CN118238622B - Vehicle-mounted power supply protection method and system based on multi-sensor monitoring - Google Patents

Abstract

The application provides a vehicle-mounted power supply protection method and system based on multi-sensor monitoring, and relates to the technical field of vehicle-mounted battery control, wherein the method comprises the following steps: performing battery clustering through a battery pack simulation model, and determining a plurality of battery neighborhoods; setting up a temperature monitoring network to obtain a temperature monitoring data set; performing anomaly identification on the vibration and noise monitoring data set based on the correlation factor to generate a temperature prediction set; performing mapping compensation on the temperature monitoring data set based on the temperature prediction set and the battery load prediction data; and carrying out heat dissipation analysis on the compensation temperature data set through a heat dissipation control model to obtain an optimal heat dissipation control decision. The application can solve the technical problem that the heat dissipation of the battery pack cannot be accurately controlled because the temperature change state of the battery pack cannot be accurately predicted due to incomplete sensing monitoring data of the battery pack and lower data analysis precision, can realize the technical aim of accurately predicting the temperature change of the battery pack, and achieves the effect of accurately controlling the heat dissipation of the battery pack.

Description

Vehicle-mounted power supply protection method and system based on multi-sensor monitoring

Technical Field

The application relates to the technical field of vehicle-mounted battery control, in particular to a vehicle-mounted power supply protection method and system based on multi-sensor monitoring.

Background

The battery pack of the new energy electric car is one of core components of the electric car and is responsible for storing and supplying electric energy to drive the motor to work, and the performance and the safety of the battery pack directly influence the endurance mileage, the use convenience and the safety of the whole electric car.

At present, when a traditional new energy electric car carries out heat dissipation of a battery pack, temperature monitoring is generally carried out according to a plurality of temperature sensors distributed inside or outside the battery pack, and then heat dissipation control is carried out according to monitoring results.

In summary, the heat dissipation method of the new energy electric car power supply system has the technical problems that the sensing and monitoring data of the battery pack are incomplete, the data analysis precision is low, the temperature change state of the battery pack cannot be accurately predicted, and the heat dissipation of the battery pack cannot be accurately controlled in a targeted manner.

Disclosure of Invention

The application aims to provide a vehicle-mounted power supply protection method and system based on multi-sensing monitoring, which are used for solving the technical problems that the existing heat dissipation method of a new energy electric car power supply system cannot accurately control heat dissipation of a battery pack due to the fact that the sensing monitoring data of the battery pack are incomplete and the data analysis precision is low, and the temperature change state of the battery pack cannot be accurately predicted.

In view of the above problems, the application provides a vehicle-mounted power supply protection method and system based on multi-sensor monitoring.

In a first aspect, the present application provides a vehicle-mounted power supply protection method based on multi-sensor monitoring, the method is implemented by a vehicle-mounted power supply protection system based on multi-sensor monitoring, wherein the method includes: performing simulation modeling based on basic data of a target battery pack to generate a battery pack simulation model; performing battery clustering through the battery pack simulation model, and determining a plurality of battery neighborhoods according to clustering results; setting up a temperature monitoring network based on the plurality of battery neighborhood layout temperature sensors, and executing temperature monitoring of the target battery pack through the temperature monitoring network to obtain a temperature monitoring data set under a preset node; collecting a vibration monitoring data set and a noise monitoring data set of the target battery pack, carrying out abnormal characteristic identification on the vibration monitoring data set and the noise monitoring data set based on preset association factors, and generating a temperature prediction set according to an identification result; acquiring battery load prediction data, and performing mapping compensation on the temperature monitoring data set based on the temperature prediction set and the battery load prediction data to obtain a compensated temperature data set; carrying out heat dissipation analysis on the compensation temperature data set through a heat dissipation control model to obtain an optimal heat dissipation control decision, wherein a heat dissipation topology network is embedded in the heat dissipation control model; and executing the heat dissipation control of the target battery pack based on the optimal heat dissipation control decision.

In a second aspect, the present application further provides a vehicle-mounted power supply protection system based on multi-sensor monitoring, for performing a vehicle-mounted power supply protection method based on multi-sensor monitoring as described in the first aspect, where the system includes: the battery pack simulation model generation module is used for carrying out simulation modeling based on basic data of a target battery pack to generate a battery pack simulation model; the battery neighborhood determining module is used for carrying out battery clustering through the battery pack simulation model and determining a plurality of battery neighborhoods according to a clustering result; the temperature monitoring network building module is used for building a temperature monitoring network based on the plurality of battery neighborhood layout temperature sensors, and executing temperature monitoring of the target battery pack through the temperature monitoring network to acquire a temperature monitoring data set under a preset node; the abnormal characteristic recognition module is used for collecting a vibration monitoring data set and a noise monitoring data set of the target battery pack, carrying out abnormal characteristic recognition on the vibration monitoring data set and the noise monitoring data set based on a preset correlation factor, and generating a temperature prediction set according to recognition results; the mapping compensation module is used for acquiring battery load prediction data, and performing mapping compensation on the temperature monitoring data set based on the temperature prediction set and the battery load prediction data to obtain a compensated temperature data set; the optimal heat dissipation control decision obtaining module is used for carrying out heat dissipation analysis on the compensation temperature data set through a heat dissipation control model to obtain an optimal heat dissipation control decision, and a heat dissipation topology network is embedded in the heat dissipation control model; and the heat dissipation control module is used for executing heat dissipation control of the target battery pack based on the optimal heat dissipation control decision.

One or more technical schemes provided by the application have at least the following technical effects or advantages:

1. Generating a battery pack simulation model by performing simulation modeling based on basic data of a target battery pack; performing battery clustering through the battery pack simulation model, and determining a plurality of battery neighborhoods according to clustering results; setting up a temperature monitoring network based on the plurality of battery neighborhood layout temperature sensors, and executing temperature monitoring of the target battery pack through the temperature monitoring network to obtain a temperature monitoring data set under a preset node; collecting a vibration monitoring data set and a noise monitoring data set of the target battery pack, carrying out abnormal characteristic identification on the vibration monitoring data set and the noise monitoring data set based on preset association factors, and generating a temperature prediction set according to an identification result; acquiring battery load prediction data, and performing mapping compensation on the temperature monitoring data set based on the temperature prediction set and the battery load prediction data to obtain a compensated temperature data set; carrying out heat dissipation analysis on the compensation temperature data set through a heat dissipation control model to obtain an optimal heat dissipation control decision, wherein a heat dissipation topology network is embedded in the heat dissipation control model; and executing the heat dissipation control of the target battery pack based on the optimal heat dissipation control decision. The comprehensive performance of the battery pack sensing monitoring data and the accuracy of data analysis can be improved, the technical aim of accurately predicting the temperature change of the battery pack is achieved, and therefore the problems of vehicle performance reduction, battery damage and the like caused by overheat of the battery pack are avoided, the heat dissipation of the battery pack is accurately controlled, and the technical effects of vehicle running quality and running safety are improved.

