Intelligent connected vehicle inverter control system

Abstract

The present invention discloses an intelligent networked automobile inverter control system, which relates to the field of inverter control technology. The control center includes a control center, wherein the control center is communicatively connected to a data acquisition module, a data processing module, an inverter control module and a network analysis module; the data acquisition module is used to generate a data unit packet and obtain the thread speed corresponding to the thread pool and then establish a corresponding transmission channel; the data processing module cleans the data unit packet, filters the data and generates a load data set, divides the load data set into normal data and abnormal data after load calculation, obtains normal data through the inverter control module for data analysis, and switches the working mode of the inverter according to the result of the data analysis; the network analysis module is used to obtain abnormal data, and after matching with a large data set stored in a cloud server, performs inverter life prediction, thereby realizing inverter control of the automobile.

Description

Technical Field

[0001] The present invention relates to the field of inverter control technology, and in particular to an inverter control system for an intelligent connected vehicle. Background Art

[0002] An inverter is an electronic device used to convert direct current into alternating current and supply it to the electric motor of an electric vehicle for driving. Inverter control can be achieved through remote monitoring, data analysis, and optimized control of the inverter, thereby improving the performance and efficiency of the vehicle.

[0003] How to detect the working status of the inverter in real time while the car is driving, collect the parameters generated by the inverter operation, adjust the inverter according to the vehicle driving related parameters to ensure safe driving of the vehicle, and analyze the parameters generated by the inverter operation to predict the life span are all issues we need to consider. Summary of the Invention

[0004] In order to solve the above problems, the purpose of the present invention is to provide an inverter control system for an intelligent connected vehicle.

[0005] The object of the present invention can be achieved by the following technical solution: an intelligent networked vehicle inverter control system includes a control center, wherein the control center is communicatively connected to a data acquisition module, a data processing module, an inverter control module, and a network analysis module;

[0006] The data acquisition module is used to collect inverter parameter information, vehicle driving information, and vehicle energy consumption information, and encapsulate them into different data unit packets and transmit them to the thread pool set by the data acquisition module. After obtaining the thread speed corresponding to the thread pool, the data unit packets are uploaded to the data processing module by establishing a corresponding transmission channel;

[0007] The data processing module includes a pre-processing unit, a load regulation unit, and an abnormality determination unit. The pre-processing unit is used to clean and filter the data unit packet and generate a load data set; the load regulation unit is used to obtain the load data set and perform load calculation; the abnormality determination unit is used to determine whether the load data set after load calculation is normal data or abnormal data.

[0008] The inverter control module is used to obtain normal data for data analysis and switch the working mode of the inverter included in the inverter control module according to the results of the data analysis;

[0009] The networked analysis module is used to obtain abnormal data and upload it to the control center. The control center sends an analysis application instruction to the cloud server to obtain the large data set stored in the cloud server. The abnormal data is matched with the large data set to perform life prediction.

[0010] Furthermore, the data unit packet encapsulation process includes:

[0011] Different encapsulation sequence codes are assigned to the inverter parameter information, vehicle driving information and vehicle energy consumption information respectively, and different data unit packages are encapsulated according to the encapsulation sequence codes. The encapsulation sequence codes include S code, P code and Q code, and the data unit packages are provided with B1 package, B2 package and B3 package; S code corresponds to B1 package, P code corresponds to B2 package, and Q code corresponds to B3 package; a thread pool is provided, and the encapsulated B1 package, B2 package and B3 package are summarized and input into the thread pool.

[0012] Furthermore, the process of establishing the transmission channel includes:

[0013] The thread pool includes different thread speeds, and the thread speed V is obtained. 线 , V 线 Set the corresponding opening threshold V 阈 , when V 线 ≥V 阈 The transmission channel is allowed to be established only when V 线 <V 阈 When transmitting the data unit packet to the data processing module, wireless transmission is adopted; the transmission channel includes a high-speed channel, a medium-speed channel and a low-speed channel; the thread speed is interval-defined to generate different speed intervals, and corresponding transmission channels are established according to the speed intervals to transmit the data unit packet.

