A fault detection system and method for new energy vehicle multi-source data fusion

Abstract

The application provides a new energy automobile multi-source data fusion fault detection system and method, a battery charge abnormality index of a battery state of charge is determined through consistency characteristics of cell voltages and gradient characteristics of battery pack currents in a new energy automobile; a temperature difference abnormality characteristic of the new energy automobile is determined through temperature rising rates of each cell measurement point and temperature difference characteristics between each cell measurement point, a battery temperature runaway risk index in the new energy automobile is determined through the temperature difference abnormality characteristic and a flow rate descending gradient in cooling liquid flow rate information; the battery charge abnormality index and the runaway risk index are dependently fused to obtain an abnormality probability of the battery in the new energy automobile, when the abnormality probability is greater than a new energy automobile battery fault threshold, a fault warning is triggered, and a fault cell is located through a characteristic contribution degree of each cell. Based on the above scheme, the fusion detection of temperature characteristics and electrical characteristics of the battery fault in the new energy automobile can be realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of new energy vehicles, and more particularly to a new energy vehicle multi-source data fusion fault detection system and method. BACKGROUND

[0002] A new energy vehicle refers to a vehicle that uses unconventional vehicle fuel as a power source, and currently mainly adopts electric power driving as the core technology route. The new energy vehicle stores electric energy through a high-capacity battery pack and drives a motor to provide power, achieving zero emission, low noise, and high energy efficiency operation. The new energy vehicle integrates advanced battery management systems, intelligent temperature control technology, and energy recovery systems, and is a key solution to energy crisis and environmental pollution, representing the future direction of the transformation of the automobile industry to electrification and intelligence.

[0003] The battery fault detection method of the traditional new energy vehicle usually only relies on a single dimension such as voltage or temperature to make a judgment, which has significant limitations. For example, simply monitoring temperature rise cannot effectively distinguish between normal fast charging caused heat and heat runaway precursors; relying only on voltage changes makes it difficult to identify potential thermal risks caused by cooling system failure, and the alarm mechanism based on simple threshold is often triggered when the fault has fully manifested (for example: a certain point temperature is extremely high or voltage drops sharply), the early warning time window is extremely short, and it is difficult to provide enough advance response time for the vehicle driver or the battery management system, resulting in delayed risk intervention and insufficient safety protection capability. Therefore, how to realize the fusion detection of the temperature characteristics and electrical characteristics of the battery fault in the new energy vehicle has become a difficult problem faced by the industry. SUMMARY

[0004] The present application provides a new energy vehicle multi-source data fusion fault detection system and method, which can realize the fusion detection of the temperature characteristics and electrical characteristics of the battery fault in the new energy vehicle.

[0005] In a first aspect, the present application provides a new energy vehicle multi-source data fusion fault detection method, comprising:

[0006] Collecting electrical operation information, cooling liquid flow information, and temperature information of each cell measurement point of the new energy vehicle through a vehicle-mounted sensor;

[0007] Extracting a consistency feature of cell voltage and a gradient feature of battery pack current from the electrical operation information, and determining a state of charge abnormality index of the battery state of charge in the new energy vehicle through the consistency feature and the gradient feature;

[0008] extract temperature rising rates of each battery cell measurement point in the new energy vehicle from the various temperature information, determine a temperature difference abnormality feature of the new energy vehicle through all the temperature rising rates and temperature differences between each battery cell measurement point, and determine a runaway risk index of the battery temperature in the new energy vehicle through the temperature difference abnormality feature and a flow rate descending gradient in the coolant flow information;

[0009] perform dependent fusion on the charge abnormality index and the runaway risk index through an operation dependency relationship between the battery temperature and the battery state of charge in the new energy vehicle, obtain an abnormality probability of the battery in the new energy vehicle, trigger a fault warning when the abnormality probability is greater than a fault threshold of the battery of the new energy vehicle, and locate a fault battery cell through a feature contribution degree of each battery cell.

[0010] In some embodiments, extracting the consistency feature of the battery cell voltage and the gradient feature of the battery pack current from the electrical operation information specifically includes:

[0011] obtaining battery cell voltage data and battery pack current data from the electrical operation information;

[0012] calculating statistical dispersion at the same sampling time according to the battery cell voltage data to obtain the consistency feature of the battery cell voltage;

[0013] performing discrete difference operation on the battery pack current data to obtain the gradient feature of the battery pack current.

[0014] In some embodiments, determining the charge abnormality index of the battery state of charge in the new energy vehicle through the consistency feature and the gradient feature specifically includes:

[0015] obtaining a state of charge feature of the new energy vehicle under normal working conditions;

[0016] calculating a decision distance between the state of charge feature and the consistency feature;

[0017] calculating a gradient distance between the state of charge feature and the gradient feature;

[0018] determining the charge abnormality index of the battery state of charge in the new energy vehicle through the decision distance and the gradient distance.

