A method, apparatus, electronic device, and storage medium for predicting device status.

Abstract

This application discloses a method, apparatus, electronic device, and storage medium for predicting equipment status. The method includes: acquiring equipment operating data to be predicted; acquiring a status prediction model, which is trained using time-series data from multiple devices; and using the status prediction model to perform feature extraction and probability distribution mapping based on the equipment operating data, outputting the predicted equipment status. This application can achieve equipment status prediction, improve the efficiency and accuracy of equipment status prediction, help predict the operating status and potential faults of equipment in advance, and avoid significant losses due to failure to detect equipment malfunctions in a timely manner. It can be widely applied in the field of industrial equipment technology.

Description

Technical Field

[0001] This application relates to the field of industrial equipment technology, and in particular to a method, apparatus, electronic device and storage medium for predicting equipment status. Background Technology

[0002] Online monitoring and predictive maintenance of industrial equipment, as a core technology in modern industrial maintenance management, plays an irreplaceable role in preventing major safety accidents. By collecting, analyzing, and interpreting relevant data on equipment operation status online, early signals of potential internal faults can be detected in a timely manner. Therefore, predicting the condition of industrial equipment is particularly necessary, as it helps staff anticipate the equipment's operating status and potential malfunctions, enabling them to respond more promptly. Summary of the Invention

[0003] The main objective of this application is to provide a device, apparatus, electronic device, and storage medium for predicting device status, which can achieve device status prediction and improve the accuracy of device status prediction.

[0004] On the one hand, embodiments of this application propose a device state prediction method, the method comprising the following steps:

[0005] Obtain the equipment operation data to be predicted;

[0006] A state prediction model is obtained, which is trained using time-series data from multiple devices;

[0007] The state prediction model is used to extract features and map probability distributions based on the device's operating data, and outputs the predicted state of the device.

[0008] In some embodiments, obtaining the state prediction model specifically includes:

[0009] Obtain the sample dataset;

[0010] A state prediction model is constructed, and the state prediction model is trained and evaluated using the sample dataset.

[0011] In some embodiments, obtaining the sample dataset specifically includes:

[0012] Collect time series data of multiple prediction devices;

[0013] The time series data of each device are normalized to obtain normalized time series data of each device;

[0014] Obtain multiple device status prediction labels;

[0015] The normalized time series data of each device are transformed to determine the target device prediction status label corresponding to each device time series data from multiple device status prediction labels.

[0016] In some embodiments, the construction of the state prediction model, and the training and evaluation of the state prediction model using the sample dataset, specifically includes:

[0017] The sample dataset is divided into a training dataset and a validation dataset;

[0018] The state prediction model is trained using the training dataset, and the state prediction model is evaluated using the validation dataset to determine the model evaluation result.

[0019] Based on the model evaluation results, determine whether to continue training the state prediction model.

[0020] In some embodiments, the state prediction model includes a first feature extraction module, a second feature extraction module, and a feature mapping module. The step of using the state prediction model to perform feature extraction and probability distribution mapping based on the device operating data, and outputting the predicted device state, specifically includes:

[0021] The device operation data is input into the state prediction model, and the first feature extraction module is used to perform feature extraction and convolution operations on the device operation data to obtain the first feature data corresponding to the device operation data.

[0022] The second feature extraction module is used to perform word embedding transformation and feature fusion on the device operation data to obtain the corresponding second feature data. Based on the first feature data and the second feature data, feature fusion is performed to obtain the corresponding feature fusion data.

[0023] The feature mapping module is used to perform probability distribution mapping on the feature fusion data, and the device state prediction label with the highest probability is determined from multiple device state prediction labels. The device state prediction label with the highest probability is then obtained as the device prediction state.

[0024] In some embodiments, the method further includes:

[0025] Obtain feature weight data, and generate a corresponding class activation map based on the feature weight data, wherein the feature weight data is obtained by the second feature extraction module when performing feature fusion on the first feature data and the second feature data.

[0026] In some embodiments, obtaining feature weight data and generating a corresponding class activation graph based on the feature weight data specifically includes:

[0027] Obtain the data matrix corresponding to the device operation data;

[0028] The feature weight data is superimposed and mapped onto the data matrix to generate the class activation graph.

[0029] On the other hand, embodiments of this application propose a device for predicting device status, the device comprising:

[0030] The first module is used to acquire the equipment operation data to be predicted.

