Electrolytic cell fault diagnosis method and device, computer equipment and storage medium

Abstract

The application relates to an electrolytic cell fault diagnosis method and device, computer equipment and a storage medium. The method comprises the following steps: obtaining an observation feature of a target electrolytic cell at a current moment and a fault posterior distribution at a previous moment; the current observation feature is used for representing the current running state of the target electrolytic cell; the fault posterior distribution at the previous moment is obtained based on a historical observation feature sequence of the target electrolytic cell and is used for representing the probability of the target electrolytic cell being in each fault state at the previous moment; according to a state transition probability model constructed in advance, the fault posterior distribution at the previous moment is predicted to obtain a fault prior prediction distribution at the current moment; and according to the current observation feature and a conditional probability generation model constructed in advance, the fault prior prediction distribution at the current moment is corrected to obtain the fault posterior distribution at the current moment, so that the online quantification of the electrolytic cell multi-fault probability is realized, and the fault diagnosis precision and reliability of the target electrolytic cell are improved.

Description

Technical Field

[0001] This application relates to the field of electrolytic cell technology, and in particular to an electrolytic cell fault diagnosis method, apparatus, computer equipment, and computer-readable storage medium. Background Technology

[0002] The proton exchange membrane electrolyzer cell (PEMEC) is a highly efficient water electrolysis device that uses a proton exchange membrane to conduct protons and decompose water into hydrogen and oxygen. Due to its high energy efficiency and net-zero emissions, it has become the most popular research subject among the six future industries of "hydrogen energy".

[0003] In related technologies, electrolytic cell fault diagnosis technology is still limited to discrete discrimination of whether a fault "occurs" and "what type of fault it is", resulting in low fault diagnosis accuracy. Summary of the Invention

[0004] Therefore, it is necessary to provide an electrolytic cell fault diagnosis method, apparatus, computer equipment, and computer-readable storage medium that can output the probability distribution of electrolytic cell faults and improve the accuracy of fault diagnosis, in order to address the above-mentioned technical problems.

[0005] In a first aspect, this application provides a method for diagnosing faults in an electrolytic cell, the method comprising:

[0006] The current observation features of the target electrolytic cell at the current moment and the fault posterior distribution at the previous moment are obtained. The current observation features are used to characterize the current operating state of the target electrolytic cell. The fault posterior distribution at the previous moment is obtained based on the historical observation feature sequence of the target electrolytic cell and is used to characterize the probability that the target electrolytic cell was in each fault state at the previous moment.

[0007] Based on the pre-built state transition probability model, the fault posterior distribution of the previous time step is predicted to obtain the fault prior prediction distribution of the current time step; the state transition probability model is used to characterize the probability that the target electrolytic cell will transition from the fault state of the previous time step to the fault state of the current time step.

[0008] Based on the current observation features and the pre-built conditional probability generation model, the prior fault prediction distribution at the current moment is corrected to obtain the posterior fault distribution at the current moment. The conditional probability generation model is used to characterize the conditional probability of collecting each observation feature given each fault state, and the posterior fault distribution at the current moment is used to characterize the probability that the target electrolytic cell is in each fault state at the current moment.

[0009] In one embodiment, the prior fault prediction distribution at the current moment is corrected based on the current observation features and a pre-built conditional probability generation model to obtain the posterior fault distribution at the current moment, including:

[0010] Obtain the physical consistency prior distribution of the target electrolytic cell in each fault state at the current moment; the physical consistency prior distribution is based on the impedance, magnetic field spatial change rate, reference impedance of each fault state, and reference magnetic field spatial change rate of each fault state of the target electrolytic cell at the current moment.

[0011] Based on the physical consistency prior distribution, the fault prior prediction distribution at the current moment is corrected to obtain the corrected fault prior prediction distribution.

[0012] Based on the current observation characteristics and the pre-constructed conditional probability generation model, the corrected fault prior prediction distribution is normalized to obtain the fault posterior distribution at the current time.

[0013] In one embodiment, before obtaining the fault prior prediction distribution at the current moment, the method further includes:

[0014] Obtain the evolution path data of the target electrolytic cell between various fault states;

[0015] A state transition probability model is constructed based on the data of each fault state and evolution path.