[0019] In some embodiments, extracting the temperature rising rates of each battery cell measurement point in the new energy vehicle from the various temperature information specifically includes:

[0020] For each battery cell measurement point in the new energy vehicle, filtering and denoising the temperature information of the battery cell measurement point to obtain denoised temperature information of the battery cell measurement point;

[0021] Extract the temperature rise rate of the battery measurement point from the denoised temperature information, and then obtain the temperature rise rate of each battery measurement point in the new energy vehicle.

[0022] In some embodiments, the temperature difference anomaly feature of the new energy vehicle is determined by all temperature rise rates and temperature difference features between each battery measurement point, and specifically includes:

[0023] Determine the temperature difference feature between each battery measurement point;

[0024] Determine the temperature difference value between each battery measurement point by all temperature rise rates;

[0025] Combine all temperature difference values based on each temperature difference feature to obtain the temperature difference anomaly feature of the new energy vehicle.

[0026] In some embodiments, the out-of-control risk index of the battery temperature in the new energy vehicle is determined by the temperature difference anomaly feature and the flow drop gradient in the cooling liquid flow information, and specifically includes:

[0027] Calculate the flow drop gradient in the cooling liquid flow information, and obtain the rated flow of the cooling liquid flow in the new energy vehicle;

[0028] Determine the flow anomaly index of the cooling liquid flow in the new energy vehicle by the flow drop gradient and the rated flow;

[0029] Determine the out-of-control risk index of the battery temperature in the new energy vehicle according to the flow anomaly index and the temperature difference anomaly feature.

[0030] In some embodiments, the abnormal probability of the battery in the new energy vehicle is obtained by dependent fusion of the charge abnormal index and the out-of-control risk index based on the operating dependency relationship between the battery temperature and the battery state of charge in the new energy vehicle, and specifically includes:

[0031] Initialize a Bayesian network model based on joint feature reasoning;

[0032] The operating dependency relationship between the battery temperature and the battery state of charge in the new energy vehicle is used as a directed edge in the Bayesian network model;

[0033] The charge abnormal index is used as a first observation node in the Bayesian network model;

[0034] The out-of-control risk index is used as a second observation node in the Bayesian network model;

[0035] Use the Bayesian network model to perform anomaly reasoning on the battery health state of the new energy vehicle to obtain the abnormal probability of the battery in the new energy vehicle.

[0036] In a second aspect, the present application provides a fault detection system for new energy vehicle multi-source data fusion, comprising a fault alarm unit, wherein the fault alarm unit comprises:

[0037] a collection module, configured to collect electrical operation information, cooling liquid flow information and temperature information of each battery cell measurement point of the new energy vehicle through a vehicle-mounted sensor;

[0038] a processing module, configured to extract consistency features of battery cell voltages and gradient features of battery pack currents from the electrical operation information, and determine a charge abnormality index of a battery state of charge in the new energy vehicle according to the consistency features and the gradient features;

[0039] the processing module is further configured to extract temperature rising rates of each battery cell measurement point in the new energy vehicle from the temperature information, determine a temperature difference abnormality feature of the new energy vehicle according to all the temperature rising rates and temperature difference features between each battery cell measurement point, and determine a risk index of a battery temperature runaway in the new energy vehicle according to the temperature difference abnormality feature and a flow rate descending gradient in the cooling liquid flow information;

[0040] an execution module, configured to perform dependent fusion on the charge abnormality index and the risk index of the battery temperature runaway in the new energy vehicle according to an operation dependence between the battery temperature and the battery state of charge, to obtain an abnormality probability of the battery in the new energy vehicle, trigger a fault warning when the abnormality probability is greater than a fault threshold of the battery of the new energy vehicle, and locate a fault battery cell according to a feature contribution degree of each battery cell.

[0041] In a third aspect, the present application provides a computer device, comprising a memory and a processor, wherein the memory is configured to store a computer program, and the processor is configured to call and run the computer program from the memory, so that the computer device executes the new energy vehicle multi-source data fusion fault detection method described above.

[0042] In a fourth aspect, the present application provides a computer readable storage medium, wherein the computer readable storage medium stores instructions or codes, and when the instructions or codes are run on a computer, the computer executes the new energy vehicle multi-source data fusion fault detection method described above.