[0031] The second module is used to acquire a state prediction model, which is trained using time series data from multiple devices.

[0032] The third module is used to perform feature extraction and probability distribution mapping based on the device operation data using the state prediction model, and output the predicted state of the device.

[0033] On the other hand, embodiments of this application propose an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the device state prediction method described above.

[0034] On the other hand, embodiments of this application propose a computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned device state prediction method.

[0035] The embodiments of this application include at least the following beneficial effects: The equipment status prediction method, apparatus, electronic device, and storage medium provided in this application acquire the operating data of the equipment to be predicted, obtain a status prediction model, and use the status prediction model to perform feature extraction and probability distribution mapping based on the equipment operating data to output the predicted equipment status. This application can realize equipment status prediction, improve the efficiency and accuracy of equipment status prediction, help to predict the working status and possible failures of equipment in advance, and avoid significant losses caused by failure to detect equipment malfunctions in a timely manner. Attached Figure Description

[0036] Figure 1 This is a flowchart of a device status prediction method provided in an embodiment of this application;

[0037] Figure 2 This is a schematic diagram of the class activation graph in an embodiment of this application;

[0038] Figure 3 This is a schematic diagram of the structure of a device state prediction device provided in an embodiment of this application;

[0039] Figure 4 This is a schematic diagram of the hardware structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0040] 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 of this application and are not intended to limit it. In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with those of this application; they are merely examples of apparatuses and methods consistent with some aspects of the embodiments of this application as detailed in the appended claims.

[0041] It is understood that the terms “first,” “second,” etc., used in this application may be used herein to describe various concepts, but unless otherwise stated, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of the embodiments of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the words “if,” “when,” or “in response to a determination” as used herein may be interpreted as “when…” or “when…” or “in response to a determination.”

[0042] As used in this application, the terms "at least one", "multiple", "each", "any", etc., "at least one" includes one, two or more, "multiple" includes two or more, "each" refers to each of the corresponding multiples, and "any" refers to any one of the multiples.

[0043] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0044] It should be noted that in all specific embodiments of this application, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is obtained first. Furthermore, the collection, use, and processing of this data comply with relevant laws, regulations, and standards. In addition, when embodiments of this application require access to sensitive personal information of users, separate permission or consent from the user is obtained through pop-ups or redirection to confirmation pages. Only after obtaining the user's separate permission or consent is the necessary user-related data required for the proper functioning of these embodiments acquired.

[0045] Before providing a detailed description of the embodiments of this application, some of the nouns and terms involved in the embodiments of this application will be explained first. The nouns and terms involved in the embodiments of this application are subject to the following interpretations.

[0046] 1) Process industries: These are industrial sectors that produce goods or semi-finished products through the conversion of matter or energy, based on continuous or batch processes such as chemical, physical, or biological processes. These industries typically handle raw materials, transforming them into final products through a series of processes, which are often irreversible. Typical characteristics of process industries include continuity, large scale, and high automation.

[0047] 2) Digital Twin: This refers to a technological concept that involves building a virtual model of a physical entity or system in a digital environment to enable monitoring, analysis, and optimization throughout its entire lifecycle. Digital twin technology utilizes sensors, IoT devices, data analytics, and machine learning to map the real-time state, behavior, and performance of a physical entity to its virtual copy, thereby enabling real-time monitoring, simulation prediction, fault diagnosis, and remote control.

[0048] 3) Transformer: It is a deep learning model architecture. Its core innovation lies in its attention-based architecture, which abandons the traditional recurrent neural network (RNN) and convolutional neural network (CNN) structures, thereby achieving more efficient and flexible model training and inference.

[0049] 4) Softmax: A commonly used activation function, widely applied in multi-class machine learning models, especially in the output layer of neural networks. It transforms a real-valued vector into a probability distribution such that each output value is between 0 and 1, and the sum of all output values ​​is 1.

[0050] Digital twins are an advanced modeling technique that creates a virtual replica of a physical entity to simulate, predict, and optimize the entity's behavior and performance in real time. With the advancement of next-generation electronic information technologies such as the Internet of Things, big data, cloud computing, and artificial intelligence, the implementation of digital twins has gradually become possible. Besides the aerospace field, digital twins are also applied in industries such as power, shipbuilding, manufacturing, and oil and gas. Particularly in the industrial process sector, digital twins are considered an effective means of achieving interactive integration between the information world and the physical world of manufacturing.

