient temperature, mileage, average load, and travel speed includes: determining a first ambient temperature range, a second ambient temperature range, and a third ambient temperature range; determining a corresponding ambient temperature range based on the outdoor ambient temperature; and establishing a vehicle-kilometer energy consumption model in the corresponding ambient temperature range that relates vehicle-kilometer energy consumption to outdoor ambient temperature, the mileage, average load, and travel speed.

[0066] In the embodiment, when determining the vehicle-kilometer energy consumption model under the corresponding ambient temperature range, the ambient temperature range is first divided, for example, into a first ambient temperature range, a second ambient temperature range, and a third ambient temperature range. In the first ambient temperature range, the outdoor ambient temperature is >22℃; in the second ambient temperature range, 18℃ ≤ outdoor temperature ≤22℃; and in the third ambient temperature range, the outdoor temperature is <18℃.

[0067] For each corresponding ambient temperature range, a vehicle-kilometer energy consumption model is established. Specifically, for each ambient temperature range, a model is trained to predict the relationship between vehicle-kilometer energy consumption, outdoor ambient temperature, mileage, average load, and travel speed. For example, the vehicle-kilometer energy consumption model for the first ambient temperature range is denoted as t22, for the second ambient temperature range as t1822, and for the third ambient temperature range as t18. This means that for the first ambient temperature range, there exists a corresponding vehicle-kilometer energy consumption model; for the second ambient temperature range, there exists a corresponding model; and for the third ambient temperature range, there exists a corresponding model. When training the vehicle-kilometer energy consumption model, only factors such as outdoor ambient temperature, average load, mileage, and travel speed are considered. A multiple regression model is used to construct the vehicle-kilometer energy consumption prediction model for the corresponding ambient temperature range. Furthermore, this vehicle-kilometer energy consumption model does not require research into energy consumption structure or factors that are difficult to quantify comprehensively, such as roads, vehicles, and drivers, making the model structure simple.

[0068] For example, the training of the vehicle-kilometer energy consumption model t22 under the first ambient temperature range is explained. Outliers in the vehicle-kilometer energy consumption model t22 are removed according to the following rules: Outliers are considered to have an absolute value greater than 2 based on the student-standardized residual. High-leverage points are considered to have a model hat value exceeding 3*p / n, where p is the number of predictor variables and n is the number of samples. Strong-influence points are considered to have a distance value greater than 4 / (nk-1), where n is the number of samples and k is the number of predictor variables.

[0069] The results of the vehicle-kilometer energy consumption model t22 show that the p-value of the F-test for the equation is <0.05, indicating that the equation is valid. The adjusted R-squared value is 0.91. Other tests for the vehicle-kilometer energy consumption model, such as linearity, normality, homogeneity of variance, and multicollinearity, all passed. In summary, the vehicle-kilometer energy consumption model t22 has a good fit. The formula for the vehicle-kilometer energy consumption model t22 is:

[0070] emp=-0.0698+0.0158*w+0.0126*zh-0.00024*mil+0.0029*lv

[0071] Where emp represents energy consumption per vehicle kilometer, w represents the outdoor ambient temperature, zh represents the average load, mil represents the mileage traveled, and lv represents the travel speed. It is understandable that the energy consumption per vehicle kilometer model t22 is trained daily based on operating data from the previous week and weeks with outdoor ambient temperatures above 22°C. Therefore, the energy consumption per vehicle kilometer model t22 trained for today's prediction may differ from the model t22 retrained for tomorrow's prediction; that is, the coefficients of the above parameters may change. Similarly, the energy consumption per vehicle kilometer under the corresponding ambient temperature range is trained based on outdoor ambient temperature, mileage traveled, average load, and travel speed to obtain energy consumption per vehicle kilometer models t1822 and t18.

[0072] In some embodiments, before determining the mean mileage, mean load, and mean travel speed corresponding to the week number to be predicted, the method further includes: obtaining all mileage, mean load, and travel speed of the tram within a predetermined time period from a historical database; calculating the mean mileage, mean load, and mean travel speed corresponding to each week number based on all mileage, mean load, and mean travel speed; and constructing a prediction dataset based on the mean mileage, mean load, mean travel speed, and predicted outdoor ambient temperature.

[0073] In this embodiment, before determining the average mileage, average load, and average travel speed corresponding to the week number to be predicted, the operation data of all tram mileage, average load, and travel speed within a predetermined time period, such as the most recent three months, are obtained from a historical database.

