Equipment abnormity diagnosis method and related device

Abstract

The invention provides an equipment abnormality diagnosis method and a related device, and relates to the field of equipment monitoring, and the method comprises the steps: firstly training a target prediction model constructed based on an auto-encoder neural network based on training data (sound features, vibration features and electrical features) corresponding to target equipment; the target training model is used to acquire the working condition data of the to-be-tested equipment in the working process, the abnormity diagnosis is performed on the to-be-tested equipment based on the working condition data, the model training is performed by using the multi-dimensional data, the abnormal state of the to-be-tested equipment can be comprehensively predicted, and the accuracy of the abnormity diagnosis is improved. Moreover, the target prediction model constructed based on the autoencoder neural network is used for predicting the abnormal condition of the to-be-detected equipment, does not need to depend on professional domain knowledge and does not depend on historical fault data, and improves the generalization capability of the scheme.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment monitoring, and in particular to an equipment abnormality diagnosis method based on an autoencoder neural network and a related device. Background Art

[0002] In the industrial field, there are many types of equipment, including rotating equipment (such as pumps, compressors, gas turbines and steam turbines) and non-rotating equipment (such as heat exchangers, distillation towers, valves, etc.). Given the differences in equipment types, the diagnosis of equipment abnormalities has traditionally been refined into multiple branches based on professional fields: abnormal diagnosis based on vibration data, abnormal diagnosis based on sound data, and abnormal diagnosis based on electrical data. These diagnostic methods all rely on the deep knowledge and rich experience of experts in the field, and require manual extraction of corresponding feature information, such as time domain and frequency domain analysis of vibration signals. However, this method is highly dependent on prior knowledge and therefore has certain limitations. More importantly, there is a lack of a good fusion mechanism between these diagnostic methods, making it difficult to fully and accurately depict the overall health status of the equipment. Summary of the invention

[0003] In view of this, an embodiment of the present invention provides a device abnormality diagnosis method and related devices based on an autoencoder neural network to accurately depict the overall health status of the device.

[0004] To achieve the above objectives, the embodiments of the present invention provide the following technical solutions:

[0005] A device abnormality diagnosis method, comprising:

[0006] Acquiring training data of a target device, wherein the training data includes sound characteristics, vibration characteristics, and electrical characteristics of the target device under normal working conditions;

[0007] The training data is used to train the initial prediction model to obtain a target prediction model, wherein the input data of the target prediction model is the training data, the output of the target prediction model is the abnormal diagnosis result, and the target prediction model is a model constructed based on an autoencoder neural network;

[0008] Acquiring working condition data of the device under test during operation, wherein the working condition data includes sound characteristics, vibration characteristics, and electrical characteristics of the device under test during operation;

[0009] The target prediction model is used to perform abnormal diagnosis on the equipment under test based on the operating condition data.

[0010] Optionally, in the above-mentioned device abnormality diagnosis method, using the training data to train the initial prediction model includes:

[0011] Normalizing the training data, and using the normalized training data to train the initial prediction model;

[0012] Using the target prediction model to perform abnormal diagnosis on the equipment under test based on the operating condition data includes:

[0013] The operating condition data is normalized, and the target prediction model is used to perform abnormal diagnosis on the equipment under test based on the normalized operating condition data.

[0014] Optionally, in the above-mentioned device abnormality diagnosis method, the sound characteristics include: sound pressure level, spectral kurtosis, spectral slope, spectral flux, and spectral centroid;

[0015] The vibration characteristics include: one-fold frequency amplitude, effective value, and peak-to-peak value;

[0016] The electrical characteristics include: current, voltage, motor efficiency, motor load rate, and temperature.

[0017] Optionally, in the above-mentioned device abnormality diagnosis method, the configuration parameters of the target device are the same as those of the device under test.

[0018] Optionally, in the above device abnormality diagnosis method, obtaining training data of the target device includes:

[0019] Get the configuration parameters of the device under test;

[0020] Determine a corresponding target device based on the configuration parameters;

[0021] The training data matching the target device is obtained from the training database.

[0022] Optionally, in the above-mentioned device abnormality diagnosis method, obtaining training data of the target device includes:

[0023] Obtaining training data collection instructions input by a user;

[0024] Based on the training data collection instruction, taking the device to be tested as a target device, acquiring sound characteristics, vibration characteristics, and electrical characteristics of the target device;

[0025] The acquired sound features, vibration features, and electrical features are used as training data for the target device.

[0026] A device for diagnosing equipment abnormality, comprising:

[0027] A training data acquisition unit, used to obtain training data of a target device, wherein the training data includes sound characteristics, vibration characteristics, and electrical characteristics of the target device under normal working conditions;

[0028] A model training unit, used to train the initial prediction model using the training data to obtain a target prediction model, wherein the input data of the target prediction model is the training data, the output of the target prediction model is the abnormal diagnosis result, and the target prediction model is a model constructed based on an autoencoder neural network;

[0029] The working condition acquisition unit of the device under test is used to obtain working condition data of the device under test during its working process, wherein the working condition data includes sound characteristics, vibration characteristics and electrical characteristics of the device under test during its working process;

[0030] A diagnosis unit is used to use the target prediction model to perform abnormality diagnosis on the equipment under test based on the operating condition data.

