CN115911623A - Battery thermal runaway diagnosis method and system of energy storage system based on acoustic signals - Google Patents
Battery thermal runaway diagnosis method and system of energy storage system based on acoustic signals Download PDFInfo
- Publication number
- CN115911623A CN115911623A CN202211414561.7A CN202211414561A CN115911623A CN 115911623 A CN115911623 A CN 115911623A CN 202211414561 A CN202211414561 A CN 202211414561A CN 115911623 A CN115911623 A CN 115911623A
- Authority
- CN
- China
- Prior art keywords
- signal
- sound signal
- mixed
- exhaust
- noise
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
- Y02E60/10—Energy storage using batteries
Landscapes
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The invention discloses a battery thermal runaway diagnosis method and system of an energy storage system based on acoustic signals, wherein the method comprises the following steps: obtaining a mixed acoustic signal in a battery energy storage system bay, the mixed acoustic signal comprising: exhaust acoustic signals, ambient noise, other noise; performing spectral subtraction on the mixed sound signal to remove ambient noise in the mixed sound signal; and constructing an exhaust sound signal identification model, and inputting the mixed sound signal without the environmental noise into the exhaust sound signal identification model for identification so as to identify the exhaust sound signal and judge whether the battery thermal runaway occurs. The method can timely and accurately diagnose the thermal runaway of the battery based on the acoustic signal.
Description
Technical Field
The invention relates to the technical field of intelligent fault diagnosis of a battery energy storage system, in particular to a battery thermal runaway diagnosis method and system based on an acoustic signal for the energy storage system.
Background
In recent years, lithium ion batteries have received wide attention from academic and industrial fields due to their advantages of high power, high energy density, long cycle life, low self-discharge rate, and the like, and are used in different scenes such as electric vehicles, distributed energy storage, large-scale energy storage, and the like. The lithium ion battery is used as an energy storage device and relates to a complex electrochemical reaction/transfer mechanism, so that high potential safety hazards exist, attention needs to be paid to various aspects such as battery management system design, energy storage system structure optimization and the like, and the safe, stable and reliable operation of the lithium ion battery in practical use is ensured. On the basis of recognizing a fault initiation mechanism of a battery system, fault diagnosis needs to be implemented at least on two layers of a single lithium ion battery and the system, so that various faults such as gradual change, sudden and the like can be effectively identified, early warning is realized, and the safety, the stability and the reliability of the actual operation of the battery system are improved.
At present, researches taking thermal characteristics, a thermal runaway mechanism, a protection and control method of an energy storage lithium battery as a core have become hot spots and key problems in the field of scientific research. At present, researchers at home and abroad have conducted a great deal of exploration on the thermal runaway of the lithium ion battery from the aspects of experimental exploration, mechanism analysis, safety protection and the like, and have preliminarily mastered the triggering mechanism, the expansion mechanism and the safety protection early warning method of the thermal runaway. The method comprises the following steps of (1) a main lithium ion battery thermal runaway theory, modeling and a basic chemical reaction; summarizing abuse behaviors, thermal safety problems and the like of the high-energy lithium ion battery; these summarized studies provide good references for the thermal runaway research of lithium ion power batteries, but the current battery thermal runaway diagnostic methods still have many defects. Most of battery thermal runaway diagnosis methods only use battery electric and thermal signals, and when the electric and thermal signals are abnormal, the battery thermal runaway cannot be restrained. Therefore, other types of signal assistance are needed to timely and accurately diagnose thermal runaway of the battery.
Aiming at the problem that the battery thermal runaway cannot be restrained in the prior art when the battery thermal runaway diagnosis method is only based on the battery electric and thermal signals and the electric and thermal signals are abnormal, an effective solution is not provided at present.
Disclosure of Invention
The embodiment of the invention provides a battery thermal runaway diagnosis method and system of an energy storage system based on acoustic signals, and aims to solve the problem that battery thermal runaway cannot be restrained in the prior art when the battery thermal runaway diagnosis method only depends on battery electric and thermal signals and the electric and thermal signals are abnormal.