[0016] In one embodiment, before obtaining the fault posterior distribution at the current moment, the method further includes:

[0017] Obtain the observed characteristics, mean vector, and covariance matrix of the target electrolytic cell under different fault states;

[0018] Based on the mean vector and covariance matrix of the observed features, a likelihood function relationship between the fault state and the observed features is established, and a conditional probability generation model is obtained.

[0019] In one embodiment, the method further includes:

[0020] Obtain the fault state with the highest probability from the fault posterior distribution at the current moment;

[0021] The fault state with the highest probability is taken as the fault judgment result of the target electrolytic cell at the current moment.

[0022] In one embodiment, the method further includes:

[0023] The target entropy value is obtained based on the fault posterior distribution at the current moment; the target entropy value is used to quantify the disorder of the fault posterior distribution at the current moment.

[0024] If the target entropy value is greater than a preset threshold, the fault identification result is determined to be abnormal.

[0025] In one embodiment, obtaining the current observed characteristics of the target electrolytic cell at the current moment includes:

[0026] Acquire the electrochemical impedance spectroscopy and electromagnetic matrix data of the target electrolytic cell at the current moment;

[0027] Feature extraction was performed on the electrochemical impedance spectroscopy data and electromagnetic matrix data to obtain impedance features and electromagnetic matrix features, respectively.

[0028] Based on impedance characteristics and electromagnetic matrix characteristics, the current observation characteristics are constructed.

[0029] Secondly, this application also provides an electrolytic cell fault diagnosis device, the device comprising:

[0030] The current observation feature acquisition module is used to acquire the current observation features of the target electrolytic cell at the current moment and the fault posterior distribution at the previous moment. The current observation features are used to characterize the current operating state of the target electrolytic cell. The fault posterior distribution at the previous moment is obtained based on the historical observation feature sequence of the target electrolytic cell and is used to characterize the probability that the target electrolytic cell was in each fault state at the previous moment.

[0031] The fault prior prediction distribution acquisition module is used to predict the fault posterior distribution of the previous time step based on the pre-built state transition probability model, and obtain the fault prior prediction distribution of the current time step. The state transition probability model is used to characterize the probability that the target electrolytic cell will transition from the fault state of the previous time step to the fault state of the current time step.

[0032] The fault posterior distribution acquisition module is used to correct the fault prior prediction distribution at the current moment based on the current observation features and the pre-built conditional probability generation model, and obtain the fault posterior distribution at the current moment. The conditional probability generation model is used to characterize the conditional probability of collecting each observation feature given each fault state, and the fault posterior distribution at the current moment is used to characterize the probability that the target electrolytic cell is in each fault state at the current moment.

[0033] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method steps of the first aspect.

[0034] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method steps of the first aspect.

[0035] The aforementioned electrolytic cell fault diagnosis method, apparatus, computer equipment, and computer-readable storage medium acquire the current observation features of the target electrolytic cell at the current moment and the fault posterior distribution at the previous moment. The current observation features characterize the current operating state of the target electrolytic cell. The fault posterior distribution at the previous moment is obtained based on the historical observation feature sequence of the target electrolytic cell and is used to characterize the probability that the target electrolytic cell was in each fault state at the previous moment. Based on a pre-constructed state transition probability model, the fault posterior distribution at the previous moment is predicted to obtain the fault prior prediction distribution at the current moment. The state transition probability model characterizes the fault posterior distribution of the target electrolytic cell from the previous moment... The probability of a fault state transitioning to the current fault state is calculated. Based on the current observation features and a pre-built conditional probability generation model, the prior fault prediction distribution at the current time is corrected to obtain the posterior fault distribution at the current time. The conditional probability generation model is used to characterize the conditional probability of each observation feature collected under each fault state, and the posterior fault distribution at the current time is used to characterize the probability that the target electrolytic cell is in each fault state at the current time. Thus, based on obtaining the current observation features and the posterior fault distribution of the previous time, this application introduces a state transition probability model and a conditional probability generation model, which can automatically update the probability of the target electrolytic cell being in each fault state, thereby improving the fault diagnosis accuracy of the target electrolytic cell. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 This is a diagram illustrating the application environment of an electrolytic cell fault diagnosis method in one embodiment.