[0031] A device for diagnosing abnormality of an equipment comprises at least one processing device and a storage device connected to the processing device, wherein:

[0032] The storage device is used to store computer programs;

[0033] The processing device is used to execute the computer program so that the device abnormality diagnosis device can implement any one of the device abnormality diagnosis methods described above.

[0034] A computer program product includes computer-readable instructions. When the computer-readable instructions are executed on a device abnormality diagnosis device, the device abnormality diagnosis device implements any one of the device abnormality diagnosis methods described above.

[0035] A computer storage medium, wherein the storage medium carries one or more computer programs, and when the one or more computer programs are executed by a device abnormality diagnosis device, the device abnormality diagnosis device can implement any one of the device abnormality diagnosis methods described above.

[0036] Based on the above technical solution, the equipment abnormality diagnosis solution provided by the embodiment of the present invention first trains the target prediction model constructed based on the autoencoder neural network based on the training data corresponding to the target device (sound features, vibration features, and electrical features), and then uses the target training model to obtain the working condition data of the device under test during operation, and performs abnormal diagnosis on the device under test based on the working condition data. By using multi-dimensional data for model training, it is possible to comprehensively predict the abnormal state of the device under test and improve the accuracy of abnormal diagnosis. In addition, the target prediction model constructed based on the autoencoder neural network is used to predict the abnormal situation of the device under test, which does not need to rely on professional field knowledge or historical fault data, thereby improving the generalization ability of this solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0038] Figure 1 A schematic diagram of a system architecture for implementing a device abnormality diagnosis method provided in an embodiment of the present application;

[0039] Figure 2 A schematic diagram of the structure of a terminal for implementing a method for diagnosing device abnormalities provided in an embodiment of the present application;

[0040] Figure 3 A schematic diagram of the structure of a server for implementing a device abnormality diagnosis method provided in an embodiment of the present application;

[0041] Figure 4 A schematic diagram of a process flow of a device abnormality diagnosis method provided in an embodiment of the present application;

[0042] Figure 5 It is a comparison chart of error and threshold;

[0043] Figure 6 This is a schematic diagram of the data fitting process during training;

[0044] Figure 7 Schematic diagram of error distribution;

[0045] Figure 8 A schematic diagram of a flow chart of a device abnormality diagnosis method provided in another embodiment of the present application;

[0046] Fig. 9 A schematic diagram of the structure of a device abnormality diagnosis device provided in an embodiment of the present application;

[0047] Fig.10 A schematic diagram of the structure of a device abnormality diagnosis device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] The following describes the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. The terms used in the implementation method section of the present application are only used to explain the specific embodiments of the present application, and are not intended to limit the present application.

[0050] The embodiments of the present application are described below in conjunction with the accompanying drawings. Those skilled in the art will appreciate that, with the development of technology and the emergence of new scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0051] The terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and need not be used to describe a specific order or sequential order. It should be understood that the terms used in this way can be interchangeable under appropriate circumstances, which is only to describe the distinction mode adopted by the objects of the same attributes when describing in the embodiments of the present application. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, so that the process, method, system, product or equipment comprising a series of units need not be limited to those units, but may include other units that are not clearly listed or inherent to these processes, methods, products or equipment.

[0052] The present application can be applied in the field of equipment detection. The present application can be, but is not limited to, applied to applications with data processing functions or terminals and cloud servers provided by cloud-side servers, etc., which will be introduced below.

[0053] See also Figure 1 , Figure 1 A schematic diagram of a system architecture for implementing a method for diagnosing device abnormalities is shown. The system may include a terminal 100 and a server 200. The server 200 may include one or more servers ( Figure 1 In the example, a server is included in the description), and the server 200 can provide the method provided by the embodiment of the present application for one or more terminals. In this solution, the terminal 100 can implement the device abnormality diagnosis method alone, and the terminal 100 and the server 200 can also implement the device abnormality diagnosis method in combination. When the two are combined, the terminal 100 is used to realize data collection of the device under test and upload the collected data to the server 200.

[0054] Among them, an application for implementing the equipment abnormality diagnosis method can be installed on the terminal 100. The above application and web page can provide an interface. The terminal 100 can receive relevant parameters (such as identification information of the equipment to be detected) entered by the user on the interactive interface and collect equipment operating condition data, and send the above parameters and data to the server 200. The server 200 can obtain processing results based on the received parameters and return the processing results to the terminal 100.

[0055] It should be understood that in some optional implementations, the terminal 100 can also complete the action of obtaining the processing result based on the received parameters by itself without the cooperation of the server, and the embodiments of the present application are not limited to this.

[0056] Next describe Figure 1 The product form of the mid-terminal 100;

[0057] The terminal 100 in the embodiment of the present application can be a mobile phone, a tablet computer, a wearable device, a vehicle-mounted device, an augmented reality (AR) / virtual reality (VR) device, a laptop computer, an ultra-mobile personal computer (UMPC), a netbook, a personal digital assistant (PDA), etc., and the embodiment of the present application does not impose any limitation on this.