[0074] Based on the acquired operational data, the average mileage, average load, and average travel speed corresponding to each week number are calculated. For example, for the operational data of the line for the most recent three months, the data is grouped according to week number, and the average mileage, average load, and average travel speed for the corresponding week number (Monday to Sunday) within the most recent three months are calculated. The average mileage, average load, and average travel speed corresponding to each week number, along with the predicted outdoor ambient temperature, form a prediction dataset. By determining the prediction dataset, when obtaining the week number to be predicted and the predicted outdoor ambient temperature, the energy consumption of the tram can be predicted based on the target vehicle-kilometer energy consumption model determined by the week number to be predicted and the predicted outdoor ambient temperature.

[0075] In some embodiments, determining the corresponding target vehicle-kilometer energy consumption model based on the predicted outdoor ambient temperature and the vehicle-kilometer energy consumption model under the corresponding ambient temperature range includes: determining the corresponding ambient temperature range where the predicted outdoor ambient temperature is located; and using the vehicle-kilometer energy consumption model corresponding to the corresponding ambient temperature range as the target vehicle-kilometer energy consumption model.

[0076] In this embodiment, when determining the predicted outdoor ambient temperature, a corresponding target vehicle-kilometer energy consumption model is determined based on the predicted outdoor ambient temperature. For example, when the predicted outdoor ambient temperature is 17°C, it is determined to be in the third ambient temperature range, and the vehicle-kilometer energy consumption model t18 is used as the target vehicle-kilometer energy consumption model; when the predicted outdoor ambient temperature is 20°C, it is determined to be in the second ambient temperature range, and the vehicle-kilometer energy consumption model t1822 is used as the target vehicle-kilometer energy consumption model; when the predicted outdoor ambient temperature is 28°C, it is determined to be in the first ambient temperature range, and the vehicle-kilometer energy consumption model t22 is used as the vehicle-kilometer energy consumption model.

[0077] In some embodiments, determining the vehicle-kilometer energy consumption value based on the target vehicle-kilometer energy consumption model and the average values ​​of the predicted outdoor ambient temperature, mileage, average load, and travel speed corresponding to the week number to be predicted includes: inputting the average values ​​of the predicted outdoor ambient temperature, mileage, average load, and travel speed corresponding to the week number to be predicted from the prediction dataset into the target vehicle-kilometer energy consumption model to obtain the vehicle-kilometer energy consumption value.

[0078] In this embodiment, when predicting vehicle-kilometer energy consumption for the next few days, the predicted outdoor ambient temperature and the average values ​​of mileage, average load, and travel speed corresponding to the week number to be predicted are input into the vehicle-kilometer energy consumption model. For example, if the week number to be predicted is Wednesday, the average values ​​of mileage, average load, and travel speed corresponding to Wednesday in the historical database are calculated, and the above average values ​​corresponding to Wednesday and the predicted outdoor ambient temperature are input into the determined target vehicle-kilometer energy consumption model. For example, 28°C is input into the target vehicle-kilometer energy consumption model for the corresponding ambient temperature range to obtain the vehicle-kilometer energy consumption value.

[0079] In some embodiments, after obtaining the vehicle-kilometer energy consumption value, the method further includes: updating the target vehicle-kilometer energy consumption model based on the vehicle-kilometer energy consumption value.

[0080] In this embodiment, the vehicle-kilometer energy consumption value can be stored in a database energy consumption prediction table for use by various applications. The calculation accuracy of the vehicle-kilometer energy consumption value is affected by the predicted outdoor ambient temperature, but the error between the actual value and the predicted value can be controlled within 10%, that is, the absolute error is basically below 0.06. It is understood that in each prediction, the vehicle-kilometer energy consumption value of the previous day is added and the model is retrained. The latest trained model is then used to predict the vehicle-kilometer energy consumption value for the following week. As the operating time increases, the training sample size increases, and the accuracy of the vehicle-kilometer energy consumption value also improves.

[0081] After acquiring the daily vehicle-kilometer energy consumption, outdoor ambient temperature, average load, mileage and travel speed of the tram, as well as the predicted outdoor ambient temperature, the average mileage, average load and average travel speed corresponding to each week number within a predetermined time period, the above operating data is stored. This facilitates problem tracing in the event of a fault and enables the optimization of the predicted vehicle-kilometer model.

