Intelligent sewage treatment monitoring system and method based on digital twin

Through digital twin technology and deep learning algorithms, the vibration signal characteristics of sewage treatment equipment are extracted, real-time monitoring and abnormal prediction of equipment status are realized, problems of low efficiency and misjudgment of manual inspections are solved, and monitoring accuracy and equipment management are improved. efficiency.

CN118445740BActive Publication Date: 2024-11-05KARAMAY WATER SUPPLY ENG TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410918026.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-07-10
Publication Date
2024-11-05
Estimated Expiration
2044-07-10

AI Technical Summary

Technical Problem

The monitoring of existing sewage treatment equipment relies on manual inspection, is inefficient and susceptible to human factors, which may lead to misjudgment and make it difficult to achieve real-time and accurate equipment status monitoring and fault prediction.

Method used

Using a smart monitoring system based on digital twins, the vibration signals of sewage treatment equipment are obtained through vibration sensors, and the signal processing is performed using wavelet transformation and deep learning algorithms. Local time-frequency semantic feature vectors and full-time domain vibration feature mode are extracted, and hash similarity is calculated. The degree and graph convolution neural network model generates monitoring results to display the working status of the device.

Benefits of technology

It realizes the capture of the full-time domain operation characteristic mode of sewage treatment equipment, improves the accuracy and efficiency of abnormal monitoring of equipment status, reduces manual misjudgment, and supports real-time monitoring and early warning.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118445740B_ABST
    Figure CN118445740B_ABST
Patent Text Reader

Abstract

The present disclosure relates to a smart sewage treatment monitoring system and method based on digital twins. The method includes: using wavelet transform to process the acquired vibration signal and then extracting local vibration semantic feature patterns to obtain a set of vibration local time-frequency semantic feature vectors; calculating the hash similarity of the set of vibration local time-frequency semantic feature vectors to obtain a local vibration time-frequency consistency topological matrix; extracting full-time domain vibration feature patterns from the set of vibration local time-frequency semantic feature vectors and the local vibration time-frequency consistency topological matrix to obtain a full-time domain vibration time-frequency topological feature matrix, and then determining the monitoring results, and displaying the monitoring results on the screen. In this way, the full-time domain operation characteristic pattern of the integrated sewage treatment equipment can be captured, thereby realizing abnormal monitoring of the working state of the integrated sewage treatment equipment.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present disclosure relates to the technical field of sewage treatment, and in particular, to a smart sewage treatment monitoring system and method based on digital twins. Background Art

[0002] With the acceleration of urbanization and the rapid development of industry, the efficiency and accuracy of sewage treatment have become increasingly important. Integrated sewage treatment equipment is a kind of equipment that integrates multiple sewage treatment process flows. It usually includes pretreatment, primary treatment, secondary treatment, deep treatment and other links, and can centrally treat sewage to meet discharge standards or reuse standards.

[0003] Ensuring the normal operation of sewage treatment equipment can prevent untreated or substandard sewage from being directly discharged into the natural environment, thus avoiding pollution of water bodies and ecosystems. Therefore, it is usually necessary to monitor and manage the use of integrated sewage treatment equipment. However, the monitoring method that relies on manual inspection and operation is inefficient and easily affected by human factors, which may cause problems such as misjudgment.

[0004] Digital twin technology is a technology that simulates, analyzes and optimizes the performance of physical entities by creating digital models of physical entities. In the field of sewage treatment, digital twins can be used to simulate the operating status of sewage treatment equipment, and optimize equipment performance and visualize fault prediction by real-time monitoring and analysis of the actual operating data of the equipment.

[0005] Therefore, we look forward to an intelligent sewage treatment monitoring system and method based on digital twins. Summary of the invention

[0006] This summary is provided to introduce concepts in a brief form that will be described in detail in the detailed description below. This summary is not intended to identify key features or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

[0007] In the first aspect, the present disclosure provides a smart sewage treatment monitoring method based on digital twins, the method comprising: obtaining a vibration signal of an integrated sewage treatment equipment collected by a vibration sensor; processing the vibration signal using a wavelet transform to obtain a vibration two-dimensional time-frequency diagram; extracting local vibration semantic feature patterns from the vibration two-dimensional time-frequency diagram to obtain a set of vibration local time-frequency semantic feature vectors; calculating the hash similarity between any two vibration local time-frequency semantic feature vectors in the set of vibration local time-frequency semantic feature vectors to obtain a local vibration time-frequency consistency topological matrix; extracting full-time-domain vibration feature patterns from the set of vibration local time-frequency semantic feature vectors and the local vibration time-frequency consistency topological matrix to obtain a full-time-domain vibration time-frequency topological feature matrix; determining monitoring results based on the full-time-domain vibration time-frequency topological feature matrix, and displaying the monitoring results on a screen.

[0008] Optionally, local vibration semantic feature pattern extraction is performed on the vibration two-dimensional time-frequency map to obtain a set of vibration local time-frequency semantic feature vectors, including: performing image block segmentation on the vibration two-dimensional time-frequency map to obtain a set of vibration local two-dimensional time-frequency maps; each vibration local two-dimensional time-frequency map in the set of vibration local two-dimensional time-frequency maps is passed through a vibration time-frequency feature extractor based on a void convolutional neural network model to obtain a set of vibration local time-frequency feature maps; each vibration local time-frequency feature map in the set of vibration local time-frequency feature maps is passed through a multi-scale semantic feature extractor to obtain a set of vibration local time-frequency semantic feature vectors.

[0009] Optionally, each vibration local time-frequency feature map in the set of vibration local time-frequency feature maps is respectively passed through a multi-scale semantic feature extractor to obtain a set of vibration local time-frequency semantic feature vectors, including: extracting global semantic feature information and local semantic feature information from the vibration local time-frequency feature maps to obtain a vibration time-frequency global semantic feature vector and a vibration time-frequency local semantic feature vector; and fusing the vibration time-frequency global semantic feature vector and the vibration time-frequency local semantic feature vector to obtain the vibration local time-frequency semantic feature vector.

[0010] Optionally, extracting global semantic feature information and local semantic feature information from the vibration local time-frequency feature map to obtain a vibration time-frequency global semantic feature vector and a vibration time-frequency local semantic feature vector, including: upsampling the vibration local time-frequency feature map to obtain an upsampled vibration local time-frequency feature map; using a vibration global semantic feature extraction module to process the upsampled vibration local time-frequency feature map to obtain a vibration time-frequency global semantic extraction feature vector; dimensionally adjusting the vibration time-frequency global semantic extraction feature vector to obtain the vibration time-frequency global semantic feature vector; using a vibration local semantic feature extraction module to process the upsampled vibration local time-frequency feature map to obtain a vibration time-frequency local semantic extraction feature vector; dimensionally adjusting the vibration time-frequency local semantic extraction feature vector to obtain the vibration time-frequency local semantic feature vector.

[0011] Optionally, the vibration global semantic feature extraction module includes a global average pooling layer, a first point convolution layer, a first batch normalization layer and a first activation layer, and the vibration local semantic feature extraction module includes a second point convolution layer, a second batch normalization layer and a second activation layer.