[0058] Figure 2 An optional hardware structure diagram of the terminal 100 is shown.

[0059] refer to Figure 2 As shown, the terminal 100 may include a radio frequency unit 110, a memory 120, an input unit 130, a display unit 140, a camera 150 (optional), an audio circuit 160 (optional), a speaker 161 (optional), a microphone 162 (optional), an earphone jack 163 (optional), a processor 170, an external interface 180, a power supply 190 and other components. Those skilled in the art will appreciate that Figure 2 These are merely examples of terminals or multi-function devices and do not constitute limitations on the terminals or multi-function devices, which may include more or fewer components than those shown in the figures, or combinations of certain components, or different components.

[0060] The input unit 130 can be used to receive input digital or character information, and generate key signal input related to the user settings and function control of the portable multifunctional device. Specifically, the input unit 130 may include a touch screen 131 (optional) and / or other input devices 132. The touch screen 131 can collect the user's touch operations on or near it (such as the user's operation on or near the touch screen using any suitable object such as fingers, joints, stylus, etc.), and drive the corresponding connection device according to a pre-set program. The touch screen can detect the user's touch action on the touch screen, convert the touch action into a touch signal and send it to the processor 170, and can receive and execute the command sent by the processor 170; the touch signal at least includes the touch point coordinate information. The touch screen 131 can provide an input interface and an output interface between the terminal 100 and the user. In addition, the touch screen can be implemented using multiple types such as resistive, capacitive, infrared and surface acoustic wave. In addition to the touch screen 131, the input unit 130 can also include other input devices. Specifically, other input devices 132 may include, but are not limited to, one or more of a physical keyboard, function keys (such as a volume control key, a switch key, etc.), a trackball, a mouse, a joystick, and the like.

[0061] Among them, the input device 132 can receive input data and the like.

[0062] The display unit 140 may be used to display information input by the user or provided to the user, various menus of the terminal 100, interactive interfaces, file displays, and / or playback of any multimedia file. In the embodiment of the present application, the display unit 140 may be used to display the interface of human-computer interaction, processing results, etc.

[0063] The memory 120 can be used to store instructions and data. The memory 120 can mainly include an instruction storage area and a data storage area. The data storage area can store various data, such as multimedia files, texts, etc.; the instruction storage area can store software units such as operating systems, applications, instructions required for at least one function, or their subsets and extensions. It can also include a non-volatile random access memory; provide the processor 170 with hardware, software and data resources including management of computing and processing equipment, and support control software and applications. It is also used for the storage of multimedia files, and the storage of running programs and applications.

[0064] The processor 170 is the control center of the terminal 100. It uses various interfaces and lines to connect various parts of the entire terminal 100. By running or executing instructions stored in the memory 120 and calling data stored in the memory 120, it executes various functions of the terminal 100 and processes data, thereby controlling the terminal device as a whole. Optionally, the processor 170 may include one or more processing units; preferably, the processor 170 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface and application program, and the modem processor mainly processes wireless communication. It is understandable that the above-mentioned modem processor may not be integrated into the processor 170. In some embodiments, the processor and the memory may be implemented on a single chip, and in some embodiments, they may also be implemented separately on separate chips. The processor 170 may also be used to generate corresponding operation control signals, send them to corresponding components of the computing and processing device, read and process data in the software, especially read and process data and programs in the memory 120, so that each functional module therein performs corresponding functions, thereby controlling the corresponding components to act according to the requirements of the instructions.

[0065] Among them, the memory 120 can be used to store software codes related to the device abnormality diagnosis method, the processor 170 can execute the steps of the device abnormality diagnosis method, and can also schedule other units (such as the above-mentioned input unit 130 and display unit 140) to implement corresponding functions.

[0066] The radio frequency unit 110 (optional) can be used for receiving and sending information or receiving and sending signals during a call, for example, after receiving the downlink information of the base station, it is sent to the processor 170 for processing; in addition, the designed uplink data is sent to the base station. Generally, the RF circuit includes but is not limited to an antenna, at least one amplifier, a transceiver, a coupler, a low noise amplifier (Low Noise Amplifier, LNA), a duplexer, etc. In addition, the radio frequency unit 110 can also communicate with network devices and other devices through wireless communication. The wireless communication can use any communication standard or protocol, including but not limited to Global System of Mobile communication (Global System of Mobile communication, GSM), General Packet Radio Service (General Packet Radio Service, GPRS), Code Division Multiple Access (Code Division Multiple Access, CDMA), Wideband Code Division Multiple Access (Wideband Code Division Multiple Access, WCDMA), Long Term Evolution (Long Term Evolution, LTE), email, Short Messaging Service (SMS), etc.

[0067] In this embodiment of the present application, the RF unit 110 can send data to the server 200 and receive processing results sent by the server 200.

[0068] It should be understood that the radio frequency unit 110 is optional and can be replaced by other communication interfaces, such as a network port.

