A plunger pump fault online monitoring method and system based on multi-modal data

Abstract

The application provides a plunger pump fault online monitoring method and system based on multi-modal data, wherein a training set is constructed, a fault classification model is trained, and a trained fault classification model is obtained; multi-modal data in a plunger pump operation process are acquired in real time and preprocessed; a first feature vector is constructed according to the preprocessed multi-modal data, and the trained fault classification model is inputted, and a first prediction result is outputted; if the prediction result is an abnormal state, the first feature vector under the abnormal state of the prediction result is analyzed and an incremental training set is constructed; the fault classification model is repeatedly trained by using the incremental training set, and a new model is obtained; the first feature vector at a current time is acquired and inputted into the new model, and a second prediction result is obtained. The application introduces an uncertainty evaluation mechanism, quantitatively analyzes uncertainty indexes of different modal data, further screens high-quality incremental samples, and enhances model updating reliability.

Description

[0001] This application relates to the field of fault monitoring technology, and in particular to an online fault monitoring method and system for plunger pumps based on multimodal data. Background Technology

[0002] In modern industrial environments, plunger pumps are widely used as critical equipment in various fields, such as petroleum, chemical, aerospace, and manufacturing. The normal operation of these devices is essential for ensuring production efficiency and safety. However, due to prolonged use or harsh operating conditions, plunger pumps may experience various failures, such as mechanical wear, seal aging, and pressure fluctuations. Therefore, developing an effective online monitoring method to detect the health status of plunger pumps in real time and predict potential failures has become an important means of ensuring safe equipment operation.

[0003] Traditional plunger pump fault diagnosis relies primarily on expert experience and periodic maintenance checks. This method is not only time-consuming and labor-intensive, but also often fails to detect early signs of faults, leading to increased maintenance costs and prolonged downtime. In recent years, with the development of machine learning technology, fault classification models based on fixed training sets have been increasingly applied to plunger pump fault monitoring. These models can identify and classify specific types of faults by analyzing characteristic parameters in historical data (such as the wavelet energy ratio of vibration signals and the dominant frequency energy density of pressure signals). Although this method significantly improves the accuracy of fault detection, it has a significant limitation: when new fault types emerge, the lack of corresponding training samples makes it difficult for the model to accurately identify new faults, thus affecting its long-term effectiveness and reliability.

[0004] Specifically, fault classification models trained on fixed training sets are typically only applicable to known fault modes. Once unforeseen new fault types appear during actual operation, the model may fail. This is because existing models cannot dynamically adapt to new fault characteristics arising from environmental changes and equipment degradation. Furthermore, traditional methods lack an understanding of the complex relationships between multimodal data (such as vibration, pressure, temperature, and acoustic emission signals), limiting their effectiveness in complex operating conditions. Therefore, effectively integrating incremental learning mechanisms to enable fault classification models to continuously learn and adapt to new fault types without losing existing knowledge has become a pressing technical challenge. Summary of the Invention

[0005] This application aims to at least partially address one of the technical problems in the related art.

[0006] Therefore, the first objective of this application is to propose an online monitoring method for plunger pump failure based on multimodal data, so as to improve the accuracy of effect evaluation.

[0007] The second objective of this application is to propose an online monitoring system for plunger pump failures based on multimodal data.

[0008] To achieve the above objectives, the first aspect of this application proposes an online monitoring method for plunger pump faults based on multimodal data, comprising: Step 1: Construct a training set and train the fault classification model to obtain the trained fault classification model; Step 2: Acquire multimodal data during the operation of the plunger pump in real time and perform preprocessing; Step 3: Based on the preprocessed multimodal data, construct the first feature vector, input it into the trained fault classification model, and output the first prediction result; Step 4: If the prediction result is an abnormal state, analyze the first feature vector under the abnormal state and construct an incremental training set. Step 5: Repeat the training of the fault classification model using the incremental training set to obtain a new model; Step 6: Obtain the first feature vector at the current time and input it into the new model to obtain the second prediction result.