[0012] Optionally, fusing the vibration time-frequency global semantic feature vector and the vibration time-frequency local semantic feature vector to obtain the vibration local time-frequency semantic feature vector, including: calculating the position-by-position element-by-element addition of the vibration time-frequency global semantic feature vector and the vibration time-frequency local semantic feature vector to obtain a multi-scale vibration time-frequency semantic fusion feature vector; and performing activation processing based on a sigmoid function on the multi-scale vibration time-frequency semantic fusion feature vector to obtain the vibration local time-frequency semantic feature vector.

[0013] Optionally, the hash similarity between any two vibration local time-frequency semantic feature vectors in the set of vibration local time-frequency semantic feature vectors is calculated to obtain a local vibration time-frequency consistency topological matrix, including: using a hash mapping function to map and transform each vibration local time-frequency semantic feature vector in the set of vibration local time-frequency semantic feature vectors to obtain a set of hash-coded vibration local time-frequency semantic feature vectors; and calculating the Chebyshev distance between any two hash-coded vibration local time-frequency semantic feature vectors in the set of hash-coded vibration local time-frequency semantic feature vectors to obtain the local vibration time-frequency consistency topological matrix.

[0014] Optionally, a full-time-domain vibration feature pattern extraction is performed on the set of vibration local time-frequency semantic feature vectors and the local vibration time-frequency consistency topological matrix to obtain a full-time-domain vibration time-frequency topological feature matrix, including: passing the set of vibration local time-frequency semantic feature vectors and the local vibration time-frequency consistency topological matrix through a full-time-domain vibration pattern feature encoder based on a graph convolutional neural network model to obtain the full-time-domain vibration time-frequency topological feature matrix.

[0015] Optionally, the monitoring results are determined based on the full-time-domain vibration time-frequency topological feature matrix, and the monitoring results are displayed on the screen, including: passing the full-time-domain vibration time-frequency topological feature matrix through a classifier-based monitoring result generator to obtain the monitoring results, and the monitoring results are used to indicate whether there is any abnormality in the working status of the integrated sewage treatment equipment.

[0016] In the second aspect, the present disclosure provides a smart sewage treatment monitoring system based on digital twins, the system comprising: a vibration signal acquisition module, used to acquire the vibration signal of the integrated sewage treatment equipment collected by the vibration sensor; a signal processing module, used to process the vibration signal using wavelet transform to obtain a vibration two-dimensional time-frequency diagram; a local vibration semantic feature pattern extraction module, used to extract the local vibration semantic feature pattern of the vibration two-dimensional time-frequency diagram to obtain a set of vibration local time-frequency semantic feature vectors; a hash similarity calculation module, used to calculate the hash similarity between any two vibration local time-frequency semantic feature vectors in the set of vibration local time-frequency semantic feature vectors to obtain a local vibration time-frequency consistency topological matrix; a full-time domain vibration feature pattern extraction module, used to extract the full-time domain vibration feature pattern of the set of vibration local time-frequency semantic feature vectors and the local vibration time-frequency consistency topological matrix to obtain a full-time domain vibration time-frequency topological feature matrix; a monitoring result determination module, used to determine the monitoring result based on the full-time domain vibration time-frequency topological feature matrix, and display the monitoring result on the screen; wherein, the vibration two The invention relates to a method for extracting local vibration semantic feature patterns from a two-dimensional time-frequency graph to obtain a set of vibration local time-frequency semantic feature vectors, comprising: performing image block segmentation on the vibration two-dimensional time-frequency graph to obtain a set of vibration local two-dimensional time-frequency graphs; passing each vibration local two-dimensional time-frequency graph in the set of the vibration local two-dimensional time-frequency graphs through a vibration time-frequency feature extractor based on a hole convolutional neural network model to obtain a set of vibration local time-frequency feature graphs; passing each vibration local time-frequency feature graph in the set of the vibration local time-frequency feature graphs through a multi-scale semantic feature extractor to obtain a set of vibration local time-frequency semantic feature vectors; wherein passing each vibration local time-frequency feature graph in the set of the vibration local time-frequency feature graphs through a multi-scale semantic feature extractor to obtain a set of vibration local time-frequency semantic feature vectors comprises: extracting global semantic feature information and local semantic feature information from the vibration local time-frequency feature graphs to obtain a vibration time-frequency global semantic feature vector and a vibration time-frequency local semantic feature vector; and fusing the vibration time-frequency global semantic feature vector and the vibration time-frequency local semantic feature vector to obtain the vibration local time-frequency semantic feature vector.

[0017] The above technical solution is adopted, after the acquired vibration signal is processed by using wavelet transform, local vibration semantic feature pattern extraction is performed to obtain a set of vibration local time-frequency semantic feature vectors; the hash similarity of the set of vibration local time-frequency semantic feature vectors is calculated to obtain a local vibration time-frequency consistency topological matrix; the full-time domain vibration feature pattern extraction is performed on the set of vibration local time-frequency semantic feature vectors and the local vibration time-frequency consistency topological matrix to obtain a full-time domain vibration time-frequency topological feature matrix, and then the monitoring results are determined and displayed on the screen. In this way, the full-time domain operation feature pattern of the integrated sewage treatment equipment can be captured, thereby realizing abnormal monitoring of the working state of the integrated sewage treatment equipment.

[0018] Other features and advantages of the present disclosure will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale.

[0020] In the attached picture: Figure 1 It is a flow chart of a smart sewage treatment monitoring method based on digital twins according to an exemplary embodiment.

[0021] Figure 2 It is a block diagram of a smart sewage treatment monitoring system based on digital twins according to an exemplary embodiment.

[0022] Figure 3 It is a schematic diagram of full-space intelligent operation according to an exemplary embodiment.

[0023] Figure 4 It is a block diagram of an electronic device according to an exemplary embodiment.

[0024] Figure 5 It is an application scenario diagram of an intelligent sewage treatment monitoring method based on digital twins according to an exemplary embodiment. DETAILED DESCRIPTION

[0025] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments described herein, which are instead provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0026] It should be understood that the various steps described in the method embodiments of the present disclosure may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this respect.

[0027] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0028] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0029] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0030] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0031] In order to solve the above problems, the present disclosure provides a digital twin-based intelligent sewage treatment monitoring system and method, which processes the acquired vibration signal using wavelet transform and then extracts the local vibration semantic feature pattern to obtain a set of vibration local time-frequency semantic feature vectors; calculates the hash similarity of the set of vibration local time-frequency semantic feature vectors to obtain a local vibration time-frequency consistency topological matrix; extracts the full-time domain vibration feature pattern of the set of vibration local time-frequency semantic feature vectors and the local vibration time-frequency consistency topological matrix to obtain a full-time domain vibration time-frequency topological feature matrix, and then determines the monitoring results and displays the monitoring results on the screen. In this way, the full-time domain operation feature pattern of the integrated sewage treatment equipment can be captured, thereby realizing abnormal monitoring of the working state of the integrated sewage treatment equipment.