[0069] The terminal 100 also includes a power supply 190 (such as a battery) for supplying power to various components. Preferably, the power supply can be logically connected to the processor 170 through a power management system, so that the power management system can manage functions such as charging, discharging, and power consumption.

[0070] The terminal 100 further includes an external interface 180 , which may be a standard Micro USB interface or a multi-pin connector, and may be used to connect the terminal 100 to communicate with other devices, or to connect a charger to charge the terminal 100 .

[0071] Although not shown, the terminal 100 may also include a flashlight, a wireless fidelity (WiFi) module, a Bluetooth module, sensors with different functions, etc., which are not described in detail here. Some or all of the methods described below may be applied in the following embodiments. Figure 2 In the terminal 100 shown.

[0072] Next describe Figure 1 The product form of the server 200;

[0073] Figure 3 A structural diagram of a server 200 is provided, such as Figure 3 As shown, the server 200 includes a bus 201, a processor 202, a communication interface 203, and a memory 204. The processor 202, the memory 204, and the communication interface 203 communicate with each other via the bus 201.

[0074] The bus 201 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0075] The processor 202 may be any one or more of a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor (MP), or a digital signal processor (DSP).

[0076] The memory 204 may include a volatile memory, such as a random access memory (RAM). The memory 204 may also include a non-volatile memory, such as a read-only memory (ROM), a flash memory, a hard disk drive (HDD), or a solid state drive (SSD).

[0077] The memory 204 may be used to store software codes related to the device abnormality diagnosis method, and the processor 202 may execute the steps of the device abnormality diagnosis method of the chip, and may also schedule other units to implement corresponding functions.

[0078] It should be understood that the above-mentioned terminal 100 and server 200 can be centralized or distributed devices, and the processors in the above-mentioned terminal 100 and server 200 (such as processor 170 and processor 202) can be hardware circuits (such as application specific integrated circuit (ASIC), field-programmable gate array (FPGA), general-purpose processor, digital signal processor (DSP), microprocessor or microcontroller, etc.), or a combination of these hardware circuits. For example, the processor can be a hardware system with an instruction execution function, such as a CPU, DSP, etc., or a hardware system without an instruction execution function, such as an ASIC, FPGA, etc., or a combination of the above-mentioned hardware systems without an instruction execution function and hardware systems with an instruction execution function.

[0079] In order to solve the above problems, the embodiment of the present application provides a device abnormality diagnosis method. The device abnormality diagnosis method of the embodiment of the present application is described in detail below with reference to the accompanying drawings.

[0080] Reference Figure 4 , Figure 4 A schematic diagram of a device abnormality diagnosis method provided in an embodiment of the present application is shown in FIG. Figure 4 As shown, a data processing method provided in an embodiment of the present application may include steps S401 to S404, and these steps are described in detail below.

[0081] Step S401: Acquire training data of a target device, wherein the training data includes sound characteristics, vibration characteristics, and electrical characteristics of the target device under normal working conditions.

[0082] When training a model for diagnosing whether a target device has a fault, it is crucial to obtain high-quality training data. For the target device, the training data can cover multiple dimensions to fully reflect the operating status and potential fault characteristics of the device. In the technical solution disclosed in this embodiment, in addition to conventional electrical feature data, the training data can also include sound features and vibration features. The following is a detailed analysis of the sound feature data, vibration feature data, and electrical feature data:

[0083] Sound feature data: A microphone or sound sensor can be used to capture the sound of the device under test when it is running. These devices should be installed in a position where they can clearly capture the sound of the target device while avoiding interference from background noise. After obtaining the sound feature data through a microphone or sound sensor, the features that can be extracted from the sound feature data may include the frequency, amplitude, timbre, etc. of the sound. These feature reagent libraries can be extracted through signal processing algorithms (such as Fourier transform). The extracted sound feature data has a significant effect on detecting mechanical faults of the target device (such as bearing wear, motor imbalance, etc.), because these faults usually cause changes in the sound of the target device when it is working.

[0084] Vibration signature data: The vibration of the target device can be measured using accelerometers or vibration sensors. These sensors should be installed at key locations on the target device to capture vibration signals during the operation of the target device. Vibration signature data may include the frequency, amplitude, phase, etc. of the vibration. These features can be extracted using vibration analysis algorithms. Vibration signature data is very effective in detecting faults in rotating components of the target device, such as motors, fans, etc. Abnormal vibrations usually indicate wear or imbalance of the components.

[0085] Electrical characteristic data: The electrical parameters of the target device can be measured using instruments such as voltmeters, ammeters, or power factor meters. These parameters may include voltage, current, power factor, harmonic content, etc. The electrical characteristic data can be used directly as input to the model, or through further processing (such as calculating statistics such as mean and standard deviation) to extract features. Electrical characteristic data has a significant effect on detecting circuit faults (such as short circuits, overloads, etc.) of the target device. These faults usually cause changes in electrical parameters.

[0086] In summary, by collecting sound feature data, vibration feature data, and electrical feature data, the operating status and potential fault characteristics of the target device can be fully reflected. These data will serve as the basis for training models to diagnose whether the device has a fault.