[0009] In some implementations, the multimodal data includes vibration signals, pressure signals, temperature signals, and acoustic emission signals.

[0010] In some implementations, the first feature vector consists of the following feature parameters: wavelet energy ratio of the vibration signal, dominant frequency energy density of the pressure signal, first-order difference mean of the temperature signal, and instantaneous frequency change rate of the acoustic emission signal.

[0011] In some implementations, if the prediction result is an abnormal state, the first feature vector of the abnormal state is analyzed and an incremental training set is constructed, including the following steps: Step 41: Call the first feature vector of the predicted abnormal state and define it as a candidate sample. Determine the consistency index between each candidate sample and the historical samples in the historical failure case library. Preset a first target threshold. Based on the relationship between the first target threshold and the average Mahalanobis distance, determine whether the current candidate sample is a new failure type sample or a duplicate sample. If it is a new failure type sample, define it as the first undetermined sample. If it is a duplicate sample, add it to the historical failure case library to update or remove historical failure cases. Step 42: Determine the similarity between each candidate sample and each historical sample in the historical fault case library, preset a second target threshold, and determine whether the current candidate sample is a first quality sample or a second quality sample based on the relationship between the second target threshold and the similarity. If the current candidate sample is a first quality sample, then the candidate sample is defined as a second undetermined sample; if the current candidate sample is a second quality sample, then the candidate sample is defined as a third undetermined sample. Step 43: If a candidate sample is defined as both the first undetermined sample and the second undetermined sample, construct an incremental training set based on the candidate sample.

[0012] In some implementations, constructing an incremental training set based on candidate samples includes the following steps: Step 431: Call the preprocessed multimodal data of the candidate samples; Step 432: Input the data of each modality into the trained fault classification model separately, and output the second prediction result; Step 433: Define at least three uncertainty indicators for the second prediction result; Step 434: Construct a ratio between every two uncertainty indicators, preset a third target threshold, and when 2 / 3 of the ratios are greater than the third target threshold, define the second prediction result of the current mode as the first uncertainty, determine the uncertainty score based on the ratio, and determine the optimization threshold based on the uncertainty score. Step 435: The candidate samples that are simultaneously defined as the first undetermined sample and the second undetermined sample are compared with the optimization threshold. If the candidate sample is less than the optimization threshold, the candidate sample is defined as an incremental sample; if the candidate sample is greater than or equal to the optimization threshold, it is discarded.

[0013] In some implementations, if the consistency index between each candidate sample and historical samples in the historical fault case library is greater than the first target threshold, the current candidate sample is defined as a new fault type sample; if the consistency index between each candidate sample and historical samples in the historical fault case library is less than or equal to the first target threshold, the current candidate sample is defined as a duplicate sample.

[0014] In some implementations, if the similarity between each candidate sample and each historical sample in the historical failure case library is greater than the second target threshold, the current candidate sample is defined as a first quality sample; if the similarity between each candidate sample and each historical sample in the historical failure case library is less than or equal to the second target threshold, the current candidate sample is defined as a second quality sample.

[0015] To achieve the above objectives, a second aspect of this application proposes an online monitoring system for plunger pump faults based on multimodal data, comprising: Fault classification model: Construct a training set and train the fault classification model to obtain the trained fault classification model; Data acquisition and preprocessing module: Connected to the fault classification model, it is used to acquire multimodal data during the operation of the plunger pump in real time and perform preprocessing. First prediction module: connected to the data acquisition and preprocessing module, used to construct the first feature vector based on the preprocessed multimodal data, input it into the trained fault classification model, and output the first prediction result; Incremental training set construction module: connected to the first prediction module, used to analyze the first feature vector when the prediction result is in an abnormal state and construct an incremental training set; Incremental training module: Connected to the incremental training set construction module, it is used to repeatedly train the fault classification model using the incremental training set to obtain a new model; Second prediction module: connected to the incremental training module, used to obtain the first feature vector at the current time and input it into the new model to obtain the second prediction result. Fault classification model: constructs a training set and trains the fault classification model to obtain the trained fault classification model. Data acquisition and preprocessing module: Connected to the fault classification model, it is used to acquire multimodal data during the operation of the plunger pump in real time and perform preprocessing. First prediction module: connected to the data acquisition and preprocessing module, used to construct the first feature vector based on the preprocessed multimodal data, input it into the trained fault classification model, and output the first prediction result; Incremental training set construction module: connected to the first prediction module, used to analyze the first feature vector when the prediction result is in an abnormal state and construct an incremental training set; Incremental training module: Connected to the incremental training set construction module, it is used to repeatedly train the fault classification model using the incremental training set to obtain a new model; The second prediction module is connected to the incremental training module. It is used to obtain the first feature vector at the current time and input it into the new model to obtain the second prediction result.