In order to achieve the above object, in one aspect, the present invention provides a method for diagnosing thermal runaway of a battery of an energy storage system based on an acoustic signal, wherein the method comprises: obtaining a mixed acoustic signal in a battery energy storage system bay, the mixed acoustic signal comprising: exhaust acoustic signals, ambient noise, other noise; performing spectral subtraction on the mixed sound signal to remove ambient noise in the mixed sound signal; and constructing an exhaust sound signal identification model, inputting the mixed sound signal without the environmental noise into the exhaust sound signal identification model for identification so as to identify the exhaust sound signal and judge whether the battery thermal runaway occurs.
Optionally, the constructing an exhaust acoustic signal recognition model includes: performing feature extraction on the exhaust sound signals to obtain a plurality of feature vectors of each exhaust sound signal; performing model training by using preset data of all feature vectors of all exhaust sound signals as a training set to obtain an initial signal recognition model; inputting the rest data of all the feature vectors of all the exhaust sound signals into the initial signal identification model as a test set for testing to obtain an output label, and if the output label is 1, judging that the initial signal identification model is the exhaust sound signal identification model; and conversely, inputting preset data of all feature vectors of all the exhaust sound signals serving as a training set into the initial signal recognition model for repeated model training.
Optionally, the inputting the mixed sound signal with the environmental noise removed into the exhaust sound signal identification model for identification to identify the exhaust sound signal includes: performing feature extraction on the mixed sound signal without the environmental noise through a Mel frequency cepstrum to obtain a plurality of feature vectors of the mixed sound signal without the environmental noise; inputting a plurality of feature vectors of the mixed sound signal without the environmental noise into the exhaust sound signal identification model for identification so as to obtain the classification of each feature vector; identifying the exhaust acoustic signal according to the classification of each feature vector.
Optionally, the acquiring the mixed acoustic signal in the battery energy storage system cabin includes: and analyzing the mixed sound signal in the battery energy storage system cabin through frequency spectrum acquisition.
Optionally, the spectral subtraction method is calculated according to the following formula:
D(w)＝P s (w)-aP n (w)
wherein, the P s (w) is a preset estimated ambient noise spectrum, P n (w) is a frequency spectrum of the input signal, i.e., a frequency spectrum of the mixed sound signal, D (w) is a frequency spectrum of the removed noise signal, P' s (w) is the frequency spectrum of the output signal, namely the frequency spectrum of the mixed sound signal after the environmental noise is removed, alpha is not less than 0, alpha is a subtraction factor, 0<β<<1, beta is a spectrum lower limit parameter.
In another aspect, the present invention provides a system for diagnosing thermal runaway of a battery of an energy storage system based on an acoustic signal, the system comprising: an acquisition unit for acquiring a mixed acoustic signal in a battery energy storage system bay, the mixed acoustic signal comprising: exhaust acoustic signals, ambient noise, other noise; a denoising unit, configured to perform a spectral subtraction on the mixed sound signal to remove ambient noise in the mixed sound signal; and the identification unit is used for constructing an exhaust sound signal identification model and inputting the mixed sound signal without the environmental noise into the exhaust sound signal identification model for identification so as to identify the exhaust sound signal and judge whether the thermal runaway of the battery occurs.
Optionally, the identification unit includes: a first extraction subunit, configured to perform feature extraction on the exhaust sound signals to obtain a plurality of feature vectors of each exhaust sound signal; the training subunit is used for performing model training by taking preset data of all feature vectors of all the exhaust sound signals as a training set to obtain an initial signal recognition model; the test subunit is configured to input the remaining data of all feature vectors of all the exhaust sound signals as a test set into the initial signal identification model for testing to obtain an output tag, and if the output tag is 1, determine that the initial signal identification model is the exhaust sound signal identification model; and conversely, inputting preset data of all feature vectors of all the exhaust sound signals serving as a training set into the initial signal recognition model for repeated model training.
Optionally, the identification unit further includes: a second extraction subunit, configured to perform feature extraction on the mixed sound signal from which the environmental noise is removed through mel-frequency cepstrum, so as to obtain a plurality of feature vectors of the mixed sound signal from which the environmental noise is removed; the recognition subunit is used for inputting a plurality of feature vectors of the mixed sound signal with the environmental noise removed into the exhaust sound signal recognition model for recognition so as to obtain the classification of each feature vector; and the judging subunit is used for identifying the exhaust sound signal according to the classification of each feature vector.
Optionally, the acquiring the mixed acoustic signal in the battery energy storage system cabin includes: and analyzing the mixed sound signal in the battery energy storage system cabin through frequency spectrum acquisition.