[0043] The technical scheme provided by the embodiments of the present application has the following beneficial effects:

[0044] The application provides a new energy vehicle multi-source data fusion fault detection system and method, wherein the electrical operation information, cooling liquid flow information and temperature information of each battery cell measurement point of the new energy vehicle are collected through the vehicle-mounted sensor; the consistency feature of the battery cell voltage and the gradient feature of the battery pack current are extracted from the electrical operation information, the abnormal charge index of the battery state of charge in the new energy vehicle is determined through the consistency feature and the gradient feature; the temperature rising rate of each battery cell measurement point in the new energy vehicle is extracted from each temperature information, the temperature difference abnormal feature of the new energy vehicle is determined through all the temperature rising rates and the temperature difference feature between each battery cell measurement point, the risk index of the battery temperature out-of-control in the new energy vehicle is determined through the temperature difference abnormal feature and the flow drop gradient in the cooling liquid flow information; the abnormal probability of the battery in the new energy vehicle is obtained by dependent fusion of the abnormal charge index and the risk index of the battery temperature out-of-control through the operation dependence between the battery temperature and the battery state of charge in the new energy vehicle, when the abnormal probability is greater than the fault threshold of the battery of the new energy vehicle, the fault warning is triggered, and the fault battery cell is located through the feature contribution degree of each battery cell.

[0045] Therefore, in the present application, when the abnormal probability is greater than the fault threshold of the new energy vehicle battery, a fault warning is triggered, and the fault cell is located through the feature contribution degree of each cell. First, the determination of the charge abnormality index can realize multi-dimensional abnormal perception of the battery electrical state, thereby providing key electrical feature input for fusion detection. By extracting the consistency feature of the cell voltage, the new energy vehicle can effectively identify the imbalance inside the battery pack. At the same time, by calculating the gradient feature of the battery pack current, the current mutation in the charging and discharging process can be sensitively captured, and external circuit failure or load abnormality can be identified. The cell voltage feature is combined with the current gradient feature and the state of charge feature under normal working conditions for distance quantization, and finally fused into the charge abnormality index. The charge abnormality index not only reflects the instantaneous abnormality of voltage and current, but also reflects the dynamic change trend, providing a quantitative and reliable electrical abnormality degree evaluation for the system. Then, the determination of the out-of-control risk index can realize multi-source abnormality fusion judgment of the battery thermal state, thereby providing key thermal feature input for fusion detection. By analyzing the temperature rise rate of each cell, the system can early identify the local overheating trend. By calculating the temperature difference feature between cells, thermal diffusion abnormalities or uneven cooling problems can be found. Combined with the cooling fluid flow rate gradient, internal heat production abnormalities and external cooling failure can be further distinguished. The temperature rise feature, temperature distribution feature and cooling system state feature are fused to finally generate the out-of-control risk index. The out-of-control risk index comprehensively reflects the overall state of the battery thermal management, which can not only capture the early signs of internal thermal runaway, but also identify the degradation of external cooling function, breaking through the limitations of traditional single temperature threshold judgment, and realizing multi-source, dynamic and correlation evaluation of thermal related faults, thereby providing sufficient thermal state information support for cross-domain fusion with electrical features. BRIEF DESCRIPTION OF DRAWINGS

[0046] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0047] Figure 1 is an exemplary flowchart of a new energy vehicle multi-source data fusion fault detection method according to some embodiments of the present application;

[0048] Figure 2 is a flowchart of determining the out-of-control risk index according to some embodiments of the present application;

[0049] Figure 3is a structural schematic diagram of a fault alarm unit according to some embodiments of the present application;

[0050] Figure 4 is a structural schematic diagram of a computer device implementing a new energy vehicle multi-source data fusion fault detection method according to some embodiments of the present application. DETAILED DESCRIPTION

[0051] In order to better understand the technical solutions of the present application, the technical solutions of the present application will be described in detail below in combination with the accompanying drawings and specific embodiments.

[0052] Reference Figure 1 The figure is an exemplary flowchart of a new energy vehicle multi-source data fusion fault detection method according to some embodiments of the present application, which mainly includes the following steps:

[0053] In step 101, the electrical operation information, cooling liquid flow information and temperature information of each measurement point of the electric core of the new energy vehicle are collected by the vehicle-mounted sensor.

[0054] It should be noted that in the present application, the electrical operation information refers to a data set reflecting the internal electrical characteristics of the new energy vehicle power battery pack in the working state, the cooling liquid flow information is a physical quantity information for quantifying the strength of the cooling liquid circulation in the battery thermal management system; and the temperature information is a measurement data set sensing the temperature condition of the electric core itself.

[0055] In a specific implementation, first, the voltage of each battery cell is measured by a high-voltage differential sensor connected to the positive and negative poles of the battery pack every fixed time interval (1 second by default), so that the set of all battery cell voltages is taken as battery cell voltage data; the battery pack current of the new energy vehicle is collected by a Hall current sensor connected in series in the main circuit every fixed time interval (1 second by default), so that the set of all battery pack currents is taken as battery pack current data, and the set of battery cell voltage data and battery pack current data is taken as the electrical operation information of the new energy vehicle; then, the cooling liquid flow rate is collected by a turbine flow sensor installed on the outlet pipeline of the cooling liquid circulation loop of the battery pack; when the cooling liquid flows through the sensor, it drives the magnetic turbine inside to rotate, and the magnet on the turbine periodically sweeps across the Hall element inside the sensor, thereby generating a series of electric pulse signals whose frequency is proportional to the flow rate; the pulse signals are transmitted to the signal processing unit of the battery management system, the counter inside the unit accurately measures the frequency of the pulses, and the pulse frequency value is converted into the real-time cooling liquid volume flow rate value by querying the frequency-flow conversion curve calibrated when the sensor was manufactured, so that the set of all real-time cooling liquid volume flow rate values is taken as the cooling liquid flow rate information of the new energy vehicle; finally, for each battery cell measurement point in the new energy vehicle, the temperature information of the battery cell measurement point is collected by a temperature sensor every fixed time interval (1 second by default), and the temperature information of each battery cell measurement point in the new energy vehicle can be obtained in the above manner.

