An intelligent management system for chemical production based on data analysis

Abstract

The present application relates to the field of intelligent management, and specifically to an intelligent management system for chemical production based on data analysis. It uses artificial intelligence technology based on deep learning to analyze the time-frequency characteristics of the operating vibration signal of the production equipment, mines the waveform semantic characteristics of the vibration signal, and performs dynamic clustering analysis on it through the attention mechanism to identify the abnormal vibration mode of the production equipment, thereby realizing intelligent detection of the operating status of the production equipment. In this way, the accuracy and robustness of abnormality detection can be effectively improved, and the complex changes in the operating status of equipment in the chemical production process can be better adapted, thereby improving production safety and efficiency.

Description

Technical Field

[0001] The present application relates to the field of intelligent management, and specifically to an intelligent management system for chemical production based on data analysis. Background Art

[0002] In the chemical production process, equipment operation status monitoring is a key link to ensure production safety and efficiency. Traditional monitoring methods mainly rely on manual regular inspections, which are inefficient and easily affected by human factors. In recent years, with the development of big data analysis and artificial intelligence technology, intelligent management systems based on data analysis have been widely used in equipment status monitoring, which can identify abnormal equipment conditions in real time and accurately, and warn of potential failures in advance, thereby reducing downtime and maintenance costs.

[0003] In this regard, patent CN116911619A discloses a chemical production intelligent management and control system based on data analysis, which monitors and analyzes the real-time operation data of all production equipment in the chemical production line to generate corresponding operation status signals, thereby realizing automatic detection of equipment anomalies. However, when analyzing and processing equipment operation data, it uses set static thresholds and predefined failure modes to make abnormal judgments, which has weak adaptability to complex working conditions and cannot effectively capture the dynamic evolution of complex equipment operation status. It may generate false alarms or missed alarms, which in turn leads to unnecessary increase in operation and maintenance costs or equipment failures not being discovered in a timely manner.

[0004] Therefore, an optimized intelligent management system for chemical production based on data analysis is expected. Summary of the invention

[0005] This application is made in consideration of the above problems. One purpose of this application is to provide a chemical production intelligent management system based on data analysis.

[0006] The embodiment of the present application provides a chemical production intelligent management system based on data analysis, including: an intelligent management and control platform, which is used to collect the operating status parameters of the k-th production equipment on the chemical production line, wherein k={1,2,…,m}, m represents the number of production equipment and m is a positive integer; and an operating parameter real-time detection module, which is used to perform real-time detection and analysis of the operating parameters of the k-th production equipment to generate an abnormal operating status warning signal; wherein the operating parameter real-time detection module includes:

[0007] An operation status monitoring unit, used for acquiring an operation status parameter of the kth production equipment collected by a sensor, wherein the operation status parameter is a vibration signal and the sensor is a vibration sensing sensor;

[0008] A vibration signal decomposition unit, used for performing CEEMDAN decomposition on the vibration signal to obtain a set of IMF components of the vibration signal;

[0009] A vibration time-frequency feature extraction unit, used to extract the vibration time-frequency features of each vibration signal IMF component in the set of vibration signal IMF components to obtain a set of vibration signal IMF component waveform semantic feature vectors;

[0010] A feature enhancement unit, used for performing attention enhancement on a set of semantic feature vectors of the IMF component waveform of the vibration signal to obtain a set of semantic feature vectors of the IMF component waveform of the attention-selective vibration signal;

[0011] An operation state abnormality detection unit is used to determine whether the operation state of the k-th production equipment is abnormal based on the set of waveform semantic feature vectors of the IMF component of the attention-selected vibration signal;

[0012] The warning signal generating unit is used to generate an abnormal operation status warning signal in response to an abnormal operation status of the kth production equipment.

[0013] For example, according to the chemical production intelligent management system based on data analysis of an embodiment of the present application, the vibration time-frequency feature extraction unit is used to:

[0014] Each vibration signal IMF component in the set of vibration signal IMF components is respectively input into a vibration time-frequency feature extractor based on a 1D-CNN model to obtain a set of waveform semantic feature vectors of the vibration signal IMF components.

[0015] For example, in the chemical production intelligent management system based on data analysis according to an embodiment of the present application, the feature enhancement unit is used to:

[0016] The set of semantic feature vectors of the IMF component waveform of the vibration signal is input into the reverse selection enhancement network based on the feature reference anchor to obtain the set of semantic feature vectors of the IMF component waveform of the attention-selected vibration signal.

[0017] For example, according to the chemical production intelligent management system based on data analysis of an embodiment of the present application, the feature enhancement unit includes:

[0018] An anchor center calculation subunit, used to calculate the anchor center of the set of semantic feature vectors of the IMF component waveform of the vibration signal to obtain an anchor center feature vector;

[0019] A semantic association calculation subunit, used to calculate the semantic latent association coefficient between each vibration signal IMF component waveform semantic feature vector in the set of vibration signal IMF component waveform semantic feature vectors and the anchor center feature vector to obtain a set of semantic latent association coefficients;

[0020] A semantic anti-correlation calculation subunit, used for calculating the reciprocal of each semantic latent correlation coefficient in the set of semantic latent correlation coefficients to obtain a set of semantic anti-correlation coefficients;

[0021] The feature reverse screening and purification subunit is used to perform feature reverse selection enhancement on the set of semantic feature vectors of the IMF component waveform of the vibration signal based on the set of semantic anti-correlation coefficients to obtain a set of semantic feature vectors of the IMF component waveform of the attention-selective vibration signal.