[0087] Furthermore, after obtaining the training data, the present application may also perform preprocessing on the training data, such as denoising, normalization, etc., to improve the training effect and generalization ability of the model.

[0088] Step S402: using the training data to train the initial prediction model to obtain a target prediction model, wherein the input data of the target prediction model is the training data, and the output of the target prediction model is an abnormality diagnosis result.

[0089] This step mainly uses training data to train the initial prediction model. During the model training process, all normal data in the training data are used as training sets, and the established initial prediction model is used for training. The specific process may include:

[0090] Data partitioning: Divide the dataset into training set, validation set, and test set. The training set is used to train the model; the validation set is used to evaluate the model performance during the training process in order to adjust the parameters; the test set is used to evaluate the final performance of the model after the training is completed.

[0091] Select model architecture: Choose a suitable machine learning or deep learning model architecture based on the complexity of the problem and the characteristics of the data. For example, the autoencoder neural network model is selected in this solution.

[0092] Set model parameters: According to the selected model architecture, set the model parameters, such as learning rate, number of network layers, number of neurons, activation function, etc. These parameters will affect the training speed and performance of the model and need to be tuned through experiments.

[0093] Model training: Use the training set and corresponding labels to train the model. The training process usually involves multiple iterations, and each iteration updates the model's parameters to minimize the prediction error.

[0094] Model evaluation: After each iteration, the validation set is used to evaluate the performance of the model. Common evaluation indicators include accuracy. Based on the evaluation results, the parameters of the model are adjusted to optimize the performance. Specifically, the evaluation process may include: By training the model with normal training data, the autoencoder learns the encoding format and reconstruction method of normal data. In the test phase, when a new data set (which may contain normal data and abnormal data) is provided to the trained autoencoder, the encoder will encode the data set according to the encoding format of normal data. Then, the decoder will try to decode the encoded data back to the original form according to the learned reconstruction method. Figure 5 is a comparison chart of error and threshold, based on Figure 5The comparison result of the error and the threshold shown can be used to preliminarily determine whether the device is abnormal. If the error continues to be greater than the threshold for a period of time, it can be determined that the device is abnormal. If the error continues to be greater than the threshold at a discrete moment and the error is less than the threshold at other moments, it can be determined that the device is normal. That is, if the error (such as the mean square error MSE) between the decoded data set and the original input data set is within a preset threshold range, then the input data set can be considered "normal" because the autoencoder can reconstruct it well. If the error exceeds this threshold, then the input data set can be considered "abnormal" because the autoencoder cannot accurately reconstruct it according to the encoding format of "normal data". Use the training set for prediction, calculate the mean square error MSE of the result and the training set, and use the error obtained at the specified percentile as the threshold to indicate whether the data set is abnormal.

[0095] Feature selection and optimization: Select the features that contribute most to model performance in each training data through methods such as feature importance analysis. Remove redundant or noisy features to improve the efficiency and accuracy of the model.

[0096] Prevent overfitting: Use regularization, dropout and other techniques to prevent the model from overfitting on the training data. Monitor the loss function and validation set performance during training to detect and deal with overfitting problems in a timely manner. Figure 6 As shown, Figure 6 It is a schematic diagram of the data fitting process during training. Figure 6 It can be seen that the training set and the validation set can be well fitted, and there is no overfitting phenomenon.

[0097] Final training: After adjusting the model parameters and feature selection, the initial prediction model is finally trained using all the training data (including the validation set) to obtain the target prediction model.

[0098] Among them, the type of model can be selected according to user needs, and further considering the diversity of industrial equipment, strong professional field knowledge, and many devices do not have historical fault data. Therefore, the prediction model used in the technical solution disclosed in this application can be a model built based on an autoencoder neural network (Autoencoder Neural Network). The autoencoder neural network is an unsupervised learning neural network model. Its core idea is to compress the input data into a low-dimensional latent space representation, and then reconstruct the original data from this latent space representation to achieve data compression and feature learning. In this patent method, an autoencoder neural network is used to encode input feature data of different dimensions into low-dimensional variables, and then decode them back to the original data. The error between the decoded data and the original data will reflect the health status of the equipment. As the error increases, the equipment degradation becomes more serious. The distribution of errors is monitored to predict whether the equipment is abnormal. The distribution of errors is as follows: Figure 7 As shown, in Figure 6 In the example, it can be seen that most of the errors are within 0.15, which means that the data within this error is likely to be normal, and this value can be used as the threshold for abnormal judgment. It can be seen that the use of autoencoder neural network does not need to rely on professional domain knowledge or historical fault data. It starts from normal data and encodes and decodes in the format of normal data, thereby realizing the diagnosis of abnormal data.

[0099] The comprehensive abnormality diagnosis method for equipment based on autoencoder neural network does not need to rely entirely on prior knowledge and fault data. Starting from normal data, multimodal data can be integrated for comprehensive diagnosis. Based on a small amount of prior knowledge, some features of vibration data (peak-to-peak value, effective value, single frequency amplitude, etc.), some features of sound data (sound pressure level, spectral kurtosis, common quantity, etc.), and some features of electrical data (current, voltage, load rate, etc.) are integrated, and unsupervised machine learning is performed using autoencoder neural network to achieve comprehensive abnormality diagnosis of equipment.