[0051] In the field of industrial equipment technology, existing digital twin technology has the following drawbacks:

[0052] 1) When using a data-driven approach to create digital twin models of industrial equipment, deep learning technology is required to learn and analyze a large amount of historical data from process industry equipment. However, the on-site systems of process industry equipment are generally in normal operation, so the data generated by the equipment is all normal data. Therefore, deep learning network models can only learn the data patterns of industrial equipment under normal conditions, lacking data under abnormal conditions. The dataset has a serious imbalance problem, which leads to the deep learning model's inability to predict abnormal states of the equipment, resulting in poor prediction performance. In existing technologies, in order to obtain abnormal data of industrial equipment, it is necessary to use the mechanistic model of the industrial equipment to simulate abnormal states. However, the equipment and mechanistic reactions involved in industrial processes are complex and numerous, making the cost of using mechanistic models for simulation too high and difficult to cover all industrial equipment.

[0053] 2) Existing deep learning network models based on data-driven digital twin technology mostly focus on local feature extraction, while there is relatively little design for multi-source heterogeneous temporal-spatial feature extraction. They lack multi-dimensional, multi-modal, and multi-feature expression methods, and the information expression is not clear enough, which limits the robust and efficient digital twin modeling design ideas.

[0054] 3) Data-driven digital twin models for process industries are constructed using deep learning technology. Although deep learning network models can learn from historical data and predict future data to obtain prediction results, they cannot derive the basis for the deep learning model's judgment of prediction results. Therefore, data-driven digital twin models for process industries lack a certain degree of interpretability.

[0055] Based on this, embodiments of this application propose a method, apparatus, electronic device, and storage medium for predicting equipment status. This method utilizes a deep learning network model to classify equipment status into different category labels based on the fluctuation range of equipment operating data, improving the quality of historical datasets. This allows the digital twin model to better learn from the data, uncovering the basis for prediction by the deep learning network model, visualizing the data that contributes most to the prediction, providing a certain degree of interpretability to the digital twin model, enhancing its modeling capabilities, and effectively improving the accuracy of data-driven digital twin models in predicting the status of industrial equipment.

[0056] Reference Figure 1 , Figure 1 This is an optional flowchart of a device state prediction method provided in an embodiment of this application. The method may include, but is not limited to, steps S101 to S103:

[0057] Step S101: Obtain the equipment operation data to be predicted;

[0058] Step S102: Obtain the state prediction model, which is trained using time series data from multiple devices;

[0059] Step S103: Using the state prediction model, feature extraction and probability distribution mapping are performed based on the equipment operation data to output the predicted state of the equipment.

[0060] In some embodiments, the state prediction model is a deep learning network model. Optionally, multi-source heterogeneous device operation data is input into the state prediction model, and the state prediction model is used to extract the temporal-spatial features of the multi-source heterogeneous data, output the predicted state of the device, and improve the accuracy of device state prediction.

[0061] In some embodiments, step S102 may include, but is not limited to, steps S201 to S202:

[0062] Step S201: Obtain the sample dataset;

[0063] Step S202: Construct a state prediction model and train and evaluate the state prediction model using a sample dataset.

[0064] In some embodiments, step S201 may include, but is not limited to, steps S301 to S304:

[0065] Step S301: Collect time series data of multiple prediction devices;

[0066] Step S302: Perform data normalization processing on the time series data of each device to obtain normalized time series data of each device;

[0067] Step S303: Obtain multiple device status prediction labels;

[0068] Step S304: Perform data transformation on the normalized time series data of each device, and determine the target device prediction status label corresponding to each device time series data from multiple device status prediction labels.

[0069] In some embodiments, the predictive equipment is process industry equipment, and time series data of multiple predictive equipment are collected, with the predictive equipment defined as X. 1 X 2 , ..., X n Among them, the prediction device X n Device time-series data generated over a specified period of time can be represented as {x} M 1,x M 2,…,x M t}, where M∈[1,n], x M tThis represents the device data of the Mth prediction device at time t, meaning that each prediction device has corresponding time series data, and n is the number of prediction devices.

[0070] Then, the time series data of each device are normalized. Optionally, the Min-Max (minimum value - maximum value) normalization method can be used for normalization, and the calculation formula is as follows:

[0071]

[0072] Where, x norm The normalized data for x, where x refers to the time series data of the Mth device mentioned above {x}. M 1,x M 2,…,x M t For each value in}, x max For the time series data of the Mth device {x M 1,x M 2,…,x M t The maximum value in}, x min The time series data of the Mth device {x M 1,x M 2,…,x M t The minimum value in}.