Optionally, the spectral subtraction method is calculated according to the following formula:
D(w)＝P s (w)-aP n (w)
wherein, the P s (w) is a preset estimated ambient noise spectrum, P n (w) is a frequency spectrum of the input signal, i.e., a frequency spectrum of the mixed sound signal, D (w) is a frequency spectrum of the removed noise signal, P' s (w) is the frequency spectrum of the output signal, namely the frequency spectrum of the mixed sound signal after the environmental noise is removed, alpha is not less than 0, alpha is a subtraction factor, 0<β<<1 and beta are lower limit parameters of the frequency spectrum.
The invention has the beneficial effects that:
the invention provides a battery thermal runaway diagnosis method and system of an energy storage system based on acoustic signals, wherein the method comprises the following steps: obtaining a mixed acoustic signal in a battery energy storage system bay, the mixed acoustic signal comprising: exhaust acoustic signals, ambient noise, other noise; performing spectral subtraction on the mixed sound signal to remove ambient noise in the mixed sound signal; and constructing an exhaust sound signal identification model, inputting the mixed sound signal without the environmental noise into the exhaust sound signal identification model for identification so as to identify the exhaust sound signal and judge whether the battery thermal runaway occurs. The method solves the problem that the battery thermal runaway cannot be restrained in the prior art when the battery thermal runaway diagnosis method is only based on the battery electric and thermal signals and the electric and thermal signals are abnormal, and can timely and accurately diagnose the battery thermal runaway based on the acoustic signal.
In order to make the aforementioned and other objects, features and advantages of the invention comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a flowchart of a method for diagnosing thermal runaway of a battery of an energy storage system based on an acoustic signal according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of a system for diagnosing battery thermal runaway of an energy storage system based on an acoustic signal according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without making any creative effort based on the embodiments in the present invention, belong to the protection scope of the present invention.
The current battery thermal runaway diagnostic method still has many defects. Most of battery thermal runaway diagnosis methods only use battery electric and thermal signals, and often when the electric and thermal signals are abnormal, the battery thermal runaway cannot be restrained. Therefore, other types of signal (e.g., acoustic signal) assistance are needed to timely and accurately diagnose thermal runaway of the battery.
The invention provides a battery thermal runaway diagnosis method of an energy storage system based on an acoustic signal, and fig. 1 is a flow chart of the battery thermal runaway diagnosis method of the energy storage system based on the acoustic signal, which is provided by the embodiment of the invention, and as shown in fig. 1, the method comprises the following steps:
step S101, obtaining a mixed sound signal in a battery energy storage system cabin, wherein the mixed sound signal comprises: exhaust acoustic signals, ambient noise, other noise;
specifically, the acquiring the mixed acoustic signal in the battery energy storage system cabin includes: and analyzing the mixed sound signal in the battery energy storage system cabin through frequency spectrum acquisition.
The environmental noise includes: air conditioning noise, battery Management System (BMS) operation noise, current noise of battery charging and discharging, noise generated by a battery conversion system (PCS) and the like, and noise of a battery energy storage system door switch.
The other noises include: a sound signal of the opening of the iron door, a collision sound and a sound of a nearby person.
Step S102, performing spectral subtraction on the mixed sound signal to remove environmental noise in the mixed sound signal;
the ambient noise signal is distributed in the range of 0-8000Hz, but mainly in the low frequency band below 3000 Hz. The exhaust acoustic signal has a large amplitude, a short duration, and an exponential decay, is mainly concentrated in the low frequency band, and overlaps with the ambient noise signal. Since the complex acoustic environment (i.e., the exhaust acoustic signal mixed with various noises) within the battery energy storage system compartment affects the identification of the safety vent acoustic signal, a noise reduction process must first be performed.
Because the ambient noise signal and the exhaust noise signal are independent from each other and the white noise is relatively stable, the spectral subtraction method can be used as a noise reduction algorithm of the mixed noise signal in the battery energy storage system cabin. The method can automatically adjust the filter coefficient according to the change of the signal characteristic so as to achieve the optimal filtering effect. Furthermore, no prior knowledge of the target signal and interference is required and the implementation is relatively simple. In addition, the signal processed by the spectral subtraction method can effectively maintain the influence and attenuation characteristics of the original signal, and better process the noise environment in the battery energy storage system cabin.
