New energy automobile battery pack monitoring method and system

Abstract

The invention provides a new energy automobile battery pack monitoring method and system. The method comprises the steps that a temperature signal and a vibration signal of a battery pack in a vehicle are collected in real time through a preset temperature sensor array and a preset piezoelectric film network; generating a corresponding temperature distribution two-dimensional cloud picture in real time according to the temperature signal based on a preset rule, generating a corresponding time-frequency matrix in real time according to the vibration signal, and performing feature fusion processing on the temperature distribution two-dimensional cloud picture and the time-frequency matrix to generate a corresponding target matrix in real time; generating a corresponding target feature vector in real time according to the target matrix, and calculating a deformation probability corresponding to the battery pack in real time according to the target feature vector; and according to the deformation probability, determining an early warning grade matched with the battery pack in real time, and according to the early warning grade, executing a corresponding early warning measure in real time. According to the invention, early warning of the battery pack can be actively completed, accurate early warning measures can be taken at the same time, and the monitoring efficiency of the battery pack is correspondingly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of new energy vehicles, and particularly relates to a method and system for monitoring a battery pack of a new energy vehicle. Background Art

[0002] With the progress of technology and the rapid development of productivity, new energy vehicles have been popularized in people's daily lives and have become one of the means of transportation for people's daily travel, correspondingly facilitating people's lives.

[0003] Among them, the overall structure of existing new energy vehicles mainly consists of a vehicle body, a battery pack, and a driving motor. Specifically, the performance of the battery pack can directly affect the performance of the new energy vehicle, so it is necessary to monitor the working state of the battery pack in real time to keep the battery pack in a stable and efficient working state, correspondingly ensuring the performance of the vehicle.

[0004] Furthermore, in the process of monitoring the battery pack in the prior art, most of them adopt a passive monitoring method, that is, after the battery pack shows an abnormality, a warning message is sent, and the function of active warning cannot be realized. In addition, in the process of collecting the working parameters of the battery pack in the prior art, it is easily interfered by the external environment, resulting in low accuracy of the collected working parameters of the battery pack, correspondingly reducing the accuracy of the monitoring results and at the same time reducing the monitoring efficiency of the battery pack. Summary of the Invention

[0005] Based on this, the purpose of the present invention is to provide a method and system for monitoring a battery pack of a new energy vehicle to solve the problems that the prior art cannot achieve active warning, and at the same time the collected battery pack parameters are easily interfered by the external environment, resulting in low monitoring efficiency of the battery pack.

[0006] The first aspect of the embodiment of the present invention proposes: A method for monitoring a battery pack of a new energy vehicle, wherein the method includes: Real-time collecting temperature signals and vibration signals of an internal battery pack of a vehicle through a preset temperature sensor array and a preset piezoelectric film network respectively; Based on a preset rule, generating a corresponding two-dimensional temperature distribution cloud map in real time according to the temperature signal, generating a corresponding time-frequency matrix in real time according to the vibration signal, and performing feature fusion processing on the two-dimensional temperature distribution cloud map and the time-frequency matrix to generate a corresponding target matrix in real time; Generating a corresponding target feature vector in real time according to the target matrix, and calculating a deformation probability corresponding to the battery pack in real time according to the target feature vector; Determining a warning level adapted to the battery pack in real time according to the deformation probability, and performing corresponding warning measures in real time according to the warning level.

[0007] The beneficial effects of the present invention are as follows: By collecting the temperature signal and vibration signal of the battery pack in real time, the real-time working state of the battery pack can be correspondingly determined. Based on this, the present invention can immediately generate corresponding two-dimensional temperature distribution cloud maps and time-frequency matrices according to the pre-set rules respectively. Based on this, the deformation probability corresponding to the current battery pack can be finally calculated in real time, and the warning level required for the current battery pack can be determined in real time according to the magnitude of the deformation probability, so that warning measures can be accurately executed, and thus the warning of the battery pack can be actively completed. At the same time, the influence of the outside world can be avoided, and the monitoring efficiency of the battery pack can be correspondingly improved.

[0008] Further, the steps of generating a corresponding two-dimensional temperature distribution cloud map according to the temperature signal and a corresponding time-frequency matrix according to the vibration signal based on the preset rules include: When the temperature signal is obtained in real time, the temperature signal is transmitted to a preset filter in real time, and the temperature signal is processed by the preset filter through a sliding window filtering process to output a corresponding filtered signal in real time; Perform a full scan on the filtered signal to detect a number of original signal values contained in the filtered signal in real time, and perform dynamic threshold elimination processing on the number of original signal values to correspondingly screen out a number of target signal values; Calculate the temperature distribution gradient corresponding to the battery pack in real time according to the number of target signal values, and generate the two-dimensional temperature distribution cloud map according to the temperature distribution gradient.