[0014] Furthermore, the process of generating the load data set includes:

[0015] Clean the data unit package, deconstruct the data unit package into several fields according to the data type, and perform binary arrangement on the fields to obtain the row coordinates X and column coordinates Y corresponding to the fields, and generate the field positioning coordinates L=<X,Y> , summarize several positioning coordinates to generate a data matrix, the data matrix consists of 0 and 1, and traverse the rows and columns of the data matrix in turn. If the traversed value is 0, it means that the data here is abnormal. If the traversed value is 1, it means that the data is normal. Data abnormalities include data duplication, data missing, and data format abnormalities; if the data is repeated, the repeated traversed values ​​are removed. If the data is missing, the missing rows and columns of the data matrix are located and the fields are filled. If the data format is abnormal, it is converted into a unified field format; obtain the data unit package after data cleaning, the data unit package corresponds to different data loads, set a load limit value, and associate different load identifiers with the data unit package according to the data load and the load limit value. Summarize several data unit packages and their corresponding load identifiers to generate a load data set.

[0016] Furthermore, the process of performing load calculation according to the load data set includes:

[0017] After obtaining the load data set, the different load identifiers included in the load data set are deconstructed to obtain the load identifiers, which include SN1, SN2 and SN3. If the obtained load identifier is SN1, the data unit package in the load data set is directly calculated locally. If the obtained load identifier is SN2, the data unit package is split into the set calculable numerical size and then calculated locally. If the obtained load identifier is SN3, the data unit package is uploaded to the cloud for calculation.

[0018] Furthermore, the process of dividing normal data and abnormal data includes:

[0019] The abnormality determination unit is provided with a standard database, which stores standard values, and obtains the standard values ​​to establish a data characteristic curve; the load data set is backed up to generate a backup data set, and a number of characteristic image points are generated according to the load data set and the backup data set respectively, and the average values ​​of the characteristic image points corresponding to the load data set and the backup data set are taken in turn to obtain the coordinates of a number of fitting points, and the number of fitting points are connected to generate a fitting curve, and the fitting curve is compared with the data characteristic curve, and the part corresponding to the overlapping curve is normal data, and the other part is abnormal data.

[0020] Furthermore, the process of obtaining normal data for data analysis and switching the inverter working mode includes:

[0021] The working modes of the inverter include independent power generation mode, emergency processing mode and energy storage mode. When the inverter temperature, inverter voltage, inverter current and inverter power in the normal data are respectively within the set standard temperature range, standard voltage range, standard current range and standard power range, a switching coefficient is generated; the initial working mode of the inverter is idle mode, and switching from idle mode to other modes is performed according to the numerical range of the switching coefficient.

[0022] Furthermore, the process of performing life prediction by the network analysis module includes:

[0023] After obtaining the abnormal data, an encrypted interactive space is synchronously established. The encrypted interactive space performs symmetric encryption, asymmetric encryption, and digital signature encryption on different parts of the abnormal data. Access to the encrypted interactive space requires identity authentication to determine whether the accessing IP is in the IP whitelist. If so, access is allowed; otherwise, access is prohibited. The abnormal data in the encrypted interactive space is transmitted to the control center, and the control center sends an analysis application instruction to the cloud server. After the cloud server obtains the analysis application instruction, it reads the large data set stored in the cloud server; the abnormal data is traversed and matched with the large data set, which includes different abnormal sub-data, and the abnormal data and several abnormal sub-data are converted into binary files.

[0024] The binary file is composed of several binary strings. Two binary strings are traversed and compared in the same order to obtain the file similarity λ. If λ is greater than or equal to the set similarity judgment value λ`, the part life form corresponding to the abnormal sub-data is obtained. If λ is less than λ`, the part life form is not obtained.

[0025] Compared with the prior art, the beneficial effects of the present invention are: after generating a data unit packet through the data acquisition module and obtaining the thread speed corresponding to the thread pool, a corresponding transmission channel is established, and the thread speed obtained by the data unit packet is based on the packet capacity of the data unit packet itself. The larger the packet capacity, the higher the thread speed in the corresponding matched thread pool, so that the data unit packets with large packet capacity and those with small packet capacity will not have a large difference in transmission speed, thereby ensuring the capacity synchronization of data acquisition; the data unit packet is cleaned, the data is filtered and a load data set is generated, and the load calculation is based on the size of the data load, and the load data set is subjected to different choices of local calculation or cloud calculation. The cloud can process data with large data load faster, thereby optimizing the computing efficiency; normal data is obtained through the inverter control module for data analysis, and the working mode of the inverter is switched according to the results of the data analysis, thereby realizing efficient driving of the car and improving the performance of the car, and by connecting to the cloud server, data similar to the current inverter abnormal data is obtained, and the corresponding parts life form is obtained, thereby realizing the life prediction of the inverter. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 This is a schematic diagram of the present invention. DETAILED DESCRIPTION