[0082] In some embodiments, obtaining the tram's energy consumption per kilometer, outdoor ambient temperature, mileage, average load, and travel speed for the previous day includes: obtaining the tram's current total battery pack voltage, current total battery pack current, first pressure value, second pressure value, current station location, terminal station location, and tram speed for the previous day; determining the energy consumption per kilometer based on the current total battery pack voltage, current total battery pack current, and mileage; determining the average load based on the first and second pressure values; and determining the travel speed based on the current station location, terminal station location, and tram speed.

[0083] In this embodiment, the vehicle-mounted terminal packages and encrypts the signals collected by the vehicle-mounted sensors and various control managers, and transmits them back to the headquarters information center in real time via a 4G network for storage. The headquarters information center decrypts and parses the data according to a specific protocol, and then stores it in separate databases and tables. The system connects to the database every morning to retrieve relevant parameter data from the previous day's tables, calculates model-related variables using certain logic, and stores them in the database's energy consumption data table.

[0084] Energy consumption per kilometer can be calculated by integrating the product of the current total voltage of the battery pack and the current total current of the battery pack over time. The formula for calculating energy consumption per kilometer is as follows:

[0085]

[0086] Where vol is the current total voltage of the battery pack, cur is the current total current of the battery pack, mil is the driving range, and epm is the energy consumption per kilometer.

[0087] The average load is calculated from the voltage values ​​detected by the first and second pressure sensors. The formula for calculating the average load is as follows:

[0088] zh = (data1 + data2) * 1.77 + 0.3

[0089] Where data1 is the first pressure value and data2 is the second pressure value.

[0090] Travel speed can be calculated by determining the travel time between any two stations based on the current station location and the destination station location, and then calculating the distance between any two stations by integrating the speed signal over time. The ratio of the distance traveled between any two stations to the travel time is the travel speed. The formula for calculating travel speed is as follows:

[0091]

[0092] Where v is the vehicle speed, t1 is the time to reach the previous station, and t2 is the time to reach the next station.

[0093] In other embodiments, such as Figure 7 The diagram shown is a comparison of the actual energy consumption per vehicle kilometer and the energy consumption per vehicle kilometer according to an embodiment of the present invention.

[0094] The solid curve represents the actual energy consumption per vehicle-kilometer of the line, while the dashed curve represents the energy consumption per vehicle-kilometer of the line. It can be seen that the two curves have basically the same trend, and the error between the actual energy consumption per vehicle-kilometer and the actual energy consumption per vehicle-kilometer is within 10% for 97% of the time.

[0095] In other embodiments, when connecting the above data to the historical database, vehicle-related parameter data, vehicle-kilometer energy consumption calculation, average load calculation, outdoor ambient temperature calculation, and travel speed calculation are obtained. The calculation results are written into the historical database, and the vehicle-kilometer energy consumption model is trained on the data in the historical database. The week number to be predicted and the predicted outdoor ambient temperature are obtained, and this process is performed daily. In particular, for weather forecasts of the next week, the closer the date, the higher the accuracy of the forecast. Therefore, when updating the predicted route vehicle-kilometer energy consumption results, the vehicle-kilometer energy consumption values ​​for the next 3 days can be updated.

[0096] The following is for reference. Figure 8 The method for predicting the vehicle-kilometer energy consumption of a tram according to an embodiment of the present invention will be illustrated by example, such as... Figure 8 The diagram shown is a flowchart of a method for predicting the vehicle-kilometer energy consumption of a tram according to an embodiment of the present invention.

[0097] Step S11: Obtain the current total voltage of the battery pack, the current total current of the battery pack, the first pressure value, the second pressure value, the current station location, the terminal station location, and the speed of the tram from the previous day.

[0098] Step S12: Based on the above data, determine the tram's energy consumption per kilometer, outdoor ambient temperature, mileage, average load, and travel speed for the previous day, and write the above data into the historical database.

[0099] Step S13: Obtain from the historical database the energy consumption per kilometer, outdoor ambient temperature, mileage, average load, and travel speed of the tram on the operating route up to the previous day.

[0100] Step S14: Determine the first ambient temperature range, the second ambient temperature range, and the third ambient temperature range.

[0101] Step S15: Determine the corresponding ambient temperature range based on the outdoor ambient temperature, and establish a vehicle-kilometer energy consumption model based on the outdoor ambient temperature, driving mileage, average load, and travel speed within the corresponding ambient temperature range.