[0009] Further, the steps of generating a corresponding two-dimensional temperature distribution cloud map according to the temperature signal and a corresponding time-frequency matrix according to the vibration signal based on the preset rules include: When the vibration signal is obtained in real time, an original vibration spectrum corresponding to the vibration signal is generated in real time, and the original vibration spectrum is processed by wavelet decomposition to generate a corresponding intermediate vibration spectrum in real time; Perform envelope extraction processing on the intermediate vibration spectrum to generate a corresponding target vibration spectrum in real time; Perform a full scan on the target vibration spectrum to detect a number of target characteristic values contained in the target vibration spectrum in real time; Perform matrix processing on the number of target characteristic values to generate the time-frequency matrix in real time, and the time-frequency matrix is unique.

[0010] Further, the steps of performing feature fusion processing on the two-dimensional temperature distribution cloud map and the time-frequency matrix to generate a corresponding target matrix in real time include: When the two-dimensional cloud map of the temperature distribution is obtained in real time, the two-dimensional cloud map of the temperature distribution is converted into a corresponding temperature gradient map in real time; When the time-frequency matrix is obtained in real time, the time-frequency matrix is converted into a corresponding vibration energy map in real time; Perform feature fusion processing on the temperature gradient map and the vibration energy map to generate the target matrix in real time.

[0011] Further, the step of performing feature fusion processing on the temperature gradient map and the vibration energy map to generate the target matrix in real time includes: When the temperature gradient map and the vibration energy map are respectively obtained, several temperature gradient values included in the temperature gradient map are detected in real time, and several vibration energy values included in the vibration energy map are detected in real time; Perform matrix processing on several temperature gradient values and several vibration energy values to generate the target matrix in real time.

[0012] Further, the step of generating a corresponding target feature vector according to the target matrix in real time includes: When the target matrix is obtained in real time, perform dimensionality reduction processing on the target matrix through a preset compressive sensing fusion algorithm to output a corresponding dimensionality reduction matrix in real time; Convert the dimensionality reduction matrix into a corresponding initial feature vector in real time, and adjust the initial dimension in the initial feature vector to a target dimension in real time to generate the target feature vector.

[0013] Further, the step of calculating the deformation probability corresponding to the battery pack according to the target feature vector in real time includes: When the target feature vector is obtained in real time, call out a target fully connected layer adapted to the target feature vector in real time; Input the target feature vector into the target fully connected layer correspondingly, and perform deformation calculation on the target feature vector through a preset neural network in the target fully connected layer to calculate the deformation probability corresponding to the battery pack in real time.

[0014] A second aspect of the embodiments of the present invention proposes: A monitoring system for a new energy vehicle battery pack, wherein the system includes: An acquisition module, configured to respectively acquire the temperature signal and the vibration signal of the battery pack inside the vehicle in real time through a preset temperature sensor array and a preset piezoelectric film network; A fusion module, configured to generate a corresponding two-dimensional cloud map of temperature distribution in real time based on a preset rule according to the temperature signal, generate a corresponding time-frequency matrix in real time according to the vibration signal, and perform feature fusion processing on the two-dimensional cloud map of temperature distribution and the time-frequency matrix to generate a corresponding target matrix in real time; A calculation module, configured to generate a corresponding target feature vector in real time according to the target matrix, and calculate a deformation probability corresponding to the battery pack in real time according to the target feature vector; An execution module, configured to determine a warning level adapted to the battery pack in real time according to the deformation probability, and execute corresponding warning measures in real time according to the warning level.

[0015] Further, the fusion module is specifically configured to: When the temperature signal is acquired in real time, transmit the temperature signal to a preset filter in real time, and perform sliding window filtering on the temperature signal through the preset filter to output a corresponding filtered signal in real time; Perform a full scan on the filtered signal to detect a plurality of original signal values included in the filtered signal in real time, and perform dynamic threshold elimination processing on the plurality of original signal values to screen out a plurality of target signal values correspondingly; Calculate a temperature distribution gradient corresponding to the battery pack in real time according to the plurality of target signal values, and generate the two-dimensional cloud map of temperature distribution according to the temperature distribution gradient.

