A method and system for filling and repairing operation data of an industrial process

Abstract

The present invention provides a method and system for filling and repairing operating data of an industrial process, and relates to the technical field of data filling. The global long-term trend of the data is captured by minimizing the nuclear norm of the circulant matrix, the local short-term fluctuation characteristics of the data are described in combination with time-adaptive Laplace regularization, and the correlation between variables is characterized by graph structure regularization with adaptive weights. The balance parameters are dynamically adjusted through an adaptive weight mechanism to adapt to different time series characteristics and optimize the interpolation effect. Finally, the alternating direction multiplier method is used to solve the optimization model and generate the filled and repaired data. At the same time, the global trend, local fluctuation and relationship between variables of the data are taken into account, which significantly improves the accuracy and reliability of data filling.

Description

Technical Field

[0001] The present application relates to the technical field of data filling, and in particular to a method and system for filling and repairing operation data of an industrial process. Background Art

[0002] In the modern industrial system, intelligent and refined management of industrial production relies heavily on real-time, accurate operational data from the production process. From a micro perspective, various sensors on the production line collect real-time data on equipment operating parameters and process indicators, providing operators with first-hand information to promptly understand equipment status and adjust production processes. From a macro perspective, the integration and analysis of large amounts of industrial data can provide strong support for companies' strategic decision-making and resource allocation, helping them gain an advantage in the fierce market competition.

[0003] However, the extreme complexity of industrial production environments presents numerous challenges to data collection. High temperatures can cause thermal stress in sensor electronics, leading to reduced measurement accuracy or even failure, resulting in data loss. Strong magnetic fields can interfere with sensor signal transmission, causing errors or loss in collected data. Strong vibrations can physically damage sensors, preventing data collection. Furthermore, industrial network instability, such as signal interference and network congestion, often leads to data transmission interruptions, further exacerbating the data loss problem.

[0004] Missing data negatively impacts industrial production in many ways. Currently, data infill and repair technology is an important means of addressing this problem. While there are many traditional data infill and repair methods, each has significant limitations.

[0005] Therefore, how to improve the accuracy and reliability of industrial data filling has become a technical problem that needs to be solved urgently. Summary of the Invention

[0006] In order to improve the accuracy and reliability of industrial data filling, the present application provides a method and system for filling and repairing operation data of an industrial process.

[0007] In the first aspect, the present application provides a method for filling and repairing the operation data of an industrial process using the following technical solutions:

[0008] A method for filling and repairing operation data of an industrial process, comprising:

[0009] Collecting partial observation data of an industrial process, wherein the partial observation data includes missing data;

[0010] Based on the minimization of the nuclear norm of the circulant matrix, a global trend model of the data is constructed, and some observation data are interpolated through the minimization operation of the nuclear norm of the circulant matrix;

[0011] A time-adaptive Laplace regularization module is introduced to dynamically adjust the local smoothing constraint weights according to the time interval of the data to capture the short-term fluctuation characteristics of some observation data;

[0012] Construct an adaptive weighted graph regularization module, which learns the correlation weights between variables through the adjacency matrix and graph structure, so that the interpolation results of related variables tend to be consistent;

[0013] Design an adaptive weighting mechanism to dynamically adjust the balance parameters between global and local trends based on the local stationarity of the data;

[0014] Integrate the global trend model, the time-adaptive Laplace regularization module, the graph structure regularization module, and the adaptive weight mechanism to construct a unified optimization problem;

[0015] The optimization problem is iteratively solved using an alternating direction multiplier method until convergence conditions are met, and the interpolated complete data is output.

[0016] Optionally, minimizing the circulant matrix nuclear norm includes:

[0017] Converting the portion of observation data into a circulant matrix form;

[0018] Fast Fourier transform is used to calculate the nuclear norm of the circulant matrix, and global trend modeling and missing data interpolation are achieved by minimizing the nuclear norm.

