Data processing method for heat treatment process

Abstract

The invention relates to the technical field of data processing, in particular to a data processing method for a heat treatment process, which comprises the following steps of: respectively acquiring various process data of the heat treatment process, and acquiring final components of each process data by utilizing an independent component analysis method so as to realize data processing for the heat treatment process; in each iteration process of the independent component analysis method, in response to the fact that the change characteristics of all the components are smaller than a preset threshold value, iteration is terminated, and all the components at the moment serve as final components; comprising the following steps: respectively calculating the stability characteristic of any data point in any component, the stability of any component, the curve similarity of two components, the separation degree of the two components and the change characteristics of all components; whether iteration is stopped or not is dynamically judged according to the change characteristics of the components, premature stopping or excessive iteration caused by fixed iteration times is avoided, and computing resources are saved.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and more specifically, to a data processing method for a heat treatment process. Background Art

[0002] The heat treatment process involves multiple parameters and steps, such as heating, insulation, cooling, etc. The parameter selection of each step directly affects the performance and quality of the final product. Due to the complexity of the heat treatment process, traditional data processing methods often find it difficult to fully and accurately capture and analyze key information in the process. During the heat treatment process, sensors collect a large amount of data, which often contains redundant information and noise, interfering with the accuracy and reliability of the data.

[0003] Independent Component Analysis (ICA) is an algorithm widely used in signal processing, image processing and other fields. Its core lies in recovering independent potential components from the observed multivariate signals. This feature gives ICA a significant advantage in processing mixed signals, and it can separate independent components directly related to the thermal treatment process, thereby removing redundancy and noise.

[0004] The patent document with the announcement number CN118094210B discloses a method for identifying the charging and discharging behavior of an energy storage system based on underdetermined blind source separation, which relates to the technical field of energy storage charging and discharging behavior identification and solves the problem of low perception level of energy storage demand response capability in the prior art. The method includes: obtaining the total power signal sequence of the energy storage system, using the ensemble empirical mode decomposition algorithm to perform signal dimension increase on the total power signal sequence, and obtaining an intrinsic mode function matrix; using the principal component analysis algorithm to analyze the intrinsic mode function matrix, estimate the number of independent energy storage signals, and obtain a non-underdetermined dimension increase signal matrix corresponding to the total power signal sequence; analyzing the independence between each non-underdetermined dimension increase signal in the non-underdetermined dimension increase signal matrix, and obtaining the corresponding independent sub-signal group matrix; performing blind source signal separation on the independent sub-signal group matrix based on the fast independent component analysis method, and obtaining the charging and discharging behavior curves of each independent energy storage in the energy storage system.

[0005] However, the above patent document does not solve the problem that the independent component analysis method needs to preset the number of iterations during use. The complexity of the heat treatment process and the variability of the environment make the heat treatment data more complex. The preset number of iterations cannot be adjusted according to the complexity of the data and the convergence speed. This may cause the algorithm to stop prematurely before achieving optimal performance, or continue to iterate after convergence, resulting in a waste of resources. Summary of the invention

[0006] In order to solve the problem that the independent component analysis method needs to preset the number of iterations during use, and the complexity of the heat treatment process and the variability of the environment make the heat treatment data more complicated, and the preset number of iterations cannot be adjusted according to the complexity of the data and the convergence speed, which may cause the algorithm to stop prematurely before reaching the optimal performance, or continue to iterate after convergence, resulting in a waste of resources, the present invention proposes a data processing method for a heat treatment process, the method comprising the following steps: The process data of each heat treatment process are obtained respectively, and the final components of each process data are obtained by using the independent component analysis method to realize data processing for the heat treatment process; in each round of iteration of the independent component analysis method, in response to the change characteristics of all components being less than a preset threshold, the components of the process data are determined to be stable, the iteration is terminated, and the components at this time are used as the final components, otherwise, the next round of iteration is entered, including: recording any component separated by the independent component analysis method as a target component; calculating the stable characteristics of any data point in the target component based on the Euclidean distance between any data point in the target component and two data points adjacent to the data point; clustering is performed according to the stable characteristics of each data point in the target component, and the stability of the target component is calculated based on the number of clusters and the shape and size of the clusters in the clustering results; curve fitting is performed on the data points in all components to obtain the fitting curve of each component, and the curve similarity of the two components is calculated based on the area of ​​the figure enclosed by the fitting curves of any two components; the separation degree of the two components is calculated based on the curve similarity and DTW similarity of the two components; based on the separation degree and stability, the change characteristics of all components are calculated.