[0073] Normalizing device time-series data can scale data features to a specific range, typically between 0 and 1, to eliminate the impact of different units of measurement between different devices on classification models, thus helping to improve the training efficiency and performance of deep learning network models.

[0074] For example, since the data values ​​of various process industry equipment fluctuate at different rates, in order to highlight outliers with larger fluctuations and reduce the interference of equipment data noise fluctuations on the model, the normalized equipment time series data corresponding to a certain prediction equipment is transformed using the following formula:

[0075]

[0076] Where x′ is the predicted state label of the target device in the normalized device time series data. is the mean of the normalized device time series data, std is the standard deviation of the normalized device time series data, thr is the custom classification threshold, and 0, 1, 2, and 3 are different device predicted state labels.

[0077] After the above data transformation and processing, the target device prediction state label corresponding to each prediction device can be obtained, which serves as the classification prediction task target of the deep learning neural network, that is, to predict the target device prediction state of the prediction device in the next moment.

[0078] In some embodiments, step S202 may include, but is not limited to, steps S401 to S403:

[0079] Step S401: Divide the sample dataset into a training dataset and a validation dataset;

[0080] Step S402: Train the state prediction model using the training dataset, evaluate the state prediction model using the validation dataset, and determine the model evaluation result.

[0081] Step S403: Based on the model evaluation results, determine whether to continue training the state prediction model.

[0082] In some embodiments, the sample dataset is divided into a training dataset and a validation dataset in a certain ratio (e.g., 4:1) for the state prediction model to be trained and evaluated.

[0083] Optionally, during model training, the cross-entropy loss function is used as the training loss function. The cross-entropy loss function is as follows:

[0084]

[0085] Where L is the cross-entropy loss value, N is the number of training samples, C is the number of device state categories, and y ij Let i be the true label of the training data sample in device state category j. Calculate the cross-entropy loss based on the prediction result and the true label, then calculate the gradient using the chain rule to update the neural network weights of the state prediction model. The iterative optimizer used for model training is the Adam optimizer, and the learning rate can usually be set to 0.001. The parameters can be adjusted according to the actual performance. After repeated iterative training and optimization until the loss converges or the preset number of training rounds is reached, the trained state prediction model is obtained.

[0086] Then, the generalization ability of the state prediction model on unseen data is evaluated by validating the dataset. The model evaluation metrics used include accuracy, precision, recall, and F1 score. The specific values ​​corresponding to each model evaluation metric are determined. When the model evaluation results show that the specific values ​​corresponding to each model evaluation metric exceed the threshold set for each model evaluation metric, the training of the state prediction model is terminated; otherwise, the training of the state prediction model continues.

[0087] In some embodiments, step S103 may include, but is not limited to, steps S501 to S503:

[0088] Step S501: Input the equipment operation data into the state prediction model, and use the first feature extraction module to perform feature extraction and convolution operations on the equipment operation data to obtain the first feature data corresponding to the equipment operation data;

[0089] Step S502: Use the second feature extraction module to perform word embedding transformation and feature fusion on the device operation data to obtain the corresponding second feature data. Based on the first feature data and the second feature data, perform feature fusion to obtain the corresponding feature fusion data.

[0090] Step S503: Use the feature mapping module to perform probability distribution mapping on the feature fusion data, determine the device state prediction label with the highest probability from multiple device state prediction labels, and obtain the device state prediction label with the highest probability as the device prediction state.

[0091] In some embodiments, the device operation data is the time series data of the target device. The first feature extraction module performs feature extraction and convolution operations on the device operation data. Specifically, the first feature extraction module first uses a one-dimensional linear transformation to extract time series features from the device operation data of the target device, mapping the original input data to a higher-dimensional feature space, as shown in the following formula:

[0092] z = xA T +b;

[0093] Where z is the first feature data obtained after feature extraction of the equipment operation data, x is the equipment operation data, and A T Let b be the weight matrix of the state prediction model, and b be the bias coefficient.

[0094] Then, a three-layer convolution operation is performed on the first feature data using a two-dimensional convolution kernel. After each convolution, a ReLU activation function is used to introduce non-linearity. At the same time, a max pooling layer is used to compress the extracted feature information. After each convolution and max pooling layer, the feature dimension is halved, and local features in the time and space dimensions (such as the first feature data mentioned above) can be extracted step by step. Finally, average pooling is used to map the local features to the same dimension.

