WO2022007330A1

WO2022007330A1 - Battery electric vehicle state monitoring method and system

Info

Publication number: WO2022007330A1
Application number: PCT/CN2020/136384
Authority: WO
Inventors: 程涛; 刘远鹏
Original assignee: 深圳技术大学
Priority date: 2020-07-08
Filing date: 2020-12-15
Publication date: 2022-01-13
Also published as: CN111775711B; CN111775711A

Abstract

A battery electric vehicle state monitoring method and system. The monitoring method comprises the following steps: if it is detected that a vehicle is in a running state, acquiring first sensing information, the first sensing information comprising the type of sensing data and the total volume of the sensing data; pre-estimating a computational amount of the first sensing information according to the type of the sensing data and the total volume of the sensing data; comparing the computational amount of the first sensing information with a preset value; and if the computational amount of the first sensing information is less than the preset value, monitoring the state of the vehicle according to the first sensing information, wherein the first sensing information is from a distributed sensor. The method directly processes an operation part having a small computational amount in a data generation area, and directly calls an edge computing end located at an edge of the Internet of Vehicles to participate in computing, thus greatly improving a computing rate of a system.

Description

Method and system for state monitoring of pure electric vehicle

technical field

The invention relates to the technical field of Internet of Vehicles, in particular to a method and system for monitoring the state of a vehicle.

Background technique

As a key national development industry, pure electric vehicles are the main trend of future automobile development. In recent years, with the introduction of relevant national policies and guidance, the development of pure electric vehicles has received widespread attention. In order to judge the reliability of pure electric vehicles, the monitoring and evaluation of various state data information of vehicles has become one of the most important links. Moreover, with the continuous advancement of technology and the changing needs, the requirements for vehicle information monitoring continue to improve and progress; the data collected by the vehicle cannot be quickly interacted and processed, resulting in a decline in system performance, which is a common occurrence in traditional pure electric vehicle condition monitoring. The problem.

technical problem

The technical problem to be solved by the present invention is to provide a pure electric vehicle state monitoring method and system, which are used to speed up the calculation speed of data monitoring.

technical solutions

In order to solve the above technical problems, the technical solution adopted in the present invention is: a method for monitoring the state of a pure electric vehicle, comprising the following steps: detecting that the vehicle is in a running state, and acquiring first sensing information, where the first sensing information includes the type of sensing data and the total amount of sensing data; estimate the calculation amount of the first sensing information according to the type of sensing data and the total amount of sensing data; compare the calculation amount of the first sensing information with the preset value; when the first sensing information is calculated When the amount is less than the preset value, the state of the vehicle is monitored according to the first sensing information; wherein the first sensing information comes from distributed sensors.

Wherein, the step of monitoring the state of the vehicle according to the first sensing information specifically includes:

obtain the confidence level of the sensor;

Weighting the first sensing information according to the confidence of the sensor to generate first monitoring information;

Wherein, the first monitoring information includes the vehicle state information.

Optionally, when the calculated amount of the first sensing information is greater than a preset value, the method includes the following steps:

Preprocessing the first sensing information to generate preprocessing information;

sending the preprocessing information to the computing cloud to obtain feedback information from the computing cloud;

According to the feedback information of the computing cloud, the state of the vehicle is detected.

Further, after the step of sending the first sensing information to the computing cloud, it further includes the following steps:

performing neural network training on the first sensing information to generate optimal weights for the sensing information;

performing multi-physical domain fusion on the first sensing information to generate multi-physical domain feedback information;

Recombining the first sensing information according to the optimal weight of the sensing information and the feedback information from multiple physical domains to generate the second monitoring information;

Wherein, the first sensing information includes information of multiple physical domains.

Further, after the step of generating multi-physical domain feedback information, it also includes:

Record the historical computing process of the edge computing terminal and obtain edge computing information;

Record the historical computing process of the computing cloud to obtain cloud computing information;

obtaining computing monitoring information according to the edge computing information and the cloud computing information;

Wherein, each of the computing clouds corresponds to at least one of the edge computing terminals;

The calculated monitoring information includes the vehicle state information.

