A method and system for constructing a digital twin engine

Abstract

The present invention belongs to the field of digital twin technology, and discloses a method and system for constructing a digital twin engine, the method comprising: providing a digital twin model construction module, acquiring the design data of the vehicle, and constructing a digital twin model of the vehicle; providing a data acquisition module, acquiring first operation data related to the vehicle, and mapping the first operation data to the digital twin model through an ID identifier; providing an operation and maintenance detection module, preprocessing the first operation data to obtain second operation data, analyzing the second operation data, and obtaining the first operation state of the vehicle; providing an interaction module, when detecting a vehicle abnormality, obtaining an event code corresponding to the vehicle abnormality, and sending the event code to the vehicle end; providing an operation management module to support the operation of the digital twin model. The technical solution of the present invention can improve the accuracy of vehicle abnormality judgment and improve the safety of vehicle operation.

Description

Technical Field

[0001] The present invention belongs to the technical field of digital twins, and in particular relates to a method and system for constructing a digital twin engine. Background Art

[0002] In recent years, with the development of new generation information technologies such as 5G, Internet of Things, cloud computing, big data, artificial intelligence and mixed reality, digital twins have achieved rapid development in both theory and application. As a virtual simulation technology, digital twins have quickly emerged in the Internet of Vehicles industry, bringing higher efficiency and better safety to the industry, and helping to improve the real-time nature of vehicle fault diagnosis.

[0003] Similar prior art includes a Chinese patent application with publication number CN117877256A, which discloses a vehicle fault prediction method and system based on digital twins, constructs a sensor connectivity group; senses and records the operating data of the target vehicle through N data sensors to obtain an operating sensor data set; obtains the operating environment parameter set of the target vehicle based on the historical operating control parameters of the target vehicle; performs digital twin analysis based on the operating sensor data set and the operating environment parameter set to construct a digital twin of the target vehicle; performs real-time monitoring of the digital twin, predicts faults based on the monitoring results, and generates a fault cause data set; performs operation analysis and optimization of the target vehicle based on the fault cause data set to improve vehicle operation efficiency. There is also a Chinese patent application with publication number CN117922538A, which discloses a hybrid vehicle measurement and control method and system based on digital twin technology, trains a vehicle digital twin model, and outputs feedback test data of the hybrid vehicle through the vehicle digital twin model; determines the data importance value through the real-time interaction data between the vehicle digital twin model and the hybrid vehicle, divides the calculation subtasks according to the data importance value, obtains subtask classification data, performs classification calculation on the subtask classification data, and obtains classification calculation results; calculates data outliers and mapping deviations according to the classification calculation results and mapping relationships, determines the vehicle operating status according to the data outliers, mapping deviations and the classification calculation results, performs vehicle early warning control, and obtains control results.

[0004] When monitoring vehicles, the above-mentioned existing technologies do not consider the accuracy of the collected operating data and the differences between different types of vehicles, resulting in inaccurate vehicle operating status. Therefore, it is an urgent problem to provide a method and system for building a digital twin engine to improve the accuracy of vehicle abnormality judgment. Summary of the invention

[0005] In response to the above-mentioned technical problems, the present invention provides a method and system for constructing a digital twin engine.

[0006] In a first aspect, the present invention provides a method for constructing a digital twin engine, the method comprising the following steps:

[0007] Provide a digital twin model building module to obtain vehicle design data and build a digital twin model of the vehicle;

[0008] Provide a data acquisition module to collect first operation data related to the vehicle and map the first operation data to the digital twin model through an ID identifier;

[0009] An operation and maintenance detection module is provided to perform a first preprocessing on the first operation data to obtain second operation data, and to analyze the second operation data to obtain a first operation state of the vehicle;

[0010] An interactive module is provided to determine whether the vehicle is abnormal based on the first operating state, and when the vehicle is abnormal, obtain an event code corresponding to the vehicle abnormality and send the event code to the vehicle end;

[0011] Provide an operation management module to support the operation of the digital twin model.

[0012] Specifically, performing a first preprocessing on the first operating data to obtain second operating data, and analyzing the second operating data to obtain the first operating state of the vehicle includes:

[0013] Step 11: Acquire a first characteristic identifier of the vehicle, and identify a driving state of the vehicle based on the first operating data;

[0014] Step 12: extract any first operating data, search the first storage module based on the first feature identifier and the driving state, obtain a first mapping coefficient corresponding to any first operating data from the first storage module, and perform standardization processing on any first operating data using the first mapping coefficient;

[0015] Step 13, repeat step 12, after traversing all the first operation data, obtain the second operation data;

[0016] Step 14: searching the second storage module based on the first feature identifier and the driving state, and obtaining a first preset model corresponding to the vehicle in the driving state from the second storage module;

[0017] Step 15: Input the second operating data into the first preset model to obtain the first operating state.