[0022] For example, according to the chemical production intelligent management system based on data analysis of an embodiment of the present application, the anchor center computing subunit is used to:

[0023] The position-wise mean vector of the set of semantic feature vectors of the IMF component waveform of the vibration signal is calculated to obtain the anchor center feature vector.

[0024] For example, according to the chemical production intelligent management system based on data analysis of an embodiment of the present application, the semantic association calculation subunit is used to:

[0025] The inner product between the semantic feature vector of the IMF component waveform of the vibration signal and the anchor center feature vector is calculated to obtain the semantic correlation factor;

[0026] Calculate the sum of the squares of each eigenvalue in the semantic feature vector of the IMF component waveform of the vibration signal and the sum of the squares of each eigenvalue in the anchor center feature vector, and calculate the square root of the sum to obtain the feature distance factor;

[0027] The semantic association factor is calculated and divided by the feature distance factor to obtain the semantic latent association coefficient.

[0028] For example, according to the chemical production intelligent management system based on data analysis according to an embodiment of the present application, the feature reverse screening and purification subunit is used to:

[0029] The set of semantic anti-correlation coefficients is normalized using the Softmax function to obtain a set of semantic anti-correlation energy coefficients;

[0030] Using each semantic anti-correlation energy coefficient in the set of semantic anti-correlation energy coefficients as a weight, weighting each vibration signal IMF component waveform semantic feature vector in the set of vibration signal IMF component waveform semantic feature vectors to obtain a set of reversely suppressed vibration signal IMF component waveform semantic feature vectors;

[0031] The positional difference between the set of semantic feature vectors of the IMF component waveform of the vibration signal and the set of semantic feature vectors of the IMF component waveform of the inversely suppressed vibration signal is calculated to obtain a set of semantic feature vectors of the IMF component waveform of the attention-selective vibration signal.

[0032] For example, according to the chemical production intelligent management system based on data analysis of an embodiment of the present application, the operation status abnormality detection unit includes:

[0033] A global feature clustering subunit, used to calculate a position-based mean vector of a set of waveform semantic feature vectors of an attention-selective vibration signal IMF component to obtain a global waveform semantic feature vector of the vibration signal;

[0034] The detection result generating subunit is used to input the global waveform semantic feature vector of the vibration signal into the real-time detection module of the operating parameters based on the classifier to obtain the detection result, and the detection result is used to indicate whether there is an abnormality in the operating status of the k-th production equipment.

[0035] For example, according to the chemical production intelligent management system based on data analysis of an embodiment of the present application, the detection result generation subunit is used to:

[0036] The global waveform semantic feature vector of the vibration signal is fully connected and encoded using the fully connected layer of the classifier-based real-time detection module of operating parameters to obtain an encoded classification feature vector;

[0037] The encoded classification feature vector is input into the Softmax classification function of the real-time detection module based on the running parameters of the classifier to obtain the detection result.

[0038] According to the chemical production intelligent management system based on data analysis of the embodiment of the present application, it uses artificial intelligence technology based on deep learning to analyze the time-frequency characteristics of the operating vibration signal of the production equipment, mines the waveform semantic characteristics of the vibration signal, and performs dynamic clustering analysis on it through the attention mechanism to identify the abnormal vibration mode of the production equipment, thereby realizing intelligent detection of the operating status of the production equipment. In this way, the accuracy and robustness of abnormality detection can be effectively improved, and the complex changes in the operating status of equipment in the chemical production process can be better adapted, thereby improving production safety and efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings of the embodiments of the present application are briefly introduced below. Obviously, the drawings described below only relate to some embodiments of the present application, and are not intended to limit the present application.

[0040] Figure 1 A schematic diagram of the application architecture of a chemical production intelligent management system based on data analysis in an embodiment of the present application is shown;

[0041] Figure 2 A schematic diagram of the structure of a chemical production intelligent management system based on data analysis in an embodiment of the present application is shown;

[0042] Figure 3 A structural schematic diagram of a real-time detection module for operating parameters of a chemical production intelligent management system based on data analysis in an embodiment of the present application is shown;

[0043] Figure 4 A flow chart of a chemical production intelligent management method based on data analysis in an embodiment of the present application is shown;

[0044] Figure 5 An application scenario diagram of a chemical production intelligent management system based on data analysis in an embodiment of the present application is shown;

[0045] Figure 6 A schematic diagram of a storage medium according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0046] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present application.

[0047] The terms used in this specification are those common terms currently widely used in the art in consideration of the functions of the present application, but these terms may vary according to the intention of a person of ordinary skill in the art, precedents, or new technologies in the art. In addition, specific terms may be selected by the applicant, and in this case, their detailed meanings will be described in the detailed description of the present application. Therefore, the terms used in the specification should not be understood as simple names, but rather as a general description based on the meaning of the terms and the present application.

[0048] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0049] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. At the same time, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0050] Figure 1 A schematic diagram of the application architecture of a chemical production intelligent management system based on data analysis in an embodiment of the present application is shown, including a server 100 and a terminal device 200.