Specifically, the step of preprocessing the first sensing information to generate preprocessing information includes:

Preprocessing the image information and sensor data, and sending the preprocessed information to the computing cloud;

Wherein, the sensing data includes vehicle body running information for monitoring the running state of the vehicle body, battery information for monitoring the battery status, and driving information for monitoring the driving device;

Wherein, the vehicle body operation information, the battery information and the driving information come from at least one sensor respectively.

Further, the step of performing neural network training on the first sensing information to generate the optimal weight of the sensing information specifically includes:

Acquire error information in the vehicle body operation information, the battery information or the drive information according to the vehicle body operation information, the battery information and the drive information;

According to the error information, the optimal weight of the sensing information is obtained;

Wherein, the optimal weight of the sensing information is the weight of each sensor in the second monitoring information.

In the above, before the step of reorganizing the first sensing information to generate the second monitoring information, the method further includes:

Apply the error information to obtain and adjust the fault threshold;

According to the fault threshold, determine whether each sensor is faulty;

Wherein, the fault threshold is a threshold when each sensor operates abnormally.

Further, after the step of obtaining computing monitoring information according to the edge computing information and the cloud computing information, the method further includes:

generating the theoretical available time of the vehicle body, the battery or the drive device according to at least one of the first monitoring information, the second monitoring information and/or the calculated monitoring information;

transfer the theoretical available time of the body, battery or drive to the client;

Acquire and record the actual available time of the body, battery or drive.

A second aspect of the present application provides a pure electric vehicle state monitoring system, including:

a sensing module, configured to detect that the vehicle is in a running state and acquire first sensing information, where the first sensing information includes the type of sensing data and the total amount of sensing data;

a calculation module, configured to estimate the calculation amount of the first sensing information according to the type of the sensing data and the total amount of the sensing data;

a judgment module, configured to compare the calculated amount of the first sensing information with the preset value;

a monitoring module, configured to monitor the state of the vehicle according to the first sensing information when the calculated amount of the first sensing information is less than a preset value;

Wherein, the first sensing information comes from distributed sensors.

beneficial effect

The beneficial effect of the present invention is: in the use of distributed sensors, there is no step of estimating the calculation amount of information. If the sensing information is directly sent to the computing cloud, when the calculation amount is too small, the data used for interaction will be stored in the The computing cloud occupies a large amount of memory, which leads to performance degradation. If the edge computing terminal is directly used to calculate the sensor information, when the data is too large, the computing time will be long, which is difficult to use in practice. In the present invention, the computing part with a small amount of calculation is directly processed in the area where the data is generated, and the edge computing terminal located at the edge of the Internet of Vehicles is directly called to participate in the calculation, thereby greatly improving the computing rate of the system.

Description of drawings

The specific structure of the present invention will be described in detail below in conjunction with the accompanying drawings

1 is a flowchart of a method for monitoring the state of a pure electric vehicle according to the first embodiment of the present invention;

Fig. 2 is the flow chart of monitoring the state of the automobile in the second embodiment of the present invention;

Fig. 3 is the flow chart of monitoring the state of the automobile in the third embodiment of the present invention;

4 is a flowchart of generating second monitoring information in a fourth embodiment of the present invention;

FIG. 5 is a flowchart of generating and judging a sensor failure in a fifth embodiment of the present invention;

6 is a flowchart of obtaining calculation monitoring information in the sixth embodiment of the present invention;

FIG. 7 is a flow chart of obtaining the corrected theoretical available time in the seventh embodiment of the present invention;

FIG. 8 is a structural block diagram of a pure electric vehicle state monitoring system in the first embodiment of the present invention.

Embodiments of the present invention

In order to describe the technical content, structural features, achieved objects and effects of the present invention in detail, the following detailed description is given in conjunction with the embodiments and the accompanying drawings.