[0018] Specifically, a data classification calculation module is also provided, which is used to generate a first storage module, including the following steps:

[0019] Step 21: extracting a first preset number of first vehicle operation data from the historical operation database, wherein the first vehicle operation data is vehicle operation data of a vehicle in a normal operation state, and the first vehicle operation data includes vehicle operation data of all driving states of vehicles with all feature identifiers;

[0020] Step 22: divide all feature identifiers into N1 groups based on the categories of feature identifiers, divide the first vehicle operation data into N1 first data sets, and divide the vehicle operation data corresponding to vehicles with feature identifiers belonging to the same group into the same first data set, wherein N1 is a positive integer greater than or equal to 1;

[0021] Step 23, extract any first data set, divide any first data set into N2 second data sets based on the feature identifier, and divide the vehicle operation data corresponding to vehicles belonging to the same feature identifier into the same second data set, wherein N2 is a positive integer greater than or equal to 1;

[0022] Step 24, extracting any second data set, and dividing any second data set into N3 third data sets based on the driving state, so as to divide the vehicle operation data belonging to the same driving state into the same third data set, wherein N3 is a positive integer greater than 1;

[0023] Step 25, extracting any third data set, and dividing any third data set into N4 fourth data sets based on the type of vehicle operation data, so as to divide data belonging to the same type of vehicle operation data into the same fourth data set, wherein N4 is a positive integer greater than 1;

[0024] Step 26, repeating steps 23 to 25, dividing all first data sets layer by layer, and respectively obtaining a second data set, a third data set, and a fourth data set of each first data set;

[0025] Step 27: extract the ith first data set, and calculate the mapping coefficient of the nth vehicle operation data of the mth driving state of the vehicle corresponding to the jth feature identifier in the ith first data set. C ijmn , and the mapping coefficients C ijmn The mapping coefficients are stored in the first storage module corresponding to the i-th first data set, the j-th feature identifier, the m-th driving state, and the n-th vehicle operation data. C ijmn The calculation formula is:

[0026] ,

[0027] ,

[0028] ,

[0029] in, AV ijmn is the first average of the nth type of vehicle operation data of the mth driving state of the vehicle corresponding to the jth feature identifier in the i-th first data set, is the second average of the nth type of vehicle operation data in the mth driving state in the i-th first data set, Nu k is the kth value in the fourth data set corresponding to the nth type of vehicle operation data of the mth driving state of the vehicle corresponding to the jth feature identifier in the ith first data set, i is a positive integer from 1 to N1, j is a positive integer from 1 to N2, m is a positive integer from 1 to N3, n is a positive integer from 1 to N4, k is a positive integer from 1 to N5, and N5 is the total number of data in the fourth data set corresponding to the nth type of vehicle operation data of the mth driving state of the vehicle corresponding to the jth feature identifier in the ith first data set;

[0030] Step 28, determine whether n is equal to N4, if not, set n=n+1, return to step 27, if yes, determine whether m is equal to N3, if not, set m=m+1, return to step 27, if yes, determine whether j is equal to N2, if not, set j=j+1, return to step 27, if yes, determine whether i is equal to N1, if not, set i=i+1, return to step 27, if yes, proceed to step 29;

[0031] Step 29: After traversing all first data sets, generate a first storage module.

[0032] Specifically, the data classification calculation module is also used to generate the second storage module:

[0033] Step 31: extracting a second preset amount of second vehicle operation data from the historical operation database, wherein the second vehicle operation data includes vehicle operation data of a normal operation state and an abnormal operation state, and the second vehicle operation data includes vehicle operation data of all driving states of vehicles with all feature identifiers;

[0034] Step 32: According to the method of step 22, the second vehicle operation data is divided into N1 fifth data sets;

[0035] Step 33: extract any fifth data set, perform standardization on any vehicle operation data of any driving state of the vehicle corresponding to any feature identifier based on the mapping coefficient corresponding to any vehicle operation data of any driving state of the vehicle, and obtain a sixth data set corresponding to any fifth data set after traversing all data in any fifth data set;

[0036] Step 34: dividing the sixth data set into N3 seventh data sets based on the driving state, so as to divide the vehicle operation data belonging to the same driving state into the same seventh data set;

[0037] Step 35: extract any seventh data set, use the data in any seventh data set as training data to perform model training, generate a preset model corresponding to any driving state in any fifth data set, and store the preset model, the specific feature identification group, and the specific driving state in the second storage module in correspondence, wherein the specific driving state is the driving state corresponding to any seventh data set, and the specific feature identification group is the feature identification group corresponding to any fifth data set;

[0038] Step 36: After traversing all fifth data sets, generate a second storage module.

[0039] Specifically, the process before step 12 includes:

[0040] Step 121: extract any first operation data and obtain identification information of any first operation data;

[0041] Step 122: extracting a preset rule corresponding to any first operation data based on the identification information, and determining whether any first operation data satisfies the preset rule. If so, determining that any first operation data meets the requirement. If not, proceeding to step 123;

[0042] Step 123: Acquire historical operation data corresponding to any first operation data within a first preset time, and generate new first operation data based on the historical operation data;

[0043] Step 124: After traversing all the first operation data, proceed to step 12.

[0044] Specifically, the preset rules include a value range and / or a collection period.

[0045] Specifically, the operation management module includes a historical data management module, which manages the first historical operation data, performs a second preprocessing on the first historical operation data and stores the data in a historical operation database, wherein the second preprocessing includes:

[0046] Step 51: extract second historical operation data corresponding to any type of vehicle operation data from the first historical operation data, and evenly divide the second historical operation data into N6 third historical operation data, where N6 is a positive integer greater than 1, and the third historical operation data includes the collection time of the operation data and the operation data value, and the number of operation data in each third historical operation data is the same;

[0047] Step 52: extract any third historical operation data, generate a corresponding data block for any third historical operation data, and divide the data block into a header, a value portion, and a tail portion, wherein the header stores basic information of any third historical operation data, the value portion stores the historical operation data after transformation, and the tail portion stores a time processing flag and a value processing flag;

[0048] Step 53: Obtain the collection time of the first operation data in any third historical operation data, the storage quantity of the operation data in any third historical operation data, and the type of the operation data in any third historical operation data, and store the collection time, storage quantity and type as basic information in the header;

[0049] Step 54: Obtain the h1th operation data in any third historical operation data, calculate the time difference between the h1th operation data and the h1-1th operation data, and replace the collection time of the h1th operation data with the time difference. After traversing the 2nd to Hth operation data, set the collection time of the 1st operation data to 0, where h1 is a positive integer from 2 to H, and H is the storage quantity.