[0027] like Figure 1 As shown, the intelligent connected vehicle inverter control system includes a control center, which is communicatively connected to a data acquisition module, a data processing module, an inverter control module, and a network analysis module;

[0028] The data acquisition module is used to collect inverter parameter information, vehicle driving information, and vehicle energy consumption information, and encapsulate them into different data unit packets and transmit them to the thread pool set by the data acquisition module. After obtaining the thread speed corresponding to the thread pool, the data unit packets are uploaded to the data processing module by establishing a corresponding transmission channel;

[0029] The data processing module includes a pre-processing unit, a load regulation unit, and an abnormality determination unit. The pre-processing unit is used to clean and filter the data unit packet and generate a load data set; the load regulation unit is used to obtain the load data set and perform load calculation; the abnormality determination unit is used to determine whether the load data set after load calculation is normal data or abnormal data.

[0030] The inverter control module is used to obtain normal data for data analysis and switch the working mode of the inverter included in the inverter control module according to the results of the data analysis;

[0031] The networked analysis module is used to obtain abnormal data and upload it to the control center. The control center sends an analysis application instruction to the cloud server to obtain the large data set stored in the cloud server. The abnormal data is matched with the large data set to perform life prediction.

[0032] Specifically, the process of collecting inverter parameter information, vehicle driving information, and vehicle energy consumption information and encapsulating them into data unit packets includes:

[0033] The inverter parameter information includes inverter temperature, inverter voltage, inverter current and inverter power; the vehicle driving information includes driving speed, driving distance and driving power; the vehicle energy consumption information includes energy consumption level, fuel consumption per 100 kilometers and engine power consumption;

[0034] Different encapsulation sequence codes are assigned to the inverter parameter information, the vehicle driving information, and the vehicle energy consumption information, respectively, and different data unit packets are encapsulated according to the encapsulation sequence codes. The encapsulation sequence codes include an S code, a P code, and a Q code. The data unit packets are provided with a B1 packet, a B2 packet, and a B3 packet.

[0035] When the obtained encapsulation sequence code is S code, the inverter parameter information is encapsulated into a B1 packet in the data unit packet;

[0036] When the obtained encapsulation sequence code is a P code, the vehicle driving information is encapsulated as a B2 package in the data unit package;

[0037] When the obtained encapsulation sequence code is a Q code, the vehicle energy consumption information is encapsulated as a B3 packet in the data unit packet;

[0038] The data acquisition module is also provided with a thread pool, which collects and encapsulates the B1 package, B2 package and B3 package and inputs them into the thread pool. The thread pool includes different thread speeds, and the data unit package establishes different transmission channels after obtaining the corresponding thread speed;

[0039] It should be noted that the speed of the thread that obtains the data unit packet is based on the packet capacity of the data unit packet itself. The larger the packet capacity, the higher the thread speed in the thread pool that is matched to it. This ensures that there will not be a huge difference in transmission speed between data unit packets with large and small packet capacities.

[0040] Specifically, the process of establishing the transmission channel includes:

[0041] Get the thread speed, marked as V 线 , V 线 Set the corresponding opening threshold, marked as V 阈 , when V 线 ≥V 阈 The transmission channel is allowed to be established only when V 线 <V 阈 When the data unit package is transmitted to the data processing module by wireless transmission;

[0042] The transmission channels include a high-speed channel, a medium-speed channel and a low-speed channel;

[0043] Defining thread speeds by intervals, generating different speed intervals after the intervals are defined, the speed intervals including a first speed interval, a second speed interval, and a third speed interval;

[0044] The corresponding relationship between the speed range and the transmission channel is as follows:

[0045] If the first speed range is obtained, a high-speed channel is established accordingly;

[0046] If the second speed range is obtained, a medium-speed channel is established accordingly;

[0047] If the third speed range is obtained, a low-speed channel is established accordingly;

[0048] The data unit packet is transmitted to the data processing module through the transmission channel for further processing;

[0049] The data processing module includes a pre-processing unit, a load adjustment unit and an abnormality determination unit;

[0050] The pre-processing unit is used to perform data cleaning and data filtering on the data unit packet and generate a load data set;

[0051] The load adjustment unit obtains a load data set to perform load calculation;

[0052] The abnormality determination unit is used to determine whether the load data set after load calculation is normal data or abnormal data;

[0053] Specifically, the process of generating a load data set through data cleaning and data filtering includes:

[0054] Obtain the data unit package collected by the data acquisition module, perform data cleaning on the data unit package, first deconstruct the data unit package into several fields according to the data type, and then perform binary arrangement on the fields, that is, arrange the fields in square rows and columns, and the arrangement order is from left to right and from top to bottom;