[0100] Therefore, in this solution, an autoencoder model can be used as the prediction model. The autoencoder model is a general neural network, and its internal structure shows a certain symmetry. It includes three layers: an input layer, which is used to receive the original input data. Assuming there are n input features, the input layer will establish n neurons; a hidden layer, also called the encoding layer, which is located between the input layer and the output layer and is responsible for encoding the input data into a low-dimensional representation. The number of neurons in the hidden layer is usually much smaller than that in the input layer to achieve data compression. Assuming the number of neurons is m, it needs to satisfy m < n; an output layer, also called the decoding layer, which receives the output of the hidden layer and tries to reconstruct the original input data. The number of neurons in the output layer is the same as that in the input layer to ensure that the reconstructed data has the same dimension as the original input. In this method, TensorFlow is used to build the neural network model, and ELU is used as the activation function. ELU is an activation function that combines the ReLU and Sigmoid functions. It has the advantages of alleviating the vanishing gradient problem, having an output mean close to 0, and reducing the computational load.

[0101] Among them, TensorFlow is a widely used open-source machine learning library, which is particularly good at building and training deep neural networks. It provides rich APIs and tools, enabling researchers and developers to easily define, train, and deploy complex neural network models. ELU (Exponential Linear Unit) is an activation function, and its mathematical expression is usually: if the input x is greater than 0, the output is x; if x is less than or equal to 0, the output is a negative value close to 0 but not 0 (specifically α(exp(x) - 1), where α is a hyperparameter, usually taking a small value). The characteristic of the ELU function is that it can alleviate the vanishing gradient problem while keeping the positive input unchanged, which helps to retain the feature information of the data.

[0102] Step S403: Obtain the working condition data during the working process of the device to be tested.

[0103] After determining the target prediction model, the working condition data generated during the working process of the device to be tested is collected in real time. The type of the working condition data needs to be consistent with the type of the training data. The working condition data also includes the sound characteristics, vibration characteristics, and electrical characteristics during the working process of the device to be tested.

[0104] Step S404: Use the target prediction model to perform anomaly diagnosis on the device to be tested based on the working condition data.

[0105] In this step, after determining the target prediction model, the target prediction model (such as an auto-encoding model) can be saved to a local file using the save method, and then the target prediction model is encapsulated as a Pipline, the input parameter of the Pipline is real-time data (operating data of the device to be tested), and the return value is the prediction result. The encapsulated target prediction model and its corresponding prediction function are packaged and uploaded to the target intelligent terminal, which is used to run the encapsulated target prediction model. The target intelligent terminal can be an industrial Internet platform algorithm online service platform. After the platform deploys the prediction algorithm, the data reported by the device can be read and predicted in real time through the industrial Internet platform, and abnormal warnings can be issued in real time. Then, an automatic diagnosis application is used to obtain the operating data of the device to be tested in real time. The type of the operating data obtained in real time needs to be consistent with the type of the training data. After obtaining the operating data, the target intelligent terminal calls the above prediction function to use the target prediction model to perform equipment diagnosis on the device to be tested. The diagnosis results are saved in the database, and abnormal warnings are issued when abnormalities are detected in the device to be tested.

[0106] After obtaining the working condition data of the device under test, the target intelligent terminal will call the prediction function and use the working condition data as the input data of the target prediction model. After obtaining the working condition data, the target prediction model will automatically analyze the working condition data and predict whether the device under test has a fault based on the working condition data. Specifically, the process may include: after obtaining the working condition data of the device under test, first perform data preprocessing, including cleaning, formatting and feature engineering, to ensure data quality and adaptability. Then, select and load the target prediction model that has been trained based on a large amount of similar equipment data. The target prediction model receives the preprocessed working condition data as input, automatically performs in-depth analysis, identifies abnormal patterns or trends, and outputs the probability of equipment failure in the future. By comparing with the preset threshold, it is determined whether the equipment is likely to fail, and the alarm mechanism is triggered to notify relevant personnel. At the same time, the target prediction model can also provide maintenance suggestions to reduce the risk of failure.

[0107] In addition, during the model monitoring process, the model performance can be continuously monitored, and new data can be collected based on the actual fault conditions to optimize the model, forming a closed-loop feedback mechanism, continuously improving the prediction accuracy and reliability, and providing a scientific basis for the preventive maintenance of the equipment under test.

[0108] The above-mentioned solution disclosed in the embodiment of the present application can comprehensively predict the abnormal state of the device under test and improve the accuracy of abnormal diagnosis by using multi-dimensional data for model training.