[0050] Step 55, storing the collection time and the operation data value corresponding to the first operation data at the relative storage position of the first operation data in the numerical part, and setting the time processing flag and the value processing flag corresponding to the first operation data to 0;

[0051] Step 56, extract the h2th operation data, determine whether the collection time of the h2th operation data is equal to the collection time of the h2-1th operation data, if not, store the collection time of the h2th operation data at the relative storage position of the collection time of the h2th operation data in the numerical part, and set the time processing flag corresponding to the h2th operation data to 0, then enter step 57, if yes, delete the collection time of the h2th operation data, do not store the collection time of the h2th operation data in the numerical part, and set the time processing flag corresponding to the h2th operation data to 1, where h2 is a positive integer from 2 to H;

[0052] Step 57, determine whether the operating data value of the h2th operating data is equal to the operating data value of the h2-1th operating data. If not, store the operating data value of the h2th operating data at the relative storage position of the operating data value of the h2th operating data in the numerical portion, and set the numerical value processing flag corresponding to the h2th operating data to 0. If so, delete the operating data value of the h2th operating data, do not store the operating data value of the h2th operating data in the numerical portion, and set the numerical value processing flag corresponding to the h2th operating data to 1;

[0053] Step 58: Process all types of vehicle operation data in the first historical operation data according to the methods of steps 51 to 57.

[0054] In a second aspect, the present invention also provides a digital twin engine construction system, which includes: a digital twin model construction module, a data acquisition module, an operation and maintenance detection module, an interaction module and an operation management module;

[0055] A digital twin model building module is used to obtain the vehicle's design data and build a digital twin model of the vehicle;

[0056] A data acquisition module, used to collect first operating data related to the vehicle, and map the first operating data to the digital twin model through an ID identifier;

[0057] An operation and maintenance detection module, configured to perform a first preprocessing on the first operation data to obtain second operation data, analyze the second operation data, and obtain a first operation state of the vehicle;

[0058] An interaction module, configured to determine whether the vehicle has an abnormality according to the first operating state, and when the vehicle has an abnormality, obtain an event code corresponding to the vehicle abnormality, and send the event code to the vehicle end;

[0059] Operation management module, used to support the operation of the digital twin model.

[0060] Compared with the prior art, the beneficial effects of the present invention are at least as follows:

[0061] 1. Obtain the vehicle’s design data and build a digital twin model of the vehicle. The digital twin model reflects the vehicle’s operating parameters in real time, enabling comprehensive monitoring and accurate simulation of the vehicle;

[0062] 2. Collect vehicle operation data in real time and pre-process the operation data. The type and driving status of the vehicle are taken into consideration when pre-processing the operation data. This method is highly targeted, improves the accuracy of data processing, and obtains the second operation data to be analyzed with high accuracy.

[0063] 3. Analyze the second operation data to obtain the first operation state of the vehicle, and select a specific preset model according to the type and driving state of the vehicle, thereby improving the rigor and accuracy of the vehicle operation state detection and improving the safety of vehicle operation;

[0064] 4. The digital twin engine constructed by the present invention is applicable to various types of vehicles and is targeted, thus enhancing the differentiated competitiveness of enterprises. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0066] Figure 1 A flowchart of a method for constructing a digital twin engine of the present invention;

[0067] Figure 2 A modular schematic diagram of a construction system of a digital twin engine of the present invention. DETAILED DESCRIPTION

[0068] In order to make the purpose, technical scheme and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. Obviously, the specific embodiments described herein are only used to explain the present invention and are part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work belong to the scope of protection of the present invention.

[0069] It should be noted that if there are descriptions involving "first", "second", etc. in the embodiments of the present invention, the descriptions of "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in the field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0070] Figure 1 FIG. 1 is a flow chart of an embodiment of a method for constructing a digital twin engine provided by the present invention, and the flow chart specifically includes the following steps:

[0071] Step 1: Provide a digital twin model construction module, obtain the vehicle design data, and build a digital twin model of the vehicle.

[0072] Based on CAD (computer-aided design) software, a geometric model of the vehicle is constructed, including accurate three-dimensional models of components such as the body, engine, transmission system, chassis, etc. Subsequently, through theories such as system dynamics and thermodynamics, the interaction relationship and behavior model between the various vehicle components are established to generate a vehicle physical model. Finally, the preprocessed data is combined with the vehicle physical model to construct a digital twin model of the vehicle.

[0073] The digital twin model of a vehicle is an information-based mapping of the vehicle. It is a virtual entity that can correspond one-to-one with each component of the vehicle to achieve coordinated changes.

[0074] Step 2: Provide a data acquisition module to collect first operating data related to the vehicle, and map the first operating data to the digital twin model through an ID identifier.

[0075] A unique ID is set for each operating data of each vehicle component. Based on the digital twin model, the operating data related to the data twin model is associated through the ID table to realize the mapping of IoT data.

[0076] Step 3: Provide an operation and maintenance detection module, perform a first preprocessing on the first operation data to obtain second operation data, analyze the second operation data, and obtain a first operation state of the vehicle.

[0077] Specifically, performing a first preprocessing on the first operating data to obtain second operating data, and analyzing the second operating data to obtain the first operating state of the vehicle includes:

[0078] Step 11: Obtain a first characteristic identifier of the vehicle, and identify the driving state of the vehicle based on the first operating data.