[0019] Optionally, the step of introducing a time-adaptive Laplace regularization module to dynamically adjust the local smoothing constraint weight according to the time interval of the data to capture the short-term fluctuation characteristics of part of the observed data includes:

[0020] Constructing a Laplace matrix, and determining the adjacent relationship between data according to the Laplace matrix;

[0021] A time weighting factor is introduced to perform weighted correction on the Laplace regularization term according to the time interval between data points;

[0022] The time weighting factor is used to adjust the smoothness constraint of the data according to the time interval, and the short-term fluctuation characteristics of part of the observation data are captured in the Laplace matrix.

[0023] Optionally, the variable similarity weight in the graph structure regularization module of the adaptive weight Defined as:

[0024]

[0025] in, is the scale factor to control the similarity, and Represents two different variables, is a normalization constant used to ensure that the sum of all weights is 1.

[0026] Optionally, the adaptive weight mechanism is used to dynamically adjust the balance parameters γ , specifically including:

[0027] Calculate the variance in a local time window as a measure of local stationarity for a given time series , in The center is In the time window of It can be calculated as:

[0028]

[0029] in, is the mean of the data in the time window;

[0030] According to the local stationary measure, adaptive weights are generated by Sigmoid function mapping:

[0031]

[0032] in, It's at the time The adaptive weight at and It is an adjustable parameter used to control the shape and range of the mapping function. The range of the Sigmoid function is between (0, 1);

[0033] Based on the adaptive weight, the balance parameter γ is dynamically adjusted to achieve a dynamic balance between global and local trend weights. Assuming that the original fixed γ value is , then the dynamically adjusted γ value The calculation method is:

[0034]

[0035] in, Indicates that when more attention is paid to the overall trend The value of .

[0036] Optionally, the step of integrating the global trend model, the time-adaptive Laplace regularization module, the graph structure regularization module, and the adaptive weight mechanism to construct a unified optimization problem includes:

[0037] The global trend model, the time-adaptive Laplace regularization module, the graph structure regularization module, and the adaptive weight mechanism are integrated to construct a unified optimization model, which is expressed as:

[0038]

[0039] Among them, the global trend data is , local trend data is , the variable relationship data is , n represents the number of variables x.

[0040] Optionally, the step of iteratively solving the optimization problem using an alternating direction multiplier method until a convergence condition is satisfied and outputting the interpolated complete data includes:

[0041] Introducing auxiliary variables ;

[0042] , , , And the optimization problem is expressed as:

[0043]

[0044] The augmented Lagrangian function is constructed as follows:

[0045]

[0046] in is the Lagrange multiplier, is the penalty parameter;

[0047] Update X:

[0048] Fix the remaining variables and solve the following optimization problem:

[0049]

[0050] renew :

[0051] Fix the remaining variables and solve the following optimization problem:

[0052]

[0053] renew :

[0054] Fix the remaining variables and solve the following optimization problem:

[0055]

[0056] renew :

[0057] Fix the remaining variables and solve the following optimization problem:

[0058]

[0059] renew :

[0060] Fix the remaining variables and solve the following optimization problem:

[0061]

[0062] Update the Lagrange multipliers:

[0063]

[0064] The above sub-problems are updated iteratively until the convergence conditions are met, and the complete interpolated data are output.

[0065] In a second aspect, the present application provides an industrial process operation data filling and repair system, comprising:

[0066] A data collection module, configured to collect partial observation data of an industrial process, wherein the partial observation data includes missing data;

[0067] The global trend module is used to build a data global trend model based on the minimization of the nuclear norm of the circulant matrix and interpolate some observation data through the minimization operation of the nuclear norm of the circulant matrix;

[0068] The short-term feature module is used to introduce a time-adaptive Laplace regularization module, dynamically adjust the local smoothing constraint weights according to the time interval of the data, and capture the short-term fluctuation characteristics of some observation data;

[0069] Regularization module, used to build a graph structure regularization module with adaptive weights. It learns the correlation weights between variables through the adjacency matrix and graph structure, so that the interpolation results of related variables tend to be consistent;

[0070] Adaptive weight module, used to design an adaptive weight mechanism to dynamically adjust the balance parameters between global and local trends according to the local stationarity of the data;

[0071] An optimization problem module, which is used to integrate the global trend model, the time-adaptive Laplace regularization module, the graph structure regularization module and the adaptive weight mechanism to construct a unified optimization problem;

[0072] The output module is used to iteratively solve the optimization problem by using the alternating direction multiplier method until the convergence condition is met, and output the complete data after interpolation.