[0016] This application provides an online fault monitoring method for plunger pumps based on multimodal data. When the prediction result indicates that the system is in an abnormal state, this invention further analyzes the first feature vector under the abnormal state in depth. By determining its consistency index and similarity with samples in the historical fault case library, it determines whether the current sample belongs to a new fault type or a duplicate sample. If it is a new fault type, it is included in the incremental training set for subsequent model updates and optimizations. In this way, this invention can not only capture newly emerging fault modes in a timely manner, but also continuously improve the generalization ability and robustness of the fault classification model, ensuring its stable performance in long-term operation. In addition, this invention also introduces an uncertainty assessment mechanism. By quantitatively analyzing the uncertainty indicators of different modal data, high-quality incremental samples are further screened, enhancing the effectiveness and reliability of model updates. Attached Figure Description

[0017] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 A schematic flowchart illustrating an online fault monitoring method for a plunger pump based on multimodal data, provided in an embodiment of this application; Figure 2 This is a structural block diagram of a plunger pump fault online monitoring method system based on multimodal data provided in an embodiment of this application; Figure 3 This is a block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0018] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0019] The following description, with reference to the accompanying drawings, describes a method, apparatus, and equipment for evaluating the effectiveness of rain enhancement operations based on intelligent prediction of natural rainfall, according to embodiments of this application.

[0020] Figure 1 This is a flowchart illustrating an online fault monitoring method for a plunger pump based on multimodal data, provided in an embodiment of this application.

[0021] It should be noted that the execution subject of the online monitoring method for plunger pump faults based on multimodal data in this application embodiment is the online monitoring system for plunger pump faults based on multimodal data in this application embodiment. The online monitoring system for plunger pump faults based on multimodal data can be configured in an electronic device so that the electronic device can perform the online monitoring function for plunger pump faults based on multimodal data.

[0022] like Figure 1 As shown, the online fault monitoring method for plunger pumps based on multimodal data includes the following steps: Step 1: Construct a training set and train the fault classification model to obtain the trained fault classification model.

[0023] The training set refers to the initial or fixed training set, composed of multimodal sensor data from the plunger pump under various operating conditions. This multimodal data includes vibration signals, pressure signals, temperature signals, and acoustic emission signals. These signals reflect the dynamic changes in the equipment due to mechanical vibration, hydraulic fluctuations, thermal effects, and material fatigue. Each set of samples corresponds to a time window (e.g., 5 seconds), and key feature parameters are extracted to form a training feature vector. Specifically, the extracted feature parameters include, but are not limited to: the wavelet energy ratio of the vibration signal, the dominant frequency energy density of the pressure signal, the first-order difference mean of the temperature signal, and the instantaneous frequency change rate of the acoustic emission signal.

[0024] Among them, the wavelet energy ratio is obtained by calculating the energy proportion of each scale after decomposing the vibration signal by discrete wavelet transform (DWT); the main frequency energy density is extracted from the main frequency component and the energy density in its neighborhood based on the fast Fourier transform (FFT) result of the pressure signal; the first-order difference mean is used to reflect the temperature change trend by calculating the average absolute value of the temperature difference between adjacent time moments; the instantaneous frequency change rate is obtained by using Hilbert-Huang transform (HHT) or short-time Fourier transform (STFT) to obtain the time-frequency distribution of the acoustic emission signal, and further extracts the slope of its instantaneous frequency change.