[0079] Step 12: extract any first operating data, search the first storage module based on the first feature identifier and the driving state, obtain the first mapping coefficient corresponding to any first operating data from the first storage module, and use the first mapping coefficient to standardize any first operating data.

[0080] Step 13: Repeat step 12, and after traversing all the first operation data, obtain the second operation data.

[0081] Step 14: searching the second storage module based on the first feature identifier and the driving state, and obtaining a first preset model corresponding to the vehicle in the driving state from the second storage module.

[0082] Step 15: Input the second operating data into the first preset model to obtain the first operating state.

[0083] Preferably, the characteristic identifier of the vehicle is the model of the vehicle.

[0084] Exemplarily, the driving state of the vehicle includes low-speed driving, high-speed driving, driving in rainy and snowy weather, etc.

[0085] The same data acquisition device may generate different degrees of fluctuation in the collected operating data under different vehicle driving conditions. The collected first operating data is adjusted based on the vehicle model and driving condition, thereby improving the accuracy of data processing.

[0086] A different preset model is generated for each driving state of each type of vehicle. When performing vehicle running state detection, the preset model corresponding to the vehicle type and vehicle driving state to be analyzed is selected to improve the precision and accuracy of vehicle running state detection.

[0087] Specifically, a data classification calculation module is also provided, which is used to generate a first storage module, including the following steps:

[0088] Step 21: extract a first preset number of first vehicle operation data from a historical operation database, wherein the first vehicle operation data is vehicle operation data of a normal operation state, and the first vehicle operation data includes vehicle operation data of all driving states of vehicles with all feature identifiers.

[0089] Step 22: divide all feature identifiers into N1 groups based on the category of feature identifiers, and divide the first vehicle operation data into N1 first data sets, and divide the vehicle operation data corresponding to vehicles with feature identifiers belonging to the same group into the same first data set, where N1 is a positive integer greater than or equal to 1.

[0090] Step 23, extract any first data set, divide any first data set into N2 second data sets based on feature identifiers, and divide vehicle operation data corresponding to vehicles belonging to the same feature identifier into the same second data set, where N2 is a positive integer greater than or equal to 1.

[0091] Step 24, extract any second data set, and divide any second data set into N3 third data sets based on the driving state, so as to divide the vehicle operation data belonging to the same driving state into the same third data set, wherein N3 is a positive integer greater than 1.

[0092] Step 25, extract any third data set, and divide any third data set into N4 fourth data sets based on the type of vehicle operation data, so as to divide data belonging to the same type of vehicle operation data into the same fourth data set, wherein N4 is a positive integer greater than 1.

[0093] Step 26: Repeat steps 23 to 25 to divide all first data sets layer by layer, and respectively obtain the second data set, the third data set and the fourth data set of each first data set.

[0094] Step 27: extract the ith first data set, and calculate the mapping coefficient of the nth vehicle operation data of the mth driving state of the vehicle corresponding to the jth feature identifier in the ith first data set. C ijmn , and the mapping coefficients C ijmn The mapping coefficients are stored in the first storage module corresponding to the i-th first data set, the j-th feature identifier, the m-th driving state, and the n-th vehicle operation data. C ijmn The calculation formula is:

[0095] ,

[0096] ,

[0097] ,

[0098] in, AV ijmn is the first average of the nth type of vehicle operation data of the mth driving state of the vehicle corresponding to the jth feature identifier in the i-th first data set, is the second average of the nth type of vehicle operation data in the mth driving state in the i-th first data set, Nu k is the kth value in the fourth data set corresponding to the nth type of vehicle operating data of the mth driving state of the vehicle corresponding to the jth feature identifier in the ith first data set, i is a positive integer from 1 to N1, j is a positive integer from 1 to N2, m is a positive integer from 1 to N3, n is a positive integer from 1 to N4, k is a positive integer from 1 to N5, and N5 is the total number of data in the fourth data set corresponding to the nth type of vehicle operating data of the mth driving state of the vehicle corresponding to the jth feature identifier in the ith first data set.

[0099] Step 28. Determine whether n is equal to N4. If not, set n=n+1 and return to step 27. If so, determine whether m is equal to N3. If not, set m=m+1 and return to step 27. If so, determine whether j is equal to N2. If not, set j=j+1 and return to step 27. If so, determine whether i is equal to N1. If not, set i=i+1 and return to step 27. If so, proceed to step 29.

[0100] Step 29: After traversing all first data sets, generate a first storage module.

[0101] The first preset number is set according to the experience of those skilled in the art or according to actual application scenarios, and the embodiments of the present application do not limit this.

[0102] Exemplarily, all feature identifiers include A3, A4, A6, A8, Q1, Q3, Q5 and Q7. All the above feature identifiers can be divided into two groups. The first group is A series models, which are sedans, including four models: A3, A4, A6 and A8. The second group is Q series models, which are SUVs, including four models: Q1, Q3, Q5 and Q7.

[0103] Exemplarily, the driving state of the vehicle includes low-speed driving and high-speed driving, and the vehicle operation data is the engine speed and temperature. The data division of the technical solution of the present invention is described by taking the first group as an example. The first data set corresponding to the first group is divided based on the feature identifier, and 4 second data sets corresponding to the models A3, A4, A6 and A8 can be obtained. Then, based on the driving state, the second data sets corresponding to the models A3, A4, A6 and A8 are divided into a third data set with a low-speed driving state and a third data set with a high-speed driving state. Finally, based on the type of vehicle operation data, each third data set is divided into a fourth data set with a speed of vehicle operation data and a fourth data set with a temperature of vehicle operation data.