2. The battery pack simulation model is constructed to perform battery clustering to obtain a plurality of battery neighborhoods, and then the temperature sensors are distributed according to the battery neighborhoods, so that the distribution accuracy and reliability of the temperature sensors can be improved, the acquisition accuracy of the battery pack temperature monitoring data can be improved, and a basis is provided for performing accurate heat dissipation control.

3. The battery pack temperature change state can be analyzed and judged from multiple dimensions, and therefore accuracy and comprehensiveness of obtaining the compensated temperature data set are further improved.

4. And the heat dissipation control model is constructed to carry out heat dissipation analysis on the compensation temperature data set to obtain an optimal heat dissipation control decision, so that heat dissipation energy consumption resources can be saved on the premise of guaranteeing heat dissipation requirements, and timeliness and accuracy of the optimal heat dissipation control decision can be improved, so that timeliness and accuracy of heat dissipation control of the battery pack are improved.

The foregoing description is only an overview of the present application, and is intended to be implemented in accordance with the teachings of the present application in order that the same may be more clearly understood and to make the same and other objects, features and advantages of the present application more readily apparent. It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the application or to delineate the scope of the application. Other features of the present application will become apparent from the description that follows.

Drawings

In order to more clearly illustrate the application or the technical solutions of the prior art, the following brief description will be given of the drawings used in the description of the embodiments or the prior art, it being obvious that the drawings in the description below are only exemplary and that other drawings can be obtained from the drawings provided without the inventive effort for a person skilled in the art.

FIG. 1 is a schematic flow chart of a vehicle-mounted power supply protection method based on multi-sensor monitoring;

FIG. 2 is a schematic flow chart of determining a plurality of battery neighbors according to a clustering result in a vehicle-mounted power supply protection method based on multi-sensor monitoring;

fig. 3 is a schematic structural diagram of a vehicle-mounted power supply protection system based on multi-sensor monitoring.

Reference numerals illustrate:

the system comprises a battery pack simulation model generation module 11, a battery neighborhood determination module 12, a temperature monitoring network building module 13, an abnormal characteristic identification module 14, a mapping compensation module 15, an optimal heat dissipation control decision obtaining module 16 and a heat dissipation control module 17.

Detailed Description

The application provides a vehicle-mounted power supply protection method and a vehicle-mounted power supply protection system based on multi-sensing monitoring, which solve the technical problems that the existing heat dissipation method of the new energy electric car power supply system has incomplete sensing monitoring data of a battery pack and low data analysis precision, so that the temperature change state of the battery pack cannot be accurately predicted, and the heat dissipation of the battery pack cannot be accurately controlled in a targeted manner. The comprehensive performance of the battery pack sensing monitoring data and the accuracy of data analysis can be improved, the technical aim of accurately predicting the temperature change of the battery pack is achieved, and therefore the problems of vehicle performance reduction, battery damage and the like caused by overheat of the battery pack are avoided, the heat dissipation of the battery pack is accurately controlled, and the technical effects of vehicle running quality and running safety are improved.

In the following, the technical solutions of the present application will be clearly and completely described with reference to the accompanying drawings, and it should be understood that the described embodiments are only some embodiments of the present application, but not all embodiments of the present application, and that the present application is not limited by the exemplary embodiments described herein. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application. It should be further noted that, for convenience of description, only some, but not all of the drawings related to the present application are shown.

Referring to fig. 1, the application provides a vehicle-mounted power supply protection method based on multi-sensor monitoring, wherein the method is applied to a vehicle-mounted power supply protection system based on multi-sensor monitoring, and the method specifically comprises the following steps:

step one: performing simulation modeling based on basic data of a target battery pack to generate a battery pack simulation model;

Specifically, the digital twin technology is a method for digitally modeling an entity, an operation process or a system in the real world and performing simulation and analysis in a virtual environment by using technologies such as a computer, 3D modeling, digital simulation and the like, and the actual scene can be highly restored by constructing a simulation twin model, so that simulation and prediction can be performed in the virtual environment, and the working efficiency is improved.

First, basic data of a target battery pack is acquired, wherein the basic data comprises information such as battery structure data, battery type, battery number, battery specification, size data, operation parameters and the like in the battery pack. And then, based on a digital twin technology, carrying out simulation modeling on the target battery pack according to the basic data in a visual simulation Platform to construct a battery pack simulation model, wherein the visual simulation Platform is a tool for carrying out simulation modeling on the target battery pack, and can select an adaptive tool for modeling according to actual conditions, wherein the common visual simulation Platform comprises SimuWorks software, VR-Platform software and the like. By constructing the battery pack simulation model, support is provided for the next step of battery simulation load operation.

Step two: performing battery clustering through the battery pack simulation model, and determining a plurality of battery neighborhoods according to clustering results;

Specifically, a plurality of battery load strategies are set, and battery pack simulation operation is performed based on the plurality of battery load strategies through the battery pack simulation model, wherein in the battery pack simulation model, each battery block is provided with a unique number identifier, and a simulated temperature data set of the battery block under a plurality of battery load scenes is obtained. And then clustering the battery blocks according to the simulated temperature data set, namely, clustering the battery blocks with temperature deviation within a preset range into the same battery neighborhood to obtain a plurality of battery neighborhoods. A plurality of battery neighborhoods are determined by battery clustering, so that support is provided for the arrangement of temperature sensors in a next battery pack, and meanwhile, the accuracy of the arrangement of the temperature sensors can be improved.

Step three: setting up a temperature monitoring network based on the plurality of battery neighborhood layout temperature sensors, and executing temperature monitoring of the target battery pack through the temperature monitoring network to obtain a temperature monitoring data set under a preset node;

Specifically, temperature sensors are distributed according to the plurality of battery neighborhoods, wherein the distribution positions of the temperature sensors are the positions with the smallest sum of the distances from the plurality of battery blocks in the battery neighborhoods, and then a temperature monitoring network is built according to the plurality of distributed temperature sensors. And under a preset node, performing temperature monitoring on the target battery pack through the temperature monitoring network, wherein the preset node is a smaller time period, and the preset node can be set by a person skilled in the art according to actual conditions, for example: and setting the preset node to be 30 seconds, namely acquiring temperature monitoring data every 30 seconds to obtain a temperature monitoring data set under the preset node, wherein the temperature monitoring data set comprises temperature monitoring data of a plurality of battery neighborhoods. By obtaining the temperature monitoring data set, support is provided for heat dissipation control of the target battery pack.