[0038] Figure 2 This is a flowchart illustrating a fault diagnosis method for an electrolytic cell in one embodiment;

[0039] Figure 3 This is a schematic diagram of the target electrolytic cell in one embodiment;

[0040] Figure 4 This is a flowchart illustrating the overall process architecture of an electrolytic cell fault diagnosis method in one embodiment;

[0041] Figure 5 This is a structural block diagram of an electrolytic cell fault diagnosis device in one embodiment;

[0042] Figure 6 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0043] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0044] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0045] The electrolytic cell fault diagnosis method provided in this application embodiment can be applied to, for example... Figure 1In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104, or it can be located in the cloud or on another network server. Terminal 102 acquires the current observation features of the target electrolytic cell at the current moment and the fault posterior distribution at the previous moment. The current observation features characterize the current operating state of the target electrolytic cell. The fault posterior distribution at the previous moment is obtained based on the historical observation feature sequence of the target electrolytic cell and is used to characterize the probability that the target electrolytic cell was in each fault state at the previous moment. According to the pre-built state transition probability model, the fault posterior distribution at the previous moment is predicted to obtain the fault prior prediction distribution at the current moment. The state transition probability model characterizes the probability that the target electrolytic cell will transition from the fault state at the previous moment to the fault state at the current moment. According to the current observation features and the pre-built conditional probability generation model, the fault prior prediction distribution at the current moment is corrected to obtain the fault posterior distribution at the current moment. The conditional probability generation model characterizes the conditional probability of collecting the current observation features given the current fault state, and the fault posterior distribution at the current moment characterizes the probability that the target electrolytic cell is in the current fault state at the current moment. Terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, drones, low-altitude aircraft, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, smart in-vehicle devices, and projection equipment. Portable wearable devices can include smartwatches, smart bracelets, and head-mounted displays. Head-mounted displays can be virtual reality (VR) devices, augmented reality (AR) devices, and smart glasses. Server 104 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0046] In one embodiment, such as Figure 2 As shown, a method for diagnosing electrolytic cell faults is provided. This embodiment applies this method to... Figure 1 Taking terminal 102 as an example, the method includes the following steps:

[0047] Step S210: Obtain the current observation features of the target electrolytic cell at the current moment and the fault posterior distribution at the previous moment at the current moment; the current observation features are used to characterize the current operating state of the target electrolytic cell; the fault posterior distribution at the previous moment at the current moment is obtained based on the historical observation feature sequence of the target electrolytic cell and is used to characterize the probability that the target electrolytic cell was in each fault state at the previous moment at the current moment.

[0048] The target electrolytic cell is the electrolytic cell for which fault diagnosis is to be performed. Specifically, the target electrolytic cell can be a proton exchange membrane electrolytic cell.

[0049] like Figure 3 As shown, the proton exchange membrane electrolyzer (i.e., electrolyzer stack) is connected to an oxygen separator, a hydrogen separator, a deionized water supply subsystem, a temperature management subsystem, and a DC power supply, working together to electrolyze water into ultrapure hydrogen and oxygen. During the operation of the proton exchange membrane electrolyzer, parameters such as flow rate and pressure are periodically sampled.

[0050] The historical observation feature sequence includes the observation features from the initial moment to the previous moment.

[0051] Among them, the observed characteristics include, but are not limited to, impedance characteristics and electromagnetic matrix characteristics.

[0052] The fault states include normal, oxygen starvation, flooding, and membrane dryness. Specifically, each fault state can be identified by a number. For example, "1" represents normal, "2" represents oxygen starvation, "3" represents flooding, and "4" represents membrane dryness.

[0053] In this embodiment, the operating status data of the target electrolytic cell at the current time t can be obtained, and feature extraction can be performed on the operating status data to obtain the current observation features. .

[0054] At the previous time t-1, after the fault diagnosis of the target electrolytic cell was completed, the fault posterior distribution at the previous time will be stored. . This represents the historical observation feature sequence from the initial time t=1 to the previous time t-1. When entering fault diagnosis at the current time, the fault posterior distribution at the previous time is a known value that has been calculated and can be directly used.

[0055] For example, if the current time t=2, then the fault posterior distribution of the previous time t=1 is... It can be determined by the initial prior distribution and the observed features at time 1. The observation likelihood, given by the conditional probability generation model, is calculated using Bayes' theorem. Specifically,

[0056]

[0057] The initial prior distribution refers to the initial probability of each fault state given based on prior experience. For example, if the initial prior distribution is... , , The observation likelihood given by the conditional probability model is , , Then the prior distribution of the fault is , , .