[0104] The calculation method of the mapping coefficient of the rotation speed of the A3 model in the first group under low-speed driving state is taken as an example to illustrate the calculation of the mapping coefficient of the technical solution of the present invention. The first average value A31 of all data in the fourth data set corresponding to the rotation speed of the vehicle operation data of the A3 model when the driving state is low-speed driving is calculated, the first average value A41 of all data in the fourth data set corresponding to the rotation speed of the vehicle operation data of the A4 model when the driving state is low-speed driving is calculated, the first average value A61 of all data in the fourth data set corresponding to the rotation speed of the vehicle operation data of the A6 model when the driving state is low-speed driving is calculated, and the first average value A81 of all data in the fourth data set corresponding to the rotation speed of the vehicle operation data of the A8 model when the driving state is low-speed driving is calculated, and then the above four first average values ​​are averaged to obtain the second average value, and the ratio of the second average value to the first average value A31 is used as the mapping coefficient of the rotation speed of the A3 model under low-speed driving state.

[0105] Preferably, new data is continuously stored in the historical operation database, and the mapping coefficients can be updated periodically.

[0106] According to the technical solution of the present invention, different mapping coefficients are generated for vehicle operation data of different models, different driving states and different categories. When performing vehicle operation state detection, targeted correction can be performed on each vehicle operation data of each driving state of each model, thereby improving the accuracy of data processing.

[0107] Specifically, the data classification calculation module is also used to generate the second storage module:

[0108] Step 31: extract a second preset amount of second vehicle operation data from the historical operation database, wherein the second vehicle operation data includes vehicle operation data of normal operation status and abnormal operation status, and the second vehicle operation data includes vehicle operation data of all driving status of vehicles with all feature identifiers.

[0109] Step 32: According to the method of step 22, the second vehicle operation data is divided into N1 fifth data sets.

[0110] Step 33, extract any fifth data set, standardize any vehicle operation data based on the mapping coefficient corresponding to any vehicle operation data of any driving state of the vehicle corresponding to any feature identifier, and after traversing all the data in any fifth data set, obtain the sixth data set corresponding to any fifth data set.

[0111] Step 34: divide the sixth data set into N3 seventh data sets based on the driving state, so as to divide the vehicle operation data belonging to the same driving state into the same seventh data set.

[0112] Step 35, extract any seventh data set, use the data in any seventh data set as training data for model training, generate a preset model corresponding to any driving state in any fifth data set, and store the preset model, the specific feature identification group, and the specific driving state in the second storage module corresponding to each other, wherein the specific driving state is the driving state corresponding to any seventh data set, and the specific feature identification group is the feature identification group corresponding to any fifth data set.

[0113] Step 36: After traversing all fifth data sets, generate a second storage module.

[0114] The second preset number is set according to the experience of those skilled in the art or according to actual application scenarios, and is not limited in the embodiments of the present application.

[0115] Exemplarily, the above-mentioned preset model is a neural network model.

[0116] Exemplarily, there are two types of vehicles, A and Q, and two driving states, low-speed driving and high-speed driving. When generating a preset model for vehicle running state detection, for A series vehicles, model M1 is generated for the low-speed driving state, and model M2 is generated for the high-speed driving state. For Q series vehicles, model M3 is generated for the low-speed driving state, and model M4 is generated for the high-speed driving state.

[0117] When performing vehicle operation status detection, the operation data is first corrected based on the vehicle model, driving state and type of operation data, and then a preset model corresponding to the vehicle model and driving state of the vehicle to be analyzed is selected for operation status detection. The vehicle operation status detection according to the technical solution of the present invention is highly targeted, has high accuracy and high versatility, and can perform high-precision operation status detection on vehicles of different models.

[0118] Specifically, the process before step 12 includes:

[0119] Step 121: extract any first operation data and obtain identification information of any first operation data.

[0120] Step 122: extract the preset rule corresponding to any first operation data based on the identification information, and determine whether any first operation data satisfies the preset rule. If yes, determine that any first operation data meets the requirements; if not, proceed to step 123.

[0121] Step 123: Obtain historical operation data corresponding to any first operation data within a first preset time, and generate new first operation data based on the historical operation data.

[0122] Step 124: After traversing all the first operation data, proceed to step 12.

[0123] The first preset time is set according to the experience of those skilled in the art or according to actual application scenarios, and the embodiments of the present application are not limited to this.

[0124] Due to problems such as data acquisition equipment failure and network failure, the first operating data obtained may be missing, delayed, or mutated. In order to ensure the accuracy of data analysis and vehicle operating status detection, the first operating data is preprocessed to improve the quality of the operating data for data analysis.

[0125] Exemplarily, generating new first operating data based on historical operating data includes taking an average value of the historical operating data as the new first operating data, or generating new first operating data based on a change rule of the historical operating data.

[0126] Specifically, the preset rules include a value range and / or a collection period.

[0127] Step 4: Provide an interactive module to determine whether the vehicle has an abnormality based on the first operating state. When an abnormality occurs in the vehicle, obtain an event code corresponding to the vehicle abnormality and send the event code to the vehicle end.

[0128] When a vehicle abnormality is determined based on the vehicle's operating status, the event code corresponding to the abnormality is obtained and sent to the vehicle side. At the same time, the abnormal position is automatically located based on the vehicle's digital twin model, and control instructions are issued in a timely manner.

[0129] Preferably, the interaction module can prompt the user through voice or graphic interface.

[0130] Step 5: Provide an operation management module to support the operation of the digital twin model.