[0025] To achieve accurate fault classification, each sample needs to be assigned a clear fault type label. Based on the common typical fault modes of plunger pumps, the following fault categories are defined and assigned integer labels: Normal state (0), Plunger wear (1), Valve plate leakage (2), Suction valve jamming (3), Discharge valve failure (4), Bearing failure (5). Labels are obtained manually by combining historical maintenance records and alarm log information.

[0026] This embodiment uses a deep neural network as the basic architecture to build a fault classification model. Considering the need for multimodal data fusion, a multi-channel convolutional neural network (CNN) combined with fully connected layers is chosen to process data streams from different modalities. The model input layer receives training feature vectors, which then pass through a feature extraction module consisting of multiple convolutional layers, batch normalization layers, and ReLU activation functions. A global average pooling layer then compresses the spatial dimensionality. Afterward, feature vectors from different modalities are concatenated in a fusion layer, and finally, the probability distribution of each fault category is output through two fully connected hidden layers. The output layer uses a softmax activation function to complete the final classification task.

[0027] During model training, the cross-entropy loss function is chosen to calculate the error value. Training stops when the error value is sufficiently small or the maximum number of iterations is reached. The maximum number of iterations is 50.

[0028] After training, the model's performance is evaluated on an independent test set to ensure its generalization ability on unknown samples. If the test set accuracy reaches 90% or higher, the model is considered to have good initial classification ability, and the trained fault classification model is obtained.

[0029] Step 2: Acquire multimodal data during the operation of the plunger pump in real time and perform preprocessing.

[0030] First, multimodal data during the operation of the plunger pump are acquired in real time and then denoised, including low-pass filtering (to remove high-frequency interference) and wavelet threshold denoising (to retain effective features while suppressing noise). Then, normalization is performed to unify the data of each channel to the [0,1] interval to eliminate the influence of dimensional differences. For data points with missing or outlier values, linear interpolation or sliding window mean substitution strategies are used for repair.

[0031] Step 3: Based on the preprocessed multimodal data, construct the first feature vector and input it into the trained fault classification model to output the first prediction result.

[0032] The preprocessed signal is divided into frames according to a fixed time window (e.g., 5 seconds), and the corresponding feature parameters are extracted to form the first feature vector that can be used as input to the model. This completes the transformation from the original signal to structured feature data, providing high-quality input for subsequent fault prediction and incremental learning. The trained fault classification model is input into the model, and the first prediction result, i.e., the fault type, is output.

[0033] Step 4: If the prediction result is an abnormal state, analyze the first feature vector under the abnormal state and construct an incremental training set.

[0034] Step 41: Call the first feature vector of the prediction result under the abnormal state and define it as a candidate sample. Determine the consistency index between each candidate sample and the historical samples in the historical failure case library. Preset the first target threshold. According to the relationship between the first target threshold and the average Mahalanobis distance, determine whether the current candidate sample is a new failure type sample or a duplicate sample. If it is a new failure type sample, it is defined as the first undetermined sample. If it is a duplicate sample, it is added to the historical failure case library to update or remove historical failure cases.

[0035] If the consistency index between a candidate sample and a historical sample in the historical fault case library is greater than the first target threshold, the current candidate sample is defined as a new fault type sample; if the consistency index between a candidate sample and a historical sample in the historical fault case library is less than or equal to the first target threshold, the current candidate sample is defined as a duplicate sample.