Step four: collecting a vibration monitoring data set and a noise monitoring data set of the target battery pack, carrying out abnormal characteristic identification on the vibration monitoring data set and the noise monitoring data set based on preset association factors, and generating a temperature prediction set according to an identification result;

specifically, vibration sensors and sound sensors are distributed in a plurality of battery neighborhoods of the target battery pack, and the target battery pack is monitored through the vibration sensors and the sound sensors to obtain a vibration monitoring data set and a noise monitoring data set.

Obtaining preset association factors, wherein the preset association factors are abnormal vibration signals and abnormal noise signals, then carrying out abnormal feature recognition on the vibration monitoring data set and the noise monitoring data set according to the preset association factors, obtaining abnormal vibration frequency and abnormal noise frequency according to recognition results, and further carrying out battery neighborhood temperature prediction according to the abnormal vibration frequency and the abnormal noise frequency to obtain a temperature prediction set. By obtaining a temperature prediction set, support is provided for mapping compensation of the next step of temperature monitoring dataset.

Step five: acquiring battery load prediction data, and performing mapping compensation on the temperature monitoring data set based on the temperature prediction set and the battery load prediction data to obtain a compensated temperature data set;

Specifically, vehicle running data are collected, wherein the vehicle running data comprise information such as temperature data in a vehicle and road condition data in front of the vehicle, then battery load prediction is carried out according to the vehicle running data to obtain battery load prediction data, and further battery load prediction trend is obtained according to battery load prediction data analysis, wherein the larger the battery load is, the faster the temperature rise speed of a battery pack is.

And carrying out mapping compensation on the temperature monitoring data set according to the temperature prediction set and the battery load prediction trend based on the mapping relation of the battery neighborhood, and obtaining a compensated temperature data set. The temperature monitoring data set is subjected to mapping compensation according to the temperature prediction set and the battery load prediction data, and the temperature change state of the battery pack can be analyzed and judged from multiple dimensions, so that the accuracy and the comprehensiveness of the compensation temperature data set are further improved.

Step six: carrying out heat dissipation analysis on the compensation temperature data set through a heat dissipation control model to obtain an optimal heat dissipation control decision, wherein a heat dissipation topology network is embedded in the heat dissipation control model;

Specifically, firstly, heat dissipation basic data of a target battery pack are obtained, wherein the heat dissipation basic data comprise heat dissipation part types, heat dissipation part positions, heat dissipation part quantity and heat dissipation operation parameters, and then a heat dissipation topology network is built according to the heat dissipation basic data, and the heat dissipation topology network is an overall heat dissipation framework of the target battery pack. And embedding the heat dissipation topological network into a heat dissipation control model to obtain the heat dissipation control model. And further inputting the compensation temperature data set into the heat dissipation control model for heat dissipation analysis to obtain a plurality of heat dissipation control decisions. And then selecting a heat dissipation control decision with the minimum comprehensive energy consumption from the plurality of heat dissipation control decisions to set as the optimal heat dissipation control, and obtaining the optimal heat dissipation control decision. By obtaining the optimal heat dissipation control decision, the heat dissipation energy consumption of the battery can be reduced and the energy utilization rate of the battery can be improved on the premise of meeting the heat dissipation requirement of the battery pack.

Step seven: and executing the heat dissipation control of the target battery pack based on the optimal heat dissipation control decision.

And controlling the heat dissipation topology network according to the optimal heat dissipation control decision to complete heat dissipation control of the target battery pack.

The vehicle-mounted power supply protection method based on multi-sensor monitoring is applied to a vehicle-mounted power supply protection system based on multi-sensor monitoring, and can solve the technical problems that the existing heat dissipation method of the new energy electric car power supply system is incapable of accurately predicting the temperature change state of a battery pack and cannot accurately control heat dissipation of the battery pack due to incomplete sensing monitoring data of the battery pack and low data analysis precision. Firstly, performing simulation modeling based on basic data of a target battery pack to generate a battery pack simulation model; then, battery clustering is carried out through the battery pack simulation model, and a plurality of battery neighborhoods are determined according to clustering results; then, temperature sensors are arranged on the basis of the plurality of battery neighborhood, a temperature monitoring network is built, and temperature monitoring of the target battery pack is executed through the temperature monitoring network, so that a temperature monitoring data set under a preset node is obtained; collecting a vibration monitoring data set and a noise monitoring data set of the target battery pack, carrying out abnormal characteristic identification on the vibration monitoring data set and the noise monitoring data set based on preset association factors, and generating a temperature prediction set according to an identification result; in addition, battery load prediction data are obtained, and mapping compensation is carried out on the temperature monitoring data set based on the temperature prediction set and the battery load prediction data to obtain a compensated temperature data set; further, carrying out heat dissipation analysis on the compensation temperature data set through a heat dissipation control model to obtain an optimal heat dissipation control decision, wherein a heat dissipation topology network is embedded in the heat dissipation control model; and finally, executing the heat dissipation control of the target battery pack based on the optimal heat dissipation control decision. The temperature sensor is arranged on the basis of the neighborhood of the battery to monitor the temperature of the battery pack, and the temperature monitoring data set is compensated through a plurality of dimensional data such as vibration monitoring, noise monitoring, battery load trend prediction and the like, so that the accuracy of obtaining the compensated temperature data set can be improved; meanwhile, a heat dissipation control model is constructed to carry out heat dissipation analysis on the compensation temperature data set, so that an optimal heat dissipation control decision is obtained, and the accuracy and the rationality of heat dissipation control decision setting can be improved; therefore, the comprehensiveness of the sensing monitoring data of the battery pack and the accuracy of data analysis can be improved, the technical aim of accurately predicting the temperature change of the battery pack is fulfilled, the problems of vehicle performance reduction, battery damage and the like caused by overheat of the battery pack are avoided, the heat dissipation of the battery pack is accurately controlled, and the technical effects of improving the running quality and the running safety of a vehicle are achieved.