[0007] By dynamically determining whether to stop iteration based on the changing characteristics of the components, premature stopping or excessive iteration caused by a fixed number of iterations is avoided, saving computing resources. By evaluating the stability of multiple dimensions and the degree of separation of the calculated components, the stability of the data can be accurately determined to ensure the efficient convergence of the algorithm, reduce unnecessary calculations, improve algorithm efficiency, and reduce energy consumption and computing resource consumption.

[0008] Furthermore, the stability characteristic satisfies the following relationship: ; In the formula, For the During the iteration process The first component The stable characteristics of the data points, For the During the iteration process The first component The Euclidean distance between a data point and its two adjacent data points is For the During the iteration process The first component The value of the data point, For the During the iteration process The first component The value of the data point, For the During the iteration process The first component The value of the data point.

[0009] Furthermore, the shape and size of the cluster satisfy the following relationship: ; In the formula, For the During the iteration process The clustering results of the components The shape and size of the clusters, For the During the iteration process The clustering results of the components The maximum value of the data points in the clusters, For the During the iteration process The clustering results of the components The minimum value of the data points in the clusters, For the During the iteration process The clustering results of the components The number of data points in a cluster.

[0010] Furthermore, the stability satisfies the following relationship: ; In the formula, For the During the iteration process The stability of the components, For the During the iteration process The clustering results of the components The first-order difference value of the shape and size of the clusters, For the During the iteration process The number of clusters in the clustering results of the components, is a natural exponential function.

[0011] By introducing the first-order difference value of the cluster shape and size and processing it with the natural exponential function, the stationarity of each component in the iteration process can be measured more accurately, which can effectively capture the slight changes in the data and improve the accuracy of the stationarity assessment.

[0012] Furthermore, the curve fitting is performed by least square method.

[0013] Furthermore, the curve similarity satisfies the following relationship: ; In the formula, For the During the iteration process The components and The similarity of the curves of the components, For the During the iteration process The components and The area of ​​the figure enclosed by the fitting curves of the components, For the During the iteration process The fitting curve of the components, For the During the iteration process The fitting curve of the components, is the initial collection time of the process data of the component item, The last collection time of the process data of the component. is the absolute value symbol.

[0014] By calculating the area difference between the fitting curves, the similarity between different components can be measured more accurately, avoiding the limitations of simple numerical comparison and effectively capturing the subtle differences between components.

[0015] Furthermore, the separation degree satisfies the following relationship: ; In the formula, For the During the iteration process The components and The degree of separation of the components For the During the iteration process The components and The similarity of the curves of the components, For the During the iteration process The components and DTW similarity of the components.

[0016] By quantifying the degree of separation, the differences between different components can be measured more accurately; by combining curve similarity and DTW similarity, the similarities between components can be effectively distinguished, thereby effectively dealing with possible noise and fluctuations in the data and improving the stability and robustness of the algorithm when processing complex data.

[0017] Furthermore, the change characteristics satisfy the following relationship: ; In the formula, for The changing characteristics of all components during the round iteration, For the During the iteration process The components and The degree of separation of the components For the During the iteration process The stability of the components, For the During the iteration process The stability of the components, For the The number of all components during the round iteration, is a natural exponential function.

[0018] By combining the degree of separation and the stability of the components, the interaction and separation effect between each pair of components can be accurately measured. In the process of multiple rounds of iterations, the change characteristics of the components are gradually optimized, making the final separation effect more accurate and suitable for complex multi-component systems. By introducing the stability of the components as a regulating factor, the deviation caused by instability or noise in the separation process can be avoided.

[0019] Furthermore, the data processing for the heat treatment process includes: taking the mean of the DTW similarities between the target component and the remaining components as the similarity of the target component, recording the component corresponding to the minimum value in the similarity as a noise component and removing it, thereby completing the data processing for the heat treatment process.