[0102] Step S16: Calculate the average mileage, average load, and average travel speed corresponding to each week number within a predetermined time period in the historical database.

[0103] Step S17: Determine the week number to be predicted.

[0104] Step S18: Determine the predicted outdoor ambient temperature corresponding to the week number to be predicted.

[0105] Step S19: Determine the target vehicle-kilometer energy consumption model corresponding to the outdoor ambient temperature to be predicted.

[0106] Step S20: Input the mean mileage, mean load, mean travel speed, and predicted outdoor ambient temperature corresponding to the week number to be predicted into the target vehicle-kilometer energy consumption model to obtain the vehicle-kilometer energy consumption value.

[0107] Step S21: Store the vehicle-kilometer energy consumption value in the historical database so that it can be retrieved when applied.

[0108] The tram's vehicle-kilometer energy consumption prediction method according to embodiments of the present invention determines a vehicle-kilometer energy consumption model within a corresponding ambient temperature range. When determining the week number to be predicted and the corresponding predicted outdoor ambient temperature, it calculates the average mileage, average load, and average travel speed corresponding to the week number. Based on the predicted outdoor ambient temperature, it determines the corresponding target vehicle-kilometer energy consumption model. Thus, it predicts the vehicle-kilometer energy consumption value based on the target model and the aforementioned averages. This method simplifies the model structure by eliminating the need to analyze the tram's operating conditions, road environment, and driving behavior. The vehicle-kilometer energy consumption prediction results are stored in a database in real time. When retrieving the vehicle-kilometer energy consumption value, the prediction result can be retrieved from the corresponding table in the database without recalculating based on the data source, greatly simplifying the data processing process and improving efficiency. Furthermore, it maintains good predictive performance even for tram operation data with insufficient running time, demonstrating high feasibility.

[0109] The following describes a tram vehicle-kilometer energy consumption prediction device according to an embodiment of the present invention.

[0110] like Figure 9 As shown, the tram's vehicle-kilometer energy consumption prediction device 2 according to an embodiment of the present invention includes a first determining module 20, a second determining module 21, a third determining module 22, a fourth determining module 23, and a prediction module 24. The first determining module 20 is used to determine the tram's historical vehicle-kilometer energy consumption, outdoor ambient temperature, mileage, average load, and travel speed. The second determining module 21 is used to determine a vehicle-kilometer energy consumption model under a corresponding ambient temperature range based on the historical vehicle-kilometer energy consumption, outdoor ambient temperature, mileage, average load, and travel speed. The vehicle-kilometer energy consumption model includes the vehicle-kilometer energy consumption under the ambient temperature range, along with the outdoor ambient temperature, average load, mileage, and travel speed. The third determining module 22 is used to obtain the week number to be predicted and the predicted outdoor ambient temperature corresponding to the week number to be predicted, and to determine the mean of the driving mileage, the mean of the average load and the mean of the travel speed in the prediction dataset corresponding to the week number to be predicted; the fourth determining module 23 is used to determine the corresponding target vehicle kilometer energy consumption model based on the predicted outdoor ambient temperature and the vehicle kilometer energy consumption model under the corresponding ambient temperature range; the prediction module 24 is used to predict the vehicle kilometer energy consumption value based on the target vehicle kilometer energy consumption model and the mean of the predicted outdoor ambient temperature, the mean of the driving mileage, the mean of the average load and the mean of the travel speed corresponding to the week number to be predicted.

[0111] The tram vehicle-kilometer energy consumption prediction device 2 according to an embodiment of the present invention determines a vehicle-kilometer energy consumption model under a corresponding ambient temperature range. When determining the week number to be predicted and the corresponding predicted outdoor ambient temperature, it determines the average values ​​of the mileage, average load, and average travel speed corresponding to the week number to be predicted. Based on the predicted outdoor ambient temperature, it determines the corresponding target vehicle-kilometer energy consumption model. Thus, it predicts the vehicle-kilometer energy consumption value based on the target vehicle-kilometer energy consumption model and the aforementioned average values. This eliminates the need to analyze the tram's operating conditions, road environment, and driving behavior, simplifying the model structure. The vehicle-kilometer energy consumption prediction results are stored in a database in real time. When retrieving the vehicle-kilometer energy consumption value, it only needs to retrieve the energy consumption prediction result from the corresponding table in the database, without recalculating based on the data source. This greatly simplifies the data processing process, improves data processing efficiency, and still provides good prediction results for tram operation data with insufficient operating time, demonstrating high feasibility.