The core of spectral subtraction is the subtraction of the noise signal spectrum (ambient noise spectrum) from the signal spectrum with noise (spectrum of the mixed sound signal).
Specifically, the spectral subtraction method is calculated according to the following formula:
D(w)＝P s (w)-aP n (w)
wherein, the P s (w) is a preset estimated ambient noise spectrum, P n (w) is a spectrum of the input signal, i.e., a spectrum of the mixed acoustic signal, D (w) is a spectrum of the removed noise signal, P' s (w) is the frequency spectrum of the output signal, namely the frequency spectrum of the mixed sound signal after the environmental noise is removed, alpha is not less than 0, alpha is a subtraction factor, 0<β<<1 and beta are lower limit parameters of the frequency spectrum.
Further, the values of α and β are determined by the SNR, which represents the energy ratio of the original signal (i.e., the mixed sound signal) to the noise (i.e., the ambient noise).
The SNR is calculated according to the following formula:
where f (n) is the original signal (i.e., the mixed sound signal);
is a noisy signal (i.e. a preset ambient noise signal); />
Representing the signal power (i.e., the mixed acoustic signal power); />
Representing the de-noising power (i.e. the power of the mixed acoustic signal after removing the noise). The greater the signal-to-noise ratio, the less noise is contained in the signal.
Step S103, constructing an exhaust sound signal identification model, inputting the mixed sound signal without the environmental noise into the exhaust sound signal identification model for identification, so as to identify the exhaust sound signal and judge whether the battery thermal runaway occurs.
There are other noises in the battery energy storage system compartment such as the sound signal of the opening of the iron door, the sound of an impact, and the sound of a nearby person. In which a part of the noise such as the sound of the opening of the iron gate is similar to the sound of the exhaust gas in the time domain diagram, has a large amplitude and an exponential decay, and the frequency distribution is also similar to the exhaust gas, so it is necessary to distinguish the exhaust gas sound signal from other noises.
Thus, the present invention designs an exhaust sound signal recognition model (i.e., classifier) capable of recognizing an exhaust sound signal.
Specifically, the constructing of the exhaust acoustic signal identification model includes:
step S1031, performing feature extraction on the exhaust sound signals to obtain a plurality of feature vectors of each exhaust sound signal;
step S1032, performing model training by using preset data of all feature vectors of all exhaust acoustic signals as a training set to obtain an initial signal recognition model;
step S1033, inputting the rest data of all feature vectors of all the exhaust sound signals serving as a test set into the initial signal recognition model for testing to obtain an output label, and if the output label is 1 (namely the input data is judged to be the exhaust sound signals), judging that the initial signal recognition model is the exhaust sound signal recognition model; otherwise, inputting preset data of all feature vectors of all exhaust sound signals serving as a training set into the initial signal recognition model to repeatedly perform model training until the output label of the model is 1, wherein the model can serve as an exhaust sound signal recognition model (namely a classifier).
The inputting the mixed sound signal from which the environmental noise is removed into the exhaust sound signal identification model for identification to identify the exhaust sound signal includes:
step S1034, performing feature extraction on the mixed sound signal without the environmental noise through a Mel frequency cepstrum to obtain a plurality of feature vectors of the mixed sound signal without the environmental noise;
specifically, the process of extracting features of the mel-frequency cepstrum includes: first, sub-frame processing is performed on a mixed sound signal from which ambient noise is removed, overlapping between frames is preserved, then a Hamming window is multiplied, and a spectrum parameter of each frame is calculated using a Fast Fourier Transform (FFT) to obtain an energy distribution in a spectrum. And setting k triangular filter banks to smooth the frequency spectrum and eliminate harmonic waves, and converting the result of filtering output into logarithmic energy of each filter bank by using a relation between the Mel domain and the linear frequency spectrum. Finally, k eigenvectors (i.e., MFCC coefficients) can be obtained by discrete cosine transform of logarithmic energy.
Wherein, the relationship between the Mel domain and the linear spectrum is as follows:
in the formula: f denotes the frequency of the mixed acoustic signal with the ambient noise removed. The above formula maps linear spectrum to Mel nonlinear spectrum based on human auditory perception, and can better extract sound features.
Step S1035, inputting a plurality of feature vectors of the mixed acoustic signal from which the environmental noise is removed into the exhaust acoustic signal recognition model for recognition, so as to obtain a classification of each feature vector;
step S1036, identifying the exhaust sound signal according to the classification of each feature vector.
