A control analysis method based on industrial Internet

Abstract

The present invention discloses a control and analysis system based on the industrial Internet, and the operation method of the system includes: constructing and managing the industrial Internet rotating machinery model; performing industrial Internet detection, early warning and clustering control on the rotating machinery; further performing industrial Internet analysis, control and management on the rotating machinery; and performing log management and early warning statistics on the rotating machinery. The construction and management of the industrial Internet rotating machinery model includes: converting the original vibration signal of the industrial Internet rotating machinery from the time domain to the frequency domain, using Fourier transform to represent it as a frequency domain signal, and using it as the input of the convolutional neural network, extracting features layer by layer by using an encoder, thereby reducing the dimension of the data features, and mapping high-dimensional data to a low-dimensional feature space through a deep autoencoder, thereby constructing a deep autoencoding prediction model. The present invention has the characteristics of high accuracy and intelligent analysis and processing.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial Internet, and specifically to a control and analysis system based on industrial Internet. Background Art

[0002] Industrial Internet is a new type of industrial development model. It is mainly composed of technologies such as artificial intelligence and big data. In recent years, it has attracted more and more attention and attention. It aims to develop industrial production, operation and other links towards informatization. Rotating machinery fault diagnosis based on Industrial Internet is one of the research hotspots at this stage. With the continuous development of technology, more and more companies have begun to apply Industrial Internet to rotating machinery fault diagnosis to improve production efficiency and work safety. At present, the functions of rotating machinery are becoming more and more diversified. Rotating machinery equipment is developing towards intelligence and digitalization, so rotating machinery equipment has a higher degree of integration and a more complex structure. At the same time, it is also necessary to cope with a complex and changeable operating environment. If any component of the equipment is damaged or fails, it may cause the entire equipment to fail to work properly, resulting in production suspension and huge economic losses for the enterprise, and more seriously, it may cause safety accidents for the enterprise personnel. Therefore, it is necessary to design a control and analysis system based on Industrial Internet with high accuracy and intelligent analysis and processing. Summary of the invention

[0003] The purpose of the present invention is to provide a control and analysis system based on industrial Internet to solve the problems raised in the above-mentioned background technology.

[0004] In order to solve the above technical problems, the present invention provides the following technical solutions: a control analysis method based on industrial Internet, comprising:

[0005] Carry out the construction and management of the industrial Internet rotating machinery model;

[0006] Conduct industrial Internet detection, early warning and cluster control of rotating machinery;

[0007] Further analyze, control and manage rotating machinery using the Industrial Internet;

[0008] Carry out log management and early warning statistics of rotating machinery.

[0009] According to the above technical solution, the construction and management of the industrial Internet rotating machinery model includes:

[0010] The original vibration signal of the industrial Internet rotating machinery is converted from the time domain to the frequency domain, represented as a frequency domain signal using Fourier transform, and used as the input of the convolutional neural network. The encoder is used to extract features layer by layer to reduce the dimension of the data features, and the high-dimensional data is mapped to a low-dimensional feature space through a deep autoencoder, thereby constructing a deep autoencoder prediction model.

[0011] According to the above technical solution, the detection, early warning and clustering control of the rotating machinery by the industrial Internet includes:

[0012] After the construction of the deep autoencoder prediction model is completed, the vibration signals of the rotating mechanical equipment components under normal working conditions are standardized to obtain sampling data, which is then substituted into the deep autoencoder prediction model for detection and early warning processing;

[0013] The rotating machinery data information is processed by hierarchical clustering method, which makes it possible to intuitively present the similarity and distance relationship between data objects.

[0014] According to the above technical solution, the further analysis, control and management of the industrial Internet for rotating machinery includes:

[0015] A multi-scale approach is used to process rotating machinery data information, dividing it into a set number of subsets, and each subset is analyzed independently;

[0016] Further reconstruct the data format control of rotating machinery data information, reorganize and store it according to the set rules and structure, make it more standardized, easier to process and analyze, and more convenient for the visualization and presentation of rotating machinery data, so that it is more intuitive and easy to understand;

[0017] After identifying and selecting sensitive features that meet privacy settings from rotating machinery data, they are clustered using an adaptive hierarchical clustering method. After determining the optimal number of features by calculating the intra-class and inter-class distances, desensitization control processing is performed.

[0018] According to the above technical solution, the log management and early warning statistics of rotating machinery include:

[0019] Record every operation of the user in the rotating machinery system, so that the administrator can monitor and store the user's access and operation through security log records, and maintain the log records;

[0020] The early warning information of rotating machinery is counted according to time, equipment, and fault type, and displayed in a table, including alarm time, fault type, and vibration value at the time of fault.