Please refer to FIG. 1 . FIG. 1 is a flowchart of a method for monitoring a state of a pure electric vehicle according to a first embodiment of the present invention. The application provides a state monitoring method for a pure electric vehicle, comprising the following steps:

Step S100, detecting that the vehicle is in a running state, and acquiring first sensing information, where the first sensing information includes the type of sensing data and the total amount of sensing data;

Step S200, estimating the calculation amount of the first sensing information according to the type of sensing data and the total amount of sensing data;

Step S300, comparing the calculated amount of the first sensing information with a preset value;

Step S400 , when the calculation amount of the first sensing information is less than the preset value, monitor the state of the vehicle according to the first sensing information; wherein the first sensing information comes from a distributed sensor.

The effect of the present invention is: in the use of distributed sensors, there is no step of estimating the calculation amount of information. If the sensing information is directly sent to the computing cloud, when the calculation amount is too small, the data used for interaction will be used in the calculation. The cloud occupies a large amount of memory, which leads to performance degradation. If the edge computing terminal is directly used to calculate the sensor information, when the data is too large, the calculation time will be long, which is difficult to use in practice. In the present invention, the computing part with a small amount of calculation is directly processed in the area where the data is generated, and the edge computing terminal located at the edge of the Internet of Vehicles is directly called to participate in the calculation, thereby greatly improving the computing rate of the system.

In this embodiment, by estimating different types of sensing data and the total amount of different sensing data, under the condition that the calculation speed is known, the required calculation amount can be obtained through simple calculation.

The first sensing information in the above includes information in each sensor under multiple physical domains, because the physical domain of the distributed sensor of the present application includes acceleration, temperature and humidity, Hall switch, voltmeter, galvanometer, motor speed, Pressure, mass, torque, etc., there is at least one sensor in each physical domain. The edge computing end of the Internet of Vehicles can also be called an edge computing node. Each edge computing node can only process data from at least one sensor in one physical domain, or can process any sensor data in multiple physical domains.

Further, the above-mentioned distributed sensors are composed of a large number of sensor nodes, which are used to sense physical states and collect and process information of objects in the coverage area. Because the number of sensors is too large, and some sensors have a poor working environment, they are easily damaged. Damage to any sensor will cause errors in these first sensing data, which will lead to false alarms or loopholes in the inspection. Therefore, it is necessary to judge the possibility of data errors. Based on this, the use of confidence data is introduced to monitor the state of the vehicle.

Please refer to FIG. 2 . FIG. 2 is a flowchart of monitoring the state of a vehicle according to the second embodiment of the present invention. Step S400, the step of monitoring the state of the vehicle according to the first sensing information, specifically includes:

Step S410, acquiring the confidence level of the sensor.

It can be understood that the source of the sensor confidence can be obtained through experimental data, through simulation training in the neural network, or through the confidence evaluation of the sensor based on the information provided by the sensor manufacturer. There are at least two ways to obtain these three sources, which can be stored in a certain module in the car, or the information can be transmitted to the car by the computing cloud.

Step S420 , weighting the first sensing information according to the confidence of the sensor to generate first monitoring information.

Wherein, the first monitoring information includes vehicle status information.

In the current technology, some people use the confidence to check the road outside the car to achieve the effect of unmanned driving, and this application applies the confidence to weight the first sensing information, there are at least three situations:

In the first case, the sensor with lower confidence has a higher weight, and the sensor with higher confidence has a lower weight. It is mainly for monitoring the sensor using the first monitoring information. When a sensor fails, it can be Quickly monitor sensors.

In the second case, the sensor with higher confidence has a higher weight, and the sensor with lower confidence has a lower weight, mainly for monitoring various structures of the car using the first monitoring information. , to minimize the impact of this sensor. It can be understood that, since the sensors in this embodiment are distributed sensors, each sensor may be distributed in various positions on the vehicle body, the battery, the driving device and even the vehicle.

In the third case, the time can be set to convert the first case and the second case according to the actual needs, which has the above two advantages at the same time, but cannot have all the advantages of both.