When the exhaust sound signal is abnormal, the battery is proved to be out of thermal runaway and is in an initial stage, the control can be carried out, and the battery can be identified through the exhaust sound signal identification model.
Fig. 2 is a system for diagnosing thermal runaway of a battery of an energy storage system based on an acoustic signal, according to an embodiment of the present invention, where the system includes:
an acquisition unit 201 for acquiring a mixed acoustic signal in a battery energy storage system compartment, the mixed acoustic signal comprising: exhaust acoustic signals, ambient noise, other noise;
specifically, the acquiring the mixed acoustic signal in the battery energy storage system cabin includes: and analyzing the mixed sound signal in the battery energy storage system cabin through frequency spectrum acquisition.
The environmental noise includes: air conditioning noise, battery Management System (BMS) operation noise, current noise of battery charging and discharging, noise generated by a battery conversion system (PCS) and the like, and noise of a battery energy storage system door switch.
The other noises include: a sound signal of the opening of the iron door, a collision sound and a sound of a nearby person.
A denoising unit 202, configured to perform spectral subtraction on the mixed sound signal to remove ambient noise in the mixed sound signal;
the ambient noise signal is distributed in the range of 0-8000Hz, but mainly in the low frequency band below 3000 Hz. The exhaust sound signal has a large amplitude, a short duration, and an exponential decay, is mainly concentrated in a low frequency band, and overlaps with the ambient noise signal. Since the complex acoustic environment (i.e., exhaust acoustic signals mixed with various noises) within the battery energy storage system compartment affects the identification of the safety vent acoustic signals, a noise reduction process must first be performed.
Because the ambient noise signal and the exhaust noise signal are independent from each other and the white noise is relatively stable, the spectral subtraction method can be used as a noise reduction algorithm of the mixed noise signal in the battery energy storage system cabin. The method can automatically adjust the filter coefficient according to the change of the signal characteristic so as to achieve the optimal filtering effect. Furthermore, no prior knowledge of the target signal and interference is required and the implementation is relatively simple. In addition, the signal processed by the spectral subtraction method can effectively maintain the influence and attenuation characteristics of the original signal, and better process the noise environment in the battery energy storage system cabin.
The core of spectral subtraction is the subtraction of the noise signal spectrum (ambient noise spectrum) from the signal spectrum with noise (spectrum of the mixed sound signal).
Specifically, the spectral subtraction method is calculated according to the following formula:
D(w)＝P s (w)-aP n (w)
wherein, the P s (w) is a preset estimated ambient noise spectrum, P n (w) is a spectrum of the input signal, i.e., a spectrum of the mixed acoustic signal, D (w) is a spectrum of the removed noise signal, P' s (w) is the frequency spectrum of the output signal, namely the frequency spectrum of the mixed sound signal after the ambient noise is removed, alpha is more than or equal to 0, alpha is a subtraction factor, 0<β<<1 and beta are lower limit parameters of the frequency spectrum.
Further, the values of α and β are determined by the SNR, which represents the energy ratio of the original signal (i.e., the mixed sound signal) to the noise (i.e., the ambient noise).
The SNR is calculated according to the following formula:
where f (n) is the original signal (i.e., the mixed sound signal);
is a noisy signal (i.e. a preset ambient noise signal); />
Representing the signal power (i.e., the mixed acoustic signal power); />
Representing the de-noising power (i.e. the power of the mixed acoustic signal after removing the noise). The greater the signal-to-noise ratio, the signal contained inThe less noise.
The identification unit 203 is configured to construct an exhaust acoustic signal identification model, and input the mixed acoustic signal from which the environmental noise is removed into the exhaust acoustic signal identification model for identification, so as to identify the exhaust acoustic signal and determine whether battery thermal runaway occurs.
There are other noises in the battery energy storage system compartment such as the sound signal of the opening of the iron door, the sound of an impact, and the sound of a nearby person. In which a part of the noise such as the sound of the opening of the iron gate is similar to the sound of the exhaust gas in the time domain diagram, has a large amplitude and an exponential decay, and the frequency distribution is also similar to the exhaust gas, so it is necessary to distinguish the exhaust gas sound signal from other noises.
Thus, the present invention designs an exhaust sound signal recognition model (i.e., classifier) capable of recognizing an exhaust sound signal.