[0131] Specifically, the operation management module includes a historical data management module, which manages the first historical operation data, performs a second preprocessing on the first historical operation data and stores the data in a historical operation database, wherein the second preprocessing includes:

[0132] Step 51: extract the second historical operation data corresponding to any type of vehicle operation data from the first historical operation data, and evenly divide the second historical operation data into N6 third historical operation data, wherein N6 is a positive integer greater than 1, and the third historical operation data includes the collection time of the operation data and the operation data value, and the number of operation data in each third historical operation data is the same.

[0133] Step 52, extract any third historical operation data, generate a corresponding data block for any third historical operation data, and divide the data block into a header, a numerical part and a tail, wherein the header stores the basic information of any third historical operation data, the numerical part stores the historical operation data after transformation, and the tail stores the time processing flag and the numerical processing flag.

[0134] Step 53: Obtain the collection time of the first operation data in any third historical operation data, the storage quantity of operation data in any third historical operation data, and the type of operation data in any third historical operation data, and store the collection time, storage quantity and type as basic information in the header.

[0135] Step 54, obtain the h1th operation data in any third historical operation data, calculate the time difference between the h1th operation data and the h1-1th operation data, and replace the collection time of the h1th operation data with the time difference. After traversing the 2nd to Hth operation data, set the collection time of the 1st operation data to 0, where h1 is a positive integer from 2 to H, and H is the storage quantity.

[0136] Step 55: store the collection time and operation data value corresponding to the first operation data at the relative storage position of the first operation data in the numerical part, and set the time processing flag and the value processing flag corresponding to the first operation data to 0 and 0.

[0137] Step 56, extract the h2th operating data, and determine whether the collection time of the h2th operating data is equal to the collection time of the h2-1th operating data. If not, store the collection time of the h2th operating data at the relative storage position of the collection time of the h2th operating data in the numerical part, and set the time processing flag corresponding to the h2th operating data to 0, and then enter step 57. If so, delete the collection time of the h2th operating data, do not store the collection time of the h2th operating data in the numerical part, and set the time processing flag corresponding to the h2th operating data to 1, where h2 is a positive integer from 2 to H.

[0138] Step 57, determine whether the operating data value of the h2th operating data is equal to the operating data value of the h2-1th operating data. If not, store the operating data value of the h2th operating data at the relative storage position of the operating data value of the h2th operating data in the numerical portion, and set the numerical processing flag corresponding to the h2th operating data to 0. If so, delete the operating data value of the h2th operating data, do not store the operating data value of the h2th operating data in the numerical portion, and set the numerical processing flag corresponding to the h2th operating data to 1.

[0139] Step 58: Process all types of vehicle operation data in the first historical operation data according to the methods of steps 51 to 57.

[0140] A large amount of operation data is generated during the operation of the vehicle. If the operation data is directly stored, it will take up a lot of storage space, resulting in high storage pressure and high cost. Therefore, when storing the data, the operation data must be processed first and then stored.

[0141] The same type of operation data has the same collection time interval, which is convenient for compression. Therefore, the operation data is classified and stored according to the type of operation data. When storing data, the third historical operation data is stored as a whole, that is, a third historical operation data set is taken as a data block, and the number of operation data in the third historical operation data (i.e., the above-mentioned storage number) is set according to the experience of those skilled in the art or according to the actual application scenario, and the embodiment of the present application does not limit this.

[0142] Preferably, the quantity of the operating data in each set of the third historical operating data may be the same or different.

[0143] Exemplarily, the types of running data include integer type and floating point type, etc.

[0144] During vehicle operation, the same type of operation data has the characteristics of periodicity in collection time and invariance in operation data value. Therefore, when storing data, the repeated values ​​can be deleted by comparing the difference between any operation data and the previous operation data of any operation data, and the time or operation data value of deletion can be marked, based on which the data storage space can be reduced. And when processing, compression can be performed only based on the difference between any operation data and the previous operation data of any operation data, which can speed up the compression speed.

[0145] Although the operation data collection is periodic, the operation data is not collected when the vehicle stops running. In this case, the same third historical operation data may contain operation data with a time interval greater than the collection time interval. Therefore, when storing, it is necessary to first compare and judge the collection time and then compare and judge the operation value.

[0146] When new operation data is collected, the new operation data is first cached. When the amount of operation data in the cache reaches a preset amount, the operation data in the cache is subjected to a second preprocessing and then stored in the historical operation database.

[0147] As a preferred technical solution of the present invention, after the second preprocessing is completed, the operating data after the second preprocessing may also be subjected to a third preprocessing:

[0148] Extract the h3th operating data whose data processing flag is 0 in the data block, perform an XOR operation on the operating data value corresponding to the h3th operating data and the operating data value corresponding to the h3-1th operating data, store the operation result as the new operating data value of the h3th operating data at the relative storage position of the operating data value of the h3th operating data in the numerical part, and set the numerical processing flag corresponding to the h3th operating data to 2.

[0149] During vehicle operation, the same type of operating data has the characteristics of periodicity in collection time and invariance in operating data values. However, the operating data values ​​will also fluctuate to a certain extent, and there is continuity between the two previous and subsequent operating data values. According to the technical solution of the present invention, the motion data values ​​can be further effectively compressed to reduce data storage space.

[0150] Preferably, the operation management module also includes an engine operation management module and a file management module.

[0151] Figure 2 FIG. 1 is a schematic diagram of a structure of an embodiment of a digital twin engine construction system provided by the present invention. Figure 2 As shown, the system includes: a digital twin model construction module 10, a data acquisition module 20, an operation and maintenance detection module 30, an interaction module 40 and an operation management module 50.