[0058] After obtaining the fault posterior distribution at t=1, the fault posterior distributions at subsequent times (such as t=2, t=3, ...) can be obtained according to subsequent steps S220~S230.

[0059] Step S220: Based on the pre-constructed state transition probability model, predict the fault posterior distribution of the previous time step at the current time step to obtain the fault prior prediction distribution at the current time step; the state transition probability model is used to characterize the probability that the target electrolytic cell will transition from the fault state of the previous time step to the fault state at the current time step.

[0060] In this embodiment of the application, the expression for the fault prior prediction distribution at the current moment is:

[0061]

[0062] in, This represents the prior prediction distribution of the fault at the current moment. Let represent the state transition probability model, let represent the fault state of the previous time step at the current time step, let represent the fault state of the current time step, and let represent the fault posterior distribution of the previous time step at the current time step.

[0063] Step S230: Based on the current observation features and the pre-built conditional probability generation model, the fault prior prediction distribution at the current moment is corrected to obtain the fault posterior distribution at the current moment. The conditional probability generation model is used to characterize the conditional probability of collecting each observation feature under each fault state, and the fault posterior distribution at the current moment is used to characterize the probability that the target electrolytic cell is in each fault state at the current moment.

[0064] In this embodiment of the application, the expression for the fault posterior distribution at the current moment is:

[0065]

[0066] in, This represents the posterior distribution of the fault at the current moment. This represents a conditional probabilistic generative model, which can represent the fault state given the current time. Collect the current observation features at the current moment. The conditional probability.

[0067] The aforementioned electrolytic cell fault diagnosis method acquires the current observation features of the target electrolytic cell at the current moment and the fault posterior distribution at the previous moment. The current observation features characterize the current operating state of the target electrolytic cell. The fault posterior distribution at the previous moment is obtained based on the historical observation feature sequence of the target electrolytic cell and is used to characterize the probability that the target electrolytic cell was in each fault state at the previous moment. Based on a pre-constructed state transition probability model, the fault posterior distribution at the previous moment is predicted to obtain the fault prior prediction distribution at the current moment. The state transition probability model characterizes the transition of the target electrolytic cell from the fault state at the previous moment to the current moment. The probability of a fault state at the current moment is obtained; based on the current observation features and the pre-constructed conditional probability generation model, the prior fault prediction distribution at the current moment is corrected to obtain the posterior fault distribution at the current moment; the conditional probability generation model is used to characterize the conditional probability of each observation feature collected under each fault state, and the posterior fault distribution at the current moment is used to characterize the probability that the target electrolytic cell is in each fault state at the current moment; thus, based on obtaining the current observation features and the posterior fault distribution of the previous moment, this application introduces a state transition probability model and a conditional probability generation model, which can automatically update the probability of the target electrolytic cell being in each fault state, thereby improving the fault diagnosis accuracy of the target electrolytic cell.

[0068] In one embodiment, obtaining the current observed characteristics of the target electrolytic cell at the current moment includes:

[0069] Step S211: Obtain the electrochemical impedance spectroscopy data and electromagnetic matrix data of the target electrolytic cell at the current moment.

[0070] Among them, electrochemical impedance spectroscopy data and electromagnetic matrix data are used to invert the fault state inside the target electrolytic cell that cannot be directly observed. Combined with conventional external sensors such as voltage, temperature, and flow rate, it is possible to infer the true internal fault state from measurable external signals.

[0071] In the embodiments of this application, electrochemical impedance spectroscopy data can be obtained by electrochemical impedance spectroscopy (EIS) technology, and electromagnetic matrix data can be acquired by Hall electromagnetic sensors.

[0072] Step S212: Feature extraction is performed on the electrochemical impedance spectroscopy data and electromagnetic matrix data to obtain impedance features and electromagnetic matrix features.

[0073] In this embodiment, a small-amplitude current disturbance is superimposed on the DC bias voltage, and the system's response at a series of frequencies is analyzed to achieve dynamic characteristic analysis. During this process, the total battery voltage... This can be represented as a function with complex exponential input:

[0074]

[0075] in, The operating point potential, This represents the angular frequency of the harmonic disturbance.