[0036] When the first prediction result output by the fault classification model is an abnormal state, the corresponding first feature vector is extracted and defined as a candidate sample. Subsequently, the set of existing sample feature vectors in the historical fault case library is called to calculate the average Mahalanobis distance between the candidate sample and all historical samples, serving as an indicator of its consistency with existing fault patterns, i.e., the consistency index. Specifically, the Mahalanobis distance, by considering the covariance structure between features, effectively reflects the degree to which the candidate sample deviates from the distribution of historical samples in the feature space; if the distance is large, it indicates that the sample differs significantly from existing fault types and may belong to a new fault type; conversely, it may be a duplicate sample of an existing type. To set the judgment criteria, this embodiment presets a first target threshold, which is determined based on the statistical characteristics of the historical sample consistency index, set as the mean of the average Mahalanobis distance of historical samples plus a standard deviation, to ensure high sensitivity for identifying new samples. If the consistency index of the current candidate sample is greater than the first target threshold, it is determined to be a new fault type sample and defined as the first pending sample; if it is less than or equal to the threshold, it is determined to be a duplicate sample. At this time, it is added to the historical fault case library for update operation, and the old sample is removed according to the timestamp or similarity retention strategy, so as to realize the dynamic maintenance and optimization of the case library.

[0037] Step 42: Determine the similarity between each candidate sample and each historical sample in the historical fault case library, preset a second target threshold, and determine whether the current candidate sample is a first quality sample or a second quality sample based on the relationship between the second target threshold and the similarity. If the current candidate sample is a first quality sample, then the candidate sample is defined as a second undetermined sample; if the current candidate sample is a second quality sample, then the candidate sample is defined as a third undetermined sample. If the similarity between each candidate sample and each historical sample in the historical failure case library is greater than the second target threshold, the current candidate sample is defined as a first-quality sample; if the similarity between each candidate sample and each historical sample in the historical failure case library is less than or equal to the second target threshold, the current candidate sample is defined as a second-quality sample.

[0038] In fact, steps 42 and 41 simultaneously analyze the first feature vector of the first predicted abnormal state. When a candidate sample is identified as abnormal, the system extracts its corresponding first feature vector and compares it one by one with existing samples in the historical fault case library to calculate the similarity. By traversing all samples in the historical fault case library, the system can obtain the similarity value between the candidate sample and each historical sample, and further calculate its mean or the average of the top K highest values ​​as the final comprehensive similarity score used for quality assessment.

[0039] The similarity distribution between historical samples in the historical fault case database and other samples is statistically analyzed, and their mean and standard deviation are calculated. The second target threshold is set as the mean plus one standard deviation. Samples that highly match existing fault types are selected to ensure that only samples with high confidence are considered high-quality samples.

[0040] If the overall similarity between a candidate sample and a sample in the historical failure case library is greater than the second target threshold, the sample is considered to be highly matched with the existing failure mode and has high reliability. It is defined as the first quality sample and further marked as the second undetermined sample. If the overall similarity of a candidate sample is less than or equal to the second target threshold, it is considered to have a low degree of matching with historical samples, and may have problems with noise interference or incomplete information. It is defined as a second quality sample and marked as a third undetermined sample.

[0041] The first quality sample represents high-quality data with high matching accuracy that can be used for subsequent incremental training; while the second quality sample is considered a low-quality sample. Although it is still retained in the system for subsequent analysis, it will not directly participate in the model update process. It needs to be further combined with the uncertainty assessment mechanism to decide whether to remove or retain it.

[0042] Step 43: If a candidate sample is defined as both the first undetermined sample and the second undetermined sample, construct an incremental training set based on the candidate sample.

[0043] Step 431: Call the preprocessed multimodal data of the candidate samples.

[0044] Step 432: Input the data of each modality into the trained fault classification model separately, and output the second prediction result.

[0045] Step 433: Define at least three uncertainty indicators for the second prediction result.

[0046] Uncertainty metrics include entropy, KL divergence, and mutual information. These metrics can reflect the model's confidence in the current sample from different perspectives.

[0047] Step 434: Construct a ratio between every two uncertainty indicators, preset a third target threshold, and when 2 / 3 of the ratios are greater than the third target threshold, define the second prediction result of the current mode as the first uncertainty, determine the uncertainty score based on the ratio, and determine the optimization threshold based on the uncertainty score.