Further, as shown in fig. 2, the second step of the present application includes:

setting N battery load strategies, wherein N is an integer greater than 1;

based on the N battery load strategies, performing simulation operation through the battery pack simulation model to obtain N simulated temperature data sets, wherein the simulated temperature data have battery number identifications;

sequentially carrying out clustering division on the N simulated temperature data sets based on a preset temperature difference threshold value to obtain N clustering neighborhood sets;

And selecting N times of overlapping battery numbers in the N clustering neighborhood sets to construct a battery neighborhood, and obtaining the plurality of battery neighborhoods.

Specifically, firstly, N battery load strategies are set, where N is an integer greater than 1, and a specific value of N can be set according to actual requirements, where the greater the value of N, the higher the accuracy of the battery pack simulation test, for example: setting N as 10; where battery load strategy refers to different battery load ratios, for example: battery load 60%, battery load 65%, etc. And then, carrying out N times of simulation operation through the battery pack simulation model by taking the N battery load strategies as constraints to obtain N simulated temperature data sets, wherein the simulated temperature data sets comprise simulated temperature data of a plurality of battery blocks of a target battery pack, namely each simulated temperature data has a battery block number identifier.

Obtaining a preset temperature difference threshold, wherein the temperature difference threshold can be set according to actual conditions by a person skilled in the art, the smaller the preset temperature difference threshold is, the higher the accuracy of the battery neighborhood setting is, and the more the number of the battery neighborhoods is, for example: the preset temperature difference threshold is set to be 0.1 ℃. And then carrying out clustering division on the N simulated temperature data sets in sequence according to the preset temperature difference threshold, namely gathering the simulated temperature data with the temperature deviation between the simulated temperature data in the simulated temperature data sets within the preset temperature difference threshold into the same clustering neighborhood to obtain N clustering neighborhood sets.

And selecting N times of overlapped battery numbers in the N clustering neighborhood sets based on the N clustering neighborhood sets, namely, after N times of testing, collecting the battery numbers with the temperature deviation in a preset temperature difference threshold value into the same battery neighborhood to obtain a plurality of battery neighborhoods, wherein each neighborhood comprises a plurality of battery blocks.

The simulation operation under various battery load strategies is carried out by constructing a battery pack simulation model, the battery blocks are clustered according to the simulation operation result, a plurality of battery neighborhoods are determined, a basis is provided for temperature sensor layout, and meanwhile, the accuracy of temperature monitoring data acquisition can be improved.

Further, the fourth step of the present application includes:

respectively arranging a vibration sensor and a sound sensor based on the plurality of battery neighborhoods;

and reading monitoring data of the vibration sensor and the sound sensor in a preset time window to obtain the vibration monitoring data set and the noise monitoring data set, wherein the preset time window is a time interval between the last adjacent node and the preset node.

Specifically, a plurality of vibration sensors and a plurality of sound sensors are respectively arranged in the plurality of battery neighbors, and the target battery pack is monitored through the plurality of vibration sensors and the plurality of sound sensors. And acquiring a preset time window, wherein the preset time window is the time interval between the last adjacent node and the preset node, and further reading vibration monitoring data and noise monitoring data in the preset time window to obtain a vibration monitoring data set and a noise monitoring data set, and each vibration monitoring data set and each noise monitoring data set correspond to a battery neighborhood. By obtaining the vibration monitoring dataset and the noise monitoring dataset, data support is provided for the recognition of the abnormal features of the next step.

Further, the fourth step of the present application includes:

taking the target battery pack as constraint, and searching and acquiring a plurality of historical battery overheat events;

Respectively carrying out vibration data identification and noise data identification based on the plurality of historical battery overheat events, and determining an associated vibration factor and an associated noise factor;

Performing abnormal characteristic recognition on the vibration monitoring data set according to the associated vibration factors, and determining an abnormal vibration frequency set;

carrying out abnormal feature recognition on the noise monitoring data set according to the associated noise factors, and determining an abnormal noise frequency set;

And carrying out temperature prediction on the abnormal vibration frequency set and the abnormal noise frequency set through a temperature prediction channel to obtain the temperature prediction set.

Specifically, the big data technology is a technology for quickly obtaining valuable information from various types of mass data, and relates to various technical means such as data acquisition, data cleaning, data mining and the like. Firstly, based on big data technology, information retrieval is carried out by taking the target battery pack as retrieval constraint, and a plurality of historical battery overheat events are obtained, wherein the historical battery overheat events comprise battery temperature monitoring data, battery vibration monitoring data, battery noise monitoring data and the like.

And then respectively identifying vibration monitoring data and noise monitoring data in the historical battery overheat events, selecting one or more abnormal vibration signals with occurrence frequency larger than a preset frequency threshold value as an associated vibration factor, wherein the abnormal vibration signals comprise parameters such as abnormal amplitude, selecting one or more abnormal noise signals with occurrence frequency larger than the preset frequency threshold value as an associated noise factor, and the abnormal noise signals comprise parameters such as noise waveform, wherein the preset frequency threshold value can be set according to actual conditions, namely, selecting abnormal characteristics with higher occurrence frequency as the associated factor, and obtaining the associated vibration factor and the associated noise factor.

Carrying out abnormal feature recognition on the vibration monitoring data set according to the associated vibration factors, and obtaining an abnormal vibration frequency set according to a feature recognition result, wherein the abnormal vibration frequency represents the frequency of occurrence of abnormal features, and the higher the abnormal vibration frequency is, the higher the battery temperature is represented; and carrying out abnormal feature recognition on the noise monitoring data set according to the associated noise factors to obtain an abnormal noise frequency set, wherein the higher the abnormal noise frequency is, the higher the battery temperature is represented.

The temperature prediction channel is a neural network model which can be subjected to iterative optimization in machine learning and is obtained through supervised training of a training data set, wherein the temperature prediction channel comprises an input layer, a plurality of hidden layers and an output layer, the input data of the input layer are abnormal vibration frequency and abnormal noise frequency, and the output data of the output layer are temperature prediction results. And acquiring a plurality of sample training data to perform supervision training on the temperature prediction channel, so as to obtain the temperature prediction channel which tends to be in a convergence state.

And then sequentially inputting the abnormal vibration frequency set and the abnormal noise frequency set into the temperature prediction channel, namely selecting a group of abnormal vibration frequency and abnormal noise frequency in the same battery neighborhood each time to input the abnormal vibration frequency and the abnormal noise frequency into the temperature prediction channel for temperature prediction, and obtaining a temperature prediction set. By obtaining the temperature prediction set, support is provided for mapping compensation of the next step temperature monitoring dataset.