[0032] The specific embodiments of the present disclosure are described in detail below with reference to the accompanying drawings.

[0033] Figure 1 is a flow chart of a smart sewage treatment monitoring method based on digital twins according to an exemplary embodiment. Figure 1As shown, the method includes: step 101, obtaining a vibration signal of the integrated sewage treatment equipment collected by a vibration sensor.

[0034] Step 102: Process the vibration signal using wavelet transform to obtain a vibration two-dimensional time-frequency diagram.

[0035] Step 103: extracting local vibration semantic feature patterns from the vibration two-dimensional time-frequency diagram to obtain a set of vibration local time-frequency semantic feature vectors.

[0036] Step 104: Calculate the hash similarity between any two vibration local time-frequency semantic feature vectors in the set of vibration local time-frequency semantic feature vectors to obtain a local vibration time-frequency consistency topological matrix.

[0037] Step 105 : extracting full-time-domain vibration characteristic patterns from the set of the vibration local time-frequency semantic feature vectors and the local vibration time-frequency consistency topological matrix to obtain a full-time-domain vibration time-frequency topological feature matrix.

[0038] Step 106: Determine monitoring results based on the full-time-domain vibration time-frequency topological characteristic matrix, and display the monitoring results on a screen.

[0039] In response to the above technical problems, the technical concept of the present application is: to learn and identify the dynamic changes inside the integrated sewage treatment equipment and the operating status characteristic information contained in the vibration signal of the integrated sewage treatment equipment through an intelligent algorithm based on deep learning, and to capture the full time domain operating characteristic pattern of the integrated sewage treatment equipment in combination with graph data coding analysis, thereby realizing abnormal monitoring of the working status of the integrated sewage treatment equipment.

[0040] Based on this, in the technical solution of the present application, first, a vibration signal of the integrated sewage treatment equipment collected by a vibration sensor is obtained. Here, the vibration signal can reflect the mechanical state of the integrated sewage treatment equipment, such as bearing wear, imbalance, looseness and other problems, and the generation of these problems usually leads to changes in the vibration mode. In an embodiment of the present application, a piezoelectric or electromagnetic sensor can be used to obtain the vibration signal of the integrated sewage treatment equipment.

[0041] Then, the vibration signal is processed using wavelet transform to obtain a two-dimensional time-frequency diagram of the vibration. Among them, wavelet transform is a signal processing technology that analyzes the local time-frequency characteristics of the signal by operating the signal with wavelet functions of different scales and positions. Compared with Fourier transform, wavelet transform has the characteristics of time-frequency localization and can provide detailed information of the signal in both time and frequency dimensions. Wavelet transform can capture the instantaneous changes of the signal and is very suitable for analyzing non-stationary signals such as vibration signals. In this way, vibration feature information such as energy distribution, frequency components, time-frequency distribution, etc. that are not obvious in the original vibration signal can be extracted from the vibration signal through wavelet transform.

[0042] In one embodiment of the present disclosure, local vibration semantic feature pattern extraction is performed on the vibration two-dimensional time-frequency map to obtain a set of vibration local time-frequency semantic feature vectors, including: performing image block segmentation on the vibration two-dimensional time-frequency map to obtain a set of vibration local two-dimensional time-frequency maps; each vibration local two-dimensional time-frequency map in the set of vibration local two-dimensional time-frequency maps is passed through a vibration time-frequency feature extractor based on a void convolutional neural network model to obtain a set of vibration local time-frequency feature maps; each vibration local time-frequency feature map in the set of vibration local time-frequency feature maps is passed through a multi-scale semantic feature extractor to obtain a set of vibration local time-frequency semantic feature vectors.

[0043] Next, the vibration two-dimensional time-frequency diagram is segmented into image blocks to obtain a set of vibration local two-dimensional time-frequency diagrams. It should be understood that in the actual application scenario of the present application, when a device fails or is abnormal, corresponding abnormal fluctuations may only appear in a specific frequency in the vibration signal. Here, the vibration two-dimensional time-frequency diagram is decomposed into smaller vibration local two-dimensional time-frequency diagrams by image block segmentation, so that the subsequent model can focus on the local characteristics of the signal, thereby having an advantage in analyzing vibration signals with local fault characteristics.

[0044] After that, each of the local two-dimensional vibration time-frequency graphs in the set of local two-dimensional vibration time-frequency graphs is respectively passed through a vibration time-frequency feature extractor based on a dilated convolutional neural network model to obtain a set of local vibration time-frequency feature graphs. Among them, dilated convolution (also called dilated convolution or dilated convolution) can process non-fixed grid data and is suitable for irregular structures in time-frequency graphs. That is, using a dilated convolutional neural network model to construct a vibration time-frequency feature extractor can capture the complex neighborhood implicit vibration characteristic patterns of each local two-dimensional vibration time-frequency graph.

[0045] Since vibration signals in the real world often show different characteristics at different time scales. For example, in mechanical vibration signals, transient impacts may appear on a finer time scale, while long-term wear may be reflected on a coarser time scale. That is, when analyzing the vibration signals of integrated sewage treatment equipment, there may be multi-scale feature distribution patterns. For example, bearing damage may produce transient impacts, which appear as sharp peaks on the fine scale of the time-frequency diagram. Another example is the periodic failure of mechanical components, such as imbalance or misalignment, which will produce repetitive patterns on the corresponding scale of the time-frequency diagram, while on a larger scale, the overall energy distribution of the vibration signal may be significantly different from the energy distribution in a healthy state. Therefore, in order to capture the vibration characteristic patterns at different scales, in the technical solution of the present application, each vibration local time-frequency feature map in the set of the vibration local time-frequency feature maps is respectively passed through a multi-scale semantic feature extractor to obtain a set of vibration local time-frequency semantic feature vectors.

[0046] In one embodiment of the present disclosure, each vibration local time-frequency feature map in the set of vibration local time-frequency feature maps is respectively passed through a multi-scale semantic feature extractor to obtain a set of vibration local time-frequency semantic feature vectors, including: extracting global semantic feature information and local semantic feature information from the vibration local time-frequency feature maps to obtain a vibration time-frequency global semantic feature vector and a vibration time-frequency local semantic feature vector; fusing the vibration time-frequency global semantic feature vector and the vibration time-frequency local semantic feature vector to obtain the vibration local time-frequency semantic feature vector.

[0047] Further, in one embodiment of the present disclosure, global semantic feature information and local semantic feature information are extracted from the vibration local time-frequency feature map to obtain a vibration time-frequency global semantic feature vector and a vibration time-frequency local semantic feature vector, including: upsampling the vibration local time-frequency feature map to obtain an upsampled vibration local time-frequency feature map; using a vibration global semantic feature extraction module to process the upsampled vibration local time-frequency feature map to obtain a vibration time-frequency global semantic extraction feature vector; dimensionally adjusting the vibration time-frequency global semantic extraction feature vector to obtain the vibration time-frequency global semantic feature vector; using a vibration local semantic feature extraction module to process the upsampled vibration local time-frequency feature map to obtain a vibration time-frequency local semantic extraction feature vector; dimensionally adjusting the vibration time-frequency local semantic extraction feature vector to obtain the vibration time-frequency local semantic feature vector.