[0051] The terminal device 200 and the server 100 can be connected via the Internet to achieve mutual communication. Optionally, the above-mentioned Internet uses standard communication technology and / or protocol. The Internet is usually the Internet, but it can also be any network, including but not limited to any combination of a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), a mobile, wired or wireless network, a private network or a virtual private network. In some embodiments, the data exchanged through the network can be represented by technologies and / or formats including Hyper Text Markup Language (HTML), Extensible Markup Language (XML), etc. In addition, conventional encryption technologies such as Secure Socket Layer (SSL), Transport Layer Security (TLS), Virtual Private Network (VPN), Internet Protocol Security (IPsec), etc. can be used to encrypt all or some links. In other embodiments, customized and / or dedicated data communication technologies can also be used to replace or supplement the above-mentioned data communication technologies.

[0052] The server 100 can provide various network services for the terminal device 200, wherein the server 100 can be a single server, a server cluster consisting of several servers, or a cloud computing center. Specifically, the server 100 may include a processor 110 (Center Processing Unit, CPU), a memory 120, an input device 130, and an output device 140, etc. The input device 130 may include a keyboard, a mouse, a touch screen, etc., and the output device 140 may include a display device, such as a liquid crystal display (Liquid Crystal Display, LCD), a cathode ray tube (Cathode Ray Tube, CRT), etc.

[0053] The memory 120 may include a read-only memory (ROM) and a random access memory (RAM), and provides the processor 110 with program instructions and data stored in the memory 120. In the embodiment of the present application, the memory 120 may be used to store the corresponding method of the chemical production intelligent management system based on data analysis in the embodiment of the present application.

[0054] The processor 110 calls the program instructions stored in the memory 120, and the processor 110 is used to execute the steps of any one of the chemical production intelligent management methods based on data analysis in the embodiments of the present application according to the obtained program instructions.

[0055] In addition, the application architecture diagram in the embodiment of the present application is intended to more clearly illustrate the technical solution in the embodiment of the present application, and does not constitute a limitation on the technical solution provided in the embodiment of the present application. Of course, for other application architectures and business applications, the technical solution provided in the embodiment of the present application is also applicable to similar problems.

[0056] The following is a non-restrictive description of the intelligent management system for chemical production based on data analysis provided according to at least one embodiment of the present application through several examples or embodiments. As described below, different features in these specific examples or embodiments can be combined with each other without conflicting with each other to obtain new examples or embodiments, and these new examples or embodiments also fall within the scope of protection of the present application.

[0057] As mentioned in the background technology above, patent CN116911619A discloses a chemical production intelligent management and control system based on data analysis, such as Figure 2 As shown, it specifically includes: an intelligent management and control platform 810, which is used to collect the operating status parameters of the kth production equipment on the chemical production line, where k={1,2,…,m}, m represents the number of production equipment and m is a positive integer; an operating parameter real-time detection module 820, which is used to perform real-time detection and analysis of the operating parameters of the kth production equipment to generate an operating safety signal or an operating warning signal; an operation summary analysis module 830, which is used to set an operation summary period of duration L1, perform an operation summary analysis on the operating status parameters of the kth production equipment to generate an unqualified operation signal or a control detection and analysis signal, and send the control detection and analysis signal to the operation control detection module via the intelligent management and control platform; an operation control detection module 840, which is used to record the moment when the kth production equipment generates an operating warning signal, record the moment when the kth production equipment changes to an operating safety state, and generate an operating qualified signal or an unqualified operation signal through control detection and analysis, and send the unqualified operation signal to the chemical production warning module via the intelligent management and control platform.

[0058] Specifically, when analyzing and processing equipment operation data, the traditional equipment anomaly identification method is to perform fault detection based on preset static thresholds and predefined fault modes. However, this fault detection method has weak adaptability to complex working conditions and may produce false alarms or missed alarms, which in turn leads to unnecessary increase in operation and maintenance costs or equipment failures not being discovered in a timely manner. In response to the above technical problems, the technical concept of the present application is to use artificial intelligence technology based on deep learning to perform time-frequency feature analysis on the operating vibration signals of production equipment, mine the waveform semantic features of the vibration signals, and perform dynamic clustering analysis on them through an attention mechanism to identify abnormal vibration modes of production equipment, thereby realizing intelligent detection of the operating status of production equipment. In this way, the accuracy and robustness of anomaly detection can be effectively improved, and the complex changes in the operating status of equipment in the chemical production process can be better adapted, thereby improving production safety and efficiency.

[0059] Based on this, Figure 3 As shown, the real-time detection module 820 of operating parameters includes: an operating status monitoring unit 821, which is used to obtain the operating status parameters of the k-th production equipment collected by the sensor, wherein the operating status parameters are vibration signals and the sensors are vibration sensing sensors; a vibration signal decomposition unit 822, which is used to perform CEEMDAN decomposition on the vibration signal to obtain a set of IMF components of the vibration signal; a vibration time-frequency feature extraction unit 823, which is used to extract the vibration time-frequency features of each vibration signal IMF component in the set of vibration signal IMF components to obtain a set of waveform semantic feature vectors of the vibration signal IMF components; a feature enhancement unit 824, which is used to perform attention enhancement on the set of waveform semantic feature vectors of the vibration signal IMF components to obtain a set of waveform semantic feature vectors of the attention-selected vibration signal IMF components; an operating status abnormality detection unit 825, which is used to determine whether the operating status of the k-th production equipment is abnormal based on the set of waveform semantic feature vectors of the attention-selected vibration signal IMF components; and an early warning signal generation unit 826, which is used to generate an operating status abnormality early warning signal in response to the existence of an abnormality in the operating status of the k-th production equipment.