[0021] According to the above technical solution, a control analysis system based on industrial Internet includes:

[0022] A control processing module is used to control rotating machinery through the industrial Internet;

[0023] Analysis and management module, used for industrial Internet analysis and management of rotating machinery;

[0024] The management and statistics module is used to perform information management and statistics on rotating machinery.

[0025] According to the above technical solution, the control processing module includes:

[0026] Model building module, used to build industrial Internet models of rotating machinery;

[0027] Detection and early warning module, used for operation detection and early warning of rotating machinery;

[0028] The cluster management module is used to perform cluster management on rotating machinery data information.

[0029] According to the above technical solution, the analysis management module includes:

[0030] A multi-scale processing module, used for multi-scale control processing of rotating machinery;

[0031] Reconstruction data module, used for industrial Internet reconstruction data processing of rotating machinery;

[0032] The feature selection module is used to perform feature selection and desensitization processing on rotating machinery.

[0033] According to the above technical solution, the management statistics module includes:

[0034] Log management module, used for log control and management of rotating machinery;

[0035] The early warning statistics module is used to collect statistics on the early warning information of rotating machinery.

[0036] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: the present invention, by providing a control processing module, an analysis management module and a management statistics module, effectively improves the analysis and management accuracy and processing efficiency of rotating machinery, avoids the generation of errors, and reduces the amount of data that needs to be analyzed and processed, thereby saving analysis and processing resources to a great extent. It can capture local features existing in signals, improve the efficiency and accuracy of detection and analysis of rotating machinery equipment, reduce data redundancy, and improve data quality, so as to enhance data accuracy and consistency. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0038] Figure 1 This is a flow chart of a control analysis method based on the industrial Internet provided in Embodiment 1 of the present invention;

[0039] Figure 2 This is a module structure diagram of a control and analysis system based on the industrial Internet provided in Example 2 of the present invention. DETAILED DESCRIPTION

[0040] 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.

[0041] Embodiment 1: Figure 1 This is a flowchart of a control analysis method based on the industrial Internet provided in the first embodiment of the present invention. This embodiment can be applied to the industrial Internet system. The method can be executed by a control analysis system based on the industrial Internet provided in the embodiment of the present invention. The system is composed of multiple software and hardware modules, such as Figure 1 As shown, the method specifically comprises the following steps:

[0042] S101. Carry out construction and management of industrial Internet rotating machinery models;

[0043] Exemplarily, in an embodiment of the present invention, the original vibration signal of the industrial Internet rotating machinery is converted from the time domain to the frequency domain, represented as a frequency domain signal using Fourier transform, and used as the input of the convolutional neural network. The encoder is used to extract features layer by layer, thereby reducing the dimension of the data features, and the high-dimensional data is mapped to a low-dimensional feature space through a deep autoencoder, thereby constructing a deep autoencoder prediction model; in this step, the frequency domain representation of the mechanical signal, that is, the spectrum, is obtained and used as a training sample, and then the sample data is converted into a value in the range of [0, 1] by normalization, the data set is divided according to a specific ratio to obtain the training set and test set of the model, and the unsupervised learning method is used to train the deep autoencoder layer by layer, thereby obtaining the number of hidden layers of the deep neural network, and for each deep autoencoder, the output of its hidden layer is used as the input of the next layer of deep autoencoder to complete the construction of the corresponding deep autoencoder prediction model, effectively improving the accuracy and processing efficiency of subsequent analysis and management, and avoiding the generation of errors.

[0044] S102. Conduct industrial Internet detection, early warning and cluster control of rotating machinery;

[0045] Exemplarily, in an embodiment of the present invention, after completing the construction of the deep autoencoding prediction model, the vibration signal of the rotating mechanical equipment component under normal working conditions is standardized to obtain the sampled data, which is then substituted into the deep autoencoding prediction model for detection and early warning processing; in this step, the sample is input into the deep autoencoding prediction model to obtain the reconstruction error vector of the sample, and the adaptive threshold is determined by the adaptive threshold algorithm. After obtaining the vibration signal of the rotating mechanical equipment, it is subjected to fast Fourier transform processing, and the result is normalized to obtain the corresponding data, which is then input into the deep autoencoding prediction model to obtain the corresponding reconstruction error, and compared with the set threshold range. If it does not meet the set threshold range, an early warning notification is immediately issued, thereby effectively improving the operating accuracy and analysis and processing efficiency of the rotating mechanical equipment.

[0046] The rotating machinery data information is processed by a hierarchical clustering method so that it can intuitively present the similarity and distance relationship between data objects. In this step, the rotating machinery data objects are clustered according to the hierarchical structure, and a tree structure is constructed according to the similarity or distance between the data objects, in which the top is a single cluster and the bottom is a separate data point. Each data point is used as a separate category, and then these categories are gradually merged until the preset number of clusters is reached or the merging threshold is reached, effectively reducing the amount of data that needs to be analyzed and processed, and greatly saving analysis and processing resources.

[0047] S103, further analyzing, controlling and managing the rotating machinery using the Industrial Internet;

[0048] Exemplarily, in an embodiment of the present invention, a multi-scale method is used to process rotating machinery data information, divide it into a set number of subsets, and perform independent analysis on each subset; in this step, multi-scale processing can help identify the characteristics of different rotating machinery equipment, and use the characteristic information obtained at multiple scales to better locate and identify the target rotating machinery equipment. At the same time, by analyzing and processing the rotating machinery vibration signals at different time and space scales, the analysis and processing performance is improved. For example, in fault diagnosis, wavelet transform can decompose the rotating machinery vibration signal into sub-bands within different frequency ranges, and each sub-band has a corresponding time and space scale, which can effectively capture the local features existing in the signal and improve the efficiency and accuracy of rotating machinery equipment detection and analysis.

[0049] The rotating machinery data information is further reconstructed for data format control, and is reorganized and stored according to set rules and structures, making it more standardized, easier to process and analyze, and more convenient for visualization and presentation of rotating machinery data, thereby making it more intuitive and easy to understand; in this step, by reconstructing the data format, the rotating machinery data can be organized and stored according to set structures and rules, making it more standardized and easy to understand and use, thereby improving the scalability and reusability of the data, reducing data redundancy, and improving data quality to enhance data accuracy and consistency.

[0050] After identifying and selecting sensitive features that meet the privacy settings from the rotating machinery data, they are clustered using an adaptive hierarchical clustering method, and after determining the optimal number of features by calculating the intra-class and inter-class distances, desensitization control processing is performed; in this step, the selected sensitive features are normalized to generate a normalized feature set, and the set is clustered using an adaptive hierarchical clustering method, which can make the desensitization processing of the rotating machinery data more efficient and accurate and avoid the generation of errors.

[0051] S104, performing log management and early warning statistics of rotating machinery;

[0052] Exemplarily, in an embodiment of the present invention, every operation of a user in an industrial Internet system is recorded, so that an administrator can monitor and store user access and operations through security log records, and maintain the log records;

[0053] The early warning information of rotating machinery is counted according to time, equipment, and fault type, and displayed in a table, including alarm time, fault type, and vibration value at the time of fault.

[0054] Embodiment 2: Embodiment 2 of the present invention provides a control analysis system based on industrial Internet. Figure 2 A schematic diagram of the module structure of a control and analysis system based on the industrial Internet provided in the second embodiment, such as Figure 2 As shown, the system includes:

[0055] A control processing module is used to control rotating machinery through the industrial Internet;

[0056] Analysis and management module, used for industrial Internet analysis and management of rotating machinery;

[0057] The management and statistics module is used to perform information management and statistics on rotating machinery.

[0058] In some embodiments of the present invention, the control processing module includes:

[0059] Model building module, used to build industrial Internet models of rotating machinery;

[0060] Detection and early warning module, used for operation detection and early warning of rotating machinery;

[0061] The cluster management module is used to perform cluster management on rotating machinery data information.

[0062] In some embodiments of the present invention, the analysis management module includes:

[0063] A multi-scale processing module, used for multi-scale control processing of rotating machinery;

[0064] Reconstruction data module, used for industrial Internet reconstruction data processing of rotating machinery;

[0065] The feature selection module is used to perform feature selection and desensitization processing on rotating machinery.

[0066] In some embodiments of the present invention, the management statistics module includes:

[0067] Log management module, used for log control and management of rotating machinery;

[0068] The early warning statistics module is used to collect statistics on the early warning information of rotating machinery.