[0056] In step 102, the consistency feature of the battery cell voltage and the gradient feature of the battery pack current are extracted from the electrical operation information, and the state of charge abnormality index of the battery in the new energy vehicle is determined based on the consistency feature and the gradient feature.

[0057] In some embodiments, the extraction of the consistency feature of the battery cell voltage and the gradient feature of the battery pack current from the electrical operation information can be achieved by the following steps:

[0058] The battery cell voltage data and the battery pack current data are obtained from the electrical operation information.

[0059] The statistical dispersion at the same sampling time is calculated based on the battery cell voltage data, and the consistency feature of the battery cell voltage is obtained.

[0060] The discrete difference operation is performed on the battery pack current data, and the gradient feature of the battery pack current is obtained.

[0061] It should be noted that, in this application, the gradient feature is a feature of the instantaneous change rate when the battery pack experiences a sharp change; the consistency feature is a statistical quantity for quantifying the difference between all battery cell voltage values at the same time, and the consistency feature is a core index for measuring the consistency of the battery pack.

[0062] In a specific implementation, first, the cell voltage data and the battery pack current data are obtained from the electrical operation information; then, the cell voltages of the cells at the same sampling time are obtained from the cell voltage data, the standard deviations of all the cell voltages at the sampling times are calculated as the statistical dispersions of the cells at the corresponding sampling times, and the mean of all the statistical dispersions is taken as the consistency feature of the cell voltages; finally, the difference between the battery pack currents at adjacent sampling times is calculated as a discrete difference value, and the mean of all the discrete difference values is taken as the gradient feature of the battery pack current.

[0063] In some embodiments, determining the charge abnormality index of the battery state of charge in the new energy vehicle through the consistency feature and the gradient feature can be implemented by the following steps:

[0064] obtaining the charge state feature of the new energy vehicle under normal working conditions;

[0065] calculating a decision distance between the charge state feature and the consistency feature;

[0066] calculating a gradient distance between the charge state feature and the gradient feature;

[0067] determining the charge abnormality index of the battery state of charge in the new energy vehicle through the decision distance and the gradient distance.

[0068] It should be noted that, in the present application, the charge abnormality index is a comprehensive evaluation value of the current health degree of the battery system; the charge state feature is a reference data set for describing the normal fluctuation range of the electrical characteristics of the battery system under the healthy state, and the charge abnormality index is a reference benchmark for judging whether the current state is abnormal; the decision distance is a quantitative distance value reflecting the degree of deviation of the current battery voltage consistency state from the normal range; and the gradient distance is a quantitative distance value for measuring the abnormal degree of the current change.

[0069] In a specific implementation, first, the voltage consistency value and the current gradient value under normal working conditions are obtained from the center console of the new energy vehicle; second, the absolute value of the difference between the voltage consistency value in the charge state feature and the consistency feature is taken as the decision distance; then, the difference between the current gradient value in the charge state feature and the gradient feature is taken as the gradient distance; finally, the mean of the normalized decision distance and the normalized gradient distance is taken as the charge abnormality index of the battery state of charge in the new energy vehicle.

[0070] In step 103, the temperature rising rate of each battery cell measuring point in the new energy vehicle is extracted from the respective temperature information, the temperature difference abnormality feature of the new energy vehicle is determined through all the temperature rising rates and the temperature difference features between the respective battery cell measuring points, and the out-of-control risk index of the battery temperature in the new energy vehicle is determined through the temperature difference abnormality feature and the flow drop gradient in the cooling liquid flow information.

[0071] In some embodiments, the temperature rising rate of each battery cell measuring point in the new energy vehicle can be extracted from the respective temperature information by the following steps:

[0072] For each battery cell measuring point in the new energy vehicle, the temperature information of the battery cell measuring point is filtered and denoised to obtain the denoised temperature information of the battery cell measuring point.

[0073] The temperature rising rate of the battery cell measuring point is extracted from the denoised temperature information, and then the temperature rising rate of each battery cell measuring point in the new energy vehicle is obtained.