[0076] Total impedance It can be represented as:

[0077]

[0078] in, Indicates high-frequency impedance. Indicates charge transfer impedance, For mass transfer impedance, It reflects the ohmic resistance between the electrolyte and the electrode, and can be expressed as the membrane water content. The function is as follows:

[0079]

[0080] in, and These represent the length and cross-sectional area of ​​the membrane electrode assembly, respectively.

[0081] in, Used to characterize electrochemical reaction kinetics. Let the reference current density be... ,but .

[0082] in, The charge storage capacity of the electrode-electrolyte interface is described. Furthermore, the impedance is presented in the frequency domain form. middle, The time derivative term is considered as Therefore, changes in the imaginary part will be introduced. Represents the gas constant. For fuel cell temperature, is Faraday's constant.

[0083] Step S213: Construct the current observation characteristics based on impedance characteristics and electromagnetic matrix characteristics.

[0084] In this embodiment of the application, a specified frequency band is used during the operation of the target electrolytic cell. impedance Defined as Combining the characteristics of the electromagnetic matrix at time t Construct observation features: The electromagnetic matrix is ​​characterized by the m-dimensional sensor parameters obtained from the Hall electromagnetic sensor at time t. .

[0085] Optionally, to improve model robustness, the observed features are normalized, as shown in the expression:

[0086]

[0087] in, , These are the statistical mean and standard deviation matrices of the observed features at all times during the training phase.

[0088] In one embodiment, before obtaining the fault prior prediction distribution at the current moment, the method further includes:

[0089] Step S221: Obtain the evolution path data of the target electrolytic cell between various fault states.

[0090] Among them, the evolution path data is used to characterize the gradual degradation of the target electrolyzer between various fault states.

[0091] Step S222: Construct a state transition probability model based on the data of each fault state and evolution path.

[0092] In this embodiment of the application, considering the asymptotic nature of electrolytic cell failures, a state transition probability model is introduced:

[0093]

[0094] Wherein, the state matrix And satisfy This state matrix is ​​used to describe the evolution path of a fault from a normal state to a degraded state.

[0095] In one embodiment, before obtaining the fault posterior distribution at the current moment, the method further includes:

[0096] Step S231: Obtain the observed characteristics, mean vector, and covariance matrix of the observed characteristics of the target electrolytic cell under different fault states.

[0097] Step S232: Based on the mean vector and covariance matrix of the observed features, establish the likelihood function relationship between the fault state and the observed features to obtain the conditional probability generation model.

[0098] In this embodiment of the application, the conditional probability generation model for the observation features generated by the fault state is as follows:

[0099]

[0100] in, Let be the mean vector of the observed features of the k-th type of fault state during the training phase. Let be the covariance matrix of the observed features of the k-th type of fault state during the training phase.

[0101] In one embodiment, the prior fault prediction distribution at the current time is corrected based on the current observation features and a pre-built conditional probability generation model to obtain the posterior fault distribution at the current time, including:

[0102] Step S310: Obtain the physical consistency prior distribution of the target electrolytic cell in each fault state at the current moment; the physical consistency prior distribution is obtained based on the impedance, magnetic field spatial change rate, reference impedance of each fault state, and reference magnetic field spatial change rate of each fault state of the target electrolytic cell at the current moment.

[0103] In this embodiment of the application, the expression for the physical consistency prior distribution is:

[0104]

[0105] in, This represents the physically consistent prior distribution of the k-th type of fault state at the current time t. The physical uniformity energy function is expressed as:

[0106]

[0107] in, and Let be the high-frequency impedance at time t and the reference impedance for the k-th type of fault state, respectively. and Let be the rate of change of the magnetic field in space at time t, and the rate of change of the reference magnetic field in space for the k-th type of fault state, respectively. and These are the weights for impedance characteristics and electromagnetic characteristics, respectively. and These are the impedance uniformity function and the magnetic field uniformity function, respectively.

[0108] Step S320: Based on the physical consistency prior distribution, correct the fault prior prediction distribution at the current moment to obtain the corrected fault prior prediction distribution.