[0048] In an embodiment of the present application, the specific process of passing each vibration local time-frequency feature map in the set of the vibration local time-frequency feature map through a multi-scale semantic feature extractor to obtain a set of vibration local time-frequency semantic feature vectors is to extract global semantic feature information and local semantic feature information from the vibration local time-frequency feature map to obtain a vibration time-frequency global semantic feature vector and a vibration time-frequency local semantic feature vector, and then fuse the vibration time-frequency global semantic feature vector and the vibration time-frequency local semantic feature vector to obtain the vibration local time-frequency semantic feature vector. Among them, the core of extracting global semantic feature information and local semantic feature information from the vibration local time-frequency feature map is to respectively use a vibration global semantic feature extraction module (including a global average pooling layer, a first point convolution layer, a first batch normalization layer and a first activation layer) and a vibration local semantic feature extraction module (including a second point convolution layer, a second batch normalization layer and a second activation layer) to further extract different scale features, thereby capturing multi-scale vibration feature patterns in different local areas of the vibration signal and adapting to the complexity of the vibration feature pattern, which helps to understand the deep meaning of the vibration signal.

[0049] Among them, the vibration global semantic feature extraction module includes a global average pooling layer, a first point convolution layer, a first batch normalization layer and a first activation layer, and the vibration local semantic feature extraction module includes a second point convolution layer, a second batch normalization layer and a second activation layer.

[0050] Furthermore, in one embodiment of the present disclosure, the vibration time-frequency global semantic feature vector and the vibration time-frequency local semantic feature vector are fused to obtain the vibration local time-frequency semantic feature vector, including: calculating the position-by-position element-by-element addition of the vibration time-frequency global semantic feature vector and the vibration time-frequency local semantic feature vector to obtain a multi-scale vibration time-frequency semantic fusion feature vector; and performing activation processing based on a sigmoid function on the multi-scale vibration time-frequency semantic fusion feature vector to obtain the vibration local time-frequency semantic feature vector.

[0051] Furthermore, the hash similarity between any two vibration local time-frequency semantic feature vectors in the set of the vibration local time-frequency semantic feature vectors is calculated to obtain a local vibration time-frequency consistency topological matrix. That is, the association between each local feature is revealed by calculating the hash similarity, which is very important for understanding the intrinsic structure and dynamic changes of complex signals. For example, the same type of bearings causing equipment failure may present a characteristic pattern with high periodicity or correlation in the vibration signal.

[0052] In one embodiment of the present disclosure, the hash similarity between any two vibration local time-frequency semantic feature vectors in the set of vibration local time-frequency semantic feature vectors is calculated to obtain a local vibration time-frequency consistency topological matrix, including: using a hash mapping function to map and transform each vibration local time-frequency semantic feature vector in the set of vibration local time-frequency semantic feature vectors to obtain a set of hash-coded vibration local time-frequency semantic feature vectors; calculating the Chebyshev distance between any two hash-coded vibration local time-frequency semantic feature vectors in the set of hash-coded vibration local time-frequency semantic feature vectors to obtain the local vibration time-frequency consistency topological matrix.

[0053] After that, the set of the vibration local time-frequency semantic feature vectors and the local vibration time-frequency consistency topological matrix are passed through the full-time domain vibration pattern feature encoder based on the graph convolutional neural network model to obtain the full-time domain vibration time-frequency topological feature matrix. Specifically, in the technical solution of the present application, each of the vibration local time-frequency semantic feature vectors is used as a node input in the graph convolutional neural network (GCNN) model, and the local vibration time-frequency consistency topological matrix describes the connection relationship between these nodes, which can be regarded as an edge input in the graph convolutional neural network model. In this way, by utilizing the integration capability of GCNN, the interaction patterns between each local vibration time-frequency feature pattern and its related adjacent nodes are integrated, reducing the high dependence on the characteristics of a single node, thereby deeply understanding the full-time domain vibration state of the vibration signal and understanding the overall behavior pattern of the signal.

[0054] In one embodiment of the present disclosure, a full-time-domain vibration feature pattern extraction is performed on the set of vibration local time-frequency semantic feature vectors and the local vibration time-frequency consistency topological matrix to obtain a full-time-domain vibration time-frequency topological feature matrix, including: passing the set of vibration local time-frequency semantic feature vectors and the local vibration time-frequency consistency topological matrix through a full-time-domain vibration pattern feature encoder based on a graph convolutional neural network model to obtain the full-time-domain vibration time-frequency topological feature matrix.

[0055] Then, the full-time-domain vibration time-frequency topological feature matrix is ​​passed through a classifier-based monitoring result generator to obtain a monitoring result, and the monitoring result is used to indicate whether there is any abnormality in the working status of the integrated sewage treatment equipment; and the monitoring result is displayed on the screen so that the operator can see the monitoring result in real time and understand the operating status of the integrated sewage treatment equipment in a timely manner.

[0056] In one embodiment of the present disclosure, monitoring results are determined based on the full-time domain vibration time-frequency topological feature matrix, and the monitoring results are displayed on a screen, including: passing the full-time domain vibration time-frequency topological feature matrix through a classifier-based monitoring result generator to obtain the monitoring results, and the monitoring results are used to indicate whether there is any abnormality in the working status of the integrated sewage treatment equipment.

[0057] In an optimized embodiment of the present application, the full-time-domain vibration time-frequency topological feature matrix is ​​passed through a monitoring result generator based on a classifier to obtain the monitoring result, including: firstly, the full-time-domain vibration time-frequency topological feature matrix is ​​expanded into a full-time-domain vibration time-frequency topological feature vector, the autocorrelation matrix of the full-time-domain vibration time-frequency topological feature vector and its own transposition is calculated, and the first and The inner product of the row vectors is used as the first The matrix value of the position is obtained to obtain a weight matrix, and then the weight matrix is ​​matrix-multiplied with the autocorrelation matrix, and then further matrix-vector multiplied with the full time-domain vibration time-frequency topological eigenvector to obtain a correction vector, and finally the correction vector is dot-multiplied with the full time-domain vibration time-frequency topological eigenvector to obtain an optimized full time-domain vibration time-frequency topological eigenvector, and the optimized full time-domain vibration time-frequency topological eigenvector is input into the classifier-based monitoring result generator to obtain the monitoring result.

[0058] The optimization steps are specifically expressed as follows: ;in, represents the full-domain vibration time-frequency topological eigenvector, represents the transpose of a vector, represents the autocorrelation matrix, and The autocorrelation matrix is ​​represented by and row vector, represents the weight matrix, The weight matrix The matrix value of the position, represents the height of the autocorrelation matrix, represents the set of real numbers, represents matrix multiplication, represents dot product, Represents the optimized full-time-domain vibration time-frequency topological eigenvector.