[0073] In a third aspect, the present application provides a computer device, comprising: a memory and a processor, wherein the processor executes the method described above when running computer instructions stored in the memory.

[0074] In a fourth aspect, the present application provides a computer-readable storage medium comprising instructions, which, when executed on a computer, enable the computer to execute the method described above.

[0075] In summary, this application has the following beneficial technical effects:

[0076] This application captures the global long-term trend of the data by minimizing the circulant matrix nuclear norm, combines time-adaptive Laplace regularization to describe the local short-term fluctuation characteristics of the data, and uses graph structure regularization with adaptive weights to characterize the correlation between variables. An adaptive weight mechanism dynamically adjusts the balance parameters to adapt to different time series characteristics and optimize the interpolation effect. Finally, the alternating direction multiplication method is used to solve the optimization model and generate the imputed and repaired data. Simultaneously considering the global trend of the data, local fluctuations, and the relationship between variables, the accuracy and reliability of data imputation are significantly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] Figure 1 It is a schematic diagram of the computer device structure of the hardware operating environment involved in the embodiment of the present application;

[0078] Figure 2 This is a flow chart of a first embodiment of the method for filling and repairing operating data of an industrial process of the present application;

[0079] Figure 3 This is a Laplace matrix diagram of the first embodiment of the method for filling and repairing operating data of an industrial process of the present application;

[0080] Figure 4 It is a structural block diagram of the first embodiment of the industrial process operation data filling and repair system of the present application. DETAILED DESCRIPTION

[0081] In order to make the purpose, technical solutions and advantages of this application more clearly understood, the present application is further described in detail below through the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0082] Reference Figure 1 , Figure 1 This is a schematic diagram of the computer device structure of the hardware operating environment involved in the embodiment of the present application.

[0083] like Figure 1As shown, the computer device may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. The communication bus 1002 is used to enable communication between these components. The user interface 1003 may include a display and an input unit, such as a keyboard. Optionally, the user interface 1003 may also include a standard wired interface or a wireless interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a wireless fidelity (Wi-Fi) interface). The memory 1005 may be a high-speed random access memory (RAM) or a stable non-volatile memory (NVM), such as a disk storage device. The memory 1005 may also be a storage device independent of the processor 1001.

[0084] Those skilled in the art will understand that Figure 1 The structure shown in the figure does not constitute a limitation on the computer device, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0085] like Figure 1 As shown, the memory 1005 as a storage medium may include an operating system, a network communication module, a user interface module, and an industrial process operation data filling and repair program.

[0086] exist Figure 1 In the computer device shown, the network interface 1004 is mainly used for data communication with the network server; the user interface 1003 is mainly used for data interaction with the user; the processor 1001 and the memory 1005 in this application can be set in the computer device, and the computer device calls the industrial process operation data filling and repair program stored in the memory 1005 through the processor 1001, and executes the industrial process operation data filling and repair method provided in the embodiment of this application.

[0087] The present application embodiment provides a method for filling and repairing the operation data of an industrial process, referring to Figure 2 , Figure 2 This is a flow chart of the first embodiment of the method for filling and repairing operating data of an industrial process of the present application.

[0088] In this embodiment, the method for filling and repairing the operation data of an industrial process includes the following steps:

[0089] Step S10: Collecting part of the observation data in the industrial process, where the part of the observation data includes missing data.

[0090] It should be noted that in industrial processes, industrial process data modeling is crucial for the optimization and regulation of industrial production processes. The data filling and repair problems can be summarized as follows:

[0091] For any part of the observation data , whose observation index set is Ω, and the goal is to Fill in missing data In this situation, represents the orthogonal projection of the support onto Ω, and represents the orthogonal projection of the support onto the complement of Ω. The description is as follows:

[0092]

[0093] in .