[0016] Further, the fusion module is specifically configured to: When the vibration signal is acquired in real time, generate an original vibration spectrum corresponding to the vibration signal in real time, and perform wavelet decomposition on the original vibration spectrum to generate a corresponding intermediate vibration spectrum in real time; Perform envelope extraction processing on the intermediate vibration spectrum to generate a corresponding target vibration spectrum in real time; Perform a full scan on the target vibration spectrum to detect a plurality of target eigenvalue included in the target vibration spectrum in real time; Perform matrix processing on the plurality of target eigenvalue to generate the time-frequency matrix in real time, and the time-frequency matrix is unique.

[0017] Further, the fusion module is specifically configured to: When the two-dimensional cloud map of temperature distribution is acquired in real time, convert the two-dimensional cloud map of temperature distribution into a corresponding temperature gradient map in real time; When the time-frequency matrix is acquired in real time, convert the time-frequency matrix into a corresponding vibration energy map in real time; Perform feature fusion processing on the temperature gradient map and the vibration energy map to generate the target matrix in real time.

[0018] Further, the fusion module is specifically configured to: When the temperature gradient map and the vibration energy map are respectively obtained, detect in real time a plurality of temperature gradient values included in the temperature gradient map, and detect in real time a plurality of vibration energy values included in the vibration energy map; Perform matrix processing on the plurality of temperature gradient values and the plurality of vibration energy values to generate the target matrix in real time.

[0019] Further, the calculation module is specifically configured to: When the target matrix is obtained in real time, perform dimensionality reduction processing on the target matrix through a preset compressive sensing fusion algorithm to output a corresponding dimensionality reduction matrix in real time; Convert the dimensionality reduction matrix into a corresponding initial feature vector in real time, and adjust the initial dimension in the initial feature vector to a target dimension in real time to generate the target feature vector.

[0020] Further, the calculation module is specifically configured to: When the target feature vector is obtained in real time, call out a target fully connected layer adapted to the target feature vector in real time; Input the target feature vector into the target fully connected layer correspondingly, and perform deformation calculation on the target feature vector through a preset neural network in the target fully connected layer to calculate the deformation probability corresponding to the battery pack in real time.

[0021] The third aspect of the embodiments of the present invention proposes: A computer includes a memory, a processor, and a computer program stored on the memory and executable on the processor. Wherein, when the processor executes the computer program, the new energy vehicle battery pack monitoring method as described above is implemented.

[0022] The fourth aspect of the embodiments of the present invention proposes: A readable storage medium stores a computer program thereon. Wherein, when the program is executed by a processor, the new energy vehicle battery pack monitoring method as described above is implemented.

[0023] The additional aspects and advantages of the present invention will be partially given in the following description, partially become obvious from the following description, or be understood through the practice of the present invention. Description of the Drawings

[0024] Figure 1 It is a flowchart of the new energy vehicle battery pack monitoring method provided by the first embodiment of the present invention; Figure 2 It is a structural block diagram of a new energy battery pack monitoring system provided by the third embodiment of the present invention.

[0025] The following specific embodiments will further illustrate the present invention in conjunction with the above-mentioned drawings. Specific Embodiments

[0026] To facilitate the understanding of the present invention, the present invention will be described more comprehensively below with reference to the relevant drawings. Several embodiments of the present invention are shown in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, these embodiments are provided to make the disclosure of the present invention more thorough and comprehensive.

[0027] It should be noted that when an element is referred to as being "fixedly provided on" another element, it can be directly on the other element or there may also be an intermediate element. When an element is considered to be "connected" to another element, it can be directly connected to the other element or there may be an intermediate element at the same time. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are only for the purpose of illustration.

[0028] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0029] Please refer to Figure 1 , which shows a new energy vehicle battery pack monitoring method provided by the first embodiment of the present invention. The new energy vehicle battery pack monitoring method provided in this embodiment can actively complete the early warning of the battery pack, and at the same time can objectively and accurately judge the early warning level of the battery pack, correspondingly improving the monitoring efficiency.

[0030] Specifically, this embodiment provides: A new energy vehicle battery pack monitoring method, specifically including the following steps: Step S10, respectively and real-time collect the temperature signal and vibration signal of the battery pack inside the vehicle through a preset temperature sensor array and a preset piezoelectric film network; Step S20, based on a preset rule, respectively and real-time generate a corresponding two-dimensional temperature distribution cloud map according to the temperature signal, generate a corresponding time-frequency matrix according to the vibration signal, and perform feature fusion processing on the two-dimensional temperature distribution cloud map and the time-frequency matrix to real-time generate a corresponding target matrix; Step S30: Generate a corresponding target feature vector in real time according to the target matrix, and calculate the deformation probability corresponding to the battery pack in real time according to the target feature vector. Step S40: Determine the warning level adapted to the battery pack in real time according to the deformation probability, and execute corresponding warning measures in real time according to the warning level.