[0112] In some embodiments, the second determining module 21 is specifically used to: determine a first ambient temperature range, a second ambient temperature range, and a third ambient temperature range; determine the corresponding ambient temperature range based on the outdoor ambient temperature; and establish a vehicle-kilometer energy consumption model based on the outdoor ambient temperature, driving mileage, average load, and travel speed within the corresponding ambient temperature range.

[0113] The following describes an embodiment of the side tram of the present invention.

[0114] like Figure 10 As shown, the tram 3 in this embodiment of the invention includes the vehicle-kilometer energy consumption prediction device 2 of the tram in the above embodiment.

[0115] According to an embodiment of the present invention, the tram 3 determines a vehicle-kilometer energy consumption model within a corresponding ambient temperature range. When determining the week number to be predicted and the corresponding predicted outdoor ambient temperature, it calculates the average mileage, average load, and average travel speed corresponding to the week number. Based on the predicted outdoor ambient temperature, it determines the corresponding target vehicle-kilometer energy consumption model. Thus, it predicts vehicle-kilometer energy consumption values ​​based on the target model and the aforementioned averages. This eliminates the need to analyze the tram's operating conditions, road environment, and driving behavior, simplifying the model structure. The predicted vehicle-kilometer energy consumption results are stored in a database in real time. When retrieving vehicle-kilometer energy consumption values, the prediction results can be retrieved from the corresponding table in the database without recalculating based on the data source, greatly simplifying the data processing process and improving efficiency. Furthermore, it maintains good predictive performance even for tram operation data with insufficient running time, demonstrating high feasibility.

[0116] The fourth aspect of the present invention is a non-transitory computer-readable storage medium storing a tram's vehicle-kilometer energy consumption prediction program. When the tram's vehicle-kilometer energy consumption prediction program is executed by a processor, it implements the tram's vehicle-kilometer energy consumption prediction method as described in the above embodiments.

[0117] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example.

[0118] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. A method for predicting the energy consumption per vehicle-kilometer of a tram, characterized in that, include: Determine the historical energy consumption per kilometer, outdoor ambient temperature, mileage, average load, and travel speed of the tram. Based on the vehicle-kilometer energy consumption, outdoor ambient temperature, mileage, average load, and travel speed, a vehicle-kilometer energy consumption model is determined for the corresponding ambient temperature range. The vehicle-kilometer energy consumption model includes a function relating the vehicle-kilometer energy consumption to the outdoor ambient temperature, the average load, the mileage, and the travel speed within the ambient temperature range. Obtain the week number to be predicted and the predicted outdoor temperature corresponding to the week number to be predicted. Based on the week number to be predicted, determine the mean of the driving mileage, the mean of the average load, and the mean of the travel speed in the prediction dataset corresponding to the week number to be predicted. The target vehicle-kilometer energy consumption model is determined based on the predicted outdoor ambient temperature and the vehicle-kilometer energy consumption model under the corresponding ambient temperature range. The vehicle-kilometer energy consumption value is predicted based on the target vehicle-kilometer energy consumption model and the predicted outdoor ambient temperature, the average mileage, the average load, and the average travel speed corresponding to the week number to be predicted. The vehicle-kilometer energy consumption model, determined based on the vehicle-kilometer energy consumption, outdoor ambient temperature, mileage, average load, and travel speed, for a given ambient temperature range includes: Determine the first ambient temperature range, the second ambient temperature range, and the third ambient temperature range; Based on the outdoor ambient temperature, a corresponding ambient temperature range is determined. Within the corresponding ambient temperature range, a vehicle-kilometer energy consumption model is established that relates the vehicle-kilometer energy consumption to the outdoor ambient temperature, the driving mileage, the average load, and the travel speed. Before determining the mean of the mileage, the mean of the average load, and the mean of the travel speed corresponding to the week number to be predicted in the prediction dataset based on the week number to be predicted, the method further includes: Obtain all mileage, average load, and travel speed of the tram within the predetermined time period from the historical database; Calculate the average mileage, average load, and average travel speed for each week number based on all the mileage, average load, and travel speed. The prediction dataset is constructed based on the mean of the mileage, the mean of the average load, and the mean of the travel speed.