[0020] The present invention has the following beneficial effects: In the independent component analysis method, the preset number of iterations cannot be adjusted according to the complexity of the data and the convergence speed, resulting in the algorithm stopping prematurely before reaching the optimal performance, or continuing to iterate after convergence, resulting in a waste of resources. By obtaining and setting appropriate convergence conditions, it is possible to ensure that the algorithm stops in time at the optimal performance, avoid inefficient iterations caused by the preset number of times, ensure that the best results are obtained, and thus improve the denoising effect; at the same time, the present invention can cope with the complex data and changeable environment in the heat treatment process, and improve the accuracy and reliability of data processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein: Figure 1 It is a step flow chart of a data processing method for a heat treatment process according to an embodiment of the present invention.

[0022] Figure 2 It is a flowchart of the steps in each round of iteration of the independent component analysis method of a data processing method for a heat treatment process according to an embodiment of the present invention. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0024] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0025] See also Figure 1 , which shows a flow chart of a data processing method for a heat treatment process provided by an embodiment of the present invention, the method comprising the following steps: S1: Obtain various process data of the heat treatment process respectively.

[0026] By installing various types of sensors to collect data in the heat treatment process, such as temperature, time, toughness, etc., different collection frequencies are set for different types of data processed. Taking temperature data as an example for analysis, implementers can set the data collection frequency according to the specific implementation situation, for example, 1Hz.

[0027] S2: Use independent component analysis to obtain the final components of each process data.

[0028] It should be noted that due to the complexity of the heat treatment process and the variability of the environment, the heat treatment data is relatively complex. When the independent component analysis is separated by iteration, the preset number of iterations cannot be adjusted according to the complexity of the data and the convergence speed. This may cause the algorithm to stop prematurely before reaching the optimal performance, or continue to iterate after convergence, which will lead to poor separation of the final components, affect the removal of noise components, and lead to unsatisfactory denoising effect of heat treatment data. The change characteristics of the components obtained during the iteration process can directly reflect the convergence of the algorithm. When the algorithm gradually converges to the optimal solution, the change amount of the obtained components will tend to stabilize or gradually decrease. By monitoring the change characteristics of the obtained components, the iterative process can be accurately controlled; when it is found that the change amount of the obtained components reaches the preset threshold, the iteration can be terminated in time, thereby ensuring the accuracy and reliability of the results, and helping to avoid errors and uncertainties caused by insufficient or excessive iterations. Because the independent component analysis method divides the data into multiple independent components when processing one-dimensional data, as the iteration proceeds, the multiple components will gradually tend to be stable, which means that the algorithm has converged and further iterations will no longer bring significant changes; at the same time, as the iteration proceeds, the degree of separation between the multiple components will gradually increase, which means that the model has effectively extracted the differences in various features, and further iterations may not lead to further improvements. Therefore, when analyzing the changing characteristics of each component during the iteration process, the stability of the components and the changing characteristics between the components can be considered. In summary, the present invention analyzes the changing characteristics of the components from the stability of the components during the iteration process and the changing characteristics between the components, and uses it as a basis for judging the termination of the iteration, thereby obtaining the noise component to achieve denoising of the heat treatment data.

[0029] During each iteration of the independent component analysis, see Figure 2 In response to the change characteristics of all components being less than the preset threshold, the components of the process data are deemed stable, the iteration is terminated, and the components at this time are taken as the final components. Otherwise, the next round of iteration is entered, including: Implementers can set the threshold according to specific implementation circumstances, for example, 0.5.

[0030] S201: Calculate the stable characteristics of any data point in any component.

[0031] Any component separated by independent component analysis is recorded as the target component; based on the Euclidean distance between any data point in the target component and two adjacent data points, the stable characteristics of any data point in the target component are calculated.

[0032] Specifically, the stability feature satisfies the following relationship: ; In the formula, For the During the iteration process The first component The stable characteristics of the data points, For the During the iteration process The first component The Euclidean distance between a data point and its two adjacent data points is For the During the iteration process The first component The value of the data point, For the During the iteration process The first component The value of the data point, For the During the iteration process The first component The value of the data point.

[0033] in, Indicates During the iteration process The first component The Euclidean distance between a data point and its two adjacent data points. The smaller the distance, the higher the stability feature of the data.

[0034] S202: Calculate the stationarity of any component.

[0035] It should be noted that when obtaining the stationarity of each component, it can be reflected through the clustering results of the data. However, in many practical applications, the original data is often affected by factors such as sensor errors, resulting in noise in the data. These noises will interfere with the effect of the clustering algorithm and make the clustering results inaccurate. The stationary features are usually calculated based on the change trend between the data, the distance between adjacent data points, etc., and can reflect the inherent stability of the data. Clustering stationary features can help remove unstable data points and noise, and focus on the actual trend and stability of the data, so that the clustering effect will be more accurate and reliable.