The above technical solution is mainly for calculating data with a small amount of calculation. Correspondingly, in step S400, when the amount of calculation of the first sensing information is greater than the preset value, if the edge computing terminal is directly used to calculate the first sensing information. For processing, it is easy to cause data overflow, resulting in calculation errors, or accidents may occur due to excessive processing time, affecting the life and health of users;

Based on this, please refer to FIG. 3 , which is a flowchart of monitoring the state of the vehicle in the third embodiment of the present invention. Step S400 also includes the following steps:

Step S430, preprocessing the first sensing information to generate preprocessing information.

When the preprocessing operation is performed at the edge computing terminal, generally no data calculation error will occur, and the preprocessing of the first sensing information can reduce the calculation amount of the terminal and ensure the data calculation speed.

In a specific embodiment, step S430 includes: preprocessing the picture information and the sensor data, and sending the preprocessed information to the computing cloud.

It should be understood that, in this specific embodiment, on the premise that the calculation amount of the first sensing information is greater than the preset value, the edge computing end not only preprocesses the picture information in the first sensing information, but also preprocesses the image information in the first sensing information. Preprocessing sensor data. It should be understood that after the image information is preprocessed, the confidence level of the estimated sensor can be calculated through the neural network in the computing cloud.

The sensing data in the above includes vehicle body operation information for monitoring vehicle body operation status, battery information for monitoring battery status, and driving information for monitoring driving device; wherein, vehicle body operation information, battery information and driving information respectively come from at least one sensor.

The driving device in this embodiment may be only one driving motor, or may be multiple driving motors.

In this specific embodiment, the three angles of vehicle body operation information, battery information and driving information are used to detect the operation state of the vehicle body, and certain redundant information can be formed to ensure the reliability of the overall information. These three angles can include multiple physical domains, so as to detect the running state of the vehicle body more accurately.

Step S440, sending the preprocessing information to the computing cloud to obtain feedback information from the computing cloud.

In this embodiment, the feedback information of the computing cloud is the processing result of the preprocessing information on the computing cloud, and after the preprocessing at the edge computing end, the computing speed of the computing cloud is relatively fast.

Step S450: Detect the state of the vehicle according to the feedback information from the computing cloud.

In one embodiment, the amount of computation required for the first sensing data is less, and when it is only for image preprocessing or only data in one physical domain needs to be processed, the computation can be performed directly at the edge computing end, without the need to transmit the data. In another embodiment, neural network training or multi-physical domain data fusion is required for the first sensing, and in this case, the first sensing data needs to be transmitted to the cloud for calculation.

It is understandable that when there are too many sensors in a physical domain, the data in the same physical domain can be processed by multiple edge computing terminals, and after these edge computing terminals process the data in this physical domain, the processed results It needs to be transmitted to the computing cloud and processed in the computing cloud.

Based on this, after step S450, the step of sending the first sensing information to the computing cloud, please refer to FIG. 4, which is a flowchart of generating the second monitoring information in the fourth embodiment of the present invention; the above method further includes the following steps: step:

Step S460: Perform neural network training on the first sensing information to generate an optimal weight of the sensing information.

Through the neural training of the first sensing information, neural network training can be performed on the relevant data of each sensing information, and the optimal weights can be used to reorganize the first sensing information to monitor vehicle information more accurately in real time.

Specifically, please refer to FIG. 5 . FIG. 5 is a flowchart of generating and judging a sensor failure in a fifth embodiment of the present invention. In step S460, it specifically includes:

Step S461: Acquire error information in the vehicle body operation information, the battery information or the drive information according to the vehicle body operation information, the battery information and the drive information.

In this embodiment, after collecting the actual vehicle body operation information, battery information and driving information, the confidence level of each sensor can be matched to determine the error between the actual information and the theoretical information, so as to more accurately judge the vehicle state.

Step S462: Obtain the optimal weight of the sensing information according to the error information.

The optimal weight of the sensing information is the weight of each sensor in the second monitoring information.