[0074] It should be noted that in this application, the temperature rising rate represents the speed of temperature change per unit time; in specific implementation, firstly, for each battery cell measuring point in the new energy vehicle, a first-order low-pass filter is applied to the temperature information of the battery cell measuring point to filter and denoise to eliminate measurement noise, so that the filtered and denoised temperature information is used as the denoised temperature information of the battery cell measuring point, and the denoised temperature information refers to the temperature data obtained after filtering by a signal processing algorithm; then, the mean value of the difference between the temperatures at adjacent collection times in the denoised temperature information is calculated as the temperature rising rate of the battery cell measuring point, and the temperature rising rate of each battery cell measuring point in the new energy vehicle can be obtained by the above method.

[0075] In some embodiments, the temperature difference abnormality feature of the new energy vehicle can be determined by all the temperature rising rates and the temperature difference features between the respective battery cell measuring points by the following steps:

[0076] The temperature difference features between the respective battery cell measuring points are determined.

[0077] The temperature rising difference values between the respective battery cell measuring points are determined through all the temperature rising rates.

[0078] All the temperature rising difference values are combined based on the respective temperature difference features to obtain the temperature difference abnormality feature of the new energy vehicle.

[0079] It should be noted that in the present application, the temperature difference abnormality feature is a comprehensive index quantifying the abnormality degree of the temperature distribution and the temperature rising behavior of the entire battery pack; the temperature rising difference value refers to the difference in temperature rising rate per unit time of different battery cells; the temperature difference feature refers to the difference degree of temperature values between different battery cell measurement points at the same time, and the temperature difference feature can be used to reflect the non-uniformity of the temperature distribution in the battery pack.

[0080] In a specific implementation, first, the temperature data of all battery cells at the same sampling time is obtained, the temperature difference value between each two battery cells is calculated to form a difference matrix, and the set of standard deviation and range in the difference matrix is extracted as the temperature difference feature at the time, the characteristic values at multiple sampling times are processed by sliding average to eliminate instantaneous fluctuation interference, and thus the temperature difference feature between the battery cell measurement points can be obtained; then, the difference between the temperature rising rates of the battery cell measurement points is calculated as the temperature rising difference value, and thus the temperature rising difference value between the battery cell measurement points can be obtained; finally, the temperature difference features at different time periods and the corresponding temperature rising difference values are normalized to eliminate the dimension influence, the normalized temperature difference features and temperature rising difference values are aligned by time and composed into a feature vector, and the second norm (i.e., the Euclidean norm) of the feature vector is calculated as the temperature difference abnormality feature of the new energy vehicle.

[0081] In some embodiments, the temperature rising difference value is determined by the temperature difference abnormality feature and the flow drop gradient in the cooling liquid flow information, and the temperature rising difference value is used to determine the out-of-control risk index of the battery temperature in the new energy vehicle. Figure 2 The figure is a flowchart for determining the out-of-control risk index in some embodiments of the present application, and the out-of-control risk index in the present embodiment can be determined by the following steps:

[0082] In step 1031, the flow drop gradient in the cooling liquid flow information is calculated, and the rated flow of the cooling liquid flow in the new energy vehicle is obtained.

[0083] In step 1032, the flow abnormality index of the cooling liquid flow in the new energy vehicle is determined by the flow drop gradient and the rated flow.

[0084] In step 1033, the out-of-control risk index of the battery temperature in the new energy vehicle is determined according to the flow abnormality index and the temperature difference abnormality feature.

[0085] It should be noted that in the present application, the risk index of out-of-control is a comprehensive risk assessment value for quantifying the probability of battery thermal runaway; the flow reduction gradient is a dynamic indicator for quantifying the severity of cooling system performance degradation or blockage; the rated flow is the standard flow value set by the new energy vehicle cooling system under normal working conditions; the flow anomaly index is a quantitative value for representing the abnormal degree of cooling liquid flow under the real-time working state of the cooling system.

[0086] In specific implementation, first, the difference between the cooling flow of adjacent sampling time in the cooling liquid flow information is calculated as the flow reduction gradient, and the rated flow of the cooling liquid flow in the new energy vehicle is obtained from the center console of the new energy vehicle; then, the ratio of the flow reduction gradient to the rated flow is taken as the flow anomaly index of the cooling liquid flow in the new energy vehicle; finally, the flow anomaly index and the temperature difference anomaly feature are standardized to ensure that they are in the same numerical order of magnitude, and the weight coefficients of the flow anomaly index and the temperature difference anomaly feature are set according to expert experience combined with historical fault data, and the weight of the temperature difference anomaly feature is usually slightly higher because it more directly reflects the precursor of thermal runaway, and the weighted sum of the flow anomaly index and the temperature difference anomaly feature is calculated as the risk index of the battery temperature in the new energy vehicle by using a linear weighting fusion algorithm.