Further, the fifth step of the present application comprises:

Carrying out power consumption analysis on the vehicle-mounted component based on the historical vehicle power consumption log, and determining frequent power consumption factors;

collecting power consumption related data according to the frequent power consumption factors, and predicting power consumption according to a collecting result to generate frequent load prediction data;

Reading vehicle navigation data, and carrying out vehicle-to-vehicle load prediction based on the vehicle navigation data to generate sudden load prediction data;

And obtaining the battery load prediction data according to the frequent load prediction data and the burst load prediction data.

Specifically, first, a plurality of historical vehicle power consumption logs including vehicle power consumption components and power consumption ratios are collected based on big data. Then, the vehicle-mounted component power consumption analysis is performed according to the plurality of historical vehicle power consumption logs, and the frequent power consumption factor is determined, wherein the frequent power consumption factor refers to a power consumption component with relatively large power consumption, for example: air conditioning systems, lighting systems, entertainment systems, and the like.

And collecting power consumption related data according to the frequent power consumption factors, for example: and then, carrying out power consumption prediction according to the power consumption related data acquisition result to obtain frequent load prediction data. For example: obtaining temperature difference data based on standard in-vehicle temperature data and in-vehicle real-time temperature data, and performing temperature adjustment load prediction according to the temperature difference data to obtain temperature adjustment load data; and acquiring the image data outside the vehicle through the vehicle-mounted image sensor, and carrying out illumination load prediction based on the image data outside the vehicle to generate illumination load data and the like.

Reading vehicle navigation data in a preset time window, wherein the preset time window is a smaller time period and can be set according to actual conditions, for example: and 3 minutes, then, carrying out vehicle-to-vehicle load prediction according to the vehicle navigation data, and generating sudden load prediction data. For example: a vehicle driving simulator can be constructed, and the vehicle navigation data is input into the vehicle driving simulator to simulate the load of a vehicle, for example: the front road section is an ascending road section, so that the load trend of the vehicle and the machine is increased. And further carrying out summation calculation on the frequent load prediction data and the burst load prediction data, and taking a summation result as the battery load prediction data. By obtaining the battery load prediction data, support is provided for mapping compensation of the next step of temperature monitoring dataset.

Further, the fifth step of the present application comprises:

configuring weight coefficients of temperature prediction data and temperature monitoring data, and carrying out weighted calculation on the temperature prediction set and the temperature monitoring data set based on the weight coefficients to obtain an updated temperature data set;

Carrying out load trend analysis based on the battery load prediction data, and determining a load trend coefficient;

and compensating the updated temperature data set according to the load trend coefficient to obtain the compensated temperature data set.

Specifically, the weight coefficients of the temperature prediction data and the temperature monitoring data are configured, wherein the weight coefficient of the temperature monitoring data is far larger than the weight coefficient of the temperature prediction data, the sum of the two weight coefficients is 1, and the client sets the weight coefficient based on the reliability of the temperature prediction data, wherein the higher the reliability is, the larger the corresponding weight coefficient is. And then, carrying out weighted calculation on the temperature prediction set and the temperature monitoring data set according to the weight coefficient to obtain an updated temperature data set.

And carrying out load trend analysis according to the battery load prediction data, wherein the larger the battery load prediction data is, the larger the trend of the battery load is, the larger the load trend coefficient is, and the load trend coefficient is determined. And further taking the load trend coefficient as weight, weighting and calculating the updated temperature data set, namely multiplying the load trend coefficient by the updated temperature data set, and taking the product of the load trend coefficient and the updated temperature data set as compensation temperature data to obtain the compensation temperature data set. By carrying out mapping compensation on the temperature monitoring data set according to the temperature prediction set and the battery load prediction data, the accuracy of obtaining the compensated temperature data set can be improved, and therefore support is provided for accurate heat dissipation control of a next step of target battery pack.

Further, the sixth step of the present application includes:

Acquiring heat dissipation basic data of the target battery pack, wherein the heat dissipation basic data comprises heat dissipation part types, heat dissipation part positions, heat dissipation part quantity and heat dissipation operation parameters;

building a heat dissipation topological network based on the heat dissipation basic data, and embedding the heat dissipation topological network into the heat dissipation control model;

generating a plurality of heat dissipation control decisions through a heat dissipation topology network of the heat dissipation control model with the aim of meeting the heat dissipation requirement of the compensation temperature data set;

and optimizing the plurality of heat dissipation control decisions to obtain the optimal heat dissipation control decision, wherein the optimal heat dissipation control decision is a heat dissipation scheme with the minimum comprehensive energy consumption.

Specifically, firstly, heat dissipation basic data of the target battery pack are obtained, wherein the heat dissipation basic data comprise heat dissipation part types, heat dissipation part positions, the number of heat dissipation parts and heat dissipation operation parameters, and the heat dissipation part types comprise heat dissipation fins, air cooling parts and liquid cooling parts; the interior of a vehicle generally comprises a plurality of cooling fins, a plurality of air cooling components and a liquid cooling component, wherein the cooling operation parameters of the air cooling component comprise parameters such as wind speed and the like, and the cooling operation parameters of the liquid cooling component comprise parameters such as liquid flow and the like.

And building a heat dissipation topological network according to the heat dissipation basic data, wherein the heat dissipation topological network is an integral heat dissipation framework of the target battery pack, and then embedding the heat dissipation topological network into a heat dissipation control model. And taking the heat dissipation requirement meeting the compensation temperature data set as a heat dissipation control purpose, and performing heat dissipation control decision enumeration through a heat dissipation topology network of the heat dissipation control model to obtain a plurality of heat dissipation control decisions.

And optimizing the plurality of heat dissipation control decisions by taking the lowest comprehensive energy consumption as an optimizing purpose, wherein the comprehensive energy consumption calculation can be obtained by building a prediction model based on a neural network to predict the energy consumption or building a twin simulation model to perform simulation operation, namely, the comprehensive energy consumption calculation is respectively performed on the plurality of heat dissipation control decisions to obtain the comprehensive energy consumption corresponding to the plurality of heat dissipation control decisions, and then the heat dissipation control decision with the lowest comprehensive energy consumption is selected to be set as an optimal heat dissipation control decision to obtain the optimal heat dissipation control decision.

Further, the seventh step of the present application includes:

Acquiring an adjacent compensation temperature data set of a next adjacent node, and performing temperature deviation mapping calculation on the compensation temperature data set based on the adjacent compensation temperature data set to obtain a plurality of temperature deviation data;

Marking temperature deviation data which are larger than a preset deviation threshold value in the plurality of temperature deviation data as abnormal temperature difference data, and obtaining a plurality of abnormal temperature difference data;

Acquiring a plurality of abnormal battery neighborhoods corresponding to the plurality of abnormal temperature difference data, and carrying out heat dissipation analysis on the plurality of abnormal temperature difference data and the plurality of abnormal battery neighborhoods through the heat dissipation control model to obtain a local heat dissipation control decision;

And carrying out optimization adjustment on the optimal heat dissipation control decision based on the local heat dissipation control decision, and executing heat dissipation control of the battery pack of the next adjacent node according to the adjustment heat dissipation control decision.