[0048] First, for each modality, three uncertainty indices are calculated for the second prediction result: entropy (E), KL divergence (KL), and mutual information (MI), and three sets of ratios are further constructed: E / KL, MI / E, and KL / MI. Then, the prediction results of the same modality under historical operating conditions are retrieved, and the historical distribution of the above three sets of ratios is statistically analyzed, calculating their mean (μ1, μ2, μ3) and standard deviation (σ1, σ2, σ3) respectively. Based on this, a third target threshold is set as the mean of the corresponding ratio plus a standard deviation (T1, T2, T3). For the current candidate sample, if any two of its ratios are simultaneously greater than the corresponding third target threshold, the prediction of this modality is considered to have high uncertainty, and it is marked as the first uncertainty state; otherwise, it is considered a low uncertainty state. To further quantify the degree of uncertainty, the sum of the absolute values ​​of the differences between the current three sets of ratios and their respective means is calculated to form an uncertainty score. Next, the mean uncertainty score for all modalities is calculated, and a weighted average is generated based on the deviation of each modality's uncertainty score from its mean, resulting in a comprehensive optimization score. The weights can be pre-set according to the importance of the modality or dynamically adjusted based on historical performance. Finally, the optimization threshold, based on the third objective threshold and plus the comprehensive optimization score, is used for subsequent screening of incremental samples.

[0049] Step 435: The candidate samples that are simultaneously defined as the first undetermined sample and the second undetermined sample are compared with the optimization threshold. If the candidate sample is less than the optimization threshold, the candidate sample is defined as an incremental sample; if the candidate sample is greater than or equal to the optimization threshold, it is discarded.

[0050] Candidate samples simultaneously defined as the first and second undetermined samples are used as evaluation objects. These candidate samples have been preliminarily screened for data points with potential incremental value through consistency index and similarity analysis. Next, their uncertainty scores are compared with the optimization threshold: if the uncertainty score is less than the optimization threshold, the candidate sample is determined to have low uncertainty, high confidence, good representativeness and stability, and can be included in the incremental training set; if the uncertainty score is greater than or equal to the optimization threshold, the sample is determined to have high prediction uncertainty or feature instability risk, which may affect the model update effect, and is therefore removed.

[0051] Step 5: Repeat the training of the fault classification model using the incremental training set to obtain a new model.

[0052] The construction of the incremental training set relies first on the preliminary sample selection mechanism. The system identifies candidate samples that are both representative of new fault types and meet high-quality standards through methods such as consistency index, similarity analysis, and uncertainty assessment. These samples typically originate from abnormal states that occur during equipment operation, providing new knowledge boundaries for the model. To ensure the quality of the incremental training set, the following strategies are employed during extraction: Uncertainty score ranking: Prioritize samples with lower uncertainty scores to ensure higher prediction confidence; Time window filtering: retain recently collected data (such as the last 3 months) to reflect the latest changes in operating conditions; Feature matching verification: Check whether the key features of the sample meet the model input requirements to avoid introducing noise or invalid data.

[0053] Adding incremental training sets directly to the initial training set can cause the dataset size to expand rapidly, impacting training efficiency and computational resource consumption. Therefore, certain data management strategies need to be adopted to control the total data volume before merging. Set a maximum capacity limit: for example, limit the entire training set to a maximum of 10,000 samples; Aging data removal: Remove fault type samples that have not appeared for a long time (such as those that have not been updated for more than a year); Redundant data compression: Use clustering algorithms (such as K-means) to remove highly similar samples and retain representative samples.

[0054] The recommended exclusion ratio is as follows: At least 20% of historical samples should be retained for each type of fault to maintain the model's memory capacity; The proportion of incremental samples should not exceed 20% of the total dataset to prevent the introduction of too much noise; Regularly (e.g., every six months or a year), conduct a comprehensive review of the dataset and adjust the retention ratio.

[0055] During the model training phase, a loss function is defined that integrates the old knowledge retention term and the new sample fitting term to balance the relationship between the two: ; and Let represent the loss functions on the original training set and the newly added samples, respectively, both using cross-entropy loss; λ∈[0.3,0.7] represents the weighting coefficients of the loss for the old samples. θ represents the regularization term, used to prevent drastic fluctuations in model parameters. θ0 represents the model parameters at the end of the last training, and γ is the regularization strength coefficient.