[0031] Specifically, in this embodiment, first of all, it should be noted that in order to actively complete the warning of the battery pack, it is necessary to obtain the real-time working state of the battery pack in real time. Among them, it should be pointed out that the real-time working state of the battery pack can be reflected by the real-time working parameters of the battery pack. Based on this, in order to collect the real-time working parameters of the battery pack in real time and accurately, the present invention will pre-set a corresponding temperature sensor array on the bottom surface of the battery pack inside the current vehicle. Among them, it should be pointed out that the temperature sensor array can form a corresponding honeycomb-shaped temperature monitoring network on the bottom surface of the current battery pack. Specifically, the honeycomb-shaped temperature monitoring network can divide the bottom surface of the battery pack into honeycomb units of 5 cm×5 cm, and nine MEMS sensors will be deployed inside each honeycomb unit, so as to comprehensively and accurately collect the temperature signal of the battery pack. Correspondingly, the present invention will also set a piezoelectric film network on the bottom surface of the battery pack, and the piezoelectric film network can comprehensively and accurately collect the vibration signal generated by the battery pack. Based on this, after respectively collecting the required temperature signal and vibration signal, the vehicle controller set inside the vehicle can immediately perform corresponding analysis and processing on the current temperature signal and the current vibration signal according to the pre-set rules. Based on this, a corresponding two-dimensional temperature distribution cloud map and a corresponding time-frequency matrix can be formed in real time for subsequent processing.

[0032] Furthermore, after obtaining the required two-dimensional temperature distribution cloud map and time-frequency matrix through the above steps respectively, corresponding quantization processing needs to be performed at this time, and finally, the warning level corresponding to the current battery pack in real time can be objectively determined according to the quantization result. Based on this, the present invention will perform feature fusion processing on the current two-dimensional temperature distribution cloud map and the current time-frequency matrix again. At the same time, a corresponding target matrix can be fused, and the current target matrix can be converted into a corresponding target feature vector in real time. On this basis, the deformation probability corresponding to the current battery pack in real time can be finally calculated according to the target feature vector, that is, the probability that the structure of the current battery pack may be deformed. Based on this, the present invention will finally determine the warning level adapted to the current battery pack in real time according to the magnitude of the deformation probability, and can finally execute the warning measures adapted to the current battery pack according to the magnitude of the warning level, so as to actively complete the warning function of the current battery pack, correspondingly greatly improving the monitoring efficiency of the battery pack.

[0033] Second Embodiment Furthermore, the steps of generating a corresponding two-dimensional cloud map of temperature distribution in real time according to the temperature signal and generating a corresponding time-frequency matrix in real time according to the vibration signal based on a preset rule include: When the temperature signal is acquired in real time, the temperature signal is transmitted to a preset filter in real time, and the temperature signal is processed by the preset filter through a sliding window filtering process to output a corresponding filtered signal in real time; Perform a full scan on the filtered signal to detect a number of original signal values included in the filtered signal in real time, and perform a dynamic threshold elimination process on the number of original signal values to screen out a number of target signal values correspondingly; Calculate the temperature distribution gradient corresponding to the battery pack in real time according to the number of target signal values, and generate the two-dimensional cloud map of temperature distribution correspondingly according to the temperature distribution gradient.

[0034] Furthermore, the steps of generating a corresponding two-dimensional cloud map of temperature distribution in real time according to the temperature signal and generating a corresponding time-frequency matrix in real time according to the vibration signal based on a preset rule include: When the vibration signal is acquired in real time, generate an original vibration spectrum corresponding to the vibration signal in real time, and perform wavelet decomposition on the original vibration spectrum to generate a corresponding intermediate vibration spectrum in real time; Perform envelope extraction on the intermediate vibration spectrum to generate a corresponding target vibration spectrum in real time; Perform a full scan on the target vibration spectrum to detect a number of target characteristic values included in the target vibration spectrum in real time; Perform matrix processing on the number of target characteristic values to generate the time-frequency matrix in real time, and the time-frequency matrix is unique.