[0095] In some embodiments, the second feature extraction module is designed based on the Transformer model. It performs word embedding transformation and feature fusion on device operation data. The second feature extraction module includes a word embedding unit, a positional encoding unit, and a multi-head attention mechanism unit. Optionally, the device operation data is first input into the word embedding unit and converted into word embedding vectors. Simultaneously, the word embedding vectors are divided into blocks, and the block-based word embedding vectors are input into the positional encoding unit. Since the Transformer model does not have built-in ability to process sequence order, positional encoding is introduced to provide positional information. The positional encoding generated by sine and cosine functions is shown in the following equation:

[0096]

[0097] Where x refers to the time series data of the Mth device mentioned above {x M 1,x M 2,…,x M t In the formula}, pos is the position of the word embedding vector after the data X is segmented, i is the dimension index, and d is the position of the word embedding vector. model It represents the dimension of the model, and PE() is the location encoding function.

[0098] Then, the position-encoded data is input into the encoder, which consists of N identical encoder layers stacked together. Each encoder layer includes a multi-head mutual attention mechanism, a feedforward fully connected network, residual connections, and layer normalization to extract the spatiotemporal global key features of the device operation data (such as the second feature data mentioned above).

[0099] The multi-head attention mechanism unit is used to fuse the local features and global key features of spatiotemporal data extracted by the above model. The multi-head attention mechanism can capture the relationship between input features at different degrees through each head, and can also flexibly adjust the weights between different heads. It has very good adaptability and flexibility, thereby improving the model's ability to model complex relationships, fuse features, and obtain feature fusion data.

[0100] In some embodiments, the feature mapping module includes a normalization layer, a fully connected layer, and a Softmax activation function. Optionally, after the above-mentioned feature fusion data is sequentially input into the normalization layer, the fully connected layer, and the Softmax activation function, the output can be mapped to the probability distribution of the predicted state labels of each device. The predicted state label of the device with the highest probability is the predicted state of the device predicted by the model.

[0101] In some embodiments, the above-described device state prediction method may further include step S601:

[0102] Step S601: Obtain feature weight data and generate a corresponding class activation map based on the feature weight data. The feature weight data is obtained by the second feature extraction module when performing feature fusion on the first feature data and the second feature data.

[0103] In step S601 of some embodiments, the class activation map is a heatmap used to show which regions in the input image contribute the most to the prediction of a specific category. It highlights the areas of interest to the model by overlaying color or brightness onto the image, obtains the data matrix corresponding to the device operating data, and maps the feature weight data onto the data matrix to generate the class activation map.

[0104] Specifically, activation graph-like techniques were used to extract the gradient weight matrix of the state prediction model during the prediction process. The gradient weight matrix is ​​the feature weight matrix output by the multi-head attention mechanism during the prediction process of the above model. The specific process is as follows:

[0105] First, calculate the prediction score y of the predicted state c of the device. c The attention weight matrix A relative to the h-th attention head in the multi-head attention mechanism unit h gradient: Then, for each attention head h, the gradients of all attention weight matrices are averaged to obtain the weight a of that attention head. h The calculation formula is as follows:

[0106]

[0107] Where Z is the size of the attention weight matrix A. Then, the gradient weight matrix L is calculated by weighted summation of the weights of each attention head, as shown in the following formula:

[0108]

[0109] ReLU is a commonly used activation function that sets negative input values ​​to 0 while leaving positive values ​​unchanged.

[0110] Finally, the gradient weight matrix L is superimposed on the data matrix, mapping it to the correlation degree of industrial equipment data. This reflects the contribution of each data value in the equipment operation data to the model prediction results, improves the transparency of the model, and gives the digital twin model a certain degree of interpretability.

[0111] In some embodiments, for example, taking the prediction of the state of water treatment-related equipment as an example, the above-described equipment state prediction method is applied to predict the equipment state of water treatment-related equipment, as follows:

[0112] The first step is to collect the equipment operation data of the target equipment. The equipment operation data records the status information of 16 sensors in the target equipment, including flow rate, pressure, liquid level, pump status, valve status, etc. The equipment operation data records the operating status data of each sensor in the equipment at different time points, including timestamp, sensor data, sensor identifier and actuator status. The sampling period for each piece of operating status data is 1 second.