Specifically, the constructing an exhaust acoustic signal recognition model in the recognition unit 203 includes:
a first extraction subunit 2031, configured to perform feature extraction on the exhaust acoustic signals to obtain a plurality of feature vectors of each exhaust acoustic signal;
a training subunit 2032, configured to perform model training using preset data of all feature vectors of all the exhaust acoustic signals as a training set, to obtain an initial signal recognition model;
a testing subunit 2033, configured to input the remaining data of all feature vectors of all the exhaust acoustic signals as a test set into the initial signal identification model for testing to obtain an output label, and if the output label is 1 (that is, it is determined that the input data is an exhaust acoustic signal), determine that the initial signal identification model is the exhaust acoustic signal identification model; otherwise, inputting preset data of all feature vectors of all exhaust acoustic signals into the initial signal recognition model as a training set to repeatedly perform model training until the output label of the model is 1, wherein the model can be used as an exhaust acoustic signal recognition model (namely a classifier).
The inputting, in the recognition unit 203, the mixed sound signal from which the environmental noise is removed into the exhaust sound signal recognition model for recognition, so as to recognize the exhaust sound signal includes:
a second extracting subunit 2034, configured to perform feature extraction on the mixed sound signal without the environmental noise through mel-frequency cepstrum to obtain a plurality of feature vectors of the mixed sound signal without the environmental noise;
specifically, the process of extracting features of the mel-frequency cepstrum includes: first, sub-frame processing is performed on a mixed acoustic signal from which ambient noise is removed, overlapping between frames is preserved, then a Hamming window is multiplied, and spectral parameters of each frame are calculated using a Fast Fourier Transform (FFT) to obtain energy distribution in a spectrum. K triangular filter banks are arranged to smooth the frequency spectrum and eliminate harmonics, and the result of the filtering output is converted into logarithmic energy of each filter bank by using a relationship between the Mel domain and the linear frequency spectrum. Finally, k eigenvectors (i.e., MFCC coefficients) can be obtained by discrete cosine transform of logarithmic energy.
Wherein, the relationship between the Mel domain and the linear spectrum is as follows:
in the formula: f denotes the frequency of the mixed acoustic signal with the ambient noise removed. The above formula is to map the linear spectrum to the Mel nonlinear spectrum based on human auditory perception, so as to better extract the sound characteristics.
A recognition subunit 2035, configured to input a plurality of feature vectors of the mixed acoustic signal with the ambient noise removed into the exhaust acoustic signal recognition model for recognition, so as to obtain a classification of each feature vector;
a judging subunit 2036, configured to identify the exhaust sound signal according to the classification of each feature vector.
When the exhaust sound signal is abnormal, the battery is proved to be out of thermal runaway and is in an initial stage, the control can be carried out, and the battery can be identified through the exhaust sound signal identification model.
The invention has the beneficial effects that:
the invention provides a battery thermal runaway diagnosis method and system of an energy storage system based on acoustic signals, wherein the method comprises the following steps: obtaining a mixed acoustic signal in a battery energy storage system bay, the mixed acoustic signal comprising: exhaust acoustic signals, ambient noise, other noise; performing spectral subtraction on the mixed sound signal to remove ambient noise in the mixed sound signal; and constructing an exhaust sound signal identification model, inputting the mixed sound signal without the environmental noise into the exhaust sound signal identification model for identification so as to identify the exhaust sound signal and judge whether the battery thermal runaway occurs. The method solves the problem that the battery thermal runaway can not be restrained in the prior art when the battery thermal runaway diagnosis method only depends on the battery electric and thermal signals which are abnormal, and can timely and accurately diagnose the battery thermal runaway based on the acoustic signals.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The principle and the implementation mode of the invention are explained by applying specific embodiments in the invention, and the description of the embodiments is only used for helping to understand the method and the core idea of the invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.
Claims (10)
1. A battery thermal runaway diagnosis method of an energy storage system based on an acoustic signal is characterized by comprising the following steps:
obtaining a mixed acoustic signal in a battery energy storage system bay, the mixed acoustic signal comprising: exhaust acoustic signals, ambient noise, other noise;
performing spectral subtraction on the mixed sound signal to remove ambient noise in the mixed sound signal;
and constructing an exhaust sound signal identification model, inputting the mixed sound signal without the environmental noise into the exhaust sound signal identification model for identification so as to identify the exhaust sound signal and judge whether the battery thermal runaway occurs.