[0055] Get the row coordinates corresponding to several fields, marked as X, and the column coordinates, marked as Y, and then get the positioning coordinates of the field, marked as L, with L =<X,Y> , the values ​​of the binary arrangement include 0 and 1, that is, the value of the position corresponding to the positioning coordinate is 0 or 1;

[0056] Aggregate several positioning coordinates to generate a data matrix. The data matrix consists of 0s and 1s. Traverse each row and column of the data matrix in turn. If the traversed value is 0, it means that the data here is abnormal. If the traversed value is 1, it means that the data is normal. Abnormal data includes data duplication, data missing, and data format abnormality.

[0057] If the data is repeated, the number of traversed values ​​is recorded as S, and S-1 repeated traversed values ​​are eliminated. If the data is missing, locate the missing rows and columns of the data matrix and fill in the fields. If the data format is abnormal, convert it to a unified field format.

[0058] Obtaining a data unit packet after data cleaning, wherein the data unit packet corresponds to different data loads, denoted as Load, and setting load limit values ​​D1, D2, D3, and D4, such that D1 < D2 < D3 < D4;

[0059] When D1≤Load≤D2, the data unit packet corresponding to the data load is associated with the load identifier SN1;

[0060] When D2<Load≤D3, the data unit packet corresponding to the data load is associated with the load identifier SN2;

[0061] When D3<Load≤D4, the data unit packet corresponding to the data load is associated with a load identifier SN3;

[0062] Aggregating a number of data unit packets and their corresponding load identifiers to generate a load data set, and transmitting the load data set to a load regulation unit;

[0063] Specifically, the process of performing load calculation according to the load data set includes:

[0064] After obtaining the load data set, deconstructing it to obtain the different load identifiers included therein;

[0065] If the load identifier obtained is SN1, the data unit package in the load data set is directly calculated locally. If the load identifier obtained is SN2, the data unit package is split into the set calculable values ​​and then locally calculated and aggregated. If the load identifier obtained is SN3, the data unit package is uploaded to the cloud for calculation.

[0066] The calculation carrier of the local calculation is the calculation program set in the load adjustment unit, and the carrier of the cloud load calculation is the cloud computer set in the cloud;

[0067] Specifically, the process of dividing normal data and abnormal data includes:

[0068] The abnormality determination unit is provided with a standard database, which stores standard values ​​of inverter parameter information, vehicle driving information, and vehicle energy consumption information. The standard values ​​include data type and data quantity. The data type is used as a horizontal parameter and the data quantity is used as a vertical parameter. The horizontal parameter and the vertical parameter are obtained, and a data characteristic curve is established in the abnormality determination unit through a set characteristic curve generation program;

[0069] Obtain a load data set after load calculation, back up the load data set to generate a backup data set, and use the load data set as an input parameter to input into a characteristic curve generation program to generate a number of characteristic image points, numbered i, i = 1, 2, 3, ..., n, where n is a natural integer;

[0070] The backup data set is also used as an input parameter, and a number of characteristic image points are generated by the characteristic curve generation program, numbered j, j = 1, 2, 3, ..., m, where m is a natural integer;

[0071] Take the average value of each characteristic image point at the corresponding position i and j in turn to obtain the coordinates of several fitting points, recorded as k, and use a smooth curve to connect several fitting points in turn to generate a fitting curve. Compare the fitting curve with the data characteristic curve. The part corresponding to the overlapping curve is normal data, and the other part is abnormal data;

[0072] Assign "Error" to abnormal data and "Normal" to normal data, transmit abnormal data to the network analysis module, and transmit normal data to the inverter control module;

[0073] Specifically, the process of obtaining normal data for data analysis and switching the inverter working mode includes:

[0074] After obtaining normal data, the working mode of the inverter is switched according to the normal data;

[0075] The working modes of the inverter include independent power generation mode, emergency processing mode and energy storage mode;

[0076] When the inverter temperature, inverter voltage, inverter current and inverter power in the normal data are respectively within the set standard temperature range, standard voltage range, standard current range and standard power range, a switching coefficient is generated, which is recorded as α;

[0077] The values ​​of the inverter temperature, the inverter voltage, the inverter current, and the inverter power are respectively recorded as A1, A2, A3, and A4, and the weight values ​​of the inverter temperature, the inverter voltage, the inverter current, and the inverter power are respectively recorded as β1, β2, β3, and β4;