[0035] Furthermore, the steps of performing feature fusion processing on the two-dimensional cloud map of temperature distribution and the time-frequency matrix to generate a corresponding target matrix in real time include: When the two-dimensional cloud map of temperature distribution is acquired in real time, convert the two-dimensional cloud map of temperature distribution into a corresponding temperature gradient map in real time; When the time-frequency matrix is acquired in real time, convert the time-frequency matrix into a corresponding vibration energy map in real time; Perform feature fusion processing on the temperature gradient map and the vibration energy map to generate the target matrix in real time.

[0036] Furthermore, the steps of performing feature fusion processing on the temperature gradient map and the vibration energy map to generate the target matrix in real time include: When the temperature gradient map and the vibration energy map are respectively obtained, several temperature gradient values included in the temperature gradient map are detected in real time, and several vibration energy values included in the vibration energy map are detected in real time; Matrix processing is performed on the several temperature gradient values and the several vibration energy values to generate the target matrix in real time.

[0037] Further, the step of generating the corresponding target feature vector according to the target matrix includes: When the target matrix is obtained in real time, dimensionality reduction processing is performed on the target matrix through a preset compressive sensing fusion algorithm to output the corresponding dimensionality reduction matrix in real time; The dimensionality reduction matrix is converted into the corresponding initial feature vector in real time, and the initial dimension in the initial feature vector is adjusted to the target dimension correspondingly to generate the target feature vector in real time.

[0038] Further, the step of calculating the deformation probability corresponding to the battery pack according to the target feature vector includes: When the target feature vector is obtained in real time, the target fully connected layer adapted to the target feature vector is retrieved in real time; The target feature vector is correspondingly input into the target fully connected layer, and the deformation calculation of the target feature vector is performed through a preset neural network in the target fully connected layer to calculate the deformation probability corresponding to the battery pack in real time.

[0039] In addition, in this embodiment, it should also be noted that after the required temperature signal and vibration signal are obtained in real time through the above steps, it is necessary to immediately perform separate parsing processes on the current two signals. It should be pointed out that since the types of signals are different, the required parsing processes will also be different. Based on this, after the required temperature signal is obtained in real time, the present invention will first transmit the temperature signal to a pre-set filter, and can simultaneously perform sliding window filtering on the current temperature signal through the filter, so as to eliminate the interference signal in the current temperature signal and synchronously output the corresponding filtered signal. The present invention will then perform a full scan on the current filtered signal and can simultaneously detect a number of original signal values contained inside the current filtered signal. It should be pointed out that since the filtering process has been performed, the number of original signal values detected in real time currently are all valid values. Based on this, in order to improve the accuracy of subsequent data processing, the present invention also needs to perform dynamic threshold elimination processing on the current number of original signal values, that is, to screen out the too large or too small values in real time, so as to form a number of corresponding target signal values, and can calculate the temperature distribution gradient corresponding to the current battery pack in real time based on the current number of target signal values, so as to form the above temperature distribution two-dimensional cloud map. Similarly, after the required vibration signal is obtained in real time, the present invention will extract the original vibration spectrum generated by the current vibration signal in real time. At the same time, wavelet decomposition processing will be performed on the current original vibration spectrum to generate a corresponding intermediate vibration spectrum. Based on this, the present invention will perform envelope extraction processing on the current intermediate vibration spectrum, so as to generate the required target vibration spectrum again. It should be pointed out that in order to facilitate the generation of the subsequent matrix, it is also necessary to perform a full scan on the current target vibration spectrum at this time, and can simultaneously detect a number of target eigenvalue values contained inside the current target vibration spectrum, so as to perform corresponding matrix processing based on the current number of target eigenvalue values, and can generate the above time-frequency matrix, so as to effectively complete the parsing of the above vibration signal and temperature signal, and can simultaneously complete the quantization processing of the current vibration signal and temperature signal, for the convenience of subsequent processing.