2. The method of claim 1, wherein the constructing an exhaust acoustic signal recognition model comprises:
performing feature extraction on the exhaust sound signals to obtain a plurality of feature vectors of each exhaust sound signal;
performing model training by using preset data of all feature vectors of all exhaust sound signals as a training set to obtain an initial signal recognition model;
inputting the rest data of all the feature vectors of all the exhaust sound signals into the initial signal identification model as a test set for testing to obtain an output label, and if the output label is 1, judging that the initial signal identification model is the exhaust sound signal identification model; and conversely, inputting preset data of all feature vectors of all exhaust sound signals serving as a training set into the initial signal recognition model for repeated model training.
3. The method of claim 1, wherein the inputting the mixed acoustic signal with the environmental noise removed into the exhaust acoustic signal identification model for identification to identify the exhaust acoustic signal comprises:
performing feature extraction on the mixed sound signal without the environmental noise through a Mel frequency cepstrum to obtain a plurality of feature vectors of the mixed sound signal without the environmental noise;
inputting a plurality of feature vectors of the mixed sound signal without the environmental noise into the exhaust sound signal identification model for identification so as to obtain the classification of each feature vector;
identifying the exhaust acoustic signal according to the classification of each feature vector.
4. The method of claim 1, wherein the acquiring a mixed acoustic signal in a battery energy storage system bay comprises:
and analyzing the mixed sound signal in the battery energy storage system cabin through frequency spectrum acquisition.
5. The method of claim 1, wherein the spectral subtraction is calculated according to the following formula:
D(w)＝P s (w)-aP n (w)
wherein, the P is s (w) is a preset estimated ambient noise spectrum, P n (w) is a spectrum of the input signal, i.e., a spectrum of the mixed acoustic signal, D (w) is a spectrum of the removed noise signal, P' s (w) is the frequency spectrum of the output signal, namely the frequency spectrum of the mixed sound signal after the environmental noise is removed, alpha is not less than 0, alpha is a subtraction factor, 0<β<<1 and beta are lower limit parameters of the frequency spectrum.
6. An energy storage system battery thermal runaway diagnostic system based on acoustic signals, comprising:
an acquisition unit for acquiring a mixed acoustic signal in a battery energy storage system bay, the mixed acoustic signal comprising: exhaust acoustic signals, ambient noise, other noise;
a denoising unit, configured to perform spectral subtraction on the mixed sound signal to remove ambient noise in the mixed sound signal;
and the identification unit is used for constructing an exhaust sound signal identification model, inputting the mixed sound signal without the environmental noise into the exhaust sound signal identification model for identification so as to identify the exhaust sound signal and judge whether the battery thermal runaway occurs.
7. The system of claim 6, wherein the identification unit comprises:
a first extraction subunit, configured to perform feature extraction on the exhaust sound signals to obtain a plurality of feature vectors of each exhaust sound signal;
the training subunit is used for performing model training by taking preset data of all feature vectors of all the exhaust sound signals as a training set to obtain an initial signal recognition model;
the testing subunit is used for inputting the rest data of all the feature vectors of all the exhaust sound signals into the initial signal identification model as a testing set to be tested to obtain an output label, and if the output label is 1, the initial signal identification model is determined to be the exhaust sound signal identification model; and conversely, inputting preset data of all feature vectors of all the exhaust sound signals serving as a training set into the initial signal recognition model for repeated model training.
8. The system of claim 6, wherein the identification unit further comprises:
a second extraction subunit, configured to perform feature extraction on the mixed sound signal from which the environmental noise is removed through mel-frequency cepstrum to obtain a plurality of feature vectors of the mixed sound signal from which the environmental noise is removed;
the recognition subunit is used for inputting a plurality of feature vectors of the mixed sound signal with the environmental noise removed into the exhaust sound signal recognition model for recognition so as to obtain the classification of each feature vector;
and the judging subunit is used for identifying the exhaust sound signal according to the classification of each feature vector.
9. The system of claim 6, wherein the acquiring the mixed acoustic signal in the battery energy storage system bay comprises:
and analyzing the mixed sound signal in the battery energy storage system cabin through frequency spectrum acquisition.