[0059] That is, here, for the image semantic features of the vibration two-dimensional time-frequency graph in the local image semantic space domain, the topological representation of the local image semantic space domain hash similarity topology is based on the global image semantic space domain. Due to the difference in different image semantic association dimensions in the local-global image semantic space domain, the out-of-distribution noise of the multi-dimensional image semantic feature association representation set caused by the difference in different image semantic association dimensions in the local-global image semantic space domain causes the class distribution noise of the entire time-domain vibration time-frequency topological feature matrix, thereby affecting the accuracy of the classification result.

[0060] In this technical solution, based on the self-correlation dimension of the full-time-domain vibration time-frequency topological feature vector as the object to be modulated, correlation expansion is performed based on the spatial sub-dimensional complexity of the high-dimensional feature space of the feature distribution, thereby introducing the decomposable correlation dimension offset of the full-time-domain vibration time-frequency topological feature vector into the heterogeneous correlation embedding space, thereby enhancing its correlation self-consistency through the joint fine-tuning of the decomposable dimension set represented by the heterogeneous correlation embedding space to reduce its class distribution noise, thereby improving the feature set of the full-time-domain vibration time-frequency topological feature vector The predetermined category harmony in the associated target classification domain is improved, thereby improving the classification training speed and the accuracy of the classification results of the full-time-domain vibration time-frequency topological feature matrix when it is classified by a classifier-based monitoring result generator.

[0061] To sum up, the above scheme is adopted to learn and identify the dynamic changes inside the equipment and the characteristic information of the operating status contained in the vibration signal of the integrated sewage treatment equipment through an intelligent algorithm based on deep learning, and the full time domain operating characteristic pattern of the integrated sewage treatment equipment is captured in combination with graph data coding analysis, thereby realizing abnormal monitoring of the working status of the integrated sewage treatment equipment.

[0062] Figure 2 FIG. 1 is a block diagram of a digital twin-based intelligent sewage treatment monitoring system according to an exemplary embodiment. Figure 2 As shown, the smart sewage treatment monitoring system 200 includes: a vibration signal acquisition module 201, which is used to obtain the vibration signal of the integrated sewage treatment equipment collected by the vibration sensor.

[0063] The signal processing module 202 is used to process the vibration signal using wavelet transform to obtain a vibration two-dimensional time-frequency diagram.

[0064] The local vibration semantic feature pattern extraction module 203 is used to extract the local vibration semantic feature pattern from the vibration two-dimensional time-frequency diagram to obtain a set of vibration local time-frequency semantic feature vectors.

[0065] The hash similarity calculation module 204 is used to calculate the hash similarity between any two vibration local time-frequency semantic feature vectors in the set of vibration local time-frequency semantic feature vectors to obtain a local vibration time-frequency consistency topology matrix.

[0066] The full-time-domain vibration characteristic pattern extraction module 205 is used to extract the full-time-domain vibration characteristic pattern from the set of the vibration local time-frequency semantic feature vectors and the local vibration time-frequency consistency topological matrix to obtain a full-time-domain vibration time-frequency topological feature matrix.

[0067] The monitoring result determination module 206 is used to determine the monitoring result based on the full-time-domain vibration time-frequency topological characteristic matrix and display the monitoring result on the screen.

[0068] Due to the decentralized and independent operation of sewage treatment equipment: the sewage treatment plants built are distributed in different cities and operate independently of each other; the existing data is not aggregated: due to the independent operation of sewage treatment plants, the actual operation data is stored locally and the superiors cannot know it; there is a lack of remote supervision mechanism: the existing supervision can only be summarized in the form of documents, and the actual operation cannot be known in time, and remote supervision cannot be achieved; emergency dispatch is difficult: due to the lack of an interconnection mechanism, data cannot be obtained in the first place. It is impossible to be on duty remotely, and it is difficult to coordinate resources and remotely command in case of emergency.

[0069] In one embodiment of the present disclosure, an integrated sewage treatment equipment is provided. Guided by the thinking of the Internet of Things, advanced geographic information technology (GIS), data collection and communication technology, sensor technology, and computer network technology are used to design a smart integrated sewage equipment visualization centralized control solution to change the operation mode of the traditional sewage treatment industry: from experience-driven to data-driven; service mode: from passive service to active service; management and control mode: from post-processing to pre-data prediction.

[0070] In this way, the following can be achieved: 1. Improving sewage treatment capacity: By updating and renovating equipment, optimizing pipeline network layout and other measures, the sewage treatment capacity can be improved to meet the needs of urban development and residents' lives.

[0071] 2. Improve sewage treatment efficiency: adopt advanced energy-saving technologies and equipment to reduce energy consumption, improve energy utilization efficiency, and achieve green and low-carbon development.

[0072] 3. Realizing remote control and automatic control of integrated sewage equipment can reduce the number of on-duty personnel and directly reduce the labor cost of sewage equipment operation and maintenance.

[0073] 4. Realize real-time video monitoring and intrusion alarm to ensure the safety of sewage equipment, improve the safety factor, and facilitate management personnel to grasp the operation status of the site anytime and anywhere.

[0074] Build a system platform based on the new generation of information technology, construct a new model of sewage equipment operation and maintenance, accelerate the improvement of sewage treatment and service levels, save energy and reduce consumption from the perspective of enterprises, improve economic benefits, and let technological progress promote the improvement of the technical level and management level of the entire industry.

[0075] like Figure 3 As shown in the figure, a full-space intelligent operation is proposed to build intelligent water resources measurement and control facilities to realize the digital collection and control of water resources and surrounding related environment; meet diverse networking needs and realize multi-network integration and application of data; guided by data value-driven, promote the specialization of internal and external data such as production, environment, time and space in sewage areas, and build a sewage area water resources data center; realize the informatization application of sewage area business from the aspects of measurement and control, scheduling, operation and maintenance, promote water resources allocation projects, and realize the reasonable allocation of water resources in sewage areas; build a multi-dimensional command and display center to grasp the overall situation of the sewage area in real time and make scientific management decisions.

[0076] Furthermore, full-space intelligent operations include: centralized equipment management: supporting multi-level asset management and equipment operation data aggregation; equipment repair, maintenance, and spare parts management.

[0077] Real-time data collection: Support IoT application layer protocols and device collection communication protocols; real-time equipment operation status monitoring.

[0078] Remote monitoring: Real-time detection of key process indicators on the web and mobile terminals; remote control to achieve unattended operation.

[0079] Real-time video monitoring: Real-time video monitoring data access at key monitoring points can be used to view real-time video of the scene or historical video within the preservation range.

[0080] Cloud configuration monitoring: HTML5-based Web configuration, drag-and-drop layout element components; model-based data services.

[0081] Abnormal alarm: support multi-level asset management, equipment operation data summary; equipment repair, maintenance, and spare parts management.

[0082] Cloud-based logic control: With the platform's low-code development capabilities, cloud-based control logic such as logic delay, switch counting, and parameter configuration can be implemented.