[0094] Step S20: Based on the minimization of the nuclear norm of the circulant matrix, a global trend model of the data is constructed, and part of the observation data is interpolated through the nuclear norm minimization operation of the circulant matrix.

[0095] It should be noted that minimizing the nuclear norm of the circulant matrix includes: converting the part of the observation data into a circulant matrix form; calculating the nuclear norm of the circulant matrix using fast Fourier transform, and realizing global trend modeling and missing data interpolation by minimizing the nuclear norm.

[0096] In practice, industrial data has low-rank characteristics and can be completed using low-rank models. The traditional low-rank model is as follows:

[0097]

[0098] in For the repaired data, is the observation data, is the observation set. The following example can help us understand this model more intuitively:

[0099] There are some observation data as shown below:

[0100]

[0101] Then its observation set Ω is Utilizing the Model The repair matrix shown below can be obtained:

[0102]

[0103] But using the definition of convex function, we can prove that the model The optimization function is a non-convex function, and this optimization problem is an NP-hard problem. Therefore, the optimization function is converted into a convex function by using the nuclear norm of the matrix to perform convex inclusion on the optimization function. The nuclear norm of the matrix is defined as follows:

[0104]

[0105] in represents the kth largest singular value of the matrix. Therefore, the optimization problem becomes:

[0106]

[0107] Using the definition of a convex function and the Courant-Fisher theorem, it can be proved that this optimization function is a convex function. However, in order to pursue the global trend of the data, this embodiment converts the nuclear norm of the data matrix into the nuclear norm of the circulant matrix of the data. The circulant matrix is defined as follows:

[0108]

[0109] Therefore, the optimization problem becomes:

[0110]

[0111] When minimizing the nuclear norm of the circulant matrix of data, since each row and column of the circulant matrix contains complete information about the data, minimizing the nuclear norm of the circulant matrix can be used to complete data interpolation and pursue the global trend of the data. In addition, the nuclear norm of the circulant matrix can be calculated using the discrete Fourier transform as follows:

[0112]

[0113] Using discrete Fourier transform can greatly improve computational efficiency.

[0114] Step S30: Introduce a time-adaptive Laplace regularization module to dynamically adjust the local smoothing constraint weights according to the time interval of the data to capture the short-term fluctuation characteristics of some observation data.

[0115] It can be understood that the step of introducing a time-adaptive Laplace regularization module, dynamically adjusting the local smoothness constraint weight according to the time interval of the data, and capturing the short-term fluctuation characteristics of part of the observation data includes: constructing a Laplace matrix, determining the adjacent relationship between the data according to the Laplace matrix; introducing a time weighting factor to perform a weighted correction on the Laplace regularization term according to the time interval between the data points; and adjusting the smoothness constraint of the data according to the time interval through the time weighting factor, so as to capture the short-term fluctuation characteristics of part of the observation data in the Laplace matrix.

[0116] In specific implementation, the traditional Laplace time regularization idea is as follows:

[0117] First, the definition of the Laplace matrix is given:

[0118]

[0119] in is the degree matrix of the graph, is the adjacency matrix of the graph. Consider Figure 3 The Laplacian matrix is shown:

[0120] Its Laplace matrix is:

[0121]

[0122] Each column in the matrix represents the adjacency of the corresponding data. 2 represents being adjacent to two data points, and -1 represents being adjacent to the data point. The first column is: It is a special form of the Laplace kernel. Here is the definition of the Laplace kernel:

[0123]

[0124] Where τ represents the degree of the graph, reflecting how many data are connected to each data. The definition of Laplace kernel and the relationship between cyclic convolution and circulant matrix are used to give the definition of Laplace regularization:

[0125]

[0126] by Figure 3 For example, its Lx matrix is:

[0127]

[0128] The second norm is equivalent to the distance between each data point and its adjacent data points. Therefore, this regularization term can be used to pursue local trends in the data and smooth interpolated data. However, in practice, the sampling time intervals may vary, and this regularization term cannot handle this situation.