[0060] Specifically, in the technical solution of the present application, first, the operating state parameters of the kth production equipment collected by the sensor are obtained, wherein the operating state parameters are vibration signals, and the sensor is a vibration sensing sensor. It should be understood that the vibration signal is a direct reflection of the operating state of the mechanical equipment and contains rich mechanical state information. When abnormal conditions such as wear, looseness, and imbalance occur inside the equipment, they are first reflected in the changes in the vibration signal. Therefore, by analyzing the vibration signal, potential equipment failures can be discovered early to avoid production interruptions and significant losses caused by sudden equipment failures. In addition, the vibration sensing sensor can collect vibration data in real time and continuously without interfering with the normal operation of the equipment. This non-invasive monitoring method will not have any impact on the production process, while ensuring the real-time and continuity of the data, which is conducive to timely detection and response to changes in equipment status.

[0061] Next, considering that in the chemical production process, the vibration signal of the equipment is often nonlinear and non-stationary, and there are many complex characteristic components. Therefore, in order to understand the characteristics and changes of the signal from different time scales, in the technical solution of the present application, the CEEMDAN algorithm is further used to decompose the vibration signal, decomposing it into multiple intrinsic mode function (IMF) components, and each vibration signal IMF component corresponds to a signal fluctuation component within a certain frequency range in the vibration signal, so as to fully reflect the vibration characteristics of different frequency components of the vibration signal. In addition, the CEEMDAN algorithm effectively suppresses the influence of noise on the decomposition results by adding adaptive noise and multiple independent decompositions and ensemble averaging, reduces the occurrence of modal aliasing, and improves the accuracy and stability of signal decomposition.

[0062] Then, a vibration time-frequency feature extractor based on the 1D-CNN model is used to extract time-frequency features of each vibration signal IMF component in the set of vibration signal IMF components. Through multi-layer stacked convolution and pooling operations, the local feature structure in the IMF components is automatically learned to capture the fluctuation information within a specific frequency range in the original vibration signal, thereby obtaining a set of waveform semantic feature vectors of the vibration signal IMF components.

[0063] Correspondingly, the vibration time-frequency feature extraction unit 823 is used to: input each vibration signal IMF component in the set of vibration signal IMF components into the vibration time-frequency feature extractor based on the 1D-CNN model to obtain a set of vibration signal IMF component waveform semantic feature vectors.

[0064] Furthermore, considering that in practical applications, vibration signals are affected by noise interference and slight changes in equipment, the set of semantic feature vectors of the IMF component waveform of the vibration signal may have uneven feature distribution and importance differences. Therefore, in order to further improve the quality of features, in the technical solution of the present application, a reverse selection enhancement network based on feature reference anchors is introduced to process the set of semantic feature vectors of the IMF component waveform of the vibration signal to highlight key features and suppress the influence of noise, thereby obtaining a set of semantic feature vectors of the IMF component waveform of the attention-selected vibration signal. Specifically, the reverse selection enhancement network based on the feature reference anchor can eliminate the offset of feature distribution through centralized processing. It dynamically allocates weights based on the feature distribution offset between the semantic feature vectors of the IMF component waveform of each vibration signal and the center of the anchor point, and strengthens the irrelevant features in the set of semantic feature vectors of the IMF component waveform of the vibration signal through the reverse selection mechanism, so as to eliminate unimportant interference features from the set of feature vectors, realize feature purification optimization and selective focusing, and thus improve the sensitivity and accuracy of anomaly detection.

[0065] Accordingly, the feature enhancement unit 824 is used to: input the set of semantic feature vectors of the IMF component waveform of the vibration signal into the reverse selection enhancement network based on the feature reference anchor to obtain a set of attention-selective semantic feature vectors of the IMF component waveform of the vibration signal.

[0066] Secondly, the position mean vector of the set of semantic feature vectors of the IMF component waveform of the attention-selected vibration signal is further calculated to summarize the overall state of the equipment operation and capture the macro trend of the equipment state change, thereby obtaining the global waveform semantic feature vector of the vibration signal. Then, the global waveform semantic feature vector of the vibration signal is input into the real-time detection module of the operating parameters based on the classifier to obtain the detection result, and the detection result is used to indicate whether the operating state of the k-th production equipment is abnormal. That is, the classification algorithm is used to realize the real-time evaluation of the equipment operation state. Specifically, the classifier model forms a classification decision boundary by learning the characteristic distribution law of the equipment operation state according to the normal state and abnormal state samples in the training data. Based on this, the trained classification model is used to perform feature pattern learning and classification reasoning on the global waveform semantic feature vector of the vibration signal, and the learned classification decision boundary can be used to judge whether the current equipment operation state is abnormal. When the detection result indicates that the operation state of the k-th production equipment is abnormal, an unqualified operation signal is generated and sent to the chemical production early warning module in real time through the intelligent management and control platform, triggering a rapid response mechanism, thereby timely discovering the abnormal behavior of the equipment, reducing false alarms and missed alarms, improving production efficiency, and ensuring the safe operation of chemical production.

[0067] Correspondingly, the feature enhancement unit 824 includes: an anchor center calculation subunit, which is used to calculate the anchor center of the set of semantic feature vectors of the vibration signal IMF component waveform to obtain the anchor center feature vector; a semantic association calculation subunit, which is used to calculate the semantic implicit association coefficient between each vibration signal IMF component waveform semantic feature vector in the set of semantic feature vectors of the vibration signal IMF component waveform and the anchor center feature vector to obtain a set of semantic implicit association coefficients; a semantic anti-association calculation subunit, which is used to calculate the inverse of each semantic implicit association coefficient in the set of semantic implicit association coefficients to obtain a set of semantic anti-association coefficients; a feature reverse screening and purification subunit, which is used to perform feature reverse selection and enhancement on the set of semantic feature vectors of the vibration signal IMF component waveform based on the set of semantic anti-association coefficients to obtain a set of attention-selective vibration signal IMF component waveform semantic feature vectors.

[0068] The anchor center calculation subunit is used to calculate the position mean vector of the set of semantic feature vectors of the vibration signal IMF component waveform to obtain the anchor center feature vector. The semantic association calculation subunit is used to calculate the inner product between the semantic feature vector of the vibration signal IMF component waveform and the anchor center feature vector to obtain the semantic association factor; calculate the sum of the squares of each eigenvalue in the semantic feature vector of the vibration signal IMF component waveform and the sum of the squares of each eigenvalue in the anchor center feature vector, and calculate the square root of the sum to obtain the feature distance factor; calculate the semantic association factor divided by the feature distance factor to obtain the semantic implicit association coefficient. The feature reverse screening and purification subunit is used to: use the Softmax function to normalize the set of semantic anti-correlation coefficients to obtain a set of semantic anti-correlation energy coefficients; use each semantic anti-correlation energy coefficient in the set of semantic anti-correlation energy coefficients as a weight to weight each vibration signal IMF component waveform semantic feature vector in the set of vibration signal IMF component waveform semantic feature vectors to obtain a set of reverse inhibition vibration signal IMF component waveform semantic feature vectors; calculate the positional difference between the set of vibration signal IMF component waveform semantic feature vectors and the set of reverse inhibition vibration signal IMF component waveform semantic feature vectors to obtain a set of attention selection vibration signal IMF component waveform semantic feature vectors.

[0069] In a specific example, the feature enhancement unit 824 is used to process the set of semantic feature vectors of the IMF component waveform of the vibration signal using the following feature distribution reverse screening formula to obtain a set of semantic feature vectors of the IMF component waveform of the attention selection vibration signal, wherein the feature distribution reverse screening formula is:

[0070] ,

[0071] ,

[0072] ,

[0073] ,

[0074] ,

[0075] ;

[0076] in, is the first in the set of semantic feature vectors of the IMF component waveform of the vibration signal The semantic feature vector of the IMF component waveform of the vibration signal, Indicates The first semantic feature vector of the IMF component waveform of the vibration signal The eigenvalues ​​at the positions, is the anchor center feature vector, Indicates the length of the semantic feature vector of the IMF component waveform of the vibration signal, Indicates The semantic latent correlation coefficient between the semantic feature vector of the IMF component waveform of the vibration signal and the anchor center feature vector, For the The semantic anti-correlation coefficient, For the The semantic anti-correlation energy coefficient, For the The semantic feature vector of the IMF component waveform of the reverse suppression vibration signal, For the The waveform semantic feature vector of IMF component of vibration signal is selected by attention.

[0077] Correspondingly, the operation status abnormality detection unit 825 includes: a global feature clustering subunit, which is used to calculate the positional mean vector of the set of waveform semantic feature vectors of the attention-selected vibration signal IMF component to obtain the global waveform semantic feature vector of the vibration signal; a detection result generation subunit, which is used to input the global waveform semantic feature vector of the vibration signal into the classifier-based operation parameter real-time detection module to obtain the detection result, and the detection result is used to indicate whether there is an abnormality in the operation status of the k-th production equipment.

[0078] Among them, the detection result generation subunit is used to: use the fully connected layer of the classifier-based operating parameter real-time detection module to fully connect the global waveform semantic feature vector of the vibration signal to obtain the encoded classification feature vector; input the encoded classification feature vector into the Softmax classification function of the classifier-based operating parameter real-time detection module to obtain the detection result.

[0079] It should be understood that the role of the classifier is to use the given categories and known training data to learn classification rules and classifiers, and then classify (or predict) unknown data. Logistic regression, SVM, etc. are often used to solve binary classification problems. For multi-class classification problems, logistic regression or SVM can also be used, but multiple binary classifications are required to form a multi-classification, but this is prone to errors and inefficient. Commonly used multi-classification methods include the Softmax classification function.