[0078] There is α=A1β1+A2β2+A3β3+A4β4;

[0079] The initial operating mode of the inverter is an idle mode, i.e., no mode is selected;

[0080] When α∈[0, 0.3), the inverter operation mode is switched from idle mode to independent power generation mode;

[0081] When α∈[0.3, 0.6), the working mode of the inverter is switched from idle mode to emergency processing mode;

[0082] When α∈[0.6, 0.95], the inverter operation mode is switched from idle mode to energy storage mode;

[0083] It should be noted that when the inverter operates in independent power generation mode, it is not connected to the power distribution network in the car. It mainly generates electricity through renewable energy such as solar cell groups, converting DC energy into AC energy to supply power to the electric vehicle's motor; when the inverter operates in grid-connected energy storage mode, it can not only convert solar power into AC energy and inject it into the power distribution network, but also store excess electricity in the battery or other energy storage devices included in the electric vehicle for subsequent use. This mode can realize the functions of self-generation, selling electricity to the grid, and using energy storage devices for power supply when needed; the emergency processing mode is to generate a fault file and upload it to the control center when a fault in the inverter power supply is detected, and the control center will perform real-time troubleshooting;

[0084] Specifically, the process of performing life prediction by the network analysis module includes:

[0085] After the network analysis module obtains the abnormal data, it synchronously establishes an encrypted interactive space. After the encrypted interactive space obtains the abnormal data, it first divides it into three parts: data header, data body and data tail;

[0086] The data header, data body, and data tail are encrypted using symmetric encryption, asymmetric encryption, and digital signature encryption respectively. Identity authentication is required to access the encrypted interactive space. The identity authentication rule is to determine whether the accessing IP is in the IP whitelist. If so, access is allowed; otherwise, access is prohibited.

[0087] Abnormal data in the encrypted interactive space is transmitted to the control center, which sends an analysis request instruction to the cloud server. After the cloud server obtains the analysis request instruction, it reads the large data set stored in the cloud server;

[0088] Traversing and matching the abnormal data with a large data set, the large data set including different abnormal sub-data, converting the abnormal data into a binary file, and also converting a plurality of abnormal sub-data included in the large data set into binary files;

[0089] The binary file is composed of a plurality of binary strings. Two binary strings are traversed and compared in parity order to obtain a file similarity, which is recorded as λ. If λ is greater than or equal to a set similarity determination value λ`, the part life form associated with the abnormal sub-data is obtained and assigned to the inverter of the electric vehicle corresponding to the abnormal data. If λ is less than λ`, the part life form is not obtained.

[0090] It should be noted that the parts life form records information related to the inverter parts life when abnormal data is generated by different types of electric vehicles. By comparing and matching the abnormal data with high similarity with the abnormal sub-data of the cloud server, the possible working life of the inverter corresponding to the abnormal data is obtained.

[0091] The above embodiments are only used to illustrate the technical method of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. An intelligent connected vehicle inverter control system, including a control center, is characterized in that: The control center is communicatively connected to a data acquisition module, a data processing module, an inverter control module, and a network analysis module; The data acquisition module is used to collect inverter parameter information, vehicle driving information, and vehicle energy consumption information, and encapsulate them into different data unit packets and transmit them to the thread pool set by the data acquisition module. After obtaining the thread speed corresponding to the thread pool, the data unit packets are uploaded to the data processing module by establishing a corresponding transmission channel; The data processing module includes a preprocessing unit, a load regulation unit and an abnormality determination unit. The preprocessing unit is used to perform data cleaning and data filtering on the data unit package and generate a load data set; the load regulation unit is used to obtain the load data set and perform load calculation; the abnormality determination unit is used to determine whether the load data set after the load calculation is normal data and abnormal data. The abnormality determination unit is provided with a standard database, which stores standard values, and obtains the standard values ​​to establish a data characteristic curve; the load data set is backed up to generate a backup data set, and a number of characteristic image points are generated according to the load data set and the backup data set respectively. The average values ​​of the characteristic image points corresponding to the load data set and the backup data set are taken in turn to obtain the coordinates of a number of fitting points, and the number of fitting points are connected to generate a fitting curve. The fitting curve is compared with the data characteristic curve, and the part corresponding to the overlapping curve is normal data, and the other part is abnormal data; The inverter control module is used to obtain normal data for data analysis and switch the working mode of the inverter included in the inverter control module according to the results of the data analysis; The network analysis module is used to obtain abnormal data and upload it to the control center. The control center sends an analysis application instruction to the cloud server to obtain the large data set stored in the cloud server. The abnormal data is matched with the large data set to perform life prediction. After obtaining the abnormal data, an encrypted interaction space is synchronously established. The encrypted interaction space performs symmetric encryption, asymmetric encryption and digital signature encryption on different parts of the abnormal data. Access to the encrypted interaction space requires identity authentication to determine whether the accessing IP is in the IP whitelist. If so, access is allowed; otherwise, access is prohibited. Abnormal data in the encrypted interactive space is transmitted to the control center, which sends an analysis request instruction to the cloud server. After the cloud server obtains the analysis request instruction, it reads the large data set stored in the cloud server; traverses and matches the abnormal data with the large data set, which includes different abnormal sub-data, and converts the abnormal data and the multiple abnormal sub-data into a binary file; The binary file is composed of several binary strings. Two binary strings are traversed and compared in the same order to obtain the file similarity λ. If λ is greater than or equal to the set similarity judgment value λ`, the part life form corresponding to the abnormal sub-data is obtained. If λ is less than λ`, the part life form is not obtained.