[0152] The digital twin model construction module 10 is used to obtain the design data of the vehicle and construct a digital twin model of the vehicle.

[0153] The data acquisition module 20 is used to collect first operating data related to the vehicle and map the first operating data to the digital twin model through an ID identifier.

[0154] The operation and maintenance detection module 30 is used to perform a first preprocessing on the first operation data to obtain second operation data, analyze the second operation data, and obtain a first operation state of the vehicle.

[0155] The interaction module 40 is used to determine whether there is an abnormality in the vehicle according to the first operating state. When an abnormality occurs in the vehicle, an event code corresponding to the abnormality is obtained and the event code is sent to the vehicle end.

[0156] The operation management module 50 is used to support the operation of the digital twin model.

[0157] It should be understood that, although each step in the flow chart of each embodiment of the present invention is shown in sequence according to the indication of the arrow, these steps are not necessarily performed in sequence according to the order indicated by the arrow. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.

[0158] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the above-mentioned program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0159] The above-mentioned embodiments only express the preferred implementation mode of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A method for constructing a digital twin engine, characterized in that: The steps include: Provide a digital twin model building module to obtain vehicle design data and build a digital twin model of the vehicle; Providing a data acquisition module to collect first operating data related to the vehicle and map the first operating data to the digital twin model through an ID identifier; providing an operation and maintenance detection module, performing a first preprocessing on the first operation data to obtain second operation data, and analyzing the second operation data to obtain a first operation state of the vehicle; Provide an interactive module, determine whether the vehicle is abnormal based on the first operating state, and when the vehicle is abnormal, obtain an event code corresponding to the vehicle abnormality, and send the event code to the vehicle end; Provide an operation management module to support the operation of the digital twin model; The performing a first preprocessing on the first operating data to obtain second operating data, and analyzing the second operating data to obtain a first operating state of the vehicle comprises: Step 11: obtaining a first characteristic identifier of the vehicle, and identifying a driving state of the vehicle based on the first operating data; Step 12: extract any first operating data, search a first storage module based on the first feature identifier and the driving state, obtain a first mapping coefficient corresponding to any first operating data from the first storage module, and perform standardization processing on any first operating data using the first mapping coefficient; Step 13, repeating step 12, after traversing all the first operation data, obtaining the second operation data; Step 14: searching a second storage module based on the first feature identifier and the driving state, and acquiring a first preset model corresponding to the vehicle in the driving state from the second storage module; Step 15: input the second operation data into the first preset model to obtain the first operation state; A data classification calculation module is also provided, which is used to generate the first storage module, comprising the following steps: Step 21: extracting a first preset number of first vehicle operation data from a historical operation database, wherein the first vehicle operation data is vehicle operation data of a vehicle in a normal operation state, and the first vehicle operation data includes vehicle operation data of all driving states of vehicles with all feature identifiers; Step 22: based on the category of the feature identifier, all the feature identifiers are divided into N1 groups, and the first vehicle operation data is divided into N1 first data sets, and the vehicle operation data corresponding to the vehicles with feature identifiers belonging to the same group are divided into the same first data set, wherein N1 is a positive integer greater than or equal to 1; Step 23, extracting any first data set, dividing any first data set into N2 second data sets based on feature identifiers, and dividing vehicle operation data corresponding to vehicles belonging to the same feature identifier into the same second data set, wherein N2 is a positive integer greater than or equal to 1; Step 24, extracting any second data set, and dividing any second data set into N3 third data sets based on the driving state, so as to divide the vehicle operation data belonging to the same driving state into the same third data set, wherein N3 is a positive integer greater than 1; Step 25, extracting any third data set, and dividing any of the third data sets into N4 fourth data sets based on the type of vehicle operation data, so as to divide data belonging to the same type of vehicle operation data into the same fourth data set, wherein N4 is a positive integer greater than 1; Step 26, repeating Step 23 to Step 25, dividing all first data sets layer by layer, and respectively obtaining the second data set, the third data set and the fourth data set of each first data set; Step 27: extract the ith first data set, and calculate the mapping coefficient of the nth vehicle operation data of the mth driving state of the vehicle corresponding to the jth feature identifier in the ith first data set. C ijmn , and the mapping coefficients C ijmn The mapping coefficients are stored in the first storage module corresponding to the i-th first data set, the j-th feature identifier, the m-th driving state, and the n-th vehicle operation data. C ijmn The calculation formula is: ， ， ， in, AV ijmn is a first average of the nth type of vehicle operation data of the mth driving state of the vehicle corresponding to the jth feature identifier in the i-th first data set, is the second average of the nth type of vehicle operation data of the mth driving state in the ith first data set, Nu k is the kth value in the fourth data set corresponding to the nth type of vehicle operation data of the mth driving state of the vehicle corresponding to the jth characteristic identifier in the i-th first data set, i is a positive integer from 1 to N1, j is a positive integer from 1 to N2, m is a positive integer from 1 to N3, n is a positive integer from 1 to N4, k is a positive integer from 1 to N5, and N5 is the total number of data in the fourth data set corresponding to the nth type of vehicle operation data of the mth driving state of the vehicle corresponding to the jth characteristic identifier in the i-th first data set; Step 28, determine whether n is equal to N4, if not, set n=n+1, return to step 27, if yes, determine whether m is equal to N3, if not, set m=m+1, return to step 27, if yes, determine whether j is equal to N2, if not, set j=j+1, return to step 27, if yes, determine whether i is equal to N1, if not, set i=i+1, return to step 27, if yes, proceed to step 29; Step 29: After traversing all first data sets, generate the first storage module.