[0109] In this embodiment of the application, the expression for the modified fault prior prediction distribution is:

[0110]

[0111] in, This represents the corrected prior fault prediction distribution. This represents the prior prediction distribution of the fault at the current moment.

[0112] Step S330: Based on the current observation features and the pre-constructed conditional probability generation model, normalize the corrected fault prior prediction distribution to obtain the fault posterior distribution at the current time.

[0113] In this embodiment of the application, the expression for the fault posterior distribution at the current moment is:

[0114]

[0115] in, This represents the fault posterior distribution at the current moment.

[0116] This application's embodiments introduce a physically consistent prior distribution as a prior constraint based on the electrolytic cell mechanism to correct the fault prior prediction distribution at the current moment. This allows the fault posterior distribution of the target electrolytic cell at the current moment to conform to the laws of electrolytic cell impedance change and electromagnetic field distribution, thereby improving the accuracy of fault state diagnosis of the target electrolytic cell.

[0117] In one embodiment, the method further includes:

[0118] Step S410: Obtain the fault state with the highest probability from the fault posterior distribution at the current moment;

[0119] Step S420: The fault state with the highest probability is taken as the fault judgment result of the target electrolytic cell at the current moment.

[0120] In this embodiment of the application, the expression for the fault determination result is:

[0121]

[0122] in, This indicates the fault diagnosis result at the current moment.

[0123] In one embodiment, the method further includes:

[0124] Step S510: Obtain the target entropy value based on the fault posterior distribution at the current moment; the target entropy value is used to quantify the disorder of the fault posterior distribution at the current moment.

[0125] In this embodiment of the application, the expression for the target entropy value is:

[0126]

[0127] in, This represents the target entropy value.

[0128] Step S520: If the target entropy value is greater than a preset threshold, the fault identification result is determined to be abnormal.

[0129] The preset threshold can be set according to actual needs.

[0130] In this embodiment, if the target entropy value is less than or equal to a preset threshold, the fault identification result is determined to be normal. If the target entropy value is greater than the preset threshold, the fault identification result is determined to be abnormal.

[0131] To facilitate understanding of the above method embodiments, as follows: Figure 4 As shown, the fault diagnosis method for the target electrolytic cell includes: acquiring electrolytic cell operating data (impedance spectrum, electromagnetic matrix, and other multi-source information); constructing an observation feature vector based on the electrolytic cell operating data; using a conditional probability generation model based on excitation constraints, combined with the observation feature vector and a Bayesian filtering framework, to obtain the fault posterior distribution at the current moment; obtaining the fault discrimination result based on the fault posterior distribution at the current moment, and calculating the uncertainty of the fault discrimination result.

[0132] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0133] Based on the same inventive concept, this application also provides an electrolytic cell fault diagnosis device for implementing the electrolytic cell fault diagnosis method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more embodiments of the electrolytic cell fault diagnosis device provided below can be found in the limitations of the electrolytic cell fault diagnosis method described above, and will not be repeated here.

[0134] In one exemplary embodiment, please refer to Figure 5 An electrolytic cell fault diagnosis device is provided, the device comprising:

[0135] The current observation feature acquisition module is used to acquire the current observation features of the target electrolytic cell at the current moment and the fault posterior distribution at the previous moment. The current observation features are used to characterize the current operating state of the target electrolytic cell. The fault posterior distribution at the previous moment is obtained based on the historical observation feature sequence of the target electrolytic cell and is used to characterize the probability that the target electrolytic cell was in each fault state at the previous moment.

[0136] The fault prior prediction distribution acquisition module is used to predict the fault posterior distribution of the previous time step based on the pre-built state transition probability model, and obtain the fault prior prediction distribution of the current time step. The state transition probability model is used to characterize the probability that the target electrolytic cell will transition from the fault state of the previous time step to the fault state of the current time step.

[0137] The fault posterior distribution acquisition module is used to correct the fault prior prediction distribution at the current moment based on the current observation features and the pre-built conditional probability generation model, and obtain the fault posterior distribution at the current moment. The conditional probability generation model is used to characterize the conditional probability of collecting each observation feature given each fault state, and the fault posterior distribution at the current moment is used to characterize the probability that the target electrolytic cell is in each fault state at the current moment.