2. The intelligent connected vehicle inverter control system according to claim 1, characterized in that: The data unit encapsulation process includes: Different encapsulation sequence codes are assigned to the inverter parameter information, vehicle driving information and vehicle energy consumption information respectively, and different data unit packages are encapsulated according to the encapsulation sequence codes. The encapsulation sequence codes include S code, P code and Q code, and the data unit packages are provided with B1 package, B2 package and B3 package; S code corresponds to B1 package, P code corresponds to B2 package, and Q code corresponds to B3 package; a thread pool is provided, and the encapsulated B1 package, B2 package and B3 package are summarized and input into the thread pool.

3. The intelligent connected vehicle inverter control system according to claim 2, characterized in that: The process of establishing the transmission channel includes: The thread pool includes different thread speeds, and the thread speed V is obtained. 线 , V 线 Set the corresponding opening threshold V 阈 , when V 线 ≥V 阈 The transmission channel is allowed to be established only when V 线 <V 阈 When transmitting the data unit packet to the data processing module, wireless transmission is adopted; the transmission channel includes a high-speed channel, a medium-speed channel and a low-speed channel; the thread speed is interval-defined to generate different speed intervals, and corresponding transmission channels are established according to the speed intervals to transmit the data unit packet.

4. The intelligent connected vehicle inverter control system according to claim 3, characterized in that: The process of generating the load data set includes: Clean the data unit package, deconstruct the data unit package into several fields according to the data type, and perform binary arrangement on the fields to obtain the row coordinates X and column coordinates Y corresponding to the fields, and generate the field positioning coordinates L=<X,Y> , summarize several positioning coordinates to generate a data matrix, the data matrix consists of 0 and 1, and traverse the rows and columns of the data matrix in turn. If the traversed value is 0, it means that the data here is abnormal. If the traversed value is 1, it means that the data is normal. Data abnormalities include data duplication, data missing, and data format abnormalities; if the data is repeated, the repeated traversed values ​​are removed. If the data is missing, the missing rows and columns of the data matrix are located and the fields are filled. If the data format is abnormal, the data format of the field is converted to a unified field format; obtain the data unit package after data cleaning, the data unit package corresponds to different data loads, set a load limit value, and associate different load identifiers with the data unit package according to the data load and the load limit value. Summarize several data unit packages and their corresponding load identifiers to generate a load data set.

5. The intelligent connected vehicle inverter control system according to claim 4, characterized in that: The process of performing load calculation based on the load data set includes: After obtaining the load data set, the different load identifiers included in the load data set are deconstructed to obtain the load identifiers, which include SN1, SN2 and SN3. If the obtained load identifier is SN1, the data unit package in the load data set is directly calculated locally. If the obtained load identifier is SN2, the data unit package is split into the set calculable numerical size and then calculated locally. If the obtained load identifier is SN3, the data unit package is uploaded to the cloud for calculation.

6. The intelligent connected vehicle inverter control system according to claim 5, characterized in that: The process of obtaining normal data for data analysis and switching the inverter's operating mode includes: The working modes of the inverter include independent power generation mode, emergency processing mode and energy storage mode. When the inverter temperature, inverter voltage, inverter current and inverter power in the normal data are respectively within the set standard temperature range, standard voltage range, standard current range and standard power range, a switching coefficient is generated; the initial working mode of the inverter is idle mode, and switching from idle mode to other modes is performed according to the numerical range of the switching coefficient.

Images