What needs to be understood is that since the state of the car may change at any time, this process should be carried out frequently. According to the error information, the edge computing end and the cloud computing end can be adjusted at any time to better monitor the vehicle.

In a further embodiment, the above method also includes the steps:

Step S463, applying the error information to acquire and adjust the fault threshold.

In this embodiment, because the error information may change to a certain extent, after the error information is generated, a corresponding new fault threshold can be obtained from the expert database.

Step S464, according to the fault threshold, determine whether each sensor is faulty.

The fault threshold is a threshold when each sensor operates abnormally.

As a result, the state of the vehicle can be better monitored to obtain a more comprehensive monitoring.

Step S470: Perform multi-physical domain fusion on the first sensing information to generate multi-physical domain feedback information.

It should be understood that when the sensor information of multiple physical domains is fused, the multi-physical domain information fusion is to synthesize the partial incomplete observations provided by multiple sensors distributed in different locations, and eliminate the possibility of information between multiple sensors. The existing redundancies and contradictions should be complemented to reduce their uncertainty to form a relatively complete and consistent perception description of the system environment, thereby improving the speed and correctness of intelligent system decision-making, planning and response, while reducing its decision-making risk.

The technology based on multi-physical domain information fusion has three advantages: It improves the monitoring accuracy of the system. Extending from the traditional single physical domain to multiple physical domains, the data fusion between the same and different physical domains reduces the interference of noise and improves the accuracy of the system.

The monitoring area of the system has been expanded. Compared with the traditional sensor layout, the extensive distribution of the edge computing side covers a larger area, and can monitor more state parameters of the vehicle.

Improve the redundancy of the system. When there is a sensor failure in the traditional pure electric vehicle condition monitoring system, the monitoring data will greatly affect the final monitoring result. The multi-physical domain information fusion technology added to this system can reasonably avoid such problems.

Step S480 , recombine the first sensing information according to the optimal weight of the sensing information and the feedback information from multiple physical domains, and generate the second monitoring information.

By using neural network training and related methods of multi-physical domain data fusion, heterogeneous sensor data is collected and trained to obtain the optimal weights, which can greatly improve the accuracy of data results and reduce measurement errors.

Further, after step S470, the step of generating multi-physical domain feedback information, please refer to FIG. 6. FIG. 6 is a flowchart of obtaining calculation monitoring information in the sixth embodiment of the present invention. The above method also includes:

Step S491 , record the historical computing process of the edge computing terminal, and obtain edge computing information.

Step S492: Record the historical computing process of the computing cloud to obtain cloud computing information.

In this embodiment, the state when the edge computing terminal calculates various data can be estimated according to the edge computing information, and the state of various data calculated by the cloud computing terminal can be estimated according to the cloud computing information, so as to analyze the whole and better plan The amount of information that should be processed by the edge computing terminal, thereby determining the preset value of the first sensing information. When some sensors are updated, the corresponding data can be directly obtained by applying this method.

Among them, each computing cloud corresponds to at least one edge computing terminal. In addition, it should be understood that the computing cloud realizes the storage and computing of big data, the management and coordination of multitasking, and the real-time interaction of the edge.

Step S493: Obtain computing monitoring information according to the edge computing information and the cloud computing information.

Computational monitoring information includes vehicle status information. In this embodiment, by calculating the monitoring information, the historical data of the vehicle can be selectively integrated, so as to realize the prediction of the state, so as to provide a basis for the maintenance and replacement of the parts of the vehicle, so as to ensure the safety of the vehicle during the driving process. Safety. At the same time, a network structure suitable for the system can be constructed, and its calculation speed is far higher than that of traditional data processing algorithms.

In a further embodiment, after step S493, please refer to FIG. 7. FIG. 7 is a flowchart of obtaining the corrected theoretical available time in the seventh embodiment of the present invention. The above method also includes:

Step S494: Generate the theoretical available time of the vehicle body, the battery or the drive device according to at least one item of the first monitoring information, the second monitoring information and/or the calculated monitoring information.