[0087] In step 104, the abnormal probability of the battery in the new energy vehicle is obtained by dependent fusion of the charge abnormality index and the risk index of out-of-control through the operational dependence between the battery temperature and the state of charge of the battery, and when the abnormal probability is greater than the fault threshold of the battery of the new energy vehicle, a fault warning is triggered, and the fault cell is located through the feature contribution degree of each cell.

[0088] In some embodiments, the abnormal probability of the battery in the new energy vehicle can be obtained by dependent fusion of the charge abnormality index and the risk index of out-of-control through the operational dependence between the battery temperature and the state of charge of the battery by using the following steps:

[0089] Initialize a Bayesian network model based on joint feature inference;

[0090] The operational dependence between the battery temperature and the state of charge of the battery in the new energy vehicle is taken as a directed edge in the Bayesian network model;

[0091] The charge abnormality index is taken as a first observation node in the Bayesian network model;

[0092] The risk index of out-of-control is taken as a second observation node in the Bayesian network model;

[0093] The abnormal probability of the battery in the new energy vehicle is obtained by using the Bayesian network model to perform abnormal inference on the battery health state of the new energy vehicle.

[0094] It should be noted that in the present application, the Bayesian network model is an uncertainty reasoning tool based on a probabilistic graphical model, which visually expresses the conditional dependency between variables by using a directed acyclic graph, and abstracts the physical dependency between the battery temperature and the state of charge (for example, the temperature rise caused by high-rate discharge) as a directed edge between nodes, and quantifies the correlation strength by a conditional probability table, and the charge abnormality index and the out-of-control risk index are injected into the corresponding observation nodes as input evidence, and the network conducts probability propagation according to Bayes' theorem: starting from the real-time evidence of the child node (observation index), the posterior probability of the parent node (battery health state) being abnormal is calculated in reverse as the abnormal probability of the battery in the new energy vehicle, thereby realizing the probabilistic fusion of multi-source heterogeneous features, and the final output abnormal probability contains both the statistical confidence under data driving and the causal cognition of the battery system operation mechanism, thereby significantly improving the reliability and interpretability of fault diagnosis.

[0095] In some embodiments, when the abnormal probability is greater than the fault threshold of the new energy vehicle battery, a fault warning is triggered, and the fault cell positioning can be realized in the following manner: when the battery abnormal probability calculated by the Bayesian network model exceeds the preset fault threshold, the system immediately triggers a fault warning and starts a fault cell positioning process: first, the warning signal is divided into different warning levels according to the severity level of the abnormal probability, and the user is notified in the form of sound, light, electricity, etc. through the human-computer interaction interface; then, the system performs fault positioning, generates various feature contribution degrees of the abnormal probability through backtracking analysis, specifically calculates the deviation degree of the voltage discreteness, temperature rise rate and other features of each cell relative to the overall abnormal value, and quantifies the feature contribution degree of each cell by using a normalization weighting algorithm, determines one or several cells with the highest contribution degree as the fault source, and finally accurately identifies the location of the fault source on the battery system topology graph, providing accurate targets for subsequent maintenance.

[0096] In addition, another aspect of the present application, in some embodiments, the present application provides a new energy vehicle multi-source data fusion fault detection system, which comprises a fault alarm unit, which is described with reference to Figure 3 The figure is a structural schematic diagram of the fault alarm unit according to some embodiments of the present application, which comprises a collection module 201, a processing module 202 and an execution module 203, which are described as follows:

[0097] The collection module 201 is mainly used for collecting the electrical operation information, cooling liquid flow information and temperature information of each cell measurement point of the new energy vehicle through the vehicle-mounted sensor in the present application;

[0098] The processing module 202 is configured to extract a consistency feature of the cell voltage and a gradient feature of the battery pack current from the electrical operation information, and determine a state of charge abnormality index of the battery in the new energy vehicle according to the consistency feature and the gradient feature.

[0099] It should be noted that the processing module 202 is further configured to extract a temperature rising rate of each cell measurement point in the new energy vehicle from each temperature information, and determine a temperature difference abnormality feature of the new energy vehicle according to all the temperature rising rates and a temperature difference feature between each cell measurement point, and determine a risk index of the battery temperature out of control in the new energy vehicle according to the temperature difference abnormality feature and a flow rate descending gradient in the coolant flow information.

[0100] The execution module 203 is configured to perform dependent fusion of the state of charge abnormality index and the risk index of the battery temperature out of control in the new energy vehicle according to an operation dependence relationship between the battery temperature and the state of charge in the new energy vehicle, to obtain an abnormality probability of the battery in the new energy vehicle, and trigger a fault warning when the abnormality probability is greater than a fault threshold of the battery in the new energy vehicle, and locate a fault cell according to a feature contribution degree of each cell.