[0036] Clustering is performed based on the stable characteristics of each data point in the target component, and the stationarity of the target component is calculated based on the number of clusters and the shape and size of the clusters in the clustering results.

[0037] Specifically, the shape and size of the cluster satisfy the following relationship: ; In the formula, For the During the iteration process The clustering results of the components The shape and size of the clusters, For the During the iteration process The clustering results of the components The maximum value of the data points in the clusters, For the During the iteration process The clustering results of the components The minimum value of the data points in the clusters, For the During the iteration process The clustering results of the components The number of data points in a cluster.

[0038] in, Indicates During the iteration process The clustering results of the components Each cluster corresponds to a data range. The smaller the data range is and the more data there are in the cluster, the denser the data is in the cluster and the smaller the shape of the cluster is. In the clustering process here, the clustering algorithm is not constrained and the clustering result can be obtained, for example, the DBSCAN clustering algorithm.

[0039] Specifically, the stability satisfies the following relationship: ; In the formula, For the During the iteration process The stability of the components, For the During the iteration process The clustering results of the components The first-order difference value of the shape and size of the clusters, For the During the iteration process The number of clusters in the clustering results of the components, is a natural exponential function.

[0040] in, Indicates During the iteration process The mean of the first-order differences of cluster shapes and sizes in the clustering results of all data stability features of a component. The smaller the value, the higher the consistency of the data stability features in the component and the higher the stability of the component.

[0041] S203: Calculate the curve similarity of the two components.

[0042] It should be noted that when obtaining the change characteristics between any two components, combining the correlation of the corresponding data of the two components and the similarity of the corresponding curves can more comprehensively evaluate the change characteristics between the two components, thus avoiding the one-sidedness that may be caused by a single indicator. For example, even if the two components have low similarity in data, if there is a strong correlation between them, this difference is not enough to indicate that the change characteristics between them are high.

[0043] Curve fitting is performed on the data points in all components to obtain the fitting curve of each component. Based on the area of ​​the figure enclosed by the fitting curves of any two components, the curve similarity of the two components is calculated.

[0044] Specifically, the curve fitting is performed by least square method.

[0045] Specifically, the curve similarity satisfies the following relationship: ; In the formula, For the During the iteration process The components and The similarity of the curves of the components, For the During the iteration process The components and The area of ​​the figure enclosed by the fitting curves of the components, For the During the iteration process The fitting curve of the components, For the During the iteration process The fitting curve of the components, is the initial collection time of the process data of the component item, The last collection time of the process data of the component. is the absolute value symbol.

[0046] in, Indicates During the iteration process The components and The smaller the area of ​​the figure enclosed by the fitting curves of the components, the higher the similarity of the two curves.

[0047] S204: Calculate the separation degree of the two components.

[0048] It should be noted that, after obtaining the similarity of the corresponding data of the two components, and using the similarity of the corresponding data of the two components and the similarity of the corresponding curves as indicators for calculating the degree of separation between the two components, the lower the similarity of the corresponding data of the two components and the smaller the similarity of the corresponding curves, the higher the degree of separation of the two components.

[0049] Based on the curve similarity and DTW similarity of any two components, the degree of separation of the two components is calculated.

[0050] Specifically, the separation degree satisfies the following relationship: ; In the formula, For the During the iteration process The components and The degree of separation of the components For the During the iteration process The components and The similarity of the curves of the components, For the During the iteration process The components and DTW similarity of the components.

[0051] S205: Calculate the variation characteristics of all components.

[0052] It should be noted that the higher the stability of the two components and the higher the degree of separation between the two components, the higher the stability of each component in this round of iteration, and the smaller the change characteristics of the components obtained in the iteration process.

[0053] Based on the degree of separation and stationarity, the variation characteristics of all components are calculated.

[0054] Specifically, the change characteristics satisfy the following relationship: ; In the formula, for The changing characteristics of all components during the round iteration, For the During the iteration process The components and The degree of separation of the components For the During the iteration process The stability of the components, For the During the iteration process The stability of the components, For the The number of all components during the round iteration, is a natural exponential function.

[0055] S3: To realize data processing for heat treatment process.

[0056] Specifically, the data processing for the heat treatment process includes: The average of the DTW similarities between the target component and