10. The system of claim 6, wherein the spectral subtraction is calculated according to the following formula:
D(w)＝P s (w)-aP n (w)
wherein, the P s (w) is a predetermined estimated ambient noise spectrum, P n (w) is a spectrum of the input signal, i.e., a spectrum of the mixed acoustic signal, D (w) is a spectrum of the removed noise signal, P' s (w) is the frequency spectrum of the output signal, namely the frequency spectrum of the mixed sound signal after the environmental noise is removed, alpha is not less than 0, alpha is a subtraction factor, 0<β<<1 and beta are lower limit parameters of the frequency spectrum.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211414561.7A CN115911623A (en) | 2022-11-11 | 2022-11-11 | Battery thermal runaway diagnosis method and system of energy storage system based on acoustic signals |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211414561.7A CN115911623A (en) | 2022-11-11 | 2022-11-11 | Battery thermal runaway diagnosis method and system of energy storage system based on acoustic signals |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115911623A true CN115911623A (en) | 2023-04-04 |
Family
ID=86475548
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211414561.7A Pending CN115911623A (en) | 2022-11-11 | 2022-11-11 | Battery thermal runaway diagnosis method and system of energy storage system based on acoustic signals |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115911623A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117162789A (en) * | 2023-11-03 | 2023-12-05 | 中国第一汽车股份有限公司 | Battery thermal safety control method, storage medium, processor and vehicle |
| CN118465693A (en) * | 2024-07-10 | 2024-08-09 | 中国矿业大学 | Lithium battery thermal runaway early acoustic signal positioning method based on time delay value |
-
2022
- 2022-11-11 CN CN202211414561.7A patent/CN115911623A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117162789A (en) * | 2023-11-03 | 2023-12-05 | 中国第一汽车股份有限公司 | Battery thermal safety control method, storage medium, processor and vehicle |
| CN118465693A (en) * | 2024-07-10 | 2024-08-09 | 中国矿业大学 | Lithium battery thermal runaway early acoustic signal positioning method based on time delay value |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112785016B (en) | New energy automobile maintenance and fault monitoring and diagnosis method based on machine learning | |
| CN115911623A (en) | Battery thermal runaway diagnosis method and system of energy storage system based on acoustic signals | |
| Su et al. | Safety warning of lithium-ion battery energy storage station via venting acoustic signal detection for grid application | |
| Jiang et al. | Bearing fault classification based on convolutional neural network in noise environment | |
| CN112885372B (en) | Intelligent diagnosis method, system, terminal and medium for power equipment fault sound | |
| CN110490071A (en) | A kind of substation's Abstraction of Sound Signal Characteristics based on MFCC | |
| CN113314144B (en) | Voice recognition and power equipment fault early warning method, system, terminal and medium | |
| CN109949823A (en) | A kind of interior abnormal sound recognition methods based on DWPT-MFCC and GMM | |
| CN111653289A (en) | Playback voice detection method | |
| CN106548786A (en) | A kind of detection method and system of voice data | |
| CN109243427A (en) | A kind of car fault diagnosis method and device | |
| Turan et al. | Monitoring Infant's Emotional Cry in Domestic Environments Using the Capsule Network Architecture. | |
| CN109658943A (en) | A kind of detection method of audio-frequency noise, device, storage medium and mobile terminal | |
| CN118364271B (en) | GIS breaker operation state monitoring method and system based on sound and image characteristics | |
| CN116778964A (en) | Power transformation equipment fault monitoring system and method based on voiceprint recognition | |
| Wang et al. | Research on a sound-based method for belt conveyor longitudinal tear detection | |
| CN116778956A (en) | Transformer acoustic feature extraction and fault identification method | |
| CN106546918A (en) | A kind of method for diagnosing faults of Hydropower Unit | |
| CN112382302A (en) | Baby cry identification method and terminal equipment | |
| Li et al. | EMD-TEO based speech emotion recognition | |
| CN117789766A (en) | Method and system for classifying discharging and bearing fault risk sounds of hydroelectric generating set | |
| CN116741200A (en) | Locomotive fan fault detection method and device | |
| Warule et al. | Hilbert-Huang Transform-Based Time-Frequency Analysis of Speech Signals for the Identification of Common Cold | |
| CN114783152A (en) | Energy storage power station fire alarm method and system based on gas-sound information fusion | |
| CN112259077B (en) | Speech recognition method, device, terminal and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |