Compressor actuating mechanism fault diagnosis method and device, medium and equipment

Abstract

The invention discloses a compressor actuating mechanism fault diagnosis method and device, a medium and equipment. The method comprises the following steps: S10, acquiring multi-dimensional data and carrying out data preprocessing; s20, performing feature fusion and dimension reduction processing; and S30, training the data after feature fusion and dimension reduction processing based on a DHK-ELM model so as to perform fault diagnosis. According to the method, the running state of the compressor actuating mechanism can be comprehensively monitored, potential fault hidden dangers can be recognized in the early stage, and economic losses and potential safety hazards caused by sudden faults are reduced. The method not only can realize accurate fault diagnosis of the compressor actuating mechanism, but also has a fault prediction function, and can predict the occurrence of potential faults based on historical data and real-time data, provide early warning and maintenance suggestions and optimize the maintenance strategy of equipment, thereby improving the reliability and safety of the equipment.

Description

Technical Field

[0001] This invention relates to the field of mechanical equipment fault detection technology, and in particular to a method, device, medium and equipment for diagnosing faults in compressor actuators. Background Technology

[0002] Compressors are widely used in various industries such as energy, chemicals, and air conditioning and refrigeration. As key equipment, their actuating mechanisms include cylinders, pistons, valves, and motors. These mechanisms bear heavy workloads, and long-term operation and complex working conditions make them prone to different types of failures. These failures not only affect the normal operation of the equipment but may also lead to production stoppages, equipment damage, or even safety accidents. Therefore, fault diagnosis and prediction of compressor actuating mechanisms are particularly important.

[0003] However, existing fault diagnosis technologies mainly rely on traditional empirical methods, model-based methods, and signal processing and pattern recognition techniques, all of which have numerous limitations. While empirical methods are simple, they depend on the operator's subjective judgment and cannot accurately handle complex and unknown fault modes. Model-based methods require a deep understanding of the equipment's operating principles, but due to the nonlinear and dynamic characteristics of the compressor's actuating mechanism, accurate modeling is often very difficult, and they are easily affected by noise and environmental changes in practical applications. Signal processing methods, such as wavelet transform and Fourier transform, are commonly used for feature extraction and can provide some fault information, but their diagnostic accuracy is limited when processing complex, multi-dimensional signals. Meanwhile, traditional pattern recognition techniques, such as support vector machines and decision trees, also suffer from high computational complexity and overfitting issues in processing high-dimensional data and training models.

[0004] Extreme Learning Machine (ELM), as an efficient machine learning method, can reduce training time and has good computational efficiency by randomly generating connection weights from the input layer to the hidden layer and quickly calculating the output layer weights using the least squares method. However, the sensitivity of ELM to input feature selection and the limitations of a single kernel function restrict its application in complex fault diagnosis. Summary of the Invention

[0005] The main objective of this invention is to provide a method, device, and medium for diagnosing faults in compressor actuators, in order to solve the technical problems in the prior art that cannot comprehensively monitor the operating status of compressor actuators and have limitations such as insufficient real-time performance, limited diagnostic accuracy, high cost, and high maintenance requirements.

[0006] To achieve the above objectives, the present invention provides a method for diagnosing faults in a compressor actuator. The method includes: S10, acquiring multi-dimensional data and performing data preprocessing; S20, performing feature fusion and dimensionality reduction processing; and S30, training the data after feature fusion and dimensionality reduction processing based on the DHK-ELM model to perform fault diagnosis.

[0007] Optionally, step S10 includes the following steps: S110, arranging vibration sensors, temperature sensors and pressure sensors at different parts of the compressor actuation mechanism; S120, acquiring multi-dimensional data based on the vibration sensors, temperature sensors and pressure sensors; S130, preprocessing the acquired multi-dimensional data.

[0008] Optionally, the multi-dimensional data includes vibration signals, temperature changes, and pressure fluctuation signals.

[0009] Optionally, step S130 includes the following steps: S1310, removing high-frequency noise from the signal based on a low-pass filter, wherein the high-frequency noise refers to sound with a frequency range of 1000Hz-20000Hz; S1320, normalizing and scaling the denoised signal.

[0010] Optionally, in step S20, feature fusion and dimensionality reduction are performed based on a deep autoencoder, wherein the deep autoencoder includes an encoder and a decoder.

[0011] Optionally, step S20 includes the following steps: S310, compressing multi-source input data into a low-dimensional latent space based on the encoder to obtain information, wherein the low-dimensional latent space refers to a spatial system with fewer than three dimensions, including one-dimensional and two-dimensional spaces, and the key feature information includes semantic representation vectors; S320, reconstructing multi-dimensional data based on the decoder to ensure the preservation of the information and the removal of redundant information.

[0012] Optionally, step S30 includes the following steps: S310, constructing a DHK-ELM model; S320, passing the feature fusion data as input to the DHK-ELM model; S330, the multi-dimensional data input to the input layer of the DHK-ELM model is processed through a multi-layer nonlinear Sigmoid activation function to automatically extract features from the input data for training, thereby performing fault diagnosis.

[0013] Furthermore, to achieve the above objectives, this application also provides a compressor actuator fault diagnosis device, the device comprising: a data acquisition module for acquiring multi-dimensional data, the data acquisition module including a vibration sensor, a temperature sensor, and a pressure sensor; a data preprocessing module for performing data preprocessing; a feature fusion module for performing feature fusion and dimensionality reduction processing; a training module for training the data after feature fusion and dimensionality reduction processing based on a DHK-ELM model; a fault diagnosis module for performing fault diagnosis; and a maintenance suggestion module for providing maintenance and suggestions for the diagnosed faults.

[0014] In addition, to achieve the above objectives, embodiments of this application also provide a computer-readable storage medium including instructions that, when executed on a computer, cause the computer to perform the compressor actuator fault diagnosis method described in any embodiment of this application.

[0015] Furthermore, to achieve the above objectives, embodiments of this application also provide a computing device, the computing device comprising: at least one processor, a memory, and an input / output unit; wherein, the memory is used to store a computer program, and the processor is used to call the computer program stored in the memory to execute the compressor actuator fault diagnosis method described in any embodiment of this application.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: The compressor actuator fault diagnosis method provided in this application uses a DHK-ELM model to train the data after feature fusion and dimensionality reduction. This enables rapid processing of high-dimensional data, accurate classification and diagnosis of fault types in the compressor actuator, and exhibits low computational complexity and high diagnostic accuracy. Furthermore, the DHK-ELM model further improves fault classification accuracy by handling complex nonlinear mapping problems through hybrid kernel functions. This method not only achieves accurate fault diagnosis of the compressor actuator but also possesses fault prediction capabilities. It can predict the occurrence of potential faults based on historical and real-time data, providing early warnings and maintenance suggestions, optimizing equipment maintenance strategies, and thereby improving equipment reliability and safety.

[0017] This method combines the advantages of deep learning and extreme learning machines. The deep hybrid kernel extreme learning machine (DHK-ELM) introduces multiple layers of nonlinear activation functions, enabling it to automatically extract deep features from high-dimensional data, making it particularly suitable for nonlinear, high-dimensional fault diagnosis tasks. Simultaneously, DHK-ELM employs hybrid kernel functions, combining various kernel functions, including Gaussian and Laplace kernels, to effectively handle complex nonlinear mapping problems. This method can process high-dimensional data quickly and efficiently with low computational complexity, significantly improving the speed and efficiency of fault diagnosis. Attached Figure Description

[0018] Figure 1 A flowchart of a compressor actuator fault diagnosis method provided in an embodiment of this application; Figure 2 This is a structural block diagram of the compressor actuation mechanism fault diagnosis device provided in the embodiments of this application; Figure 3 A flowchart illustrating the feature extraction process of the compressor actuator fault diagnosis method provided in this application embodiment; Figure 4 This is a schematic diagram of the structure of the medium provided in the embodiments of this application; Figure 5 A schematic diagram of the structure of a computing device provided in an embodiment of this application.

[0019] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0020] It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of the application. Rather, these embodiments are provided to make the disclosure more thorough and complete, and to fully convey the scope of the disclosure to those skilled in the art.

[0021] To address the aforementioned technical problems, embodiments of this application provide a method for diagnosing faults in a compressor actuator. This method can be executed by a computer, such as... Figure 1 As shown, the method may include the following steps: S10: Acquire multi-dimensional data and perform data preprocessing.

[0022] In an exemplary embodiment, step S10 may include the following steps: S110, arranging vibration sensors, temperature sensors and pressure sensors at different parts of the compressor actuation mechanism; S120, acquiring multi-dimensional data based on the vibration sensors, temperature sensors and pressure sensors; S130, preprocessing the acquired multi-dimensional data.

[0023] In an exemplary embodiment, the multidimensional data includes vibration signals, temperature changes, and pressure fluctuation signals.

[0024] Specifically, vibration sensors, temperature sensors, and pressure sensors are installed in different parts of the compressor's actuating mechanism to monitor the equipment's operating status in real time. These sensors can collect multi-dimensional data such as vibration signals, temperature changes, and pressure fluctuations at high frequency, ensuring the comprehensiveness and real-time nature of the data.

[0025] In an exemplary embodiment, step S130 may include the following steps: S1310, removing high-frequency noise from the signal based on a low-pass filter; S1320, normalizing and scaling the denoised signal.

[0026] High-frequency noise refers to sound with a frequency range of 1000Hz-20000Hz.

[0027] Specifically, the raw data collected by each sensor typically contains a large amount of noise and redundant information, which may affect the accuracy of the model if used directly for fault diagnosis. Therefore, this exemplary embodiment uses a low-pass filter to effectively remove high-frequency noise from the signal during the data preprocessing stage, ensuring the purity of the signal. Subsequently, normalization and feature scaling are performed to map the data collected by different sensors to a unified dimensional range, eliminating the impact of scale differences between data on subsequent processing.

[0028] Furthermore, a filter is a tool used to process signals, selectively allowing certain frequency components to pass while suppressing or filtering out other unwanted frequency components. Noise is typically distributed within a specific frequency range of a signal or exhibits high-frequency jitter; therefore, by designing appropriate filters, noise components can be reduced, preserving the cleaned signal. Since low-pass filters allow low-frequency signals to pass while blocking high-frequency signals, they are used to remove high-frequency noise from signals.

[0029] S20 performs feature fusion and dimensionality reduction.

[0030] In an exemplary embodiment, in step S20, feature fusion and dimensionality reduction are performed based on a deep autoencoder, wherein the deep autoencoder includes an encoder and a decoder.

[0031] In an exemplary embodiment, step S20 may include the following steps: S210, compressing multi-source input data into a low-dimensional latent space based on the encoder to obtain key feature information (such as semantic representation vectors); S220, reconstructing multi-dimensional data based on the decoder to ensure the retention of the key feature information and the removal of redundant information.

[0032] Low-dimensional potential space refers to spatial systems with fewer than three dimensions, typically including one-dimensional (such as a line) and two-dimensional (such as a plane) spaces.

[0033] Specifically, the feature fusion process employs cutting-edge data fusion and feature extraction technologies such as deep autoencoders to deeply integrate and reduce the dimensionality of multi-source data. Specifically, the deep autoencoder is used to automatically learn and extract high-dimensional nonlinear feature representations of the data. The encoder compresses multi-source input data into a low-dimensional latent space, and then the decoder reconstructs the original data, ensuring the preservation of key feature information and the removal of redundant information. The deep autoencoder, through its multi-layer hidden layer structure, can effectively capture and represent complex multi-dimensional fault feature information, improving the depth and breadth of feature fusion. The feature extraction flowchart is as follows: Figure 3 As shown.

[0034] Figure 3 In this context, the residual signal refers to the difference between the original signal and the predicted or estimated signal. Faults in the compressor's actuating mechanism include, but are not limited to, lubrication system faults, valve faults, and motor system faults. Normal state sensors include vibration sensors, temperature sensors, and pressure sensors that are free from any faults. Specifically, residual signal 1 refers to the difference between the original signal acquired by the vibration sensor and the predicted or estimated signal; residual signal 2 refers to the difference between the original signal acquired by the pressure sensor and the predicted or estimated signal; and residual signal 3 refers to the difference between the original signal acquired by the temperature sensor and the predicted or estimated signal.

[0035] In addition, the feature fusion step also includes optimizing feature representation using regularization techniques and activation functions to further enhance the feature recognition ability and the model's generalization performance, thereby significantly improving the accuracy of fault diagnosis and the overall reliability of the system.

[0036] S30 uses the DHK-ELM model to train the data after feature fusion and dimensionality reduction, thereby performing fault diagnosis.

[0037] In an exemplary embodiment, step S30 may include the following steps: S310, constructing a DHK-ELM model; S320, passing the feature fusion data as input to the DHK-ELM model; S330, inputting the multi-dimensional data into the input layer of the DHK-ELM model, and processing it through a multi-layer nonlinear Sigmoid activation function to automatically extract deep features from the input data for training, thereby performing fault diagnosis.

[0038] Specifically, the steps for constructing the DHK-ELM model are as follows: Step 1: Prepare data Sequence data: such as time series data or text data.

[0039] Preprocessing: Perform necessary preprocessing, such as standardization and normalization, which are all standard techniques.

[0040] Step 2: Design the DHK network Input layer: Receives sequence data.

[0041] Hidden layers: Use recursive structures, such as LSTM (Long Short-Term Memory) or GRU (Gated Recurrent Unit), to process sequential data.

[0042] Output layer: Designed as needed. For example, it can be a softmax layer for classification problems and a linear layer for regression problems.

[0043] Step 3: Design the ELM network Input layer: Connected to the output layer of DHK.

[0044] Hidden layer: Weights are randomly initialized, and the number of hidden units can usually be selected as needed.

[0045] Output layer: Calculate the output weights using the least squares method or other optimization algorithms.

[0046] Step 4: Combine DHK and ELM Forward propagation: First, the input data is processed through the DHK network, and then the processed output is used as the input of ELM.

[0047] Training: First train the DHK network, then fix its parameters and train only the output weights of the ELM network.

[0048] Step 5: Testing and Optimization Testing: Evaluate the model's performance using a test set.

[0049] Optimization: Improve model performance by adjusting network structure, parameters, or optimization algorithms as needed.

[0050] Furthermore, during training, the DHK-ELM (Deep Hybrid Kernel Extreme Learning Machine) model employs a hybrid kernel function strategy to handle the nonlinear relationships in the compressor actuator data. Traditional Extreme Learning Machines (ELM) use only a single kernel function, while DHK-ELM, by combining multiple kernel functions (Gaussian and Laplace kernels), effectively captures the complex nonlinear features in compressor actuator fault signals. This hybrid kernel function design enhances the model's adaptability to multi-dimensional and multi-type fault data, improving the accuracy of identifying different fault modes. In addition, the hybrid kernel function can map the original input data to a high-dimensional space, enabling the model to perform more accurate classification in a higher-dimensional space.

[0051] Furthermore, during training, the feature data, after undergoing multiple nonlinear transformations, is passed to the output layer. The model uses the least squares method to optimize the weights of the output layer, minimizing the error between the actual output and the target output. Because DHK-ELM randomly generates the connection weights from the input layer to the hidden layer and directly calculates the weights of the output layer, the training process is more efficient than traditional deep neural network training, and can be completed in a shorter time. This characteristic enables DHK-ELM to quickly adapt to real-time monitoring systems, diagnose the fault types of compressor actuators in real time, improve the real-time performance and accuracy of fault detection, and perform fault diagnosis using the trained model.

[0052] In the DHK-ELM model, the kernel function plays a crucial role. Hybrid kernel technology combines the advantages of multiple kernel functions, better adapting to different types of data. By introducing hybrid kernel technology, the model can more accurately describe the relationships between data, improving the accuracy of modeling and prediction.

[0053] Furthermore, the hybrid kernel function used in the DHK-ELM algorithm consists of a Gaussian kernel function and a Laplace kernel function. The Gaussian kernel function and the Laplace kernel function are linearly combined through weight parameters to simultaneously capture the local smoothness characteristics and sharp change features in the data, thereby enhancing the model's adaptability to complex nonlinear relationships in the input data and improving the accuracy and reliability of fault diagnosis.

[0054] Potential malfunctions in compressor actuators include lubrication system failures, piston jamming, valve malfunctions, motor overheating, motor system failures, poor lubrication, abnormal pressure, abnormal vibration, and unstable speed. These malfunctions typically affect compressor operating efficiency and can even lead to equipment damage or shutdown. The fault diagnosis algorithm of this invention can accurately identify these fault types and automatically determine the cause and location of the fault based on real-time collected vibration, temperature, and pressure signals, thereby improving the accuracy and real-time nature of fault diagnosis and ensuring the safe and stable operation of the compressor actuator.

[0055] The compressor actuator fault diagnosis method provided in this application uses a DHK-ELM model to train the data after feature fusion and dimensionality reduction. This enables rapid processing of high-dimensional data, accurate classification and diagnosis of fault types in the compressor actuator, and exhibits low computational complexity and high diagnostic accuracy. Furthermore, the DHK-ELM model further improves fault classification accuracy by handling complex nonlinear mapping problems through hybrid kernel functions. This method not only achieves accurate fault diagnosis of the compressor actuator but also possesses fault prediction capabilities. It can predict the occurrence of potential faults based on historical and real-time data, providing early warnings and maintenance suggestions, optimizing equipment maintenance strategies, and thereby improving equipment reliability and safety.

[0056] Based on the above embodiments, refer to Figure 2 Another embodiment of this application also provides a compressor actuator fault diagnosis device, which may include the following modules: Data acquisition module 210 is used to acquire multi-dimensional data. The data acquisition module 210 includes a vibration sensor, a temperature sensor and a pressure sensor. Data preprocessing module 220 is used to perform data preprocessing; The feature fusion module 230 is used for feature fusion and dimensionality reduction processing; Training module 240 is used to train the data after feature fusion and dimensionality reduction based on the DHK-ELM model; Fault diagnosis module 250 is used for fault diagnosis; The maintenance suggestion module 260 is used to maintain and suggest solutions for diagnosed faults.

[0057] In an exemplary embodiment, the data acquisition module 210 can also be used to collect and organize historical operating data and fault records of the compressor actuator, thereby establishing a comprehensive and detailed historical database. This data covers various operating parameters, including vibration, temperature, pressure, speed, fault occurrence time, and their corresponding fault types, aiming to provide rich foundational information for subsequent fault prediction. Through in-depth analysis of historical data, the system can identify the typical characteristics and development trends of different fault types, laying a solid foundation for the construction of predictive models.

[0058] In an exemplary embodiment, the fault diagnosis module 250 is used to analyze the real-time collected operating data using a trained DHK-ELM model, identify abnormal patterns in the current operating state, and thus perform fault diagnosis.

[0059] Specifically, this module uses a trained DHK-ELM model to efficiently process and accurately classify real-time data, thereby detecting abnormal features similar to historical failure modes and assessing their potential failure risks.

[0060] Based on the above embodiments, this application also provides a computer-readable storage medium, see reference. Figure 4The computer-readable storage medium shown is an optical disc 50, on which a computer program (i.e., a program product) is stored. When the computer program is run by a processor, it implements the steps described in the above method implementation, such as: S10, acquiring multi-dimensional data and performing data preprocessing; S20, performing feature fusion and dimensionality reduction processing; S30, training the data after feature fusion and dimensionality reduction processing based on the DHK-ELM model, thereby performing fault diagnosis. The specific implementation methods of each step will not be repeated here.

[0061] It should be noted that examples of the computer-readable storage medium may also include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other optical and magnetic storage media, which will not be elaborated here.

[0062] Furthermore, based on the above embodiments, this application also provides a computing device. Figure 5 A block diagram is shown of an exemplary computing device 60 suitable for implementing embodiments of the present application. The computing device 60 may be a computer system or a server. Figure 5 The computing device 60 shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0063] like Figure 5 As shown, the components of computing device 60 may include, but are not limited to: one or more processors or processing units 601, system memory 602, and bus 603 connecting different system components (including system memory 602 and processing unit 601).

[0064] The computing device 60 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the computing device 60, including volatile and non-volatile media, removable and non-removable media.

[0065] System memory 602 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 6021 and / or cache memory 6022. Computing device 60 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, ROM 6023 may be used to read and write non-removable, non-volatile magnetic media (…). Figure 5 (Not shown in the image, usually referred to as "hard drive"). Although not shown in... Figure 5The diagram illustrates that a disk drive for reading and writing to removable non-volatile disks (e.g., "floppy disks") and an optical disk drive for reading and writing to removable non-volatile optical disks (e.g., CD-ROMs, DVD-ROMs, or other optical media) can be provided. In these cases, each drive can be connected to a bus 603 connecting different system components via one or more data media interfaces. The system memory 602 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the embodiments of this application.

[0066] A program / utility 6025 having a set (at least one) of program modules 6024 may be stored, for example, in system memory 602, and such program modules 6024 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data, each or some combination of these examples may include an implementation of a network environment. Program modules 6024 typically perform the functions and / or methods described in the embodiments of this application.

[0067] The computing device 60 can also communicate with one or more external devices 604 (such as a keyboard, pointing device, display, etc.). This communication can be performed via input / output (I / O) interface 605. Furthermore, the computing device 60 can also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via network adapter 606. Figure 5 As shown, network adapter 606 communicates with other modules of computing device 60 (such as processing unit 601, etc.) via bus 603, which connects different system components. It should be understood that, although... Figure 5 Other hardware and / or software modules may be used in conjunction with computing device 60, as not shown in the diagram.

[0068] The processing unit 601 executes various functional applications and data processing by running programs stored in the system memory 602. For example, in S10, it acquires multi-dimensional data and performs data preprocessing; in S20, it performs feature fusion and dimensionality reduction processing; and in S30, it trains the data after feature fusion and dimensionality reduction based on the DHK-ELM model to perform fault diagnosis. It should be noted that although several units / modules or sub-units / sub-modules of the compressor actuation mechanism fault diagnosis device are mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, according to the embodiments of this application, the features and functions of two or more units / modules described above can be embodied in one unit / module. Conversely, the features and functions of one unit / module described above can be further divided and embodied by multiple units / modules.

[0069] In the description of this application, it should be noted that the terms "first", "second", and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0070] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0071] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the coupling or direct coupling or communication connection shown or discussed may be through some communication interface; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0072] The units described 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.

[0073] In addition, 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.

[0074] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, 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 a portion 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 several 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 described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0075] Finally, it should be noted that the above-described embodiments are merely specific implementations of this application, used to illustrate the technical solutions of this application, and not to limit them. The protection scope of this application is not limited thereto. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features, within the technical scope disclosed in this application. Such modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be covered within the protection scope of this application. Therefore, the protection scope of this application should be determined by the protection scope of the claims.

[0076] Furthermore, although the operations of the method of this application are described in a specific order in the accompanying drawings, this does not require or imply that these operations must be performed in that specific order, or that all the operations shown must be performed to achieve the desired result. Additionally or alternatively, certain steps may be omitted, multiple steps may be combined into one step, and / or one step may be broken down into multiple steps.

Claims

1. A method for diagnosing faults in a compressor actuator, characterized in that, The method includes the following steps: S10, acquire multi-dimensional data and perform data preprocessing; S20 performs feature fusion and dimensionality reduction processing; S30 uses the DHK-ELM model to train the data after feature fusion and dimensionality reduction, thereby performing fault diagnosis.

2. The compressor actuator fault diagnosis method according to claim 1, characterized in that, Step S10 includes the following steps: S110, vibration sensors, temperature sensors and pressure sensors are respectively arranged in different parts of the compressor actuation mechanism; S120, acquire multi-dimensional data based on the vibration sensor, the temperature sensor and the pressure sensor; S130, preprocess the acquired multi-dimensional data.

3. The compressor actuator fault diagnosis method according to claim 2, characterized in that, The multi-dimensional data includes vibration signals, temperature changes, and pressure fluctuation signals.

4. The compressor actuator fault diagnosis method according to claim 2, characterized in that, Step S130 includes the following steps: S1310, removes high-frequency noise from the signal based on a low-pass filter, wherein the high-frequency noise refers to sound with a frequency range of 1000Hz-20000Hz; S1320 performs normalization and feature scaling on the denoised signal.

5. The compressor actuator fault diagnosis method according to claim 1, characterized in that, In step S20, feature fusion and dimensionality reduction are performed based on a deep autoencoder, which includes an encoder and a decoder.

6. The compressor actuator fault diagnosis method according to claim 5, characterized in that, Step S20 includes the following steps: S210, based on the encoder, multi-source input data is compressed into a low-dimensional latent space to obtain key feature information. The low-dimensional latent space refers to a spatial system with fewer than three dimensions, including one-dimensional and two-dimensional spaces. The key feature information includes semantic representation vectors. S220, based on the decoder, reconstruct multi-dimensional data to ensure the retention of key feature information and the removal of redundant information.

7. The compressor actuator fault diagnosis method according to claim 1, characterized in that, Step S30 includes the following steps: S310, Construct the DHK-ELM model; S320, the feature fusion data will be used as input and passed to the DHK-ELM model; The S330 DHK-ELM model takes the multi-dimensional data as input to its input layer and processes it through a multi-layer nonlinear Sigmoid activation function. It automatically extracts features from the input data for training, thereby performing fault diagnosis.

8. A fault diagnosis device for a compressor actuator, characterized in that, include: A data acquisition module is used to acquire multi-dimensional data, and the data acquisition module includes a vibration sensor, a temperature sensor, and a pressure sensor. The data preprocessing module is used to perform data preprocessing. The feature fusion module is used for feature fusion and dimensionality reduction. The training module is used to train the data after feature fusion and dimensionality reduction based on the DHK-ELM model; The fault diagnosis module is used for fault diagnosis. The maintenance suggestion module is used to provide maintenance and suggestions for diagnosed faults.

9. A computer-readable storage medium, characterized in that, It includes instructions that, when executed on a computer, cause the computer to perform the data quality audit model construction method as described in any one of claims 1-7.

10. A computing device, characterized in that, The computing device includes: At least one processor, memory, and input / output unit; The memory is used to store computer programs, and the processor is used to call the computer programs stored in the memory to execute the compressor actuator fault diagnosis method according to any one of claims 1-7.

Images