[0040] Further, after separately obtaining the required two-dimensional temperature distribution cloud map and time-frequency matrix through the above steps, corresponding feature fusion processing needs to be immediately carried out. Specifically, for the convenience of subsequent implementation, the present invention will first convert the current time-frequency matrix into a corresponding vibration energy map in real time. Similarly, the present invention will convert the current two-dimensional temperature distribution cloud map into a corresponding temperature gradient map in real time. At the same time, for the convenience of subsequent fusion, specific numerical values need to be obtained in real time. Based on this, the present invention will detect in real time a number of temperature gradient values included in the current temperature gradient map. Similarly, the present invention will also detect in real time a number of vibration energy values included in the current vibration energy map. Based on this, immediately performing matrix processing on the current number of temperature gradient values and a number of vibration energy values can form the required target matrix. On this basis, in order for the present invention to finally determine the deformation probability corresponding to the battery pack inside the current vehicle, the current target matrix needs to be parsed again. Specifically, in order to effectively reduce the data processing volume and correspondingly shorten the data processing time, the present invention will first perform corresponding dimensionality reduction processing on the current target matrix through an existing compressive sensing fusion algorithm and can output a corresponding dimensionality reduction matrix. Based on this, the current dimensionality reduction matrix is then converted into a corresponding initial feature vector in real time, and the initial dimension in the current initial feature vector is adjusted to the target dimension. It should be noted that the dimension of the initial feature vector is 256, and correspondingly, the dimension of the target feature vector is 8. Based on this, for the convenience of subsequent calculations, the present invention will also call out in real time a target fully connected layer adapted to the current target feature vector. It should be noted that a corresponding neural network is deployed inside the target fully connected layer. Based on this, only by inputting the current target feature vector into the inside of the current target fully connected layer, the neural network can perform corresponding deformation calculations on the current target feature vector, and the deformation probability corresponding to the current battery pack can be calculated in real time. It should be noted that during the actual judgment process, it is also necessary to explain that before starting to judge the deformation probability, the present invention will first judge whether the temperature and vibration of the current battery pack have mutated. Specifically, if so, it can be preliminarily judged that the current battery pack has an abnormality. Based on this, it is necessary to judge in real time whether the deformation probability of the current battery pack is greater than 0.7. Specifically, if so, the current battery pack will be immediately subjected to a three-level protection. Correspondingly, if not, the current battery pack will continue to maintain a secondary response. In addition, if it is judged in real time that the deformation probability of the battery pack is less than 0.7, three-level protection is not required, and the vibration and temperature of the current battery pack are continuously monitored for mutation, so as to objectively and accurately complete the active warning of the battery pack and be able to actively take corresponding protection measures, thereby greatly improving the monitoring efficiency of the battery pack.

[0041] Please refer to Figure 2, the third embodiment of the present invention provides: A monitoring system for a new energy vehicle battery pack, wherein the system includes: An acquisition module, configured to respectively and real-time collect the temperature signal and the vibration signal of the battery pack inside the vehicle through a preset temperature sensor array and a preset piezoelectric film network; A fusion module, configured to generate a corresponding two-dimensional temperature distribution cloud map in real time according to the temperature signal and a corresponding time-frequency matrix in real time according to the vibration signal based on a preset rule, and perform feature fusion processing on the two-dimensional temperature distribution cloud map and the time-frequency matrix to generate a corresponding target matrix in real time; A calculation module, configured to generate a corresponding target feature vector in real time according to the target matrix, and calculate a deformation probability corresponding to the battery pack in real time according to the target feature vector; An execution module, configured to determine a warning level adapted to the battery pack in real time according to the deformation probability, and execute corresponding warning measures in real time according to the warning level.

[0042] Further, the fusion module is specifically configured to: When the temperature signal is obtained in real time, transmit the temperature signal to a preset filter in real time, and perform sliding window filtering processing on the temperature signal through the preset filter to output a corresponding filtered signal in real time; Perform a full scan on the filtered signal to detect a plurality of original signal values included in the filtered signal in real time, and perform dynamic threshold elimination processing on the plurality of original signal values to screen out a plurality of target signal values correspondingly; Calculate a temperature distribution gradient corresponding to the battery pack in real time according to the plurality of target signal values, and generate the two-dimensional temperature distribution cloud map according to the temperature distribution gradient correspondingly.

[0043] Further, the fusion module is specifically configured to: When the vibration signal is obtained in real time, generate an original vibration spectrum corresponding to the vibration signal in real time, and perform wavelet decomposition processing on the original vibration spectrum to generate a corresponding intermediate vibration spectrum in real time; Perform envelope extraction processing on the intermediate vibration spectrum to generate a corresponding target vibration spectrum in real time; Perform a full scan on the target vibration spectrum to detect a plurality of target feature values included in the target vibration spectrum in real time; Perform matrix processing on the plurality of target feature values to generate the time-frequency matrix in real time, and the time-frequency matrix is unique.

[0044] Further, the fusion module is specifically configured to: When the two-dimensional cloud map of the temperature distribution is obtained in real time, the two-dimensional cloud map of the temperature distribution is converted into a corresponding temperature gradient map in real time; When the time-frequency matrix is obtained in real time, the time-frequency matrix is converted into a corresponding vibration energy map in real time; Perform feature fusion processing on the temperature gradient map and the vibration energy map to generate the target matrix in real time.

[0045] Further, the fusion module is specifically used for: When the temperature gradient map and the vibration energy map are respectively obtained, several temperature gradient values included in the temperature gradient map are detected in real time, and several vibration energy values included in the vibration energy map are detected in real time; Perform matrix processing on several temperature gradient values and several vibration energy values to generate the target matrix in real time.

[0046] Further, the calculation module is specifically used for: When the target matrix is obtained in real time, perform dimensionality reduction processing on the target matrix through a preset compressive sensing fusion algorithm to output a corresponding dimensionality reduction matrix in real time; Convert the dimensionality reduction matrix into a corresponding initial feature vector in real time, and adjust the initial dimension in the initial feature vector to a target dimension in real time to generate the target feature vector.

[0047] Further, the calculation module is specifically used for: When the target feature vector is obtained in real time, call out the target fully connected layer adapted to the target feature vector in real time; Input the target feature vector into the target fully connected layer correspondingly, and perform deformation calculation on the target feature vector through a preset neural network in the target fully connected layer to calculate the deformation probability corresponding to the battery pack in real time.

[0048] The fourth embodiment of the present invention provides a computer, including a memory, a processor, and a computer program stored on the memory and executable on the processor. Wherein, when the processor executes the computer program, the new energy vehicle battery pack monitoring method described above is implemented.

[0049] The fifth embodiment of the present invention provides a readable storage medium, on which a computer program is stored. Wherein, when the program is executed by a processor, the new energy vehicle battery pack monitoring method described above is implemented.

[0050] In summary, the new energy vehicle battery pack monitoring method and system provided by the above embodiments of the present invention can actively complete the early warning of the battery pack, and at the same time can objectively and accurately judge the early warning level of the battery pack, correspondingly improving the monitoring efficiency.

[0051] It should be noted that the above-mentioned various modules can be functional modules or program modules, and can be implemented either by software or by hardware. For the modules implemented by hardware, the above-mentioned various modules can be located in the same processor; or the above-mentioned various modules can also be located in different processors in any combined form.

[0052] The logic and / or steps represented in the flowchart or described in other ways herein, for example, can be considered as a definite sequence list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch instructions from the instruction execution system, apparatus, or device and execute the instructions), or in combination with these instruction execution systems, apparatus, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.

[0053] More specific examples (non-exhaustive list) of computer-readable media include the following: an electrical connection portion with one or more wirings (electronic device), a portable computer diskette (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpretation, or other appropriate processing as necessary, and then stored in a computer memory.

[0054] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one or a combination of the following well-known technologies in the art can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc.

[0055] In the description of this specification, the description referring to terms such as "one embodiment", "some embodiments", "examples", "specific examples", or "some examples", etc. means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in a suitable manner in any one or more embodiments or examples.

[0056] The above-described embodiments merely represent several implementation manners of the present invention. Their descriptions are relatively specific and detailed, but should not be construed as limiting the scope of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the appended claims.

Claims

1. A monitoring method for a new energy vehicle battery pack, characterized in that, The method includes: Real-time collecting the temperature signal and vibration signal of the in-vehicle battery pack through a preset temperature sensor array and a preset piezoelectric film network respectively; Based on preset rules, generating a corresponding two-dimensional temperature distribution cloud map in real time according to the temperature signal, generating a corresponding time-frequency matrix in real time according to the vibration signal, and performing feature fusion processing on the two-dimensional temperature distribution cloud map and the time-frequency matrix to generate a corresponding target matrix in real time; Generating a corresponding target feature vector in real time according to the target matrix, and calculating a deformation probability corresponding to the battery pack in real time according to the target feature vector; Determining a warning level adapted to the battery pack in real time according to the deformation probability, and performing corresponding warning measures in real time according to the warning level.

2. The monitoring method for a new energy vehicle battery pack according to claim 1, characterized in that: The steps of generating a corresponding two-dimensional temperature distribution cloud map in real time according to the temperature signal and generating a corresponding time-frequency matrix in real time according to the vibration signal based on preset rules include: When the temperature signal is obtained in real time, transmitting the temperature signal to a preset filter in real time, and performing sliding window filtering processing on the temperature signal through the preset filter to output a corresponding filtered signal in real time; Performing a full scan on the filtered signal to detect a plurality of original signal values contained in the filtered signal in real time, and performing dynamic threshold elimination processing on the plurality of original signal values to screen out a plurality of target signal values correspondingly; Calculating a temperature distribution gradient corresponding to the battery pack in real time according to the plurality of target signal values, and generating the two-dimensional temperature distribution cloud map according to the temperature distribution gradient correspondingly.