[0080] In a preferred embodiment, the detection result generating subunit includes: an optimization secondary subunit, which is used to optimize the global waveform semantic feature vector of the vibration signal to obtain an optimized global waveform semantic feature vector of the vibration signal; a fully connected encoding secondary subunit, which is used to use the fully connected layer of the classifier-based operating parameter real-time detection module to fully connect the optimized vibration signal global waveform semantic feature vector to obtain an optimized encoded classification feature vector; a classification secondary subunit, which is used to input the optimized encoded classification feature vector into the Softmax classification function of the classifier-based operating parameter real-time detection module to obtain a detection result.

[0081] In the technical solution of the present application, each vibration signal IMF component waveform semantic feature vector in the set of vibration signal IMF component waveform semantic feature vectors respectively represents the waveform semantic features of different vibration signal IMF components in the vibration signal of the k-th production equipment. When the set of vibration signal IMF component waveform semantic feature vectors is input into the anchor-centered feature distribution reverse selection enhancement network, the anchor-centered feature distribution reverse selection enhancement network uses the anchor center of the set of vibration signal IMF component waveform semantic feature vectors to perform feature reverse screening to strengthen the effective signal components and suppress the invalid signal components, which can improve the significance of the signal distribution in the set of attention-selected vibration signal IMF component waveform semantic feature vectors. However, at the same time, this will also amplify the differences between the local feature distributions of the set of attention-selected vibration signal IMF component waveform semantic feature vectors as a whole feature distribution, thereby causing the vibration signal global waveform semantic feature vector to have a more significant sequence aggregation offset, so it is expected to further improve the global aggregation regression comprehensibility of the vibration signal global waveform semantic feature vector, thereby improving the accuracy of the detection results obtained by inputting the real-time detection module of the operating parameters based on the classifier.

[0082] Based on this, in a preferred embodiment, the global waveform semantic feature vector of the vibration signal is input into a real-time detection module of operating parameters based on a classifier to obtain a detection result, which specifically includes:

[0083] Subtract the zero norm of the global waveform semantic feature vector of the vibration signal from the length of the global waveform semantic feature vector of the vibration signal to obtain the global waveform semantic isolated zero value of the vibration signal;

[0084] Calculate the logarithm of the square of the global waveform semantic isolated zero value of the vibration signal and the sum of the global waveform semantic isolated zero value of the vibration signal to obtain the global waveform semantic superposition information value of the vibration signal;

[0085] Calculate a power function with each eigenvalue of the vibration signal global waveform semantics eigenvector as the base and the difference of the vibration signal global waveform semantics isolated zero value minus one as the exponent, and perform a point multiplication with the vibration signal global waveform semantics superposition information value to obtain a first vibration signal global waveform semantics intermediate vector;

[0086] After performing a dot multiplication on the vibration signal global waveform semantic feature vector and the difference between the vibration signal global waveform semantic isolated zero value minus one, the dot multiplication is then performed on the reciprocal of the vibration signal global waveform semantic isolated zero value to obtain a second vibration signal global waveform semantic intermediate vector;

[0087] Calculate an exponential function with a natural constant as the base and each eigenvalue of the second vibration signal global waveform semantic intermediate vector as the exponent to obtain a third vibration signal global waveform semantic intermediate vector;

[0088] Performing point addition of the first vibration signal global waveform semantic intermediate vector and the third vibration signal global waveform semantic intermediate vector to obtain an optimized vibration signal global waveform semantic feature vector;

[0089] And the optimized global waveform semantic feature vector of the vibration signal is input into a real-time detection module of operating parameters based on the classifier to obtain the detection result.

[0090] That is, the global waveform semantic feature vector of the vibration signal is recorded as The optimization process is expressed as:

[0091] ,

[0092] ,

[0093] in, Represents the global waveform semantic feature vector of the vibration signal, represents the zero norm of a vector, Represents the length of the global waveform semantic feature vector of the vibration signal, Indicates the isolated zero value of the global waveform semantics of the vibration signal, It is the point multiplication by position. is added by position, It represents the power function that calculates the eigenvalues ​​of each position in the eigenvector with the difference of the global waveform semantic isolated zero value of the vibration signal minus one as the exponent. Represents the optimized global waveform semantic feature vector of the vibration signal.

[0094] In this way, the high-dimensional feature manifold of the global waveform semantic feature vector of the vibration signal is represented by the eigenvalue of the feature set as the vector field of the aggregation dimension, and the superposition value of the vector field of the global waveform semantic feature vector of the vibration signal at the isolated zero position is used as the order information to fix the local position of the eigenvalue of its feature set, and a bias for the reversibility of the feature regression distribution field of the global waveform semantic feature vector of the vibration signal is added as a reward to achieve the mapping target tracking of the regression distribution of the global waveform semantic feature vector of the vibration signal for the eigenvalue position, so that the feature set of the global waveform semantic feature vector of the vibration signal can perceive the mapping migration to the aggregate distribution, thereby improving the accuracy of the detection result obtained by the real-time detection module of the operating parameters of the global waveform semantic feature vector of the vibration signal based on the classifier by improving the global aggregation regression comprehensibility of the global waveform semantic feature vector of the vibration signal.