[0109] In this embodiment, in order to further ensure the training effect and accuracy of the prediction model, the above-mentioned use of the training data to train the initial prediction model includes: using a normalization model to normalize the training data, and using the normalized training data to train the initial prediction model. Normalization operation plays a vital role in neural networks. It can not only improve the training efficiency and performance of the model, but also enhance the stability and generalization ability of the model. The maximum and minimum normalization model is used in this method. Correspondingly, the above-mentioned use of the target prediction model to perform abnormal diagnosis on the equipment under test based on the operating condition data specifically includes: using a normalization model to normalize the operating condition data, and using the target prediction model to perform abnormal diagnosis on the equipment under test based on the normalized operating condition data.

[0110] When it is necessary to use a normalization model to normalize the training data and the operating condition data, it is necessary to uniformly encapsulate the normalization model and the target prediction model, and upload them to the target intelligent terminal.

[0111] In this embodiment, the types of equipment to be monitored are different, and the types of training data and operating condition data are different. The specific types of data used as training data and operating condition data can be determined according to the type of equipment. In the technical solution disclosed in this embodiment, the present application also discloses the specific contents of some common sound characteristics, vibration characteristics and electrical characteristics. Specifically, the sound characteristics include: sound pressure level, spectral kurtosis, spectral slope, spectral flux, and spectral centroid; the vibration characteristics include: one-fold frequency amplitude, effective value, peak-to-peak value; the electrical characteristics include: current, voltage, motor efficiency, motor load rate, and temperature.

[0112] In this embodiment, devices of the same type have different configuration parameters. The operating data of the devices under test with different configuration parameters under the same operating conditions may be different. Therefore, in order to ensure the reliability of the constructed target prediction model, in this embodiment, the configuration parameters of the target device are the same as those of the device under test.

[0113] In this embodiment, the training database may contain training data of multiple devices. In this embodiment, when obtaining training data, the configuration parameters of the device to be tested may be obtained first, and then the corresponding target device may be determined based on the configuration parameters, and then training data matching the target device may be obtained from the training database.

[0114] In this embodiment, the training data can also be obtained by the device under test. In this case, the training data is the operating condition data of the device under test when it works normally in the historical data of the device under test. These operating condition input data are used as training data to train the initial prediction model.

[0115] In this embodiment, when there are slight abnormalities in the device under test, if these abnormalities have no effect on the operation of the device, the user can also consider that the device is in a normal state at this time, and use the working condition data page of the device in these states as training data. Therefore, in this embodiment, the user can freely select which working conditions of the device to use as training data. For details, see Figure 8 , the process may include:

[0116] Step S801: Obtaining a training data collection instruction input by a user.

[0117] The user can actively decide which operating conditions of the device under test are normal operating conditions. When the user determines that the device under test is in a normal operating condition, the user can enter a training data collection instruction into the system.

[0118] Step S802: Based on the training data collection instruction, the device to be tested is used as a target device, and the sound characteristics, vibration characteristics, and electrical characteristics of the target device are obtained.

[0119] After receiving the training data collection instruction, the system determines that the working condition of the device under test is normal at this time (regardless of whether the device under test is abnormal), and then collects the working condition data of the device under test.

[0120] Step S803: Using the acquired sound features, vibration features, and electrical features as training data for the target device.

[0121] In this embodiment, corresponding to the above method, a device for diagnosing equipment abnormality is also disclosed. For the specific working contents of each unit in the device, please refer to the contents of the above method embodiment.

[0122] The following is a description of an apparatus for diagnosing equipment abnormality provided by an embodiment of the present invention. The apparatus for diagnosing equipment abnormality described below and the apparatus for diagnosing equipment abnormality described above can be referenced to each other.

[0123] See also Fig. 9 , Fig. 9 This is a schematic diagram of the structure of a device abnormality diagnosis device provided in an embodiment of the present application. Fig. 9 As shown, the device abnormality diagnosis device comprises:

[0124] A training data acquisition unit 10 is used to obtain training data of a target device, wherein the training data includes sound characteristics, vibration characteristics, and electrical characteristics of the target device under normal working conditions;

[0125] A model training unit 20, used to train the initial prediction model using the training data to obtain a target prediction model, wherein the input data of the target prediction model is the training data, and the output of the target prediction model is an abnormality diagnosis result;

[0126] The device under test working condition acquisition unit 30 is used to obtain working condition data of the device under test during its working process, wherein the working condition data includes sound characteristics, vibration characteristics and electrical characteristics of the device under test during its working process;

[0127] The diagnosis unit 40 is used to perform abnormality diagnosis on the equipment under test based on the operating condition data by using the target prediction model.

[0128] The present application also provides a device for diagnosing abnormalities. Fig.10 As shown, it shows a schematic diagram of the structure of the device abnormality diagnosis device suitable for implementing the embodiment of the present application. The device abnormality diagnosis device in the embodiment of the present application may include but is not limited to fixed terminals such as mobile phones, laptops, PDAs (personal digital assistants), PADs (tablet computers), desktop computers, etc. Fig.10 The device abnormality diagnosis device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0129] like Fig.10 As shown, the device abnormality diagnosis device may include a processing device (such as a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage device 608 to a random access memory (RAM) 603. When the device abnormality diagnosis device is powered on, various programs and data required for the operation of the device abnormality diagnosis device are also stored in the RAM 603. The processing device 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0130] Typically, the following devices may be connected to the I / O interface 605: an input device 606 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 608 including, for example, a memory card, a hard disk, etc.; and a communication device 609. The communication device 609 may allow the device abnormality diagnosis device to communicate with other devices wirelessly or by wire to exchange data. Although Figure 6 The device abnormality diagnosis device having various devices is shown, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0131] Also provided in an embodiment of the present application is a computer program product including computer-readable instructions. When the computer-readable instructions are executed on a device abnormality diagnosis device, the device abnormality diagnosis device implements any device abnormality diagnosis method provided in an embodiment of the present application.