[0129] To address the problem that traditional Laplace regularization only applies to uniform time intervals (all variables are sampled at the same frequency), this paper introduces a time weighting factor to improve Laplace regularization. The basic idea is as follows:

[0130] Assume that the time interval between adjacent data points in the time series is , we design the time weighting factor to perform weighted correction on the Laplace regularization term as follows:

[0131]

[0132] in The time weighting factor is used. When the sampling interval is small, the weight is increased, forcing the adjacent points to be smoother. When the sampling interval is large, the weight is decreased, allowing for greater temporal variation. Using the time weighting factor, data with short time intervals are more strongly constrained, while data with long time intervals are allowed to fluctuate more, improving the smoothness and accuracy of the interpolation.

[0133] Step S40: Construct a graph structure regularization module with adaptive weights, learn the correlation weights between variables through the adjacency matrix and graph structure, and make the interpolation results of related variables tend to be consistent.

[0134] It should be noted that the variable similarity weight in the graph structure regularization module of the adaptive weight Defined as:

[0135]

[0136] in, is the scale factor to control the similarity, and Represents two different variables, is a normalization constant used to ensure that the sum of all weights is 1.

[0137] It is understandable that, in order to address the problem that traditional methods ignore the interrelationships between variables during interpolation and independently interpolate each series, the method proposed in this embodiment uses a graph-based variable relationship representation with adaptive weights to ensure that related variables have similar interpolation results. The specific ideas are as follows:

[0138] The regularization term uses variables It represents the similarity of variables and is defined as follows:

[0139]

[0140] in is the scale factor to control the similarity, is a normalization constant used to ensure that the sum of all weights is 1 (i.e.: ).

[0141] The weight-adaptive graph regularization term is thus defined as:

[0142]

[0143] if The larger the value of , the higher the similarity between variables i and j, and vice versa. The model can better capture the relationship between variables. For larger variable pairs, the model tends to give similar imputed values because they are closely related in the graph structure and are more likely to have similar change trends or characteristics in a practical sense. n represents the number of variables X.

[0144] Step S50: Design an adaptive weight mechanism to dynamically adjust the balance parameters between the global trend and the local trend according to the local stationarity of the data.

[0145] It is understandable that the adaptive weight mechanism is used to dynamically adjust the balance parameters γ , specifically including: calculating the variance in the local time window as a measure of local stationarity, for a given time series , in The center is In the time window of It can be calculated as:

[0146]

[0147] in, is the mean of the data in the time window;

[0148] According to the local stationary measure, adaptive weights are generated by Sigmoid function mapping:

[0149]

[0150] in, It's at the time The adaptive weight at and is an adjustable parameter used to control the shape and range of the mapping function. The range of the Sigmoid function is between (0, 1). Based on the adaptive weight, the balance parameter γ is dynamically adjusted to achieve a dynamic balance between the global and local trend weights. Assuming that the original fixed γ value is , then the dynamically adjusted γ value The calculation method is:

[0151]

[0152] in, Indicates that when more attention is paid to the overall trend The value of .

[0153] Step S60: Integrate the global trend model, the time-adaptive Laplace regularization module, the graph structure regularization module and the adaptive weight mechanism to construct a unified optimization problem.

[0154] It should be noted that the global trend model, the time-adaptive Laplace regularization module, the graph structure regularization module and the adaptive weight mechanism are integrated to construct a unified optimization model, which is expressed as:

[0155]

[0156] Among them, the global trend data is , local trend data is , the variable relationship data is , n represents the number of variables X.

[0157] Step S70: using the alternating direction multiplier method to iteratively solve the optimization problem until the convergence condition is met, and outputting the complete interpolated data.

[0158] The step of iteratively solving the optimization problem using the alternating direction multiplier method until a convergence condition is met and outputting the interpolated complete data includes:

[0159] Introducing auxiliary variables ;

[0160] , , , And the optimization problem is expressed as:

[0161]

[0162] The augmented Lagrangian function is constructed as follows:

[0163]

[0164] in is the Lagrange multiplier, is the penalty parameter;

[0165] Update X:

[0166] Fix the remaining variables and solve the following optimization problem:

[0167]

[0168] renew :

[0169] Fix the remaining variables and solve the following optimization problem:

[0170]

[0171] renew :

[0172] Fix the remaining variables and solve the following optimization problem:

[0173]

[0174] renew :

[0175] Fix the remaining variables and solve the following optimization problem:

[0176]

[0177] renew :

[0178] Fix the remaining variables and solve the following optimization problem:

[0179]

[0180] Update the Lagrange multipliers:

[0181]

[0182] The above sub-problems are updated iteratively until the convergence conditions are met, and the complete interpolated data are output.