[0095] Based on the above embodiments, see Figure 4 , which is a flow chart of a chemical production intelligent management method based on data analysis in an embodiment of the present application. For example, the chemical production intelligent management method based on data analysis can be executed by a server, which can be Figure 1 The server 100 shown in FIG. Figure 4 As shown, according to the embodiment of the present application, the intelligent management method for chemical production based on data analysis includes the following steps: S510, obtaining the operating status parameter of the kth production equipment collected by the sensor, wherein the operating status parameter is a vibration signal and the sensor is a vibration sensing sensor; S520, performing CEEMDAN decomposition on the vibration signal to obtain a set of IMF components of the vibration signal; S530, extracting the vibration time-frequency characteristics of each vibration signal IMF component in the set of vibration signal IMF components to obtain a set of vibration signal IMF component waveform semantic feature vectors; S540, performing attention enhancement on the set of vibration signal IMF component waveform semantic feature vectors to obtain a set of attention-selected vibration signal IMF component waveform semantic feature vectors; S550, determining whether there is an abnormality in the operating status of the kth production equipment based on the set of attention-selected vibration signal IMF component waveform semantic feature vectors; S560, generating an abnormal operating status warning signal in response to the existence of an abnormality in the operating status of the kth production equipment.

[0096] Here, those skilled in the art can understand that the specific operations of each step in the above-mentioned chemical production intelligent management method based on data analysis have been referred to above. Figure 2 to Figure 3The description of the chemical production intelligent management system 800 based on data analysis has been introduced in detail, and therefore, its repeated description will be omitted.

[0097] Figure 5 FIG. 1 is an application scenario diagram of a chemical production intelligent management system based on data analysis according to an embodiment of the present application. Figure 5 As shown, in this application scenario, first, the operating status parameters of the kth production equipment collected by the sensor are obtained (for example, Figure 5 D), where the operating state parameter is a vibration signal, and the sensor is a vibration sensing sensor. Then, the operating state parameter is input to a server (for example, Figure 5 In S) shown in , the server can use a chemical production intelligent management algorithm based on data analysis to process the operating status parameters to determine whether there is an abnormality in the operating status of the kth production equipment, and then generate an operating status abnormality warning signal in response to the abnormality in the operating status of the kth production equipment.

[0098] Based on the above embodiments, another exemplary embodiment of an electronic device is also provided in the embodiments of the present application. In some possible implementations, the electronic device in the embodiments of the present application may include a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor can implement the steps of the chemical production intelligent management method based on data analysis in the above embodiments when executing the program.

[0099] For example, in the case of electronic equipment Figure 1 Taking the server 100 in the example as an example, the processor in the electronic device is the processor 110 in the server 100, and the memory in the electronic device is the memory 120 in the server 100.

[0100] An embodiment of the present application also provides a computer-readable storage medium. Figure 6 1 shows a schematic diagram of a computer-readable storage medium 1000 according to an embodiment of the present application. Figure 6 As shown, a computer executable instruction 1001 is stored on a computer readable storage medium 1000. When the computer executable instruction 1001 is executed by a processor, the intelligent management method for chemical production based on data analysis according to the embodiment of the present application described with reference to the above figures can be executed. Computer readable storage media include, but are not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and / or cache memory (cache), etc. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0101] The embodiment of the present application also provides a computer program product or computer program, which includes computer executable instructions, and the computer executable instructions are stored in a computer-readable storage medium. The processor of the computer device reads the computer executable instructions from the computer-readable storage medium, and the processor executes the computer executable instructions, so that the computer device executes the chemical production intelligent management method based on data analysis according to the embodiment of the present application.

[0102] Those skilled in the art will appreciate that the contents disclosed in this application may be subject to various variations and improvements. For example, the various devices or components described above may be implemented by hardware, or by software, firmware, or a combination of some or all of the three.

[0103] In addition, although the present application makes various references to certain units in the system according to embodiments of the present application, any number of different units can be used and run on the client and / or server. The units are only illustrative, and different aspects of the system and method can use different units.

[0104] Those skilled in the art will appreciate that all or part of the steps in the above method can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, such as a read-only memory, a disk or an optical disk. Optionally, all or part of the steps in the above embodiment can also be implemented using one or more integrated circuits. Accordingly, each module / unit in the above embodiment can be implemented in the form of hardware or in the form of software function modules. The present application is not limited to any particular form of combination of hardware and software.

[0105] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art to which this application belongs. It should also be understood that terms such as those defined in common dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an idealized or extremely formal sense, unless explicitly defined as such herein.

[0106] The above is an explanation of the present application and should not be considered as a limitation thereof. Although several exemplary embodiments of the present application have been described, it will be readily understood by those skilled in the art that many modifications may be made to the exemplary embodiments without departing from the novel teachings and advantages of the present application. Therefore, all of these modifications are intended to be included within the scope of the present application as defined by the claims. It should be understood that the above is an explanation of the present application and should not be considered to be limited to the specific embodiments disclosed, and modifications to the disclosed embodiments and other embodiments are intended to be included within the scope of the appended claims. The present application is defined by the claims and their equivalents.