[0113] The second step involves using the operating status data of each sensor from T+1 to T+16 seconds as input to form a 16*16 matrix. The state prediction model is then used to predict the predicted state of the target device at T+17 seconds. A one-dimensional linear transformation is then performed on the matrix data to transform it into a feature vector of length 128, enhancing the feature representation and helping the model capture the complex relationships in the historical data of the target device.

[0114] The third step involves local feature extraction on the 16*16 matrix data. Three convolutional layers are applied sequentially to the input data, followed by ReLU activation after each convolution. Max pooling is then used to reduce the dimensionality of the features, extracting higher-level feature representation vectors and gradually extracting key local features. Finally, an average pooling layer represents the features as a first feature vector of length 128. Next, global feature extraction is performed on the 16*16 matrix data. First, the input data is segmented into word embeddings. Then, positional encoding information is added to the word embedding vectors before inputting them into the Transformer encoder, where a second feature vector of length 128 is extracted.

[0115] The fourth step is to input the extracted first and second feature vectors into the feature fusion module, use a multi-head attention mechanism to capture the different degrees of relationship between the input features, flexibly adjust the weights between different features, and fuse the output into a third feature vector of length 128 (as described above in the feature fusion data).

[0116] The fifth step is to normalize the third feature vector, then pass it through a fully connected layer and Softmax activation, and finally output the final device prediction state after multiple iterations of training.

[0117] The sixth step involves using activation graph techniques to obtain the gradient weight matrix in the state prediction model, and then superimposing the gradient weight matrix onto the data matrix corresponding to the device operating data, such as... Figure 2As shown, in the class activation map obtained after overlay, the warm-colored area (such as the data in the red box) is the data area that the prediction model pays more attention to and also the data area that contributes the most to the prediction. The cool-colored area (such as the data in the green box) is the data area that the prediction model pays less attention to and the data area that contributes the least to the prediction. This improves the transparency of the prediction model and gives the model a certain degree of interpretability.

[0118] Reference Figure 3 , Figure 3 This is an optional structural diagram of a device for predicting device status provided in an embodiment of this application. This device can be used to implement the above-described device status prediction method and may include:

[0119] The first module is used to acquire the equipment operation data to be predicted.

[0120] The second module is used to acquire the state prediction model, which is trained using time series data from multiple devices.

[0121] The third module is used to perform feature extraction and probability distribution mapping based on the equipment operation data using the state prediction model, and output the predicted state of the equipment.

[0122] It is understood that the content of the above method embodiments is applicable to the present device embodiments. The specific functions implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0123] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the aforementioned device state prediction method. This electronic device can be any smart terminal, including a tablet computer.

[0124] It is understood that the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0125] Please see Figure 4 , Figure 4 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes:

[0126] The processor 901 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application.

[0127] The memory 902 can be implemented as a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 902 can store the operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 902 and is called and executed by the processor 901 using the device state prediction method of the embodiments of this application.

[0128] The input / output interface 903 is used to implement information input and output;

[0129] The communication interface 904 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0130] Bus 905 transmits information between various components of the device (e.g., processor 901, memory 902, input / output interface 903, and communication interface 904);

[0131] The processor 901, memory 902, input / output interface 903, and communication interface 904 are connected to each other within the device via bus 905.

[0132] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described device state prediction method.

[0133] It is understood that the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0134] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0135] This application provides a method, apparatus, electronic device, and storage medium for predicting equipment status. It acquires the operating data of the equipment to be predicted, obtains a status prediction model, and uses the status prediction model to perform feature extraction and probability distribution mapping based on the operating data, outputting the predicted equipment status. This application enables equipment status prediction, improving its efficiency and accuracy, and helps to anticipate the operating status and potential faults of equipment in advance, avoiding significant losses due to failure to detect equipment malfunctions in a timely manner.

[0136] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0137] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0138] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0139] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0140] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0141] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0142] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0143] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0144] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0145] It should be recognized that embodiments of the present invention can be implemented or carried out by computer hardware, a combination of hardware and software, or by computer instructions stored in a non-transitory computer-readable storage medium. The method can be implemented in a computer program using standard programming techniques, including a non-transitory computer-readable storage medium configured with a computer program, wherein the storage medium is configured such that the computer operates in a specific and predefined manner—according to the methods and drawings described in the specific embodiments. Each program can be implemented in a high-level procedural or object-oriented programming language to communicate with the computer system. However, if desired, the program can be implemented in assembly or machine language. In any case, the language can be a compiled or interpreted language. Furthermore, for this purpose, the program can run on a programmed application-specific integrated circuit (ASIC).