Specifically, after the heat dissipation control of the target battery pack is performed based on the optimal heat dissipation control decision, that is, when the next adjacent node performs the heat dissipation control of the target battery pack, firstly, an adjacent compensation temperature data set of the next adjacent node is obtained, and then, temperature deviation mapping calculation is performed on the adjacent compensation temperature data set and the compensation temperature data set, that is, temperature deviation data of each adjacent node of the battery neighborhood is calculated, so as to obtain a plurality of temperature deviation data.

Acquiring a preset deviation threshold value, wherein the preset deviation threshold value can be set automatically based on actual conditions, judging a plurality of temperature deviation data according to the preset deviation threshold value, and marking the temperature deviation data which is larger than the preset deviation threshold value in the plurality of temperature deviation data as abnormal temperature difference data to obtain a plurality of abnormal temperature difference data. And further acquiring a plurality of abnormal battery neighborhoods corresponding to the plurality of abnormal temperature difference data, inputting the plurality of abnormal temperature difference data and the plurality of abnormal battery neighborhoods into the heat dissipation control model, and carrying out heat dissipation control analysis aiming at meeting the heat dissipation requirements of the plurality of abnormal temperature difference data to obtain a local heat dissipation control decision, wherein the local heat dissipation control decision is a heat dissipation control decision which meets the heat dissipation requirements of the plurality of abnormal temperature difference data and has the lowest comprehensive energy consumption required by adjusting the optimal heat dissipation control decision.

And finally, carrying out optimization adjustment on the optimal heat dissipation control decision according to the local heat dissipation control decision, namely, only carrying out control adjustment on heat dissipation components related in the local heat dissipation control decision, and executing heat dissipation control of the battery pack of the next adjacent node according to the heat dissipation adjustment control decision.

By performing temperature difference calculation of adjacent nodes and locally adjusting the optimal heat dissipation control decision according to the temperature difference calculation result, the analysis time for adjusting the heat dissipation control decision can be saved, the consumption of calculation resources is reduced, and the timeliness and the accuracy of heat dissipation control of the battery pack are improved.

In summary, the vehicle-mounted power supply protection method based on multi-sensor monitoring provided by the application has the following technical effects:

1. The temperature sensor is arranged on the basis of the neighborhood of the battery to monitor the temperature of the battery pack, and the temperature monitoring data set is compensated through a plurality of dimensional data such as vibration monitoring, noise monitoring, battery load trend prediction and the like, so that the accuracy of obtaining the compensated temperature data set can be improved; meanwhile, a heat dissipation control model is constructed to carry out heat dissipation analysis on the compensation temperature data set, so that an optimal heat dissipation control decision is obtained, and the accuracy and the rationality of heat dissipation control decision setting can be improved; therefore, the comprehensiveness of the sensing monitoring data of the battery pack and the accuracy of data analysis can be improved, the technical aim of accurately predicting the temperature change of the battery pack is fulfilled, the problems of vehicle performance reduction, battery damage and the like caused by overheat of the battery pack are avoided, the heat dissipation of the battery pack is accurately controlled, and the technical effects of improving the running quality and the running safety of a vehicle are achieved.

2. The battery pack simulation model is constructed to perform battery clustering to obtain a plurality of battery neighborhoods, and then the temperature sensors are distributed according to the battery neighborhoods, so that the distribution accuracy and reliability of the temperature sensors can be improved, the acquisition accuracy of the battery pack temperature monitoring data can be improved, and a basis is provided for performing accurate heat dissipation control.

3. By carrying out mapping compensation on the temperature monitoring data set according to the temperature prediction set and the battery load prediction data, the temperature change state of the battery pack can be analyzed and judged from multiple dimensions, so that the accuracy and the comprehensiveness of obtaining the compensation temperature data set are further improved.

4. And the heat dissipation control model is constructed to carry out heat dissipation analysis on the compensation temperature data set to obtain an optimal heat dissipation control decision, so that heat dissipation energy consumption resources can be saved on the premise of guaranteeing heat dissipation requirements, and timeliness and accuracy of the optimal heat dissipation control decision can be improved, so that timeliness and accuracy of heat dissipation control of the battery pack are improved.

In a second embodiment, based on the same concept as the vehicle-mounted power supply protection method based on multi-sensor monitoring in the foregoing embodiment, the present application further provides a vehicle-mounted power supply protection system based on multi-sensor monitoring, referring to fig. 3, where the system includes:

the battery pack simulation model generation module 11 is used for performing simulation modeling based on basic data of a target battery pack to generate a battery pack simulation model;

The battery neighborhood determining module 12 is used for performing battery clustering through the battery pack simulation model, and determining a plurality of battery neighborhoods according to a clustering result;

The temperature monitoring network building module 13 is used for building a temperature monitoring network based on the temperature sensors distributed in the plurality of battery neighborhoods, and performing temperature monitoring of the target battery pack through the temperature monitoring network to obtain a temperature monitoring data set under a preset node;

The abnormal feature recognition module 14 is configured to collect a vibration monitoring data set and a noise monitoring data set of the target battery pack, perform abnormal feature recognition on the vibration monitoring data set and the noise monitoring data set based on a preset correlation factor, and generate a temperature prediction set according to a recognition result;

the mapping compensation module 15 is configured to obtain battery load prediction data, and perform mapping compensation on the temperature monitoring dataset based on the temperature prediction set and the battery load prediction data to obtain a compensated temperature dataset;

The optimal heat dissipation control decision obtaining module 16, where the optimal heat dissipation control decision obtaining module 16 is configured to perform heat dissipation analysis on the compensated temperature dataset through a heat dissipation control model to obtain an optimal heat dissipation control decision, and a heat dissipation topology network is embedded in the heat dissipation control model;

and a heat dissipation control module 17, wherein the heat dissipation control module 17 is configured to execute heat dissipation control of the target battery pack based on the optimal heat dissipation control decision.