[0132] A computer-readable storage medium is also provided in an embodiment of the present application. The storage medium carries one or more computer programs. When the one or more computer programs are executed by the device abnormality diagnosis device, the device abnormality diagnosis device can implement any device abnormality diagnosis method provided in the embodiment of the present application.

[0133] It should also be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the drawings of the device embodiments provided by the present application, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines.

[0134] Through the description of the above implementation mode, the technicians in the field can clearly understand that the present application can be implemented by means of software plus necessary general hardware, and of course, it can also be implemented by special hardware including special integrated circuits, special CPUs, special memories, special components, etc. In general, all functions completed by computer programs can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be various, such as analog circuits, digital circuits or special circuits. However, for the present application, software program implementation is a better implementation mode in more cases. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a readable storage medium, such as a computer floppy disk, a U disk, a mobile hard disk, a ROM, a RAM, a disk or an optical disk, etc., including a number of instructions to enable a computer device (which can be a personal computer, a training device, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0135] In the above embodiments, all or part of the embodiments may be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments may be implemented in the form of a computer program product.

[0136] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from a website site, a computer, a training device, or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, training device, or data center. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a training device, a data center, etc. that includes one or more available media integrations. The available medium may be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)), etc.

Claims

1. A method for diagnosing equipment abnormality, characterized in that: include: Acquiring training data of a target device, wherein the training data includes sound characteristics, vibration characteristics, and electrical characteristics of the target device under normal working conditions; The training data is used to train the initial prediction model to obtain a target prediction model, wherein the input data of the target prediction model is the training data, the output of the target prediction model is the abnormal diagnosis result, and the target prediction model is a model constructed based on an autoencoder neural network; Acquiring working condition data of the device under test during operation, wherein the working condition data includes sound characteristics, vibration characteristics, and electrical characteristics of the device under test during operation; The target prediction model is used to perform abnormal diagnosis on the equipment under test based on the operating condition data.

2. The device abnormality diagnosis method according to claim 1, characterized in that: The initial prediction model is trained using the training data, including: Normalizing the training data, and using the normalized training data to train the initial prediction model; Using the target prediction model to perform abnormal diagnosis on the equipment under test based on the operating condition data includes: The operating condition data is normalized, and the target prediction model is used to perform abnormal diagnosis on the equipment under test based on the normalized operating condition data.

3. The device abnormality diagnosis method according to claim 1, characterized in that: The sound characteristics include: sound pressure level, spectral kurtosis, spectral slope, spectral flux, and spectral centroid; The vibration characteristics include: one-fold frequency amplitude, effective value, and peak-to-peak value; The electrical characteristics include: current, voltage, motor efficiency, motor load rate, and temperature.

4. The device abnormality diagnosis method according to claim 1, characterized in that: The configuration parameters of the target device are the same as those of the device under test.

5. The device abnormality diagnosis method according to claim 1, characterized in that: Obtaining training data for the target device includes: Get the configuration parameters of the device under test; Determine a corresponding target device based on the configuration parameters; The training data matching the target device is obtained from the training database.

6. The device abnormality diagnosis method according to claim 1, characterized in that: Obtain training data for the target device, including: Obtaining training data collection instructions input by a user; Based on the training data collection instruction, taking the device to be tested as a target device, acquiring sound characteristics, vibration characteristics, and electrical characteristics of the target device; The acquired sound features, vibration features, and electrical features are used as training data for the target device.

7. A device for diagnosing equipment abnormality, characterized in that: include: A training data acquisition unit, used to obtain training data of a target device, wherein the training data includes sound characteristics, vibration characteristics, and electrical characteristics of the target device under normal working conditions; A model training unit, used to train the initial prediction model using the training data to obtain a target prediction model, wherein the input data of the target prediction model is the training data, the output of the target prediction model is the abnormal diagnosis result, and the target prediction model is a model constructed based on an autoencoder neural network; The working condition acquisition unit of the device under test is used to obtain working condition data of the device under test during its working process, wherein the working condition data includes sound characteristics, vibration characteristics and electrical characteristics of the device under test during its working process; A diagnosis unit is used to use the target prediction model to perform abnormality diagnosis on the equipment under test based on the operating condition data.

8. A device for diagnosing abnormality of equipment, characterized in that: The method comprises at least one processing device and a storage device connected to the processing device, wherein: The storage device is used to store computer programs; The processing device is used to execute the computer program so that the device abnormality diagnosis device can implement the device abnormality diagnosis method as described in any one of claims 1 to 6.

Images