Claims

1. A chemical production intelligent management system based on data analysis, comprising: An intelligent management and control platform, used for collecting operating status parameters of the kth production equipment on a chemical production line, wherein k={1,2,…,m}, m represents the number of production equipment and m is a positive integer; and an operating parameter real-time detection module, used for performing real-time detection and analysis of operating parameters of the kth production equipment to generate an abnormal operating status warning signal; characterized in that the operating parameter real-time detection module includes: An operation status monitoring unit, used for acquiring an operation status parameter of the k-th production equipment collected by a sensor, wherein the operation status parameter is a vibration signal and the sensor is a vibration sensing sensor; A vibration signal decomposition unit, used for performing CEEMDAN decomposition on the vibration signal to obtain a set of IMF components of the vibration signal; A vibration time-frequency feature extraction unit, used to extract the vibration time-frequency features of each vibration signal IMF component in the set of vibration signal IMF components to obtain a set of vibration signal IMF component waveform semantic feature vectors; A feature enhancement unit, used for performing attention enhancement on the set of semantic feature vectors of the IMF component waveform of the vibration signal to obtain a set of semantic feature vectors of the IMF component waveform of the attention-selective vibration signal; An operation state abnormality detection unit is used to determine whether the operation state of the k-th production equipment is abnormal based on the set of semantic feature vectors of the IMF component waveform of the attention selection vibration signal; an early warning signal generating unit, configured to generate an abnormal operation state early warning signal in response to an abnormal operation state of the k-th production equipment; The vibration time-frequency feature extraction unit is used to: Inputting each vibration signal IMF component in the set of vibration signal IMF components into a vibration time-frequency feature extractor based on a 1D-CNN model to obtain a set of waveform semantic feature vectors of the vibration signal IMF components; The feature enhancement unit is used to: Inputting the set of semantic feature vectors of the IMF component waveform of the vibration signal into a reverse selection enhancement network based on a feature reference anchor to obtain the set of semantic feature vectors of the IMF component waveform of the attention selection vibration signal; The feature enhancement unit comprises: An anchor center calculation subunit, used to calculate the anchor center of the set of semantic feature vectors of the IMF component waveform of the vibration signal to obtain an anchor center feature vector; A semantic association calculation subunit, used for calculating the semantic latent association coefficient between each vibration signal IMF component waveform semantic feature vector in the set of vibration signal IMF component waveform semantic feature vectors and the anchor center feature vector to obtain a set of semantic latent association coefficients; A semantic anti-correlation calculation subunit, used for calculating the reciprocal of each semantic latent correlation coefficient in the set of semantic latent correlation coefficients to obtain a set of semantic anti-correlation coefficients; The feature reverse screening and purification subunit is used to perform feature reverse selection enhancement on the set of semantic feature vectors of the IMF component waveform of the vibration signal based on the set of semantic anti-correlation coefficients to obtain the set of semantic feature vectors of the IMF component waveform of the attention selection vibration signal.

2. The chemical production intelligent management system based on data analysis according to claim 1 is characterized in that: The anchor center computing subunit is used to: The position-based mean vector of the set of semantic feature vectors of the IMF component waveform of the vibration signal is calculated to obtain the anchor center feature vector.

3. The chemical production intelligent management system based on data analysis according to claim 2 is characterized in that: The semantic association calculation subunit is used to: Calculating the inner product between the semantic feature vector of the vibration signal IMF component waveform and the anchor center feature vector to obtain a semantic association factor; Calculating the sum of the squares of each eigenvalue in the semantic feature vector of the IMF component waveform of the vibration signal and the sum of the squares of each eigenvalue in the anchor center feature vector, and calculating the square root of the sum to obtain a feature distance factor; The semantic association factor is calculated and divided by the feature distance factor to obtain the semantic latent association coefficient.

4. The chemical production intelligent management system based on data analysis according to claim 3 is characterized in that: The characteristic reverse screening and purification subunit is used for: Using a Softmax function to normalize the set of semantic anti-correlation coefficients to obtain a set of semantic anti-correlation energy coefficients; Using each semantic anti-correlation energy coefficient in the set of semantic anti-correlation energy coefficients as a weight, weighting each vibration signal IMF component waveform semantic feature vector in the set of vibration signal IMF component waveform semantic feature vectors to obtain a set of reversely suppressed vibration signal IMF component waveform semantic feature vectors; The positional difference between the set of semantic feature vectors of the IMF component waveform of the vibration signal and the set of semantic feature vectors of the IMF component waveform of the inversely suppressed vibration signal is calculated to obtain the set of semantic feature vectors of the IMF component waveform of the attention-selective vibration signal.

5. The chemical production intelligent management system based on data analysis according to claim 4 is characterized in that: The abnormal operation state detection unit comprises: A global feature clustering subunit, used for calculating a position mean vector of a set of waveform semantic feature vectors of the attention-selective vibration signal IMF component to obtain a global waveform semantic feature vector of the vibration signal; The detection result generating subunit is used to input the global waveform semantic feature vector of the vibration signal into the real-time detection module of the operating parameters based on the classifier to obtain the detection result, and the detection result is used to indicate whether there is an abnormality in the operating status of the k-th production equipment.

6. The chemical production intelligent management system based on data analysis according to claim 5 is characterized in that: The detection result generating subunit is used for: Using the fully connected layer of the classifier-based real-time detection module for operating parameters, the global waveform semantic feature vector of the vibration signal is fully connected to obtain a coded classification feature vector; The encoded classification feature vector is input into the Softmax classification function of the classifier-based operating parameter real-time detection module to obtain the detection result.

Images