[0101] The above describes an example of the fault detection system and method for new energy vehicle multi-source data fusion provided by the embodiments of the present application in detail. It can be understood that the corresponding device includes a hardware structure and / or software module corresponding to each function to achieve the above functions. Those skilled in the art should easily realize that the units and algorithm steps of each example described in combination with the embodiments disclosed in the present application can be realized in the form of hardware or a combination of hardware and computer software. Whether a certain function is executed in hardware or computer software driven hardware depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0102] In some embodiments, the present application further provides a computer device, which includes a memory and a processor, the memory is configured to store a computer program, and the processor is configured to call and run the computer program from the memory, so that the computer device executes the new energy vehicle multi-source data fusion fault detection method described above.

[0103] In some embodiments, with reference to Figure 4 The dashed line in the figure indicates that the unit or the module is optional, and the figure is a structural schematic diagram of a computer device for implementing the new energy vehicle multi-source data fusion fault detection method according to the embodiments of the present application. The new energy vehicle multi-source data fusion fault detection method described in the above embodiments can be implemented byFigure 4 The computer device shown can be implemented by a terminal device or a server or a chip, and can include at least one processor 301, a memory 302 and at least one communication unit 305.

[0104] The processor 301 can be a general-purpose processor or a special-purpose processor. For example, the processor 301 can be a central processing unit (CPU), which can be used to control the computer device, execute a software program, and process data of the software program. The computer device can further include a communication unit 305 to realize input (reception) and output (transmission) of signals.

[0105] For example, the computer device can be a chip, and the communication unit 305 can be an input and / or output circuit of the chip, or the communication unit 305 can be a communication interface of the chip. The chip can be a component of a terminal device or a network device or other device.

[0106] For another example, the computer device can be a terminal device or a server, and the communication unit 305 can be a transceiver of the terminal device or the server, or the communication unit 305 can be a transceiver circuit of the terminal device or the server.

[0107] One or more memories 302 can be included in the computer device, and programs 304 can be stored on the memories 302. The programs 304 can be run by the processor 301 to generate instructions 303, so that the processor 301 executes the methods described in the above method embodiments according to the instructions 303. Alternatively, data (such as a target review model) can also be stored in the memory 302. Alternatively, the processor 301 can also read the data stored in the memory 302. The data can be stored in the same storage address as the programs 304, or the data can be stored in different storage addresses from the programs 304.

[0108] The processor 301 and the memory 302 can be separately arranged, or can be integrated together, for example, integrated on a system on chip (SOC) of the terminal device.

[0109] It should be understood that each step of the above method embodiments can be accomplished by logic circuits in the form of hardware or instructions in the form of software in the processor 301, which can be a CPU, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof.

[0110] Those skilled in the art will appreciate that embodiments of the application can be devised for a method, a system, or a computer program product. Accordingly, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application can take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage devices, etc.) embodying computer-readable program code.

[0111] For example, in some embodiments, the present application also provides a computer-readable storage medium, wherein instructions or codes are stored in the computer-readable storage medium, and when the instructions or codes are run on a computer, the computer is caused to perform the new energy vehicle multi-source data fusion fault detection method described above.

[0112] Although preferred embodiments of the application have been described, those skilled in the art will appreciate that additional modifications and variations to the preferred embodiments can be made without departing from the spirit and scope of the application. Accordingly, it is intended that the appended claims be interpreted as including all such alterations and modifications as fall within the scope of the application.

[0113] Obviously, various modifications and changes can be made to the present application by those skilled in the art without departing from the spirit and scope of the present application. Thus, it is intended that the present application cover the modifications and changes as fall within the scope of the claims and their equivalents.

Claims

1. A fault detection method for new energy vehicles using multi-source data fusion, characterized in that, Includes the following steps: The vehicle-mounted sensors collect electrical operation information, coolant flow information, and temperature information at various battery cell measurement points of the new energy vehicle. The consistency characteristics of cell voltage and the gradient characteristics of battery pack current are extracted from the electrical operation information, and the charge anomaly index of battery state of charge in new energy vehicles is determined by the consistency characteristics and the gradient characteristics. The temperature rise rate of each cell measurement point in the new energy vehicle is extracted from various temperature information. The temperature difference abnormality characteristics of the new energy vehicle are determined by all temperature rise rates and temperature difference characteristics between various cell measurement points. The runaway risk index of battery temperature in the new energy vehicle is determined by the temperature difference abnormality characteristics and the flow rate decrease gradient in the coolant flow information. By integrating the abnormal charge index and the runaway risk index through the operational dependency relationship between battery temperature and battery state of charge in new energy vehicles, the abnormal probability of the battery in new energy vehicles is obtained. When the abnormal probability is greater than the fault threshold of the new energy vehicle battery, a fault warning is triggered, and the faulty battery cell is located by the characteristic contribution of each battery cell. Specifically, extracting the gradient features of the battery pack current from the electrical operation information includes: Obtain battery pack current data from the electrical operation information; Discrete difference operation is performed on the battery pack current data to obtain the gradient characteristics of the battery pack current; Specifically, determining the runaway risk index of battery temperature in new energy vehicles through the abnormal temperature characteristics and the flow rate descent gradient in the coolant flow rate information includes: Calculate the flow rate descent gradient in the coolant flow rate information and obtain the rated flow rate of coolant in new energy vehicles; The abnormal flow indicators of coolant flow in new energy vehicles are determined by the flow rate descent gradient and the rated flow rate. The runaway risk index of battery temperature in new energy vehicles is determined based on the abnormal flow rate index and the abnormal temperature difference characteristics.