[0083] Intelligent controller: supports industrial equipment data collection, intelligent algorithm calculation, data processing, decision-making and automatic control.

[0084] Interlocking logic control: Provides open logic programming capabilities to flexibly realize the linkage of single and multiple devices such as logic interlocking and PID closed-loop control.

[0085] Intelligent algorithm and precise control: The intelligent controller realizes on-demand adjustment and precise control.

[0086] There are three stages in process monitoring. In the first stage, a physical treatment system is used as the first step in water purification, which can intercept floating objects in sewage, such as fibers, pulp and other fine suspended objects. In the second stage, a biological treatment system is used to use biological treatment methods: the soluble pollutants in sewage are used as food for microorganisms to achieve the purpose of purification. In the third stage, a chemical treatment system is used, which is the last link in urban sewage treatment and is an advanced sewage treatment measure. Tertiary treatment is generally carried out in a disinfection pool.

[0087] Furthermore, water quality IoT monitoring can monitor various indicators of sewage treatment in real time, such as water quality, water volume, gas concentration, etc., so as to promptly discover problems in the treatment process and take corresponding measures to avoid excessive pollutant emissions. Sensor monitoring can obtain real-time operation data of sewage treatment plants, provide decision support for managers, and achieve efficient sewage treatment.

[0088] Among them, the liquid levels of regulating tanks, sedimentation tanks, water collection tanks, aeration tanks, etc. are displayed in different processes. Different pH values ​​show the acid-base changes of sewage, acid-base sewage neutralization in regulating tanks, etc. In different process processes, the flow of sewage and sludge changes, and the inflow and outflow changes. Monitor the redox potential in water quality. Reveal the oxidizability of water bodies and water decomposition capacity. COD is used to monitor the treatment effect to ensure the normal operation of sewage treatment facilities and meet the discharge standards.

[0089] Equipment monitoring monitors and manages the use of integrated sewage treatment equipment and instruments, including usage status, operating time, fault analysis and comprehensive maintenance records, and provides equipment and instrument maintenance suggestions and standardized operating procedures to ensure the normal use and maintenance of equipment and instruments.

[0090] Specifically including: operation data, equipment operation time, operation speed, load conditions, temperature, pressure and other parameters. It can reflect the operation status and performance of the equipment; electrical data, including equipment current, voltage, power, resistance and other parameters. These data can reflect the electrical performance and power consumption of the equipment; sound data, including equipment sound frequency, sound intensity, sound characteristics and other parameters. These data can reflect the acoustic performance of the equipment; graphic data, including equipment photos, videos, infrared images, etc. These data can reflect the appearance, structure, thermal characteristics, etc. of the equipment; environmental data, including temperature, humidity, air pressure, light and other parameters of the equipment's surrounding environment. These data can reflect the environmental conditions of the equipment; vibration data, including equipment vibration amplitude, frequency, phase and other parameters. These data can reflect the vibration of the equipment; other data, including equipment flow, concentration, weight, size and other parameters. These data can reflect the working conditions and process of the equipment.

[0091] In this way, real-time monitoring can be achieved to realize remote control of alarm warnings, for example: the data collected from the actual operation of on-site equipment and facilities and process flows can be visualized, and when abnormal situations occur, alarms, statistics and intelligent alarms can be analyzed comprehensively in real time to provide support for management decisions. Improve equipment management and full-process tracking and recording requirements, for example: achieve full HD, full-network monitoring and management of the entire sewage plant, and the visual information system strengthens information communication between departments, and intuitively displays the daily operation of the sewage plant, equipment maintenance information, and maintenance work content. Integrated unified management platform, for example: digital twin technology creates a three-dimensional scene, real-time display of process units and electrical equipment, and integrates control systems, monitoring systems, operation and maintenance management systems, etc. into an integrated platform, so that the intelligent operation, management, and service of the sewage plant can be more systematic and complete.

[0092] In one embodiment of the present application, a smart integrated sewage visualization centralized control platform is provided. The homepage workbench displays the overall situation and previews the overall operation of the sewage treatment plant project on the platform; the map gis visualizes the location of the sewage treatment plant map, links to the project dashboard, and can also search for projects and quickly enter the project. The left side displays the monthly average inflow and outflow, water cost statistics and reagent cost statistics, 24-hour inflow and outflow, 24-hour inflow and outflow pH curves, and water quality level monitoring; the right side displays the gateways, projects, signal point statistics, alarm signal point statistics, reagent usage statistics, and video monitoring screens connected to the platform; at the bottom is the data monitoring of water quality parameters.

[0093] Project management is divided into three parts: project management, project documents, and progress dashboard. Project management: custom add, view, modify, and delete projects; support uploading and maintaining project information, uploading project site pictures, project leader information, etc. Project documents: manage documents in the project, support querying document materials for a single project in a certain period of time; also support manual uploading of materials to the platform. Progress dashboard: can view the status distribution of the project, project progress monitoring, the top of the project progress, as well as the progress details, progress feedback, progress details of each project, and information synchronization between the site and the headquarters, so that the headquarters can better understand the progress on site.

[0094] Hazardous waste records: Record hazardous waste products reported to the sewage treatment plant, support uploading site photos, and query hazardous waste reporting records for a certain period of time. Site hygiene records: Support uploading site photos for the convenience of headquarters to view; also support exporting Excel files for archiving.

[0095] Pollution discharge standards: By monitoring the PH value, sodium ion and other data of sewage discharged from the sewage plant, you can compare it with international standards to see whether it meets the discharge requirements. Pollution discharge standard monitoring: Real-time monitoring of sewage discharge, by binding the platform to relevant signal points and corresponding national emission standards, the platform can automatically determine whether the sewage meets the emission standards.

[0096] Operation management is divided into four parts: operation visualization, real-time data, cloud configuration editing and operation reports. Operation visualization: Visualization displays the overall process of the project site in the form of a configuration diagram, and real-time data monitoring of each point is clearer and more intuitive. The left side displays the data of today's alarm, cumulative alarm, daily water inflow, cumulative water inflow, daily water outflow, cumulative water outflow, chemical use, and cumulative chemical use statistics; project site pictures and video real-time monitoring screens. The top displays the project basic report, water inflow and outflow report, chemical statistical report, and water outflow energy consumption report; signal point parameter setting and remote control device switch.

[0097] Real-time data: Displays real-time data collected at the project site, and supports searching and viewing specific data of signal points of a sewage treatment plant. Cloud configuration editing: There are three forms of 2D visual editor, 3D visual editor, and VR visual editor. The 2D visual editor is combined by dragging, pulling, and dragging, and the operation is simple and convenient. The 3D visual editor is based on the real scene of the site for 3D modeling, which truly restores the scene. The VR visual editor can shoot 360° scenes to create a real experience of online and offline environments. Supports binding signal points on the configuration diagram to facilitate real-time viewing of real-time data on the site.

[0098] Operation reports include: Basic report: supports customized selection of signal points to query report data. Water quality report: supports viewing the data of the project's inlet and outlet water quality signal points. Inlet and outlet water report: statistics on water inflow, water outflow, and energy consumption data. Statistics report: supports customized selection of multiple signal points, and statistics on their average, maximum (minimum) value, median, range, and mode. Drug statistics report: statistics on drug usage, water-to-drug ratio, drug cost, and other data for different time periods of each project. Historical data report: supports querying detailed data of each signal point in any historical time period. Daily operation report: report template.

[0099] Process control is mainly divided into two parts: timing mode and intelligent linkage. Timing mode: The platform can set the timing on and off functions for certain signal points of the sewage treatment plant according to on-site needs, such as setting its execution cycle, execution time period, number of times issued and interval, and then bind the corresponding signal points. Intelligent linkage mode: The range value of the conditional execution of a certain signal point can be set on the platform to remotely control the start and stop status of the on-site equipment and the operating frequency of the equipment.

[0100] Production management is divided into four parts: water quality management, patrol management, attendance management and data reporting. Water quality management test template: custom test template, online generation of test records, and statistical reports; patrol management: the system automatically generates patrol tasks according to the template, and automatically issues them according to the schedule, realizing online visualization of production site patrol tasks. Attendance management: add teams for scheduling, automatically generate attendance records and statistics. Data reporting: operation and maintenance personnel can temporarily upload production process data on site.

[0101] Cost management is divided into five parts: drug management, cost statistics, inventory management, waste management and electricity price setting. Cost statistics: supports statistics on drug costs, electricity costs, water costs and total costs for a single or multiple projects at different times of the day, month and year. Inventory management: realizes full-cycle management of material inventory, including warehousing, outbound and scrapping. Waste management: realizes online management of waste treatment process for easy traceability. Electricity price setting: supports querying electricity prices for any time period, and can also manually set the unit price of electricity.

[0102] Alarm management includes: Real-time alarm monitoring: You can monitor the device status in real time through the platform. Alarm history: Query historical alarm information based on date, alarm status, alarm level, and alarm content. Alarm message settings: Support custom settings for alarm level, message push frequency, time interval, and notification range. Support custom settings for alarm signal point high alarm status on and off, and alarm upper and lower limits.

[0103] Operation management is divided into two parts: operation report and knowledge base. The knowledge base includes document management, professional knowledge base, FAQ management and skills training. Operation report: Generate project operation reports in weekly, monthly and annual formats. The report content includes water quality statistics, water inflow and outflow statistics, work order statistics, signal point alarm statistics, etc. Support download of project application report pdf file. Document management: Support uploading and downloading documents. Professional knowledge base: You can add and download knowledge base content. FAQ management: Support adding articles and classifying articles.

[0104] System management is divided into six parts: menu management, role permissions, APP management, about us, operation logs and user management. Menu management: custom add, modify, delete items. Role permissions: different roles have different permission levels, set user role permissions, control user menus, button permissions, custom upload and maintain organization project information, upload project site pictures, project leader information. Operation log: you can view the user's operation records on the platform through the operation log, such as operation time, operation content, operation equipment, operator IP, etc., to facilitate problem tracing. User management: add, modify, delete menus, user operation button permissions.

[0105] Reference below Figure 4, which shows a schematic diagram of the structure of an electronic device 600 suitable for implementing the embodiment of the present disclosure. The terminal device in the embodiment of the present disclosure may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle terminals (such as vehicle navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 4 The electronic 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 disclosure.

[0106] like Figure 4 As shown, the electronic device 600 may include a processing device (e.g., 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 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the electronic device 600 are also stored. 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.

[0107] 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 magnetic tape, a hard disk, etc.; and a communication device 609. The communication device 609 may allow the electronic device 600 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 4 The electronic device 600 is shown with various devices, 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.

[0108] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device 609, or installed from the storage device 608, or installed from the ROM 602. When the computer program is executed by the processing device 601, the above-mentioned functions defined in the method of the embodiment of the present disclosure are executed.

[0109] It should be noted that the computer-readable medium disclosed above may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0110] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0111] The computer-readable medium may be included in the electronic device, or may exist independently without being installed in the electronic device.

[0112] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including, but not limited to, object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0113] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0114] The modules involved in the embodiments described in the present disclosure may be implemented by software or hardware. The name of a module does not limit the module itself in some cases. For example, a test parameter acquisition module may also be described as a "module for acquiring device test parameters corresponding to a target device".

[0115] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0116] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0117] Figure 5 FIG. 1 is an application scenario diagram of a smart sewage treatment monitoring method based on digital twins according to an exemplary embodiment. Figure 5 As shown, in this application scenario, first, a vibration signal of the integrated sewage treatment equipment collected by a vibration sensor is obtained (for example, Figure 5 Then, the obtained vibration signal is input to a server (for example, Figure 5 In S) shown in the figure, the server is capable of processing the vibration signal based on the smart sewage treatment monitoring algorithm of the digital twin to determine the monitoring results and display the monitoring results on the screen.

[0118] The above description is only a preferred embodiment of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present disclosure is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in the present disclosure (but not limited to) by each other to form a technical solution.

[0119] In addition, although each operation is described in a specific order, this should not be understood as requiring these operations to be performed in the specific order shown or in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details are included in the above discussion, these should not be interpreted as limiting the scope of the present disclosure. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

[0120] Although the subject matter has been described in language specific to structural features and / or method logic actions, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. On the contrary, the specific features and actions described above are merely example forms of implementing the claims. Regarding the device in the above embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated here.

Claims

1. A smart sewage treatment monitoring method based on digital twins, characterized in that: include: Acquire vibration signals of the integrated sewage treatment equipment collected by the vibration sensor; Processing the vibration signal using wavelet transform to obtain a vibration two-dimensional time-frequency diagram; Extracting local vibration semantic feature patterns from the vibration two-dimensional time-frequency graph to obtain a set of vibration local time-frequency semantic feature vectors; Calculating the hash similarity between any two vibration local time-frequency semantic feature vectors in the set of vibration local time-frequency semantic feature vectors to obtain a local vibration time-frequency consistency topological matrix; Performing full-time-domain vibration feature pattern extraction on the set of the vibration local time-frequency semantic feature vectors and the local vibration time-frequency consistency topological matrix to obtain a full-time-domain vibration time-frequency topological feature matrix; Determine the monitoring result based on the full-time-domain vibration time-frequency topological characteristic matrix, and display the monitoring result on the screen; The step of extracting a local vibration semantic feature pattern from the vibration two-dimensional time-frequency graph to obtain a set of vibration local time-frequency semantic feature vectors includes: Performing image block segmentation on the vibration two-dimensional time-frequency map to obtain a set of vibration local two-dimensional time-frequency maps; Each of the vibration local two-dimensional time-frequency graphs in the set of vibration local two-dimensional time-frequency graphs is respectively passed through a vibration time-frequency feature extractor based on a hole convolutional neural network model to obtain a set of vibration local time-frequency feature graphs; Each vibration local time-frequency feature map in the set of vibration local time-frequency feature maps is respectively passed through a multi-scale semantic feature extractor to obtain a set of vibration local time-frequency semantic feature vectors; The steps of respectively passing each vibration local time-frequency feature map in the set of vibration local time-frequency feature maps through a multi-scale semantic feature extractor to obtain a set of vibration local time-frequency semantic feature vectors include: Extracting global semantic feature information and local semantic feature information from the vibration local time-frequency feature map to obtain a vibration time-frequency global semantic feature vector and a vibration time-frequency local semantic feature vector; Fusion of the vibration time-frequency global semantic feature vector and the vibration time-frequency local semantic feature vector to obtain the vibration local time-frequency semantic feature vector; Wherein, fusing the vibration time-frequency global semantic feature vector and the vibration time-frequency local semantic feature vector to obtain the vibration local time-frequency semantic feature vector includes: Calculate the vibration time-frequency global semantic feature vector and the vibration time-frequency local semantic feature vector, and add them element by element according to position to obtain a multi-scale vibration time-frequency semantic fusion feature vector; The multi-scale vibration time-frequency semantic fusion feature vector is activated based on a sigmoid function to obtain the vibration local time-frequency semantic feature vector.