According to the discussion in the above embodiment, the first monitoring information can detect not only the state of the sensor, but also the state of each structure of the vehicle; and the second monitoring information is obtained after neural network calculation and multi-physical domain information fusion It is the operation in the computing cloud. Through the redundant information of multiple sensors and the operation of a large amount of data, the monitoring area of the system is expanded and the redundancy of the system is improved. When there is a sensor failure, multi-physical domain information fusion The technology can reasonably avoid the problems caused by sensor failure. The main function of calculating the monitoring information master is embodied in steps S491 to S493, and will not be repeated here. It can be understood that the available time of the vehicle body, the battery or the drive device can be predicted based on the theoretical available time of the corresponding vehicle structure after a certain component of each vehicle structure is damaged.

The operation in step S494 requires a large amount of data, so it is generally performed in the computing cloud. Based on this, this embodiment further includes: step S495 , transmitting the theoretical available time of the vehicle body, battery or driving device to the client. In this step, the estimated time is sent to the client, so that the client can predict the available time of the above three automobile structures, so as to avoid the situation that the automobile cannot run after a certain structure of the automobile is damaged.

Step S496: Acquire and record the real available time of the vehicle body, battery or drive device.

It is understandable that the above theoretical usable time of the vehicle body, battery or drive device is obtained from prediction, because in actual circumstances, the actual usable time may not match the theoretical usable time. After recording this time, it can be used for adjustment. The theoretical available time can also be used for recording.

Please refer to FIG. 8 . FIG. 8 is a structural block diagram of a state monitoring system for a pure electric vehicle according to the first embodiment of the present invention. A second aspect of the present application provides a pure electric vehicle state monitoring system, including:

The sensing module 100 is used for detecting that the vehicle is in a running state and acquiring first sensing information, where the first sensing information includes the type of sensing data and the total amount of sensing data;

The calculation module 200 is used for estimating the calculation amount of the first sensing information according to the type of sensing data and the total amount of sensing data;

The judgment module 300 is used for comparing the calculated amount of the first sensing information with a preset value;

A monitoring module 400, configured to monitor the state of the vehicle according to the first sensing information when the calculated amount of the first sensing information is less than a preset value;

Wherein, the first sensing information comes from distributed sensors.

The above modules are essentially virtual modules, carrying the methods in the above embodiments. The above modules can be combined with any actual product. Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be implemented by instructing relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium. During execution, the processes of the embodiments of the above-mentioned methods may be included. The storage medium may be a magnetic disk, an optical disk, a read-only memory (Read-Only Memory, ROM), or a random access memory (Random Access Memory, RAM) or the like.

The present application also provides an electronic terminal, including: a memory, a processor, and a computer program stored in the memory and running on the processor, when the processor executes the computer program, the computer program described in any of the foregoing embodiments is implemented. Loop detection method. The above method is implemented when the processor executes the software program. It should be noted that the electronic terminals in the embodiments of the present invention include but are not limited to user equipment such as mobile phones, mobile computers, tablet computers, personal digital assistants, media players, smart TVs, smart watches, smart glasses, and smart bracelets.

It should be noted that the functions of the functional modules of the electronic terminal in the embodiments of the present invention can be specifically implemented according to the methods in the above method embodiments, and the specific implementation process can refer to the relevant descriptions of the above method embodiments, which will not be repeated here.

The above descriptions are only the embodiments of the present invention, and are not intended to limit the scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the description and drawings of the present invention, or directly or indirectly applied to other related technologies Fields are similarly included in the scope of patent protection of the present invention.