[0138] In one embodiment, the prior fault prediction distribution at the current moment is corrected based on the current observation features and a pre-built conditional probability generation model to obtain the posterior fault distribution at the current moment, including:

[0139] Obtain the physical consistency prior distribution of the target electrolytic cell in each fault state at the current moment; the physical consistency prior distribution is based on the impedance, magnetic field spatial change rate, reference impedance of each fault state, and reference magnetic field spatial change rate of each fault state of the target electrolytic cell at the current moment.

[0140] Based on the physical consistency prior distribution, the fault prior prediction distribution at the current moment is corrected to obtain the corrected fault prior prediction distribution.

[0141] Based on the current observation characteristics and the pre-constructed conditional probability generation model, the corrected fault prior prediction distribution is normalized to obtain the fault posterior distribution at the current time.

[0142] In one embodiment, before obtaining the fault prior prediction distribution at the current moment, the method further includes:

[0143] Obtain the evolution path data of the target electrolytic cell between various fault states;

[0144] A state transition probability model is constructed based on the data of each fault state and evolution path.

[0145] In one embodiment, before obtaining the fault posterior distribution at the current moment, the method further includes:

[0146] Obtain the observed characteristics, mean vector, and covariance matrix of the target electrolytic cell under different fault states;

[0147] Based on the mean vector and covariance matrix of the observed features, a likelihood function relationship between the fault state and the observed features is established, and a conditional probability generation model is obtained.

[0148] In one embodiment, the method further includes:

[0149] Obtain the fault state with the highest probability from the fault posterior distribution at the current moment;

[0150] The fault state with the highest probability is taken as the fault judgment result of the target electrolytic cell at the current moment.

[0151] In one embodiment, the method further includes:

[0152] The target entropy value is obtained based on the fault posterior distribution at the current moment; the target entropy value is used to quantify the disorder of the fault posterior distribution at the current moment.

[0153] If the target entropy value is greater than a preset threshold, the fault identification result is determined to be abnormal.

[0154] In one embodiment, obtaining the current observed characteristics of the target electrolytic cell at the current moment includes:

[0155] Acquire the electrochemical impedance spectroscopy and electromagnetic matrix data of the target electrolytic cell at the current moment;

[0156] Feature extraction was performed on the electrochemical impedance spectroscopy data and electromagnetic matrix data to obtain impedance features and electromagnetic matrix features, respectively.

[0157] Based on impedance characteristics and electromagnetic matrix characteristics, the current observation characteristics are constructed.

[0158] Each module in the aforementioned electrolytic cell fault diagnosis device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0159] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 6As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When executed by the processor, the computer program implements a method for diagnosing electrolytic cell faults. The display unit is used to form a visually visible image and can be a display screen, projection device, or virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0160] Those skilled in the art will understand that Figure 6 The structures shown are merely block diagrams of some structures related to the present application and do not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than shown in the figures, or combine certain components, or have different component arrangements. In one embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, which, when executed by the processor, causes the processor to perform the steps of the above-described electrolytic cell fault diagnosis method. The steps of the electrolytic cell fault diagnosis method described here may be steps from one of the electrolytic cell fault diagnosis methods in the various embodiments described above.

[0161] In one embodiment, a computer-readable storage medium is provided, storing a computer program that, when executed by a processor, causes the processor to perform the steps of the electrolytic cell fault diagnosis method described above. The steps of the electrolytic cell fault diagnosis method described here may be steps from one of the electrolytic cell fault diagnosis methods in the various embodiments described above.

[0162] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, causes the processor to perform the steps of the electrolytic cell fault diagnosis method described above. The steps of the electrolytic cell fault diagnosis method described here may be steps from one of the electrolytic cell fault diagnosis methods in the various embodiments described above.

[0163] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0164] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0165] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0166] The above embodiments are merely illustrative of several implementation methods of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for diagnosing faults in an electrolytic cell, characterized in that, The method includes: The current observation features of the target electrolytic cell at the current moment and the fault posterior distribution at the previous moment are obtained. The current observation features are used to characterize the current operating state of the target electrolytic cell. The fault posterior distribution at the previous moment is obtained based on the historical observation feature sequence of the target electrolytic cell and is used to characterize the probability that the target electrolytic cell was in each fault state at the previous moment. Based on the pre-constructed state transition probability model, the fault posterior distribution of the previous time step at the current time step is predicted to obtain the fault prior prediction distribution at the current time step; the state transition probability model is used to characterize the probability that the target electrolytic cell will transition from the fault state of the previous time step at the current time step to the fault state at the current time step. Based on the current observation features and the pre-constructed conditional probability generation model, the prior fault prediction distribution at the current moment is corrected to obtain the posterior fault distribution at the current moment. The conditional probability generation model is used to characterize the conditional probability of collecting each observation feature given each fault state, and the posterior fault distribution at the current moment is used to characterize the probability that the target electrolytic cell is in each fault state at the current moment.

