Remote diagnosis system and method for upstream and downstream temperature sensors of diesel vehicle scr

Abstract

The present application relates to the technical field of digital information transmission, and discloses a diesel vehicle SCR upstream and downstream temperature sensor remote diagnosis system and method, wherein a vehicle-mounted communication terminal in the system obtains running data of the vehicle in a preset time period from a vehicle ECU, uploads the running data to a remote data platform, and then the remote data platform carries out rationality diagnosis on the SCR upstream measured temperature according to the running data and an upstream temperature prediction model to obtain a first diagnosis result, and then the remote data platform carries out rationality diagnosis on the SCR downstream measured temperature according to the running data, the first diagnosis result and a downstream temperature prediction model to obtain a second diagnosis result, so that real-time diagnosis of the SCR upstream and downstream temperature sensors of the vehicle is realized, and the situation that the sensor measured value is unreasonable can be identified in time, and the cheating behavior of the diesel vehicle for reducing the use amount of injected urea can be found as soon as possible.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of digital information transmission, and in particular to a diesel vehicle SCR upstream and downstream temperature sensor remote diagnosis system and method. BACKGROUND

[0002] Vehicle aftertreatment system cheating (also known as aftertreatment system tampering) is a common vehicle violation phenomenon. Strictly speaking, vehicle aftertreatment system cheating is also a kind of vehicle fault, but since vehicle aftertreatment system cheating is caused by the user, it is more difficult to identify. The sub-project DIAS of the EU H2020 project conducted a very detailed investigation on the phenomenon of vehicle aftertreatment system cheating. The survey results show that at present, most of the cheating for heavy-duty diesel vehicles is cheating on the SCR (Selective Catalytic Reduction) system, and the purpose of cheating is mainly to reduce the use of urea injection to greatly save vehicle costs. Reducing urea injection often leads to a decrease in the catalytic conversion efficiency of nitrogen oxides (NOx) in the exhaust, which will cause the vehicle emissions to deteriorate seriously during actual driving.

[0003] However, vehicle emission cheating can only be discovered during the annual emission inspection process, and there is no detection means for such cheating behavior in the prior art, resulting in the discovery of vehicle emission cheating not being timely, and the vehicle emission cheating having a great negative impact on the environment.

[0004] Therefore, the present application is proposed. SUMMARY

[0005] In order to solve the above technical problems, the present application provides a diesel vehicle SCR upstream and downstream temperature sensor remote diagnosis system and method, so as to realize the timely detection of diesel vehicle emission cheating behavior, and thereby avoid the influence of the serious deterioration of vehicle emissions caused by such cheating behavior on the environment in a timely manner.

[0006] The present application provides a diesel vehicle SCR upstream and downstream temperature sensor remote diagnosis system, which comprises an SCR upstream temperature sensor, an SCR downstream temperature sensor, a vehicle ECU, a vehicle-mounted communication terminal and a remote data platform, the vehicle-mounted communication terminal communicates with the vehicle ECU through CAN, and the remote data platform communicates with the vehicle-mounted communication terminal through 4G / 5G, wherein;

[0007] The vehicle-mounted communication terminal is used for obtaining running data of the vehicle in a preset time period from the vehicle ECU, and uploading the running data to the remote data platform, wherein the running data comprises vehicle speed, engine indicated torque percentage, engine friction torque percentage, engine speed, fuel flow, air intake, coolant temperature, SCR upstream measurement temperature and SCR downstream measurement temperature.

[0008] The remote data platform is used for receiving the running data, determining an SCR upstream predicted temperature based on the running data and a pre-trained upstream temperature prediction model, performing rationality diagnosis on the SCR upstream measurement temperature according to the SCR upstream predicted temperature to obtain a first diagnosis result, determining an SCR downstream predicted temperature according to the running data, the first diagnosis result and a pre-trained downstream temperature prediction model, and performing rationality diagnosis on the SCR downstream measurement temperature according to the SCR downstream predicted temperature to obtain a second diagnosis result.

[0009] The embodiment of the present application provides a diesel vehicle SCR upstream and downstream temperature sensor remote diagnosis method, and the method comprises the following steps:

[0010] Obtaining running data, wherein the running data comprises vehicle speed, engine indicated torque percentage, engine friction torque percentage, engine speed, fuel flow, air intake, coolant temperature, SCR upstream measurement temperature and SCR downstream measurement temperature;

[0011] Determining an SCR upstream predicted temperature based on the running data and a pre-trained upstream temperature prediction model, and performing rationality diagnosis on the SCR upstream measurement temperature according to the SCR upstream predicted temperature to obtain a first diagnosis result;

[0012] Determining an SCR downstream predicted temperature according to the running data, the first diagnosis result and a pre-trained downstream temperature prediction model, and performing rationality diagnosis on the SCR downstream measurement temperature according to the SCR downstream predicted temperature to obtain a second diagnosis result.

[0013] The embodiment of the present application has the following technical effects:

[0014] The diesel vehicle SCR upstream and downstream temperature sensor remote diagnosis system provided by the embodiment of the present application, wherein the vehicle-mounted communication terminal obtains the operation data of the vehicle in a preset time period from the vehicle ECU, uploads the operation data to the remote data platform, and then the remote data platform determines the SCR upstream predicted temperature and reasonably diagnoses the SCR upstream measured temperature according to the operation data and the pre-trained upstream temperature prediction model, to obtain the first diagnosis result, and then the remote data platform determines the SCR downstream predicted temperature and reasonably diagnoses the SCR downstream measured temperature according to the operation data, the first diagnosis result and the pre-trained downstream temperature prediction model, to obtain the second diagnosis result, which realizes the real-time diagnosis of the measured values of the SCR upstream and downstream temperature sensors of the vehicle, identifies the unreasonable condition of the sensor measured values in time, discovers the cheating behavior of the diesel vehicle for reducing the injection urea usage as soon as possible, avoids the influence of the cheating behavior on the environment in time, and greatly reduces the negative influence of the cheating of the diesel vehicle aftertreatment system, especially the cheating of the SCR system, on the environment. BRIEF DESCRIPTION OF DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the specific embodiments or the prior art, the following will briefly introduce the drawings needed to be used in the specific embodiments or the prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can obtain other drawings according to these drawings without any creative effort.

[0016] Figure 1 is a structural schematic diagram of a diesel vehicle SCR upstream and downstream temperature sensor remote diagnosis system provided by the embodiment of the present application;

[0017] Figure 2 is a flowchart of a diesel vehicle SCR upstream and downstream temperature sensor remote diagnosis method provided by the embodiment of the present application. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solutions and advantages of the present application more clear, the technical solutions of the present application will be described clearly and completely. Obviously, the described embodiments are only some of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without any creative effort are within the scope of the present application.

[0019] Before the diesel vehicle SCR upstream and downstream temperature sensor remote diagnosis system provided by the embodiment of the present application is described in detail, the technical problems solved by the system are described.

[0020] The exhaust temperature is one of the key factors affecting the urea injection amount, and when the exhaust temperature is low, the urea injection amount is correspondingly reduced or even stopped, and the upstream and downstream temperature sensors of the SCR are the data input for the SCR system to obtain the exhaust temperature. Therefore, there are cheating methods for the upstream and downstream temperature sensors of the SCR, such as artificially raising or completely pulling out the upstream and downstream temperature sensors by using a gasket, so that the measured exhaust temperature is lower than the actual exhaust temperature, so as to achieve the purpose of reducing the use amount of injected urea and greatly saving the vehicle cost.

[0021] In 2018, the Ministry of Ecology and Environment issued GB 17691-2018 “Heavy-duty Diesel Vehicle Emission Limit and Measurement Method (China Phase VI)”, which requires that from the sixth stage of China, heavy-duty diesel vehicles are required to install a remote data acquisition terminal (vehicle communication terminal) throughout their life and send data to the environmental supervision department in real time.

[0022] On this basis, the embodiment of the present application can obtain data streams in real time through digital information transmission technology between the vehicle and the remote data platform, and then analyze the data streams in real time to find out the aftertreatment cheating behavior of the vehicle as soon as possible, so as to timely control the related vehicle and greatly reduce the negative impact of cheating of the aftertreatment system of the heavy-duty diesel vehicle, especially the SCR system, on the environment.

[0023] Figure 1 It is a structural schematic diagram of a diesel vehicle SCR upstream and downstream temperature sensor remote diagnosis system provided by the embodiment of the present application. Referring to Figure 1 The diesel vehicle SCR upstream and downstream temperature sensor remote diagnosis system includes an SCR upstream temperature sensor 11, an SCR downstream temperature sensor 12, a vehicle ECU 13, a vehicle communication terminal 14 and a remote data platform 15. The vehicle communication terminal 14 communicates with the vehicle ECU 13 through CAN, and the remote data platform 15 communicates with the vehicle communication terminal 14 through 4G / 5G communication.

[0024] The vehicle communication terminal 14 is used to obtain the running data of the vehicle in a preset time period from the vehicle ECU 13, and upload the running data to the remote data platform 15, wherein the running data includes vehicle speed, engine indicated torque percentage, engine friction torque percentage, engine speed, fuel flow, air intake, coolant temperature, SCR upstream measurement temperature and SCR downstream measurement temperature.

[0025] The remote data platform 15 is configured to receive the operation data, determine the SCR upstream predicted temperature based on the operation data and a pre-trained upstream temperature prediction model, perform rationality diagnosis on the SCR upstream measured temperature according to the SCR upstream predicted temperature to obtain a first diagnosis result, determine the SCR downstream predicted temperature according to the operation data, the first diagnosis result and a pre-trained downstream temperature prediction model, and perform rationality diagnosis on the SCR downstream measured temperature according to the SCR downstream predicted temperature to obtain a second diagnosis result.

[0026] In the embodiment of the present application, the vehicle ECU 13 can obtain the SCR upstream measured temperature collected by the SCR upstream temperature sensor 11 in real time and the SCR downstream measured temperature collected by the SCR downstream temperature sensor 12 in real time. The SCR upstream measured temperature can be the inlet measured temperature of the SCR system, and the SCR downstream measured temperature can be the outlet measured temperature of the SCR system.

[0027] In addition, the vehicle ECU 13 can also obtain data collected by other sensors in real time, such as the vehicle speed collected by the speed sensor, the engine speed collected by the engine sensor, the engine friction torque percentage, etc. After obtaining the operation data, the vehicle ECU 13 can send the operation data to the vehicle communication terminal 14 in real time through the CAN (Controller Area Network).

[0028] Further, the vehicle communication terminal 14 can send the operation data of the vehicle within a preset time period to the remote data platform 15 in a preset format corresponding to the remote data platform 15. The operation data can be composed of data of nine dimensions, including the vehicle speed, the engine indicated torque percentage, the engine friction torque percentage, the engine speed, the fuel flow, the intake air volume, the coolant temperature, the SCR upstream measured temperature and the SCR downstream measured temperature. The first seven dimensions of data are parameters affecting the actual temperature of the upstream and downstream of the vehicle SCR, and the purpose of obtaining the first seven dimensions of data is to realize the prediction of the upstream and downstream temperatures of the SCR by combining all the parameters affecting the actual temperature of the upstream and downstream of the vehicle SCR, so as to ensure the accuracy of the temperature prediction result.

[0029] Further, the remote data platform 15 can determine the SCR upstream predicted temperature according to the operation data and a pre-trained upstream temperature prediction model, and determine the SCR downstream predicted temperature according to the operation data, the pre-trained downstream temperature prediction model and the first diagnosis result of the SCR upstream measured temperature.

[0030] The upstream temperature prediction model and the downstream temperature prediction model can be models corresponding to the engine type of the vehicle with the operation data, that is, in the embodiment of the application, different engine types can have corresponding upstream temperature prediction models and downstream temperature prediction models respectively, and the remote data platform 15 can select a model corresponding to the engine type of the vehicle to be predicted from all the models before prediction.

[0031] In a specific embodiment, the remote data platform 15 is specifically used for:

[0032] The vehicle speed, the engine indicated torque percentage, the engine friction torque percentage, the engine speed, the fuel flow, the intake air volume, and the coolant temperature in the operation data are input into the upstream temperature prediction model to obtain the SCR upstream predicted temperature output by the upstream temperature prediction model; the first mean absolute error between the SCR upstream predicted temperature and the SCR upstream measured temperature in a preset time period is calculated, and if the first mean absolute error is greater than a first preset threshold, the first diagnosis result is determined to be that the upstream measured temperature is unreasonable.

[0033] Specifically, the remote data platform 15 can input the data of the first 7 dimensions in the operation data in the preset time period into the upstream temperature prediction model to obtain the predicted value in the preset time period output by the upstream temperature prediction model, that is, the SCR upstream predicted temperature of the vehicle in the preset time period.

[0034] Further, the remote data platform 15 can calculate the mean absolute error (MAE) between the SCR upstream predicted temperature and the SCR upstream measured temperature in the preset time period, and compare the calculated first mean absolute error with a first preset threshold, wherein the first preset threshold can be a critical error value between the predicted value and the measured value of the upstream temperature.

[0035] If the first mean absolute error is greater than the first preset threshold, it indicates that the SCR upstream measured temperature is unreasonable, and the SCR upstream temperature sensor has cheating behavior, otherwise, it indicates that the SCR upstream measured temperature is reasonable, and the SCR upstream temperature sensor does not have cheating behavior. Through the above embodiment, the remote data platform realizes the rationality diagnosis of the upstream temperature to detect whether the SCR upstream temperature sensor has cheating behavior.

[0036] In the embodiment of the present application, the remote data platform 15 first performs rationality diagnosis on the SCR upstream measurement temperature, and then further performs rationality diagnosis on the SCR downstream measurement temperature. The purpose of such arrangement is to combine the rationality diagnosis result of the upstream temperature to further diagnose the downstream temperature, thereby improving the diagnosis accuracy of the downstream temperature. Of course, the rationality diagnosis can also be performed on the SCR upstream measurement temperature and the SCR downstream measurement temperature at the same time.

[0037] In another specific embodiment, the remote data platform 15 is specifically configured to: input the vehicle speed, the engine indicated torque percentage, the engine friction torque percentage, the engine speed, the fuel flow, the intake air volume, and the coolant temperature in the running data into the downstream temperature prediction model, and input the SCR upstream predicted temperature or the SCR upstream measurement temperature into the downstream temperature prediction model to obtain the SCR downstream predicted temperature output by the downstream temperature prediction model; and calculate a second average absolute error between the SCR downstream predicted temperature and the SCR downstream measurement temperature in a preset time period, and determine that the second diagnosis result is that the downstream measurement temperature is not reasonable if the second average absolute error is greater than a second preset threshold.

[0038] Specifically, when the remote data platform 15 predicts the SCR downstream temperature, the first 7 dimensions of data in the running data in the preset time period can be input into the downstream temperature prediction model, and the SCR upstream predicted temperature or the SCR upstream measurement temperature can be input into the downstream temperature prediction model. In the embodiment of the present application, the purpose of inputting the SCR upstream predicted temperature or the SCR upstream measurement temperature into the downstream temperature prediction model is that the SCR downstream temperature is affected by the upstream temperature, so the upstream temperature can be combined to predict the downstream temperature, thereby improving the prediction accuracy of the downstream temperature.

[0039] Optionally, the remote data platform 15 is further configured to: input the SCR upstream predicted temperature into the downstream temperature prediction model in the case that the first diagnosis result is that the upstream measurement temperature is not reasonable, and input the SCR upstream measurement temperature into the downstream temperature prediction model in the case that the first diagnosis result is that the upstream measurement temperature is reasonable.

[0040] That is, when the first diagnosis result is that the upstream measurement temperature is not reasonable, the remote data platform 15 inputs the first 7 dimensions of data and the SCR upstream predicted temperature into the downstream temperature prediction model, and when the first diagnosis result is that the upstream measurement temperature is reasonable, the remote data platform 15 inputs the first 7 dimensions of data and the SCR upstream measurement temperature into the downstream temperature prediction model.

[0041] The purpose of the arrangement is that if the first diagnosis result is that the upstream measured temperature is unreasonable, it indicates that the SCR upstream measured temperature is inaccurate, at this time, the more accurate SCR upstream predicted temperature can be input into the downstream temperature prediction model to ensure the accuracy of the prediction result; if the first diagnosis result is that the upstream measured temperature is reasonable, it indicates that the SCR upstream measured temperature is accurate, at this time, the accurate SCR upstream measured temperature can be input into the downstream temperature prediction model to ensure the accuracy of the prediction result.

[0042] Further, after obtaining the downstream predicted temperature of the SCR, the remote data platform 15 can calculate the average absolute error between the downstream predicted temperature of the SCR and the downstream measured temperature of the SCR in a preset time period, compare the calculated second average absolute error with a second preset threshold, where the second preset threshold can be a preset error threshold between the predicted value and the measured value of the downstream temperature.

[0043] If the second average absolute error is greater than the second preset threshold, it indicates that the downstream measured temperature of the SCR is unreasonable, and the SCR downstream temperature sensor has cheating behavior, otherwise, it indicates that the downstream measured temperature of the SCR is reasonable, and the SCR downstream temperature sensor does not have cheating behavior. Through the above embodiment, the remote data platform realizes the rationality diagnosis of the downstream temperature to detect whether the SCR downstream temperature sensor has cheating behavior.

[0044] In the above embodiment, the first preset threshold and the second preset threshold can be artificially set or calculated according to a verification set composed of training sample data.

[0045] In an example, the remote data platform 15 is further configured to:

[0046] obtain a first verification error of the upstream temperature prediction model on the verification set and a second verification error of the downstream temperature prediction model on the verification set, determine the first preset threshold based on the first verification error and a first correction coefficient, in a case where the first diagnosis result is that the upstream measured temperature is unreasonable, determine the second preset threshold based on the second verification error and a second correction coefficient, and in a case where the first diagnosis result is that the upstream measured temperature is reasonable, determine the second preset threshold based on the second verification error and a third correction coefficient, where the second correction coefficient is less than the third correction coefficient.

[0047] The training sample data of the upstream temperature can constitute a training set and a verification set, and the training sample data of the downstream temperature can also constitute a training set and a verification set, the training of the model can be completed on the corresponding training set, and then the verification of the model can be completed on the verification set after the training is completed.

[0048] Specifically, after the upstream temperature prediction model and the downstream temperature prediction model are trained, the corresponding verification set can be predicted by the upstream temperature prediction model, and then the first verification error can be calculated according to the upstream measured temperature of the training sample data in the verification set and the upstream predicted temperature output by the model. Moreover, the corresponding verification set can be predicted by the downstream temperature prediction model, and then the second verification error can be calculated according to the downstream measured temperature of the training sample data in the verification set and the downstream predicted temperature output by the model.

[0049] Further, the product of the first verification error and the first correction coefficient can be taken as the first preset threshold. For example, the first preset threshold can be calculated by the following formula: ; wherein, is the first correction coefficient, is the first verification error, is the first preset threshold.

[0050] Moreover, the remote data platform 15 can take the product of the second verification error and the second correction coefficient as the second preset threshold in the case that the upstream measured temperature is unreasonable, and take the product of the second verification error and the third correction coefficient as the second preset threshold in the case that the upstream measured temperature is reasonable. For example, the following formula can be referred to:

[0051] , ;

[0052] wherein, is the second preset threshold, , are the second correction coefficient and the third correction coefficient respectively, is less than , is the second verification error.

[0053] For example, the first verification error is 12℃, the second verification error is 15℃, the first correction coefficient, the second correction coefficient and the third correction coefficient are 1.5, 1.5 and 2.0 respectively, then the first preset threshold can be 18℃, and the second preset threshold can be 22.5℃ or 30℃.

[0054] It should be noted that the purposes of calculating the second preset threshold by using the second correction coefficient and the third correction coefficient respectively in the case of reasonable and unreasonable upstream measurement temperature are that: since the input to the downstream temperature prediction model is the SCR upstream temperature measurement value when the upstream measurement temperature is reasonable, and the error of the SCR upstream temperature measurement value is small, therefore, the error of the predicted value output by the downstream temperature prediction model at this time is also small, and a smaller second preset threshold can be used for comparison. However, when the upstream measurement temperature is unreasonable, the input to the downstream temperature prediction model is the SCR upstream temperature prediction value, and the prediction value itself has a certain deviation, which may cause the error of the predicted value output by the downstream temperature prediction model to be larger, and therefore a larger second preset threshold can be used for comparison, so as to avoid the case of misdiagnosis of downstream unreasonable due to upstream prediction error.

[0055] In the embodiment of the present application, the training process of the model is exemplarily illustrated by taking the upstream temperature prediction model and the downstream temperature prediction model of one engine type as an example. The training of the upstream temperature prediction model and the downstream temperature prediction model can be divided into the following steps:

[0056] (1) select multiple vehicles with normal SCR upstream temperature sensors and SCR downstream temperature sensors belonging to the same engine type, obtain the operation data of each vehicle to form a training sample data set; (2) data cleaning is performed on the training sample data set; (3) extract the vehicle speed, engine indicated torque percentage, engine friction torque percentage, engine speed, fuel flow, intake air volume, coolant temperature and SCR upstream measured temperature, a total of 8-dimensional data sequences to form a first data set, and perform standard deviation normalization processing on the first data set, so that each dimensional data sequence is linearly scaled to 0-1, and the standardization processing rule formed in the first data set is saved as a first scaler; extract the vehicle speed, engine indicated torque percentage, engine friction torque percentage, engine speed, fuel flow, intake air volume, coolant temperature, SCR upstream measured temperature and SCR downstream measured temperature, a total of 9-dimensional data sequences to form a second data set, and perform standard deviation normalization processing on the second data set, so that each dimensional data sequence is linearly scaled to 0-1, and the standardization processing rule formed by training the second data set is saved as a second scaler; (4) divide the first data set and the second data set into training sets and validation sets respectively; (5) based on the LightGBM algorithm, an upstream temperature prediction model and a downstream temperature prediction model are constructed, the training sets of the first data set and the second data set are respectively input into the upstream temperature prediction model and the downstream temperature prediction model, and a tuning algorithm is used to automatically optimize the key parameters of the model within a given range; (6) after obtaining the optimal parameter value, input the LightGBM model, and after verification by the validation set, form the final upstream temperature prediction model and downstream temperature prediction model, and save the models respectively; (7) calculate the first validation error and the second validation error of the upstream temperature prediction model and the downstream temperature prediction model on the validation set respectively.

[0057] The tuning algorithm includes but is not limited to Bayesian optimization algorithm, genetic algorithm, grid search, black widow optimization algorithm, etc. For example, the filtering parameter optimization configuration is performed by the Bayesian optimization algorithm. The optimization search range of each parameter of the model in the Bayesian optimization algorithm search is shown in Table 1 below:

[0058] Table 1 Optimization range of each parameter in the model

[0059]

[0060] In order to further improve the diagnostic accuracy of the remote data platform 15, in a specific embodiment, the remote data platform 15 is further configured to:

[0061] The received operation data is data cleaned, the vehicle speed, the engine indicated torque percentage, the engine friction torque percentage, the engine speed, the fuel flow, the intake air volume and the coolant temperature in the cleaned operation data are extracted to form first feature data, the first feature data is subjected to deviation standardization processing by a first scaler, and the first feature data subjected to the deviation standardization processing is input into the upstream temperature prediction model; and

[0062] The vehicle speed, the engine indicated torque percentage, the engine friction torque percentage, the engine speed, the fuel flow, the intake air volume and the coolant temperature in the cleaned operation data are extracted, and the SCR upstream predicted temperature or the SCR upstream measured temperature is extracted to form second feature data, the second feature data is subjected to deviation standardization processing by a second scaler, and the first feature data subjected to the deviation standardization processing is input into the downstream temperature prediction model.

[0063] That is, the remote data platform 15 can perform data cleaning on the received operation data before prediction by using the upstream temperature prediction model, so as to eliminate abnormal data. Further, the first feature data is formed by extracting the data of the first 7 dimensions in the operation data, and the first feature data is subjected to deviation standardization processing by the first scaler, so as to realize normalization of the first feature data, and then the processed first feature data is input into the upstream temperature prediction model.

[0064] The remote data platform 15 can also extract the data of the first 7 dimensions in the operation data before prediction by using the downstream temperature prediction model, combine the SCR upstream predicted temperature or the SCR upstream measured temperature to form the second feature data, and perform deviation standardization processing on the second feature data by the second scaler, so as to realize normalization of the second feature data, and then input the processed second feature data into the upstream and downstream temperature prediction models. Wherein, the remote data platform 15 can form the second feature data by using the data of the first 7 dimensions and the SCR upstream predicted temperature when the first diagnostic result is unreasonable, and form the second feature data by using the data of the first 7 dimensions and the SCR upstream measured temperature when the first diagnostic result is reasonable.

[0065] In the above manner, the normalization of the data input into the model is realized, so that each dimension of data is linearly scaled to 0-1, and the prediction accuracy of the model is further improved.

[0066] Optionally, the remote data platform 15 is further configured to: for each dimension of data in the received operation data, sort the data sequence according to the data size, determine the 75% quantile value and the 25% quantile value from the sorting result, determine the data screening range according to the 75% quantile value and the 25% quantile value, and eliminate the data outside the data screening range in the data sequence, so as to perform data cleaning on the received operation data.

[0067] Specifically, the method of data cleaning on the operation data is: arranging the data sequence of each dimension data according to the data size, selecting the 75% quantile value and the 25% quantile value from the arranged sequence, determining the upper and lower limits of the data screening range through the 75% quantile value and the 25% quantile value, and then eliminating the data outside the upper and lower limits in the data sequence. For example, the data screening range determined by the 75% quantile value Q3 and the 25% quantile value Q1 is: [Q1-1.5(Q3-Q1), Q3+1.5(Q3-Q1)].

[0068] Through the above embodiment, the cleaning of abnormal data in the operation data is realized. Specifically, the data cleaning can be performed before the trained model performs prediction, or the data input into the model can be cleaned during the training process to eliminate abnormal data and avoid the influence of invalid data on the prediction result of the model, thereby further ensuring the prediction accuracy of the model.

[0069] In an optional embodiment, the system provided by the embodiment of the present application further includes a WEB end, and the remote data platform 15 communicates with the WEB end through TCP / IP; the remote data platform 15 is further configured to send the first diagnosis result and the second diagnosis result to the WEB end; and the WEB end is configured to display the first diagnosis result and the second diagnosis result of each vehicle, and display the first diagnosis result and the second diagnosis result of all vehicles under each vehicle type.

[0070] Specifically, the remote data platform 15 can send the first diagnosis result and the second diagnosis result to the WEB end for display. The WEB end can display the first diagnosis result and the second diagnosis result of each vehicle, and also display the first diagnosis result and the second diagnosis result of all vehicles under each vehicle type. For the first diagnosis result and the second diagnosis result of all vehicles under each vehicle type, the WEB end can display the statistical result, such as in the form of a pie chart.

[0071] In the above embodiment, the first diagnosis result and the second diagnosis result of each vehicle, and the first diagnosis result and the second diagnosis result of all vehicles under each vehicle type are displayed through the WEB end, which facilitates the user to determine the vehicle to be controlled based on the displayed content.

[0072] Optionally, the WEB end is further configured to generate an alarm vehicle list based on the first diagnosis result and the second diagnosis result of each vehicle; display the alarm vehicle list, or for each vehicle in the alarm vehicle list, generate corresponding alarm prompt information and send it to the vehicle-mounted communication terminal of the vehicle, so that the vehicle-mounted communication terminal displays the alarm prompt information on the vehicle instrument panel.

[0073] That is, the WEB end can determine that the first diagnostic result is that the upstream measured temperature is unreasonable, or the second diagnostic result is that the downstream measured temperature is unreasonable, obtain an alarm vehicle list, and then display the alarm vehicle list, or issue a corresponding alarm prompt information to each vehicle in the alarm vehicle list to display on the vehicle instrument panel, and then prompt the vehicle owner to detect cheating behavior and adjust the SCR upstream and downstream temperature sensors in time. Of course, the alarm prompt information can also be displayed through the vehicle voice interaction system or the multimedia interface, which is not limited.

[0074] In the above embodiment, the WEB end visualizes and displays the diagnostic results and the alarm vehicle list, which facilitates timely processing of vehicles that need to be controlled, timely prompts the cheating vehicle owner to stop cheating behavior, and greatly reduces the negative impact of heavy diesel vehicles with such cheating behavior on the environment.

[0075] The present application has the following technical effects: The diesel vehicle SCR upstream and downstream temperature sensor remote diagnosis system provided by the embodiment of the present application, the vehicle-mounted communication terminal obtains the running data of the vehicle in a preset time period from the vehicle ECU, uploads it to the remote data platform, and then the remote data platform determines the SCR upstream predicted temperature according to the running data and the pre-trained upstream temperature prediction model, and diagnoses the rationality of the SCR upstream measured temperature to obtain the first diagnostic result. Then, the remote data platform determines the SCR downstream predicted temperature according to the running data, the first diagnostic result, and the pre-trained downstream temperature prediction model, and diagnoses the rationality of the SCR downstream measured temperature to obtain the second diagnostic result, which realizes real-time diagnosis of the measured values of the SCR upstream and downstream temperature sensors of the vehicle, timely identifies the situation that the sensor measured value is unreasonable, remotely and accurately diagnoses whether the in-use diesel vehicle has cheating behaviors such as artificially raising or completely pulling out the temperature sensor, makes the urea injection amount reduce or completely stop, causes the nitrogen oxide (NOx) emission to deteriorate seriously, discovers the cheating behavior of the diesel vehicle to reduce the use amount of injected urea as soon as possible, avoids the influence of such cheating behavior on the environment, and greatly reduces the negative impact of diesel vehicle aftertreatment system, especially the cheating of SCR system, on the environment.

[0076] Figure 2 The diesel vehicle SCR upstream and downstream temperature sensor remote diagnosis method provided by the embodiment of the present application is a flowchart. The diesel vehicle SCR upstream and downstream temperature sensor remote diagnosis method can be executed by the remote data platform in the diesel vehicle SCR upstream and downstream temperature sensor remote diagnosis system, and is suitable for remotely diagnosing each diesel vehicle to determine whether it has cheating behavior. As shown in Figure 2 The method comprises the following steps:

[0077] S210, acquire running data, wherein the running data comprises vehicle speed, engine indicated torque percentage, engine friction torque percentage, engine speed, fuel flow, intake air volume, coolant temperature, SCR upstream measurement temperature and SCR downstream measurement temperature.

[0078] S220, determine the SCR upstream predicted temperature based on the running data and the pre-trained upstream temperature prediction model, perform rationality diagnosis on the SCR upstream measurement temperature according to the SCR upstream predicted temperature, and obtain a first diagnosis result.

[0079] S230, determine the SCR downstream predicted temperature according to the running data, the first diagnosis result and the pre-trained downstream temperature prediction model, perform rationality diagnosis on the SCR downstream measurement temperature according to the SCR downstream predicted temperature, and obtain a second diagnosis result.

[0080] The present application has the following technical effects: according to the running data and the pre-trained upstream temperature prediction model, the SCR upstream predicted temperature is determined and the rationality diagnosis is performed on the SCR upstream measurement temperature, and the first diagnosis result is obtained, and then, according to the running data, the first diagnosis result and the pre-trained downstream temperature prediction model, the SCR downstream predicted temperature is determined and the rationality diagnosis is performed on the SCR downstream measurement temperature, and the second diagnosis result is obtained, the real-time diagnosis of the measurement value of the SCR upstream and downstream temperature sensors of the vehicle is realized, the unreasonable condition of the sensor measurement value is identified in time, the cheating behavior of the diesel vehicle for reducing the injection of urea is found as soon as possible, the influence of the serious deterioration of the vehicle emission caused by such cheating behavior on the environment is avoided in time, and the negative influence of the cheating of the diesel vehicle aftertreatment system, especially the SCR system, on the environment is greatly reduced.

[0081] It should be noted that the terms used in the present application are only for describing specific embodiments, and are not intended to limit the scope of the present application. As shown in the specification of the present application, unless the context clearly indicates otherwise, the words "one", "a", "an" and / or "the" do not specifically refer to the singular, but also include the plural. The terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or includes elements inherent to such process, method or device. Without more limitation, the element defined by the statement "including one" does not exclude the presence of another identical element in the process, method or device including the element.

[0082] It should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", and the like, indicate an orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are used only to facilitate the description of the present application and simplify the description, and do not indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application. Unless otherwise specifically defined and limited, the terms "mount", "connect", "connect" and the like should be broadly understood, for example, can be fixedly connected, can also be detachably connected, or integrally connected; can be mechanically connected, can also be electrically connected; can be directly connected, can also be indirectly connected through an intermediate medium; can be internal communication of two elements. For those of ordinary skill in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0083] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit it; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the technical solutions of the embodiments of the present application.

Claims

1. A diesel vehicle SCR upstream and downstream temperature sensor remote diagnostic system characterized by, The system comprises an SCR upstream temperature sensor, an SCR downstream temperature sensor, a vehicle ECU, a vehicle communication terminal and a remote data platform, the vehicle communication terminal communicates with the vehicle ECU through CAN, and the remote data platform communicates with the vehicle communication terminal through 4G / 5G, wherein The vehicle communication terminal is configured to obtain running data of the vehicle within a preset time period from the vehicle ECU, and upload the running data to the remote data platform, wherein the running data comprises vehicle speed, engine indicated torque percentage, engine friction torque percentage, engine speed, fuel flow, intake air volume, coolant temperature, SCR upstream measured temperature and SCR downstream measured temperature; The remote data platform is configured to receive the running data, determine an SCR upstream predicted temperature based on the running data and a pre-trained upstream temperature prediction model, perform rationality diagnosis on the SCR upstream measured temperature according to the SCR upstream predicted temperature to obtain a first diagnosis result, determine an SCR downstream predicted temperature according to the running data, the first diagnosis result and a pre-trained downstream temperature prediction model, and perform rationality diagnosis on the SCR downstream measured temperature according to the SCR downstream predicted temperature to obtain a second diagnosis result; The remote data platform is further configured to: In a case where the first diagnosis result is that the upstream measured temperature is not reasonable, input the SCR upstream predicted temperature into the downstream temperature prediction model, and in a case where the first diagnosis result is that the upstream measured temperature is reasonable, input the SCR upstream measured temperature into the downstream temperature prediction model.

2. The system of claim 1, wherein, The system further comprises a WEB terminal, and the remote data platform communicates with the WEB terminal through TCP / IP; The remote data platform is further configured to send the first diagnosis result and the second diagnosis result to the WEB terminal; The WEB terminal is configured to display the first diagnosis result and the second diagnosis result of each vehicle, and display the first diagnosis result and the second diagnosis result of all vehicles of each vehicle type.

3. The system of claim 2, wherein, The WEB terminal is further configured to generate an alarm vehicle list based on the first diagnosis result and the second diagnosis result of each vehicle, display the alarm vehicle list, or generate corresponding alarm prompt information for each vehicle in the alarm vehicle list and send the alarm prompt information to the vehicle communication terminal of the vehicle, so that the vehicle communication terminal displays the alarm prompt information on the vehicle instrument panel.

4. The system of claim 1, wherein, The remote data platform is specifically configured to: input the vehicle speed, the engine indicated torque percentage, the engine friction torque percentage, the engine speed, the fuel flow, the intake air volume and the coolant temperature in the running data into the upstream temperature prediction model to obtain the SCR upstream predicted temperature output by the upstream temperature prediction model; calculate a first average absolute error between the SCR upstream predicted temperature and the SCR upstream measured temperature in the preset time period, and determine that the first diagnosis result is that the upstream measured temperature is unreasonable if the first average absolute error is greater than a first preset threshold.

5. The system of claim 4, wherein, The remote data platform is specifically used for: inputting the vehicle speed, the engine indicated torque percentage, the engine friction torque percentage, the engine speed, the fuel flow, the intake air amount, and the coolant temperature in the running data into the downstream temperature prediction model, and inputting the SCR upstream predicted temperature or the SCR upstream measured temperature into the downstream temperature prediction model to obtain an SCR downstream predicted temperature output by the downstream temperature prediction model; calculate a second average absolute error between the SCR downstream predicted temperature and the SCR downstream measured temperature in the preset time period, and determine that the second diagnosis result is that the downstream measured temperature is unreasonable if the second average absolute error is greater than a second preset threshold.

6. The system of claim 5, wherein, The remote data platform is further used for: obtaining a first validation error of the upstream temperature prediction model on a validation set and a second validation error of the downstream temperature prediction model on the validation set, and determining the first preset threshold based on the first validation error and a first correction coefficient; determining the second preset threshold based on the second validation error and a second correction coefficient in a case where the first diagnosis result is that the upstream measured temperature is unreasonable, and determining the second preset threshold based on the second validation error and a third correction coefficient in a case where the first diagnosis result is that the upstream measured temperature is reasonable, wherein the second correction coefficient is less than the third correction coefficient.

7. The system of claim 5, wherein, The remote data platform is further used for: performing data cleaning on the received running data, extracting vehicle speed, engine indicated torque percentage, engine friction torque percentage, engine speed, fuel flow, intake air amount, and coolant temperature in the cleaned running data to form first feature data, performing deviation standardization processing on the first feature data by a first scaler, and inputting the first feature data after the deviation standardization processing into the upstream temperature prediction model; and extracting vehicle speed, engine indicated torque percentage, engine friction torque percentage, engine speed, fuel flow, intake air amount, and coolant temperature in the cleaned running data, and the SCR upstream predicted temperature or the SCR upstream measured temperature to form second feature data, performing deviation standardization processing on the second feature data by a second scaler, and inputting the first feature data after the deviation standardization processing into the downstream temperature prediction model.

8. The system of claim 7, wherein, The remote data platform is further used for: sorting data sequences according to data sizes for each dimension of data in the received running data, determining 75% quantile values and 25% quantile values from a sorting result, determining a data screening range according to the 75% quantile values and the 25% quantile values, and removing data outside the data screening range in the data sequences to clean the received running data.

9. A diesel vehicle SCR upstream and downstream temperature sensor remote diagnosis method, characterized by, The method comprises: acquiring running data, wherein the running data comprises vehicle speed, engine indicated torque percentage, engine friction torque percentage, engine speed, fuel flow, intake air volume, coolant temperature, SCR upstream measurement temperature and SCR downstream measurement temperature; determining an SCR upstream predicted temperature based on the running data and a pre-trained upstream temperature prediction model, performing rationality diagnosis on the SCR upstream measurement temperature according to the SCR upstream predicted temperature, and obtaining a first diagnosis result; determining an SCR downstream predicted temperature based on the running data, the first diagnosis result and a pre-trained downstream temperature prediction model, performing rationality diagnosis on the SCR downstream measurement temperature according to the SCR downstream predicted temperature, and obtaining a second diagnosis result; wherein in the case that the first diagnosis result is that the upstream measurement temperature is not reasonable, the SCR upstream predicted temperature is input to the downstream temperature prediction model, and in the case that the first diagnosis result is that the upstream measurement temperature is reasonable, the SCR upstream measurement temperature is input to the downstream temperature prediction model.

Images