[0183] In a specific implementation, the method further includes: classifying the risk level of the data according to the padded and repaired data, and outputting a classification result; the risk level includes high risk, medium risk and low risk, which respectively correspond to different data fluctuation ranges.

[0184] In specific implementations, this embodiment proposes a method for filling and repairing industrial process operation data that considers global and local characteristics. This method simultaneously pursues global and local trends in the data based on circulant matrix nuclear norm minimization and time-adaptive Laplace regularization. Furthermore, a graph-structured variable relationship representation (graph structure regularization) with adaptive weights and an adaptive weight mechanism are added to consider the relationships between variables and dynamically adjust balance parameters.

[0185] Circulant matrix nuclear norm minimization is used to perform data imputation and capture the global trend of the data.

[0186] Local trends in data: Using a graph-based Laplace matrix to capture the adjacent relationships between data points, we construct a Laplace regularization term. However, this only works with data with a fixed sampling frequency. Therefore, we introduce a time-adaptive Laplace regularization module to address inconsistent sampling frequencies across variables. This module modifies the Laplace regularization by introducing a time-weighting factor.

[0187] Graph-based variable relationship representation module: Utilizes graph structure and adjacency matrix to describe the similarity weights between variables, making the interpolation results of related variables tend to be similar.

[0188] Balance parameter adaptive weight mechanism module: dynamically adjust the balance parameters according to the local stability of the data γ , to achieve a dynamic balance between global and local trend weights.

[0189] Finally, a unified optimization problem is constructed and solved using the alternating direction method of multipliers (ADMM).

[0190] This implementation captures the global long-term trends of the data through minimization of the circulant matrix nuclear norm, combines time-adaptive Laplace regularization to describe the local short-term fluctuations of the data, and utilizes graph regularization with adaptive weights to characterize the correlations between variables. An adaptive weighting mechanism dynamically adjusts the balance parameters to adapt to different time series characteristics and optimize the interpolation effect. Finally, the alternating direction multiplication method is used to solve the optimization model and generate the imputed and repaired data. This method simultaneously considers the global trends, local fluctuations, and intervariable relationships of the data, significantly improving the accuracy and reliability of data imputation.

[0191] In addition, an embodiment of the present application also proposes a computer-readable storage medium, on which a program for filling and repairing the operating data of an industrial process is stored. When the program for filling and repairing the operating data of an industrial process is executed by a processor, the steps of the method for filling and repairing the operating data of an industrial process as described above are implemented.

[0192] Reference Figure 4 , Figure 4 This is a structural block diagram of the first embodiment of the industrial process operation data filling and repair system of the present application.

[0193] like Figure 4 As shown, the industrial process operation data filling and repair system proposed in the embodiment of the present application includes:

[0194] A data collection module 10 is used to collect part of the observation data of the industrial process, wherein the part of the observation data includes missing data;

[0195] A global trend module 20 is used to construct a data global trend model based on the minimization of the nuclear norm of the circulant matrix, and to interpolate part of the observation data through the minimization operation of the nuclear norm of the circulant matrix;

[0196] The short-term feature module 30 is used to introduce a time-adaptive Laplace regularization module to dynamically adjust the local smoothing constraint weight according to the time interval of the data to capture the short-term fluctuation characteristics of some observation data;

[0197] Regularization module 40 is used to construct a graph structure regularization module with adaptive weights, which learns the correlation weights between variables through the adjacency matrix and graph structure, so that the interpolation results of related variables tend to be consistent;

[0198] Adaptive weight module 50, used to design an adaptive weight mechanism to dynamically adjust the balance parameters between global and local trends according to the local stationarity of the data;

[0199] An optimization problem module 60 is used to integrate the global trend model, the time-adaptive Laplace regularization module, the graph structure regularization module, and the adaptive weight mechanism to construct a unified optimization problem;

[0200] The output module 70 is used to iteratively solve the optimization problem using the alternating direction multiplier method until the convergence condition is met, and output the complete interpolated data.