Further, the battery neighborhood determination module 12 in the system is further configured to:

setting N battery load strategies, wherein N is an integer greater than 1;

based on the N battery load strategies, performing simulation operation through the battery pack simulation model to obtain N simulated temperature data sets, wherein the simulated temperature data have battery number identifications;

sequentially carrying out clustering division on the N simulated temperature data sets based on a preset temperature difference threshold value to obtain N clustering neighborhood sets;

And selecting N times of overlapping battery numbers in the N clustering neighborhood sets to construct a battery neighborhood, and obtaining the plurality of battery neighborhoods.

Further, the abnormal feature identification module 14 in the system is further configured to:

respectively arranging a vibration sensor and a sound sensor based on the plurality of battery neighborhoods;

and reading monitoring data of the vibration sensor and the sound sensor in a preset time window to obtain the vibration monitoring data set and the noise monitoring data set, wherein the preset time window is a time interval between the last adjacent node and the preset node.

Further, the abnormal feature identification module 14 in the system is further configured to:

taking the target battery pack as constraint, and searching and acquiring a plurality of historical battery overheat events;

Respectively carrying out vibration data identification and noise data identification based on the plurality of historical battery overheat events, and determining an associated vibration factor and an associated noise factor;

Performing abnormal characteristic recognition on the vibration monitoring data set according to the associated vibration factors, and determining an abnormal vibration frequency set;

carrying out abnormal feature recognition on the noise monitoring data set according to the associated noise factors, and determining an abnormal noise frequency set;

And carrying out temperature prediction on the abnormal vibration frequency set and the abnormal noise frequency set through a temperature prediction channel to obtain the temperature prediction set.

Further, the mapping compensation module 15 in the system is also configured to:

Carrying out power consumption analysis on the vehicle-mounted component based on the historical vehicle power consumption log, and determining frequent power consumption factors;

collecting power consumption related data according to the frequent power consumption factors, and predicting power consumption according to a collecting result to generate frequent load prediction data;

Reading vehicle navigation data, and carrying out vehicle-to-vehicle load prediction based on the vehicle navigation data to generate sudden load prediction data;

And obtaining the battery load prediction data according to the frequent load prediction data and the burst load prediction data.

Further, the mapping compensation module 15 in the system is also configured to:

configuring weight coefficients of temperature prediction data and temperature monitoring data, and carrying out weighted calculation on the temperature prediction set and the temperature monitoring data set based on the weight coefficients to obtain an updated temperature data set;

Carrying out load trend analysis based on the battery load prediction data, and determining a load trend coefficient;

and compensating the updated temperature data set according to the load trend coefficient to obtain the compensated temperature data set.

Further, the optimal heat dissipation control decision obtaining module 16 in the system is further configured to:

Acquiring heat dissipation basic data of the target battery pack, wherein the heat dissipation basic data comprises heat dissipation part types, heat dissipation part positions, heat dissipation part quantity and heat dissipation operation parameters;

building a heat dissipation topological network based on the heat dissipation basic data, and embedding the heat dissipation topological network into the heat dissipation control model;

generating a plurality of heat dissipation control decisions through a heat dissipation topology network of the heat dissipation control model with the aim of meeting the heat dissipation requirement of the compensation temperature data set;

and optimizing the plurality of heat dissipation control decisions to obtain the optimal heat dissipation control decision, wherein the optimal heat dissipation control decision is a heat dissipation scheme with the minimum comprehensive energy consumption.

Further, the heat dissipation control module 17 in the system is further configured to:

Acquiring an adjacent compensation temperature data set of a next adjacent node, and performing temperature deviation mapping calculation on the compensation temperature data set based on the adjacent compensation temperature data set to obtain a plurality of temperature deviation data;

Marking temperature deviation data which are larger than a preset deviation threshold value in the plurality of temperature deviation data as abnormal temperature difference data, and obtaining a plurality of abnormal temperature difference data;

Acquiring a plurality of abnormal battery neighborhoods corresponding to the plurality of abnormal temperature difference data, and carrying out heat dissipation analysis on the plurality of abnormal temperature difference data and the plurality of abnormal battery neighborhoods through the heat dissipation control model to obtain a local heat dissipation control decision;

And carrying out optimization adjustment on the optimal heat dissipation control decision based on the local heat dissipation control decision, and executing heat dissipation control of the battery pack of the next adjacent node according to the adjustment heat dissipation control decision.

In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the difference from other embodiments, so that a multi-sensor monitoring-based vehicle-mounted power supply protection method and a specific example in the first embodiment are equally applicable to a multi-sensor monitoring-based vehicle-mounted power supply protection system in the first embodiment, and by the foregoing detailed description of a multi-sensor monitoring-based vehicle-mounted power supply protection method, those skilled in the art can clearly know that a multi-sensor monitoring-based vehicle-mounted power supply protection system in the first embodiment is omitted herein for brevity of the specification. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant points refer to the description of the method section.

The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

It will be apparent to those skilled in the art that various modifications and variations can be made to the present application without departing from the spirit or scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the present application and the equivalent techniques thereof, the present application is also intended to include such modifications and variations.

Claims (8)

1. The vehicle-mounted power supply protection method based on multi-sensor monitoring is characterized by comprising the following steps of:

performing simulation modeling based on basic data of a target battery pack to generate a battery pack simulation model;

Performing battery clustering through the battery pack simulation model, and determining a plurality of battery neighborhoods according to clustering results;

Setting up a temperature monitoring network based on the plurality of battery neighborhood layout temperature sensors, and executing temperature monitoring of the target battery pack through the temperature monitoring network to obtain a temperature monitoring data set under a preset node;

Collecting a vibration monitoring data set and a noise monitoring data set of the target battery pack, carrying out abnormal characteristic identification on the vibration monitoring data set and the noise monitoring data set based on preset association factors, and generating a temperature prediction set according to an identification result;

acquiring battery load prediction data, and performing mapping compensation on the temperature monitoring data set based on the temperature prediction set and the battery load prediction data to obtain a compensated temperature data set;

Carrying out heat dissipation analysis on the compensation temperature data set through a heat dissipation control model to obtain an optimal heat dissipation control decision, wherein a heat dissipation topology network is embedded in the heat dissipation control model;

Executing heat dissipation control of the target battery pack based on the optimal heat dissipation control decision;

the battery clustering is performed through the battery pack simulation model, and a plurality of battery neighborhoods are determined according to the clustering result, including:

setting N battery load strategies, wherein N is an integer greater than 1;

based on the N battery load strategies, performing simulation operation through the battery pack simulation model to obtain N simulated temperature data sets, wherein the simulated temperature data have battery number identifications;

sequentially carrying out clustering division on the N simulated temperature data sets based on a preset temperature difference threshold value to obtain N clustering neighborhood sets;

And selecting N times of overlapping battery numbers in the N clustering neighborhood sets to construct a battery neighborhood, and obtaining the plurality of battery neighborhoods.