3. The monitoring method for a new energy vehicle battery pack according to claim 2, characterized in that: The steps of generating a corresponding two-dimensional temperature distribution cloud map in real time according to the temperature signal and generating a corresponding time-frequency matrix in real time according to the vibration signal based on preset rules include: When the vibration signal is obtained in real time, generating an original vibration spectrum corresponding to the vibration signal in real time, and performing wavelet decomposition processing on the original vibration spectrum to generate a corresponding intermediate vibration spectrum in real time; Performing envelope extraction processing on the intermediate vibration spectrum to generate a corresponding target vibration spectrum in real time; Performing a full scan on the target vibration spectrum to detect a plurality of target eigenvalue contained in the target vibration spectrum in real time; Performing matrix processing on the plurality of target eigenvalue to generate the time-frequency matrix in real time, and the time-frequency matrix is unique.

4. The monitoring method for a new energy vehicle battery pack according to claim 3, characterized in that: The steps of performing feature fusion processing on the two-dimensional temperature distribution cloud map and the time-frequency matrix to generate a corresponding target matrix in real time include: When the two-dimensional temperature distribution cloud map is obtained in real time, converting the two-dimensional temperature distribution cloud map into a corresponding temperature gradient map in real time; When the time-frequency matrix is obtained in real time, converting the time-frequency matrix into a corresponding vibration energy map in real time; Performing feature fusion processing on the temperature gradient map and the vibration energy map to generate the target matrix in real time.

5. The monitoring method for a new energy vehicle battery pack according to claim 4, characterized in that: The steps of performing feature fusion processing on the temperature gradient map and the vibration energy map to generate the target matrix in real time include: When the temperature gradient map and the vibration energy map are respectively obtained, several temperature gradient values included in the temperature gradient map are detected in real time, and several vibration energy values included in the vibration energy map are detected in real time; Matrix processing is performed on the several temperature gradient values and the several vibration energy values to generate the target matrix in real time.

6. The monitoring method for a new energy vehicle battery pack according to claim 1, characterized in that: The step of generating the corresponding target eigenvector in real time according to the target matrix includes: When the target matrix is obtained in real time, dimensionality reduction processing is performed on the target matrix by a preset compressive sensing fusion algorithm to output the corresponding dimensionality reduction matrix in real time; The dimensionality reduction matrix is converted into the corresponding initial eigenvector in real time, and the initial dimension in the initial eigenvector is adjusted to the target dimension correspondingly to generate the target eigenvector in real time.

7. The monitoring method for a new energy vehicle battery pack according to claim 6, characterized in that: The step of calculating the deformation probability corresponding to the battery pack in real time according to the target eigenvector includes: When the target eigenvector is obtained in real time, the target fully connected layer adapted to the target eigenvector is retrieved in real time; The target eigenvector is input into the target fully connected layer correspondingly, and the deformation calculation of the target eigenvector is performed by a preset neural network in the target fully connected layer to calculate the deformation probability corresponding to the battery pack in real time.

8. A monitoring system for a new energy vehicle battery pack, characterized in that, The system includes: An acquisition module, configured to respectively acquire the temperature signal and the vibration signal of the in-vehicle battery pack in real time through a preset temperature sensor array and a preset piezoelectric film network; A fusion module, configured to generate a corresponding two-dimensional temperature distribution cloud map in real time according to the temperature signal, generate a corresponding time-frequency matrix in real time according to the vibration signal, and perform feature fusion processing on the two-dimensional temperature distribution cloud map and the time-frequency matrix to generate a corresponding target matrix in real time; A calculation module, configured to generate a corresponding target eigenvector in real time according to the target matrix, and calculate the deformation probability corresponding to the battery pack in real time according to the target eigenvector; An execution module, configured to determine the warning level adapted to the battery pack in real time according to the deformation probability, and execute corresponding warning measures in real time according to the warning level.

9. A computer, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the new energy vehicle battery pack monitoring method according to any one of claims 1 to 7.

10. A readable storage medium, having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the new energy vehicle battery pack monitoring method according to any one of claims 1 to 7.

Images