2. The method according to claim 1, characterized in that: The data classification calculation module is also used to generate the second storage module: Step 31: extracting a second preset amount of second vehicle operation data from the historical operation database, wherein the second vehicle operation data includes vehicle operation data of a normal operation state and an abnormal operation state, and the second vehicle operation data includes vehicle operation data of all driving states of vehicles with all feature identifiers; Step 32: According to the method of step 22, the second vehicle operation data is divided into N1 fifth data sets; Step 33: extract any fifth data set, perform standardization processing on any vehicle operation data of any driving state of the vehicle corresponding to any feature identifier based on the mapping coefficient corresponding to any vehicle operation data of any driving state of the vehicle, and obtain a sixth data set corresponding to any fifth data set after traversing all the data in any fifth data set; Step 34: dividing the sixth data set into N3 seventh data sets based on the driving state, so as to divide the vehicle operation data belonging to the same driving state into the same seventh data set; Step 35: extract any seventh data set, use the data in any of the seventh data sets as training data to perform model training, generate a preset model corresponding to any driving state in any of the fifth data sets, and store the preset model, a specific feature identification group, and a specific driving state in the second storage module in correspondence, wherein the specific driving state is the driving state corresponding to any of the seventh data sets, and the specific feature identification group is the feature identification group corresponding to any of the fifth data sets; Step 36: After traversing all fifth data sets, generate the second storage module.

3. The method according to claim 1, characterized in that The step 12 includes: Step 121: extract any of the first operating data and obtain identification information of any of the first operating data; Step 122: extracting a preset rule corresponding to any of the first operating data based on the identification information, and determining whether any of the first operating data satisfies the preset rule. If so, determining that any of the first operating data meets the requirement; if not, proceeding to step 123; Step 123: Acquire historical operation data corresponding to any first operation data within a first preset time, and generate new first operation data based on the historical operation data; Step 124: After traversing all the first operation data, proceed to step 12.

4. The method according to claim 3, characterized in that The preset rules include a value range and / or a collection period.

5. The method according to claim 1, characterized in that The operation management module includes a historical data management module, the historical data management module manages first historical operation data, performs second preprocessing on the first historical operation data and stores the data in a historical operation database, wherein the second preprocessing includes: Step 51: extracting second historical operation data corresponding to any type of vehicle operation data from the first historical operation data, and evenly dividing the second historical operation data into N6 third historical operation data, wherein N6 is a positive integer greater than 1, and the third historical operation data includes the collection time of the operation data and the operation data value, and the number of operation data in each third historical operation data is the same; Step 52: extract any third historical operation data, generate a corresponding data block for any third historical operation data, and divide the data block into a header, a value portion, and a tail portion, wherein the header stores basic information of any third historical operation data, the value portion stores the historical operation data after transformation, and the tail portion stores a time processing flag and a value processing flag; Step 53: Obtain the collection time of the first piece of operation data in any of the third historical operation data, the storage quantity of the operation data in any of the third historical operation data, and the type of the operation data in any of the third historical operation data, and store the collection time, storage quantity, and type as the basic information in the header; Step 54: Obtain the h1th operation data in any of the third historical operation data, calculate the time difference between the h1th operation data and the h1-1th operation data, and replace the collection time of the h1th operation data with the time difference. After traversing the 2nd to Hth operation data, set the collection time of the 1st operation data to 0, where h1 is a positive integer from 2 to H, and H is the storage quantity; Step 55, storing the collection time and the operation data value corresponding to the first operation data at the relative storage position of the operation data of the first item in the numerical part, and setting the time processing flag and the value processing flag corresponding to the first operation data to 0; Step 56, extract the h2th running data, determine whether the collection time of the h2th running data is equal to the collection time of the h2-1th running data, if not, store the collection time of the h2th running data at the relative storage position of the h2th running data collection time in the numerical part, and set the time processing flag corresponding to the h2th running data to 0, then enter step 57, if yes, delete the h2th running data collection time, do not store the h2th running data collection time in the numerical part, and set the h2th running data collection time processing flag to 1, where h2 is a positive integer from 2 to H; Step 57, determine whether the operating data value of the operating data described in Article h2 is equal to the operating data value of the operating data described in Article h2-1. If not, store the operating data value of the operating data described in Article h2 at the relative storage position of the operating data value of the operating data described in Article h2 in the numerical portion, and set the numerical value processing flag corresponding to the operating data described in Article h2 to 0. If so, delete the operating data value of the operating data described in Article h2, do not store the operating data value of the operating data described in Article h2 in the numerical portion, and set the numerical value processing flag corresponding to the operating data described in Article h2 to 1; Step 58: Process all types of vehicle operation data in the first historical operation data according to the method of steps 51 to 57.

6. A digital twin engine construction system, used to implement the method according to any one of claims 1 to 5, characterized in that: include: Digital twin model construction module, data acquisition module, operation and maintenance detection module, interaction module and operation management module; The digital twin model building module is used to obtain the design data of the vehicle and build a digital twin model of the vehicle; The data acquisition module is used to collect first operating data related to the vehicle, and map the first operating data to the digital twin model through an ID identifier; The operation and maintenance detection module is used to perform a first preprocessing on the first operation data to obtain second operation data, analyze the second operation data, and obtain a first operation state of the vehicle; The interaction module is used to determine whether the vehicle has an abnormality according to the first operating state, and when the vehicle has an abnormality, obtain an event code corresponding to the vehicle abnormality, and send the event code to the vehicle end; The operation management module is used to support the operation of the digital twin model.

Images