[0056] By designing the loss function described above, new knowledge from newly added samples can be effectively learned while preserving the existing model knowledge. During actual training, gradient descent or other optimization algorithms are used to minimize this loss function, thereby updating the model parameters.

[0057] Finally, the system uses multiple evaluation metrics to comprehensively evaluate the new model to determine whether it is superior to the old model. Commonly used evaluation metrics include: accuracy, precision, recall, F1 score, and AUC. If the new model outperforms the old model on at least three evaluation metrics, it is considered to have better performance and can replace the old model for subsequent fault classification tasks.

[0058] Step 6: Obtain the first feature vector at the current time and input it into the new model to obtain the second prediction result.

[0059] like Figure 2 As shown, this embodiment also discloses an online monitoring system for plunger pump faults based on multimodal data, used to execute the above-described online monitoring method for plunger pump faults based on multimodal data, including: Fault classification model: Construct a training set and train the fault classification model to obtain the trained fault classification model; Data acquisition and preprocessing module: Connected to the fault classification model, it is used to acquire multimodal data during the operation of the plunger pump in real time and perform preprocessing. First prediction module: connected to the data acquisition and preprocessing module, used to construct the first feature vector based on the preprocessed multimodal data, input it into the trained fault classification model, and output the first prediction result; Incremental training set construction module: connected to the first prediction module, used to analyze the first feature vector when the prediction result is in an abnormal state and construct an incremental training set; Incremental training module: Connected to the incremental training set construction module, it is used to repeatedly train the fault classification model using the incremental training set to obtain a new model; The second prediction module is connected to the incremental training module. It is used to obtain the first feature vector at the current time and input it into the new model to obtain the second prediction result.

[0060] To implement the above embodiments, this application also proposes an electronic device. Please see [link to relevant documentation]. Figure 3 , Figure 3 This is a schematic diagram of the structure of the electronic device provided in an embodiment of this application. For example... Figure 3 As shown, the electronic device 500 includes: a processor 501 and a memory 502 communicatively connected to the processor 501; the memory 502 stores computer-executable instructions; the processor 501 executes the computer-executable instructions stored in the memory to implement the method provided in the foregoing embodiments.

[0061] To implement the above embodiments, this application also proposes a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the methods provided in the foregoing embodiments.

[0062] To implement the above embodiments, this application also proposes a computer program product, including a computer program that, when executed by a processor, implements the methods provided in the foregoing embodiments.

[0063] The collection, storage, use, processing, transmission, provision, and disclosure of user personal information involved in this application all comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0064] It should be noted that personal information collected from users should be used for legitimate and reasonable purposes and should not be shared or sold outside of these legitimate uses. Furthermore, such collection / sharing should only be conducted after receiving the user's informed consent, including but not limited to notifying the user to read the user agreement / user notice and sign an agreement / authorization that includes authorization of relevant user information before the user uses the function. In addition, any necessary steps must be taken to protect and safeguard access to such personal information data and ensure that others with access to personal information data comply with their privacy policies and procedures.

[0065] This application is intended to provide an implementation scheme for users to selectively prevent the use or access to their personal information data. Specifically, this disclosure is intended to provide hardware and / or software to prevent or block access to such personal information data. Once personal information data is no longer needed, risks can be minimized by restricting data collection and deleting data. Furthermore, where applicable, such personal information is de-identified to protect user privacy.

[0066] In the foregoing descriptions of the embodiments, the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0067] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0068] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0069] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0070] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0071] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0072] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0073] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of this application.