2. The method of claim 1, wherein the acquiring the vibration monitoring dataset and the noise monitoring dataset of the target battery pack further comprises, before:

respectively arranging a vibration sensor and a sound sensor based on the plurality of battery neighborhoods;

and reading monitoring data of the vibration sensor and the sound sensor in a preset time window to obtain the vibration monitoring data set and the noise monitoring data set, wherein the preset time window is a time interval between the last adjacent node and the preset node.

3. The method of claim 1, wherein the identifying abnormal characteristics of the vibration monitoring dataset and the noise monitoring dataset based on a preset correlation factor, generating a temperature prediction set based on the identification result, comprises:

taking the target battery pack as constraint, and searching and acquiring a plurality of historical battery overheat events;

Respectively carrying out vibration data identification and noise data identification based on the plurality of historical battery overheat events, and determining an associated vibration factor and an associated noise factor;

Performing abnormal characteristic recognition on the vibration monitoring data set according to the associated vibration factors, and determining an abnormal vibration frequency set;

carrying out abnormal feature recognition on the noise monitoring data set according to the associated noise factors, and determining an abnormal noise frequency set;

And carrying out temperature prediction on the abnormal vibration frequency set and the abnormal noise frequency set through a temperature prediction channel to obtain the temperature prediction set.

4. The method of claim 1, wherein the map compensating the temperature monitoring dataset based on the temperature prediction set and the battery load prediction data further comprises:

Carrying out power consumption analysis on the vehicle-mounted component based on the historical vehicle power consumption log, and determining frequent power consumption factors;

collecting power consumption related data according to the frequent power consumption factors, and predicting power consumption according to a collecting result to generate frequent load prediction data;

Reading vehicle navigation data, and carrying out vehicle-to-vehicle load prediction based on the vehicle navigation data to generate sudden load prediction data;

And obtaining the battery load prediction data according to the frequent load prediction data and the burst load prediction data.

5. The method of claim 1, wherein mapping the temperature monitoring dataset based on the temperature prediction set, battery load prediction data, to obtain a compensated temperature dataset comprises:

configuring weight coefficients of temperature prediction data and temperature monitoring data, and carrying out weighted calculation on the temperature prediction set and the temperature monitoring data set based on the weight coefficients to obtain an updated temperature data set;

Carrying out load trend analysis based on the battery load prediction data, and determining a load trend coefficient;

and compensating the updated temperature data set according to the load trend coefficient to obtain the compensated temperature data set.

6. The method of claim 5, wherein the performing a heat dissipation analysis on the compensated temperature dataset by a heat dissipation control model to obtain an optimal heat dissipation control decision, the heat dissipation control model having a heat dissipation topology network embedded therein, comprises:

Acquiring heat dissipation basic data of the target battery pack, wherein the heat dissipation basic data comprises heat dissipation part types, heat dissipation part positions, heat dissipation part quantity and heat dissipation operation parameters;

building a heat dissipation topological network based on the heat dissipation basic data, and embedding the heat dissipation topological network into the heat dissipation control model;

generating a plurality of heat dissipation control decisions through a heat dissipation topology network of the heat dissipation control model with the aim of meeting the heat dissipation requirement of the compensation temperature data set;

and optimizing the plurality of heat dissipation control decisions to obtain the optimal heat dissipation control decision, wherein the optimal heat dissipation control decision is a heat dissipation scheme with the minimum comprehensive energy consumption.

7. The method of claim 1, wherein the performing the thermal dissipation control of the target battery pack based on the optimal thermal dissipation control decision further comprises:

Acquiring an adjacent compensation temperature data set of a next adjacent node, and performing temperature deviation mapping calculation on the compensation temperature data set based on the adjacent compensation temperature data set to obtain a plurality of temperature deviation data;

Marking temperature deviation data which are larger than a preset deviation threshold value in the plurality of temperature deviation data as abnormal temperature difference data, and obtaining a plurality of abnormal temperature difference data;

Acquiring a plurality of abnormal battery neighborhoods corresponding to the plurality of abnormal temperature difference data, and carrying out heat dissipation analysis on the plurality of abnormal temperature difference data and the plurality of abnormal battery neighborhoods through the heat dissipation control model to obtain a local heat dissipation control decision;

And carrying out optimization adjustment on the optimal heat dissipation control decision based on the local heat dissipation control decision, and executing heat dissipation control of the battery pack of the next adjacent node according to the adjustment heat dissipation control decision.

8. An on-board power protection system based on multisensory monitoring, characterized by the steps for implementing the method of any one of claims 1 to 7, said system comprising:

The battery pack simulation model generation module is used for carrying out simulation modeling based on basic data of a target battery pack to generate a battery pack simulation model;

the battery neighborhood determining module is used for carrying out battery clustering through the battery pack simulation model and determining a plurality of battery neighborhoods according to a clustering result;

The temperature monitoring network building module is used for building a temperature monitoring network based on the plurality of battery neighborhood layout temperature sensors, and executing temperature monitoring of the target battery pack through the temperature monitoring network to acquire a temperature monitoring data set under a preset node;

the abnormal characteristic recognition module is used for collecting a vibration monitoring data set and a noise monitoring data set of the target battery pack, carrying out abnormal characteristic recognition on the vibration monitoring data set and the noise monitoring data set based on a preset correlation factor, and generating a temperature prediction set according to recognition results;

the mapping compensation module is used for acquiring battery load prediction data, and performing mapping compensation on the temperature monitoring data set based on the temperature prediction set and the battery load prediction data to obtain a compensated temperature data set;

The optimal heat dissipation control decision obtaining module is used for carrying out heat dissipation analysis on the compensation temperature data set through a heat dissipation control model to obtain an optimal heat dissipation control decision, and a heat dissipation topology network is embedded in the heat dissipation control model;

the heat dissipation control module is used for executing heat dissipation control of the target battery pack based on the optimal heat dissipation control decision;

the battery neighborhood determination module is further configured to:

setting N battery load strategies, wherein N is an integer greater than 1;

based on the N battery load strategies, performing simulation operation through the battery pack simulation model to obtain N simulated temperature data sets, wherein the simulated temperature data have battery number identifications;

sequentially carrying out clustering division on the N simulated temperature data sets based on a preset temperature difference threshold value to obtain N clustering neighborhood sets;

And selecting N times of overlapping battery numbers in the N clustering neighborhood sets to construct a battery neighborhood, and obtaining the plurality of battery neighborhoods.