2. The intelligent sewage treatment monitoring method based on digital twin according to claim 1 is characterized in that: Extracting global semantic feature information and local semantic feature information from the vibration local time-frequency feature map to obtain a vibration time-frequency global semantic feature vector and a vibration time-frequency local semantic feature vector, including: Upsampling the vibration local time-frequency feature map to obtain an upsampled vibration local time-frequency feature map; Using a vibration global semantic feature extraction module to process the upsampled vibration local time-frequency feature map to obtain a vibration time-frequency global semantic extraction feature vector; Performing dimension adjustment on the vibration time-frequency global semantic extraction feature vector to obtain the vibration time-frequency global semantic feature vector; Using a vibration local semantic feature extraction module to process the upsampled vibration local time-frequency feature map to obtain a vibration time-frequency local semantic extraction feature vector; The dimension of the vibration time-frequency local semantic extraction feature vector is adjusted to obtain the vibration time-frequency local semantic feature vector.

3. The intelligent sewage treatment monitoring method based on digital twin according to claim 2 is characterized in that: The vibration global semantic feature extraction module includes a global average pooling layer, a first point convolution layer, a first batch normalization layer and a first activation layer, and the vibration local semantic feature extraction module includes a second point convolution layer, a second batch normalization layer and a second activation layer.

4. The intelligent sewage treatment monitoring method based on digital twin according to claim 3 is characterized in that: Calculating the hash similarity between any two vibration local time-frequency semantic feature vectors in the set of vibration local time-frequency semantic feature vectors to obtain a local vibration time-frequency consistency topological matrix, including: Using a hash mapping function to map and transform each vibration local time-frequency semantic feature vector in the set of vibration local time-frequency semantic feature vectors to obtain a set of hash-coded vibration local time-frequency semantic feature vectors; The Chebyshev distance between any two hash-coded vibration local time-frequency semantic feature vectors in the set of hash-coded vibration local time-frequency semantic feature vectors is calculated to obtain the local vibration time-frequency consistency topological matrix.

5. The intelligent sewage treatment monitoring method based on digital twin according to claim 4 is characterized in that: The full-time-domain vibration feature pattern extraction is performed on the set of the vibration local time-frequency semantic feature vectors and the local vibration time-frequency consistency topological matrix to obtain a full-time-domain vibration time-frequency topological feature matrix, including: The set of the vibration local time-frequency semantic feature vectors and the local vibration time-frequency consistency topological matrix are passed through a full-time domain vibration pattern feature encoder based on a graph convolutional neural network model to obtain the full-time domain vibration time-frequency topological feature matrix.

6. The intelligent sewage treatment monitoring method based on digital twin according to claim 5 is characterized in that: The monitoring result is determined based on the full-time-domain vibration time-frequency topological characteristic matrix, and the monitoring result is displayed on the screen, including: The full-time-domain vibration time-frequency topological feature matrix is ​​passed through a classifier-based monitoring result generator to obtain the monitoring result, and the monitoring result is used to indicate whether there is an abnormality in the working state of the integrated sewage treatment equipment.

7. A smart sewage treatment monitoring system based on digital twins, characterized in that: include: A vibration signal acquisition module, used to acquire the vibration signal of the integrated sewage treatment equipment collected by the vibration sensor; A signal processing module, used for processing the vibration signal using wavelet transform to obtain a vibration two-dimensional time-frequency diagram; A local vibration semantic feature pattern extraction module is used to extract the local vibration semantic feature pattern from the vibration two-dimensional time-frequency diagram to obtain a set of vibration local time-frequency semantic feature vectors; A hash similarity calculation module, used for calculating the hash similarity between any two vibration local time-frequency semantic feature vectors in the set of vibration local time-frequency semantic feature vectors to obtain a local vibration time-frequency consistency topological matrix; A full-time-domain vibration characteristic pattern extraction module is used to extract the full-time-domain vibration characteristic pattern from the set of the vibration local time-frequency semantic feature vectors and the local vibration time-frequency consistency topological matrix to obtain a full-time-domain vibration time-frequency topological feature matrix; A monitoring result determination module, used to determine the monitoring result based on the full-time-domain vibration time-frequency topological characteristic matrix, and display the monitoring result on a screen; The step of extracting a local vibration semantic feature pattern from the vibration two-dimensional time-frequency graph to obtain a set of vibration local time-frequency semantic feature vectors includes: Performing image block segmentation on the vibration two-dimensional time-frequency map to obtain a set of vibration local two-dimensional time-frequency maps; Each of the vibration local two-dimensional time-frequency graphs in the set of vibration local two-dimensional time-frequency graphs is respectively passed through a vibration time-frequency feature extractor based on a hole convolutional neural network model to obtain a set of vibration local time-frequency feature graphs; Each vibration local time-frequency feature map in the set of vibration local time-frequency feature maps is respectively passed through a multi-scale semantic feature extractor to obtain a set of vibration local time-frequency semantic feature vectors; The steps of respectively passing each vibration local time-frequency feature map in the set of vibration local time-frequency feature maps through a multi-scale semantic feature extractor to obtain a set of vibration local time-frequency semantic feature vectors include: Extracting global semantic feature information and local semantic feature information from the vibration local time-frequency feature map to obtain a vibration time-frequency global semantic feature vector and a vibration time-frequency local semantic feature vector; Fusion of the vibration time-frequency global semantic feature vector and the vibration time-frequency local semantic feature vector to obtain the vibration local time-frequency semantic feature vector; Wherein, fusing the vibration time-frequency global semantic feature vector and the vibration time-frequency local semantic feature vector to obtain the vibration local time-frequency semantic feature vector includes: Calculate the vibration time-frequency global semantic feature vector and the vibration time-frequency local semantic feature vector, and add them element by element according to position to obtain a multi-scale vibration time-frequency semantic fusion feature vector; The multi-scale vibration time-frequency semantic fusion feature vector is activated based on a sigmoid function to obtain the vibration local time-frequency semantic feature vector.