Claims

1. A method for online monitoring of plunger pump faults based on multimodal data, characterized in that, Includes the following steps: Step 1: Construct a training set and train the fault classification model to obtain the trained fault classification model; Step 2: Acquire multimodal data during the operation of the plunger pump in real time and perform preprocessing; Step 3: Based on the preprocessed multimodal data, construct the first feature vector, input it into the trained fault classification model, and output the first prediction result; Step 4: If the prediction result is an abnormal state, analyze the first feature vector under the abnormal state and construct an incremental training set, including the following steps: Step 41: Call the first feature vector of the predicted abnormal state and define it as a candidate sample. Determine the consistency index between each candidate sample and historical samples in the historical fault case library. Preset a first target threshold. Based on the relationship between the first target threshold and the average Mahalanobis distance, determine whether the current candidate sample is a new fault type sample or a duplicate sample. If it is a new fault type sample, define it as the first undetermined sample. If it is a duplicate sample, add it to the historical fault case library to update or remove historical fault cases. If the consistency index between each candidate sample and historical samples in the historical fault case library is greater than the first target threshold, define the current candidate sample as a new fault type sample. If the consistency index between each candidate sample and historical samples in the historical fault case library is less than or equal to the first target threshold, define the current candidate sample as a duplicate sample. Step 42: Determine the similarity between each candidate sample and each historical sample in the historical fault case library. Preset a second target threshold. Based on the relationship between the second target threshold and the similarity, determine whether the current candidate sample is a first-quality sample or a second-quality sample. If the current candidate sample is a first-quality sample, define it as a second undetermined sample. If the current candidate sample is a second-quality sample, define it as a third undetermined sample. If the similarity between each candidate sample and each historical sample in the historical fault case library is greater than the second target threshold, define the current candidate sample as a first-quality sample. If the similarity between each candidate sample and each historical sample in the historical fault case library is less than or equal to the second target threshold, define the current candidate sample as a second-quality sample. Step 43: If a candidate sample is defined as both the first undetermined sample and the second undetermined sample, construct an incremental training set based on the candidate sample. The process of constructing an incremental training set based on candidate samples includes the following steps: Step 431: Call the preprocessed multimodal data of the candidate samples; Step 432: Input the data of each modality into the trained fault classification model separately, and output the second prediction result; Step 433: Define at least three uncertainty indicators for the second prediction result; Step 434: Construct a ratio between every two uncertainty indicators, preset a third target threshold, and when 2 / 3 of the ratios are greater than the third target threshold, define the second prediction result of the current mode as the first uncertainty, determine the uncertainty score based on the ratio, and determine the optimization threshold based on the uncertainty score. Step 435: The candidate samples that are simultaneously defined as the first undetermined sample and the second undetermined sample are compared with the optimization threshold. If the candidate sample is less than the optimization threshold, the candidate sample is defined as an incremental sample; if the candidate sample is greater than or equal to the optimization threshold, it is discarded. Step 5: Repeat the training of the fault classification model using the incremental training set to obtain a new model; Step 6: Obtain the first feature vector at the current time and input it into the new model to obtain the second prediction result.

2. The method according to claim 1, characterized in that, The multimodal data includes: vibration signals, pressure signals, temperature signals, and acoustic emission signals.

3. The method according to claim 2, characterized in that, The first feature vector consists of the following feature parameters: wavelet energy ratio of the vibration signal, dominant frequency energy density of the pressure signal, first-order difference mean of the temperature signal, and instantaneous frequency change rate of the acoustic emission signal.

4. A plunger pump fault online monitoring system based on multimodal data, used to execute the plunger pump fault online monitoring method based on multimodal data as described in any one of claims 1-3, characterized in that, include: Fault classification model: Construct a training set and train the fault classification model to obtain the trained fault classification model; Data acquisition and preprocessing module: Connected to the fault classification model, it is used to acquire multimodal data during the operation of the plunger pump in real time and perform preprocessing. First prediction module: connected to the data acquisition and preprocessing module, used to construct the first feature vector based on the preprocessed multimodal data, input it into the trained fault classification model, and output the first prediction result; Incremental training set construction module: connected to the first prediction module, used to analyze the first feature vector when the prediction result is in an abnormal state and construct an incremental training set; Incremental training module: Connected to the incremental training set construction module, it is used to repeatedly train the fault classification model using the incremental training set to obtain a new model; The second prediction module is connected to the incremental training module. It is used to obtain the first feature vector at the current time and input it into the new model to obtain the second prediction result.

Images