2. The method as described in claim 1, characterized in that, Extracting the consistency features of cell voltage from the electrical operation information specifically includes: Obtain cell voltage data from the electrical operation information; The statistical dispersion at the same sampling time is calculated based on the cell voltage data to obtain the consistency characteristics of the cell voltage.

3. The method as described in claim 1, characterized in that, The determination of the charge anomaly index of the battery state of charge in new energy vehicles through the consistency feature and the gradient feature specifically includes: To obtain the state of charge characteristics of new energy vehicles under normal operating conditions; Calculate the decision distance between the state of charge feature and the consistency feature; Calculate the gradient distance between the charged state feature and the gradient feature; The charge anomaly index of the battery state of charge in new energy vehicles is determined by the decision distance and the gradient distance.

4. The method as described in claim 1, characterized in that, The specific extraction of the temperature rise rate from various temperature information at each battery cell measurement point in a new energy vehicle includes: For each cell measurement point in a new energy vehicle, the temperature information of the cell measurement point is filtered and denoised to obtain the denoised temperature information of the cell measurement point. The temperature rise rate of the battery cell measurement points is extracted from the denoised temperature information, and then the temperature rise rate of each battery cell measurement point in the new energy vehicle is obtained.

5. The method as described in claim 1, characterized in that, The temperature difference anomaly characteristics of new energy vehicles are determined by analyzing all temperature rise rates and temperature difference characteristics between measurement points of each battery cell. Specifically, these include: Determine the temperature difference characteristics between the measurement points of each battery cell; The temperature rise difference between measurement points of each cell was determined by all the temperature rise rates. Based on the characteristics of each temperature difference, all the temperature difference values ​​are combined using a second-order paradigm to obtain the abnormal temperature difference characteristics of new energy vehicles.

6. The method as described in claim 1, characterized in that, By fusing the charge anomaly index and the runaway risk index through the operational dependency relationship between battery temperature and battery state of charge in new energy vehicles, the specific anomaly probability of batteries in new energy vehicles is obtained, including: Initialize a Bayesian network model based on joint feature inference; The operational dependency between battery temperature and battery state of charge in new energy vehicles is used as a directed edge in a Bayesian network model. The charge anomaly index is used as the first observation node in the Bayesian network model; The runaway risk index is used as the second observation node in the Bayesian network model; A Bayesian network model is used to infer anomalies in the battery health status of new energy vehicles, and the probability of anomalies in the batteries of new energy vehicles is obtained.

7. A fault detection system for multi-source data fusion in new energy vehicles, comprising a fault alarm unit, wherein the system employs the method described in any one of claims 1 to 6 for fault detection of multi-source data fusion in new energy vehicles, characterized in that... The fault alarm unit includes: The data acquisition module is used to collect electrical operation information, coolant flow information, and temperature information at various battery cell measurement points of new energy vehicles through on-board sensors. The processing module is used to extract the consistency characteristics of cell voltage and the gradient characteristics of battery pack current from the electrical operation information, and to determine the charge anomaly index of battery state of charge in new energy vehicles through the consistency characteristics and the gradient characteristics. The processing module is also used to extract the temperature rise rate of each cell measurement point in the new energy vehicle from the various temperature information, determine the temperature difference abnormality characteristics of the new energy vehicle by all the temperature rise rates and the temperature difference characteristics between each cell measurement point, and determine the runaway risk index of battery temperature in the new energy vehicle by the temperature difference abnormality characteristics and the flow rate decrease gradient in the coolant flow information. The execution module is used to fuse the charge anomaly index and the runaway risk index by the operational dependency relationship between battery temperature and battery state of charge in the new energy vehicle to obtain the anomaly probability of the battery in the new energy vehicle. When the anomaly probability is greater than the fault threshold of the new energy vehicle battery, a fault warning is triggered, and the faulty battery cell is located by the characteristic contribution of each battery cell.

8. A computer device, characterized in that, The computer device includes a memory and a processor. The memory is used to store computer programs, and the processor is used to call and run the computer programs from the memory, so that the computer device executes the fault detection method for multi-source data fusion of new energy vehicles according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions or code that, when executed on a computer, cause the computer to implement the fault detection method for multi-source data fusion of new energy vehicles as described in any one of claims 1 to 6.

Images