[0201] It should be understood that the above is only an example and does not constitute any limitation to the technical solution of the present application. In specific applications, technicians in this field can make settings as needed, and the present application does not impose any restrictions on this.

[0202] This implementation captures the global long-term trends of the data through minimization of the circulant matrix nuclear norm, combines time-adaptive Laplace regularization to describe the local short-term fluctuations of the data, and utilizes graph regularization with adaptive weights to characterize the correlations between variables. An adaptive weighting mechanism dynamically adjusts the balance parameters to adapt to different time series characteristics and optimize the interpolation effect. Finally, the alternating direction multiplication method is used to solve the optimization model and generate the imputed and repaired data. This method simultaneously considers the global trends, local fluctuations, and intervariable relationships of the data, significantly improving the accuracy and reliability of data imputation.

[0203] It should be noted that the workflow described above is merely illustrative and does not limit the scope of protection of this application. In actual applications, technicians in this field can select part or all of it according to actual needs to achieve the purpose of this embodiment scheme, and no restrictions are imposed here.

[0204] In addition, for technical details not fully described in this embodiment, please refer to the method for filling and repairing the operating data of the industrial process provided in any embodiment of the present application, and will not be repeated here.

[0205] In addition, it should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or system comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or system. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or system comprising the element.

[0206] The serial numbers of the above embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0207] Through the above description of the embodiments, those skilled in the art will clearly understand that the above-mentioned embodiments and methods can be implemented using software plus the necessary general-purpose hardware platform. Of course, hardware can also be used, but in many cases the former is a more preferred embodiment. Based on this understanding, the technical solution of this application, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as read-only memory (ROM) / RAM, a magnetic disk, or an optical disk) and includes several instructions for enabling a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods of each embodiment of this application. The above are only preferred embodiments of this application and do not limit the scope of the patent application. Any equivalent structure or equivalent process transformation made using the contents of this application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the scope of patent protection of this application.

Claims

1. A method for filling and repairing operation data of an industrial process, characterized in that: include: Collecting partial observation data of an industrial process, wherein the partial observation data includes missing data; Based on the minimization of the nuclear norm of the circulant matrix, a global trend model of the data is constructed, and some observation data are interpolated through the minimization operation of the nuclear norm of the circulant matrix; A time-adaptive Laplace regularization module is introduced to dynamically adjust the local smoothing constraint weights according to the time interval of the data to capture the short-term fluctuation characteristics of some observation data; Construct an adaptive weighted graph regularization module, which learns the correlation weights between variables through the adjacency matrix and graph structure, so that the interpolation results of related variables tend to be consistent; Design an adaptive weighting mechanism to dynamically adjust the balance parameters between global and local trends based on the local stationarity of the data; Integrate the global trend model, the time-adaptive Laplace regularization module, the graph structure regularization module, and the adaptive weight mechanism to construct a unified optimization problem; The optimization problem is iteratively solved by using an alternating direction multiplier method until a convergence condition is met, and the complete interpolated data is output; Among them, the adaptive weight mechanism is used to dynamically adjust the balance parameters γ , specifically including: Calculate the variance in a local time window as a measure of local stationarity for a given time series , in The center is In the time window of It can be calculated as: in, is the mean of the data in the time window; According to the local stationary measure, adaptive weights are generated by Sigmoid function mapping: in, It's at the time The adaptive weight at and It is an adjustable parameter used to control the shape and range of the mapping function. The range of the Sigmoid function is between (0, 1); Based on the adaptive weight, the balance parameter γ is dynamically adjusted to achieve a dynamic balance between global and local trend weights. Assuming that the original fixed γ value is , then the dynamically adjusted γ value The calculation method is: in, Indicates that when more attention is paid to the overall trend The value of The steps of integrating the global trend model, the time-adaptive Laplace regularization module, the graph structure regularization module, and the adaptive weight mechanism to construct a unified optimization problem include: The global trend model, the time-adaptive Laplace regularization module, the graph structure regularization module, and the adaptive weight mechanism are integrated to construct a unified optimization model, which is expressed as: Among them, the global trend data is , local trend data is , the variable relationship data is , n represents the number of variables x.