2. The method for predicting the vehicle-kilometer energy consumption of a tram according to claim 1, characterized in that, Based on the predicted outdoor ambient temperature and the vehicle-kilometer energy consumption model under the corresponding ambient temperature range, the corresponding target vehicle-kilometer energy consumption model is determined, including: Determine the corresponding ambient temperature range in which the predicted outdoor ambient temperature falls; The vehicle-kilometer energy consumption model corresponding to the corresponding ambient temperature range is used as the target vehicle-kilometer energy consumption model.

3. The method for predicting the vehicle-kilometer energy consumption of a tram according to claim 1, characterized in that, The vehicle-kilometer energy consumption value is predicted based on the target vehicle-kilometer energy consumption model and the predicted outdoor ambient temperature, the average mileage, the average load, and the average travel speed corresponding to the week number to be predicted, including: The predicted outdoor ambient temperature, the average mileage corresponding to the week number to be predicted, the average load, and the average travel speed in the predicted dataset are input into the target vehicle-kilometer energy consumption model to obtain the vehicle-kilometer energy consumption value.

4. The method for predicting the vehicle-kilometer energy consumption of a tram according to claim 1, characterized in that, After obtaining the vehicle-kilometer energy consumption value, it also includes: The target vehicle-kilometer energy consumption model is updated based on the stated vehicle-kilometer energy consumption value.

5. The method for predicting the vehicle-kilometer energy consumption of a tram according to claim 1, characterized in that, Obtain historical data on tram energy consumption per kilometer, outdoor ambient temperature, mileage, average load, and travel speed, including: The current total voltage of the battery pack of the tram, the current total current of the battery pack, the first pressure value, the second pressure value, the current station position, the terminal station position, and the speed of the tram are obtained. The vehicle-kilometer energy consumption is determined based on the current total voltage of the battery pack, the current total current of the battery pack, and the driving mileage. The average load is determined based on the first pressure value and the second pressure value; The travel speed is determined based on the current station location, the destination station location, and the vehicle speed.

6. A device for predicting the energy consumption per kilometer of a tram, characterized in that, include: The first determination module is used to determine the tram's historical energy consumption per kilometer, outdoor ambient temperature, mileage, average load, and travel speed. The second determining module is used to determine a vehicle-kilometer energy consumption model under a corresponding ambient temperature range based on the vehicle-kilometer energy consumption, outdoor ambient temperature, mileage, average load, and travel speed. The vehicle-kilometer energy consumption model includes a correspondence function between the vehicle-kilometer energy consumption and the outdoor ambient temperature, average load, mileage, and travel speed under the ambient temperature range. Specifically, the second determining module is used to: determine a first ambient temperature range, a second ambient temperature range, and a third ambient temperature range; determine a corresponding ambient temperature range based on the outdoor ambient temperature; and establish a vehicle-kilometer energy consumption model between the vehicle-kilometer energy consumption and the outdoor ambient temperature, mileage, average load, and travel speed under the corresponding ambient temperature range. The third determining module is used to obtain the week number to be predicted and the predicted outdoor temperature corresponding to the week number, and to determine the average mileage, average load, and average travel speed in the prediction dataset corresponding to the week number to be predicted based on the week number to be predicted. Before determining the average mileage, average load, and average travel speed in the prediction dataset corresponding to the week number to be predicted based on the week number to be predicted, the third determining module is further used to: obtain all mileage, average load, and travel speed of the tram within a predetermined time period from a historical database; calculate the average mileage, average load, and average travel speed corresponding to each week number based on all mileage, average load, and travel speed; and construct the prediction dataset based on the average mileage, average load, and average travel speed. The fourth determining module is used to determine the corresponding target vehicle-kilometer energy consumption model based on the predicted outdoor ambient temperature and the vehicle-kilometer energy consumption model under the corresponding ambient temperature range. The prediction module is used to predict the vehicle-kilometer energy consumption value based on the target vehicle-kilometer energy consumption model and the predicted outdoor ambient temperature, the average driving mileage, the average average load, and the average travel speed corresponding to the week number to be predicted.

7. A tram, characterized in that, include: The tram vehicle-kilometer energy consumption prediction device as described in claim 6.

8. A non-transitory computer-readable storage medium, characterized in that, The computer-readable storage medium stores a vehicle-kilometer energy consumption prediction program for a tram, which, when executed by a processor, implements the vehicle-kilometer energy consumption prediction method for a tram as described in any one of claims 1-5.

Images