[0146] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0147] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A method for predicting equipment status, characterized in that, The method includes the following steps: Obtain the equipment operation data to be predicted; the equipment operation data includes flow rate, pressure, liquid level, pump status, and valve status; A state prediction model is obtained, which is trained using time-series data from multiple devices; The state prediction model is used to perform feature extraction and probability distribution mapping based on the device operation data, and outputs the predicted state of the device. The state prediction model specifically includes: Collect time series data of multiple prediction devices; The time series data of each device are normalized to obtain normalized time series data of each device; Obtain multiple device status prediction labels; Data transformation is performed on the normalized time series data of each device to determine the target device prediction state label corresponding to each time series data of each device from multiple device state prediction labels; Construct a state prediction model and train and evaluate the state prediction model using a sample dataset; The state prediction model includes a first feature extraction module, a second feature extraction module, and a feature mapping module. The process of using the state prediction model to extract features and map probability distributions based on the device's operating data, and outputting the predicted device state, specifically includes: The device operation data is input into the state prediction model, and the first feature extraction module is used to perform feature extraction and convolution operations on the device operation data to obtain the first feature data corresponding to the device operation data. The second feature extraction module is used to perform word embedding transformation and feature fusion on the device operation data to obtain the corresponding second feature data. Based on the first feature data and the second feature data, feature fusion is performed to obtain the corresponding feature fusion data. The feature mapping module is used to perform probability distribution mapping on the feature fusion data, and the device state prediction label with the highest probability is determined from multiple device state prediction labels. The device state prediction label with the highest probability is then obtained as the device prediction state.

2. The equipment status prediction method according to claim 1, characterized in that, The construction of the state prediction model, and the training and evaluation of the state prediction model using the sample dataset, specifically includes: The sample dataset is divided into a training dataset and a validation dataset; The state prediction model is trained using the training dataset, and the state prediction model is evaluated using the validation dataset to determine the model evaluation result. Based on the model evaluation results, determine whether to continue training the state prediction model.

3. The equipment status prediction method according to claim 1, characterized in that, The method further includes: Obtain feature weight data, and generate a corresponding class activation map based on the feature weight data, wherein the feature weight data is obtained by the second feature extraction module when performing feature fusion on the first feature data and the second feature data.

4. The equipment status prediction method according to claim 3, characterized in that, The step of obtaining feature weight data and generating a corresponding class activation graph based on the feature weight data specifically includes: Obtain the data matrix corresponding to the device operation data; The feature weight data is superimposed and mapped onto the data matrix to generate the class activation graph.

5. A device for predicting equipment status, characterized in that, The device includes: The first module is used to acquire the equipment operation data to be predicted; the equipment operation data includes flow rate, pressure, liquid level, pump status, and valve status; The second module is used to acquire a state prediction model, which is trained using time series data from multiple devices. The third module is used to perform feature extraction and probability distribution mapping based on the device operation data using the state prediction model, and output the predicted state of the device. The state prediction model specifically includes: Collect time series data of multiple prediction devices; The time series data of each device are normalized to obtain normalized time series data of each device; Obtain multiple device status prediction labels; Data transformation is performed on the normalized time series data of each device to determine the target device prediction state label corresponding to each time series data of each device from multiple device state prediction labels; Construct a state prediction model and train and evaluate the state prediction model using a sample dataset; The state prediction model includes a first feature extraction module, a second feature extraction module, and a feature mapping module. The process of using the state prediction model to extract features and map probability distributions based on the device's operating data, and outputting the predicted device state, specifically includes: The device operation data is input into the state prediction model, and the first feature extraction module is used to perform feature extraction and convolution operations on the device operation data to obtain the first feature data corresponding to the device operation data. The second feature extraction module is used to perform word embedding transformation and feature fusion on the device operation data to obtain the corresponding second feature data. Based on the first feature data and the second feature data, feature fusion is performed to obtain the corresponding feature fusion data. The feature mapping module is used to perform probability distribution mapping on the feature fusion data, and the device state prediction label with the highest probability is determined from multiple device state prediction labels. The device state prediction label with the highest probability is then obtained as the device prediction state.

6. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the device state prediction method according to any one of claims 1 to 4.

7. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the device state prediction method according to any one of claims 1 to 4.

Images