2. The method for filling and repairing the operation data of an industrial process according to claim 1, characterized in that: The circulant matrix nuclear norm minimization includes: Converting the portion of observation data into a circulant matrix form; Fast Fourier transform is used to calculate the nuclear norm of the circulant matrix, and global trend modeling and missing data interpolation are achieved by minimizing the nuclear norm.

3. The method for filling and repairing the operation data of an industrial process according to claim 1, characterized in that: The step of introducing a time-adaptive Laplace regularization module to dynamically adjust the local smoothing constraint weight according to the time interval of the data to capture the short-term fluctuation characteristics of some observation data includes: Constructing a Laplace matrix, and determining the adjacent relationship between data according to the Laplace matrix; A time weighting factor is introduced to perform weighted correction on the Laplace regularization term according to the time interval between data points; The time weighting factor is used to adjust the smoothness constraint of the data according to the time interval, and the short-term fluctuation characteristics of part of the observation data are captured in the Laplace matrix.

4. The method for filling and repairing the operation data of an industrial process according to claim 1, characterized in that: The variable similarity weights in the adaptive weighted graph structure regularization module Defined as: in, is the scale factor to control the similarity, and Represents two different variables, is a normalization constant used to ensure that the sum of all weights is 1.

5. The method for filling and repairing the operation data of an industrial process according to claim 1, characterized in that: The step of iteratively solving the optimization problem using the alternating direction multiplier method until a convergence condition is met and outputting the interpolated complete data includes: Introducing auxiliary variables ; , , , And the optimization problem is expressed as: The augmented Lagrangian function is constructed as follows: in is the Lagrange multiplier, is the penalty parameter; Update X: Fix the remaining variables and solve the following optimization problem: renew : Fix the remaining variables and solve the following optimization problem: renew : Fix the remaining variables and solve the following optimization problem: renew : Fix the remaining variables and solve the following optimization problem: renew : Fix the remaining variables and solve the following optimization problem: Update the Lagrange multipliers: The above optimization problem is iteratively updated until the convergence condition is met, and the complete interpolated data is output.

6. A system for filling and repairing operation data of an industrial process, characterized in that: Executing the method according to claim 1, comprising: A data collection module, configured to collect partial observation data of an industrial process, wherein the partial observation data includes missing data; The global trend module is used to build a data global trend model based on the minimization of the nuclear norm of the circulant matrix and interpolate some observation data through the minimization operation of the nuclear norm of the circulant matrix; The short-term feature module is used to introduce a time-adaptive Laplace regularization module, dynamically adjust the local smoothing constraint weights according to the time interval of the data, and capture the short-term fluctuation characteristics of some observation data; Regularization module, used to build a graph structure regularization module with adaptive weights. It learns the correlation weights between variables through the adjacency matrix and graph structure, so that the interpolation results of related variables tend to be consistent; Adaptive weight module, used to design an adaptive weight mechanism to dynamically adjust the balance parameters between global and local trends according to the local stationarity of the data; An optimization problem module, which is used to integrate the global trend model, the time-adaptive Laplace regularization module, the graph structure regularization module and the adaptive weight mechanism to construct a unified optimization problem; The output module is used to iteratively solve the optimization problem by using the alternating direction multiplier method until the convergence condition is met, and output the complete interpolated data.

7. A computer device, characterized in that: The device comprises: a memory and a processor, wherein the processor executes the method according to any one of claims 1 to 5 when running computer instructions stored in the memory.

8. A computer-readable storage medium, characterized in that The method comprises instructions which, when executed on a computer, cause the computer to execute the method according to any one of claims 1 to 5.

Images