Claims

A method for monitoring the state of a pure electric vehicle, comprising the following steps:

Detecting that the car is in a running state, and acquiring first sensing information, where the first sensing information includes the type of sensing data and the total amount of sensing data;

Estimating the calculation amount of the first sensing information according to the type of the sensing data and the total amount of the sensing data;

comparing the calculated amount of the first sensing information with the preset value;

When the calculated amount of the first sensing information is less than a preset value, monitor the state of the vehicle according to the first sensing information;

Wherein, the first sensing information comes from distributed sensors.
The method for monitoring the state of a pure electric vehicle according to claim 1, wherein the step of monitoring the state of the vehicle according to the first sensing information specifically includes:

obtain the confidence level of the sensor;

Weighting the first sensing information according to the confidence of the sensor to generate first monitoring information;

Wherein, the first monitoring information includes the vehicle state information.
The method for monitoring the state of a pure electric vehicle according to claim 2, wherein when the calculated amount of the first sensing information is greater than a preset value, the method comprises the following steps:

Preprocessing the first sensing information to generate preprocessing information;

sending the preprocessing information to the computing cloud to obtain feedback information from the computing cloud;

According to the feedback information of the computing cloud, the state of the vehicle is detected.
The method for monitoring the state of a pure electric vehicle according to claim 3, wherein after the step of sending the first sensing information to the computing cloud, the method further comprises the following steps:

performing neural network training on the first sensing information to generate optimal weights for the sensing information;

performing multi-physical domain fusion on the first sensing information to generate multi-physical domain feedback information;

Recombining the first sensing information according to the optimal weight of the sensing information and the feedback information from multiple physical domains to generate the second monitoring information;

Wherein, the first sensing information includes information of multiple physical domains.
The state monitoring method for pure electric vehicles according to claim 4, wherein after the step of generating feedback information from multiple physical domains, the method further comprises:

Record the historical computing process of the edge computing terminal and obtain edge computing information;

Record the historical computing process of the computing cloud to obtain cloud computing information;

obtaining computing monitoring information according to the edge computing information and the cloud computing information;

Wherein, each of the computing clouds corresponds to at least one of the edge computing terminals;

The calculated monitoring information includes the vehicle state information.
The method for monitoring the state of a pure electric vehicle according to claim 4, wherein the step of preprocessing the first sensing information to generate the preprocessing information comprises:

Preprocessing the image information and sensor data, and sending the preprocessed information to the computing cloud;

Wherein, the sensing data includes vehicle body running information for monitoring the running state of the vehicle body, battery information for monitoring the battery status, and driving information for monitoring the driving device;

Wherein, the vehicle body operation information, the battery information and the driving information come from at least one sensor respectively.
The method for monitoring the state of a pure electric vehicle according to claim 6, wherein the step of performing neural network training on the first sensing information to generate the optimal weight of the sensing information specifically includes:

Acquire error information in the vehicle body operation information, the battery information or the drive information according to the vehicle body operation information, the battery information and the drive information;

According to the error information, the optimal weight of the sensing information is obtained;

Wherein, the optimal weight of the sensing information is the weight of each sensor in the second monitoring information.
The method for monitoring the state of a pure electric vehicle according to claim 7, wherein before the step of reorganizing the first sensing information to generate the second monitoring information, the method further comprises:

Apply the error information to obtain and adjust the fault threshold;

According to the fault threshold, determine whether each sensor is faulty;

Wherein, the fault threshold is a threshold when each sensor operates abnormally.
The method for monitoring the state of a pure electric vehicle according to claim 5, wherein after the step of obtaining the computing monitoring information according to the edge computing information and the cloud computing information, the method further comprises:

generating the theoretical available time of the vehicle body, the battery or the drive device according to at least one of the first monitoring information, the second monitoring information and/or the calculated monitoring information;

transfer the theoretical available time of the body, battery or drive to the client;

Acquire and record the actual available time of the body, battery or drive.
A pure electric vehicle state monitoring system, comprising:

a sensing module, configured to detect that the vehicle is in a running state and acquire first sensing information, where the first sensing information includes the type of sensing data and the total amount of sensing data;

a calculation module, configured to estimate the calculation amount of the first sensing information according to the type of the sensing data and the total amount of the sensing data;

a judgment module, configured to compare the calculated amount of the first sensing information with the preset value;

a monitoring module, configured to monitor the state of the vehicle according to the first sensing information when the calculated amount of the first sensing information is less than a preset value;

Wherein, the first sensing information comes from distributed sensors.