2. The method according to claim 1, characterized in that, The step of correcting the prior fault prediction distribution at the current moment based on the current observation features and a pre-built conditional probability generation model to obtain the posterior fault distribution at the current moment includes: Obtain the physical consistency prior distribution of the target electrolytic cell in each fault state at the current time; the physical consistency prior distribution is obtained based on the impedance of the target electrolytic cell at the current time, the spatial rate of change of the magnetic field, the reference impedance of each fault state, and the spatial rate of change of the reference magnetic field of each fault state. Based on the physical consistency prior distribution, the fault prior prediction distribution at the current moment is corrected to obtain the corrected fault prior prediction distribution. Based on the current observation features and the pre-constructed conditional probability generation model, the corrected fault prior prediction distribution is normalized to obtain the fault posterior distribution at the current moment.

3. The method according to claim 1, characterized in that, Before obtaining the fault prior prediction distribution at the current moment, the method further includes: Obtain the evolution path data of the target electrolytic cell between the various fault states; Based on the fault states and the evolution path data, the state transition probability model is constructed.

4. The method according to claim 1, characterized in that, Before obtaining the fault posterior distribution at the current moment, the method further includes: Obtain the observed characteristics of the target electrolytic cell under different fault states, the mean vector of the observed characteristics, and the covariance matrix of the observed characteristics; Based on the mean vector and covariance matrix of the observed features, a likelihood function relationship between the fault state and the observed features is established to obtain the conditional probability generation model.

5. The method according to any one of claims 1 to 4, characterized in that, The method further includes: From the fault posterior distribution at the current moment, obtain the fault state with the highest probability; The fault state with the highest probability is taken as the fault judgment result of the target electrolytic cell at the current moment.

6. The method according to claim 5, characterized in that, The method further includes: Based on the fault posterior distribution at the current moment, a target entropy value is obtained; the target entropy value is used to quantify the disorder of the fault posterior distribution at the current moment. If the target entropy value is greater than a preset threshold, the fault identification result is determined to be abnormal.

7. The method according to any one of claims 1 to 4, characterized in that, The acquisition of the current observation characteristics of the target electrolytic cell at the current moment includes: Acquire the electrochemical impedance spectroscopy data and electromagnetic matrix data of the target electrolytic cell at the current time; Feature extraction is performed on the electrochemical impedance spectroscopy data and the electromagnetic matrix data to obtain impedance features and electromagnetic matrix features, respectively. The current observation features are constructed based on the impedance characteristics and the electromagnetic matrix characteristics.

8. A fault diagnosis device for an electrolytic cell, characterized in that, The device includes: The current observation feature acquisition module is used to acquire the current observation features of the target electrolytic cell at the current moment and the fault posterior distribution at the previous moment. The current observation features are used to characterize the current operating state of the target electrolytic cell. The fault posterior distribution at the previous moment is obtained based on the historical observation feature sequence of the target electrolytic cell and is used to characterize the probability that the target electrolytic cell was in each fault state at the previous moment. The fault prior prediction distribution acquisition module is used to predict the fault posterior distribution of the previous time step based on a pre-built state transition probability model, and obtain the fault prior prediction distribution of the current time step; the state transition probability model is used to characterize the probability that the target electrolytic cell will transition from the fault state of the previous time step to the fault state of the current time step. The fault posterior distribution acquisition module is used to correct the fault prior prediction distribution at the current time based on the current observation features and the pre-built conditional probability generation model, and obtain the fault posterior distribution at the current time; the conditional probability generation model is used to characterize the conditional probability of collecting each observation feature given each fault state, and the fault posterior distribution at the current time is used to characterize the probability that the target electrolytic cell is in each fault state at the current time.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

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

Images