[0069] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0070] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A control analysis method based on industrial Internet, characterized in that: include: Carry out the construction and management of the industrial Internet rotating machinery model; Conduct industrial Internet detection, early warning and cluster control of rotating machinery; Further analyze, control and manage rotating machinery using the Industrial Internet; Carry out log management and early warning statistics of rotating machinery; The construction and management of the industrial Internet rotating machinery model includes: The original vibration signal of the industrial Internet rotating machinery is converted from the time domain to the frequency domain, and represented as a frequency domain signal using Fourier transform. This signal is used as the input of the convolutional neural network. The encoder is used to extract features layer by layer to reduce the dimension of the data features, and the high-dimensional data is mapped to a low-dimensional feature space through a deep autoencoder, thereby constructing a deep autoencoder prediction model. In this step, the frequency domain representation of the mechanical signal, i.e., the spectrum, is obtained and used as a training sample. Then, the sample data is converted into a value in the range of [0, 1] by normalization, and the data set is divided according to a specific ratio to obtain the training set and test set of the model. The deep autoencoder is trained layer by layer using an unsupervised learning method to obtain the number of hidden layers of the deep neural network. For each deep autoencoder, the output of its hidden layer is used as the input of the next layer of deep autoencoder to complete the construction of the corresponding deep autoencoder prediction model, which effectively improves the accuracy and processing efficiency of subsequent analysis and management and avoids the generation of errors. The detection, early warning and clustering control of the rotating machinery by the industrial Internet include: After the construction of the deep autoencoder prediction model is completed, the vibration signals of the rotating mechanical equipment components under normal working conditions are standardized to obtain sampling data, which is then substituted into the deep autoencoder prediction model for detection and early warning processing; In this step, the sample is input into the deep autoencoder prediction model to obtain the reconstruction error vector of the sample, and the adaptive threshold is determined by the adaptive threshold algorithm. After the vibration signal of the rotating mechanical equipment is obtained, the fast Fourier transform is performed on it, and the result is normalized to obtain the corresponding data, which is then input into the deep autoencoder prediction model to obtain the corresponding reconstruction error, and the set threshold range is compared. If it does not meet the set threshold range, an early warning notification is immediately issued, which effectively improves the operating accuracy and analysis and processing efficiency of the rotating mechanical equipment; The rotating machinery data information is processed by hierarchical clustering method, so that the similarity and distance relationship between data objects can be presented intuitively; In this step, the rotating machinery data objects are clustered according to the hierarchical structure, and a tree structure is constructed according to the similarity or distance between the data objects, where the top is a single cluster and the bottom is a separate data point. Each data point is used as a separate category, and then these categories are gradually merged until the preset number of clusters is reached or the merging threshold is reached, effectively reducing the amount of data that needs to be analyzed and processed, and greatly saving analysis and processing resources; The further analysis, control and management of the rotating machinery through the Industrial Internet includes: A multi-scale approach is used to process rotating machinery data information, dividing it into a set number of subsets, and each subset is analyzed independently; In this step, multi-scale processing can help identify the characteristics of different rotating machinery and equipment, and the characteristic information obtained at multiple scales can be used to better locate and identify the target rotating machinery and equipment. At the same time, the analysis and processing performance can be improved by analyzing and processing the rotating machinery vibration signals at different time and space scales. In fault diagnosis, wavelet transform can decompose the rotating machinery vibration signal into sub-bands within different frequency ranges, and each sub-band has a corresponding time and space scale, which can effectively capture the local features in the signal and improve the efficiency and accuracy of rotating machinery and equipment detection and analysis. Further reconstruct the data format control of rotating machinery data information, reorganize and store it according to the set rules and structure, make it more standardized, easier to process and analyze, and more convenient for the visualization and presentation of rotating machinery data, so that it is more intuitive and easy to understand; In this step, by reconstructing the data format, the rotating machinery data can be organized and stored according to the set structure and rules, making it more standardized and easy to understand and use, thereby improving the scalability and reusability of the data, reducing data redundancy, improving data quality, and enhancing data accuracy and consistency; After identifying and selecting sensitive features that meet privacy settings from rotating machinery data, an adaptive hierarchical clustering method is used to cluster them, and after calculating the intra-class and inter-class distances to determine the optimal number of features, desensitization control processing is performed; In this step, the selected sensitive features are normalized to generate a normalized feature set, which is clustered using an adaptive hierarchical clustering method, which can make the desensitization of rotating machinery data more efficient and accurate and avoid the generation of errors.

2. According to claim 1, a control analysis method based on industrial Internet is characterized in that: The log management and early warning statistics of rotating machinery include: Record every operation of the user in the rotating machinery system, so that the administrator can monitor and store the user's access and operation through security log records, and maintain the log records; The early warning information of rotating machinery is counted according to time, equipment, and fault type, and displayed in a table, including alarm time, fault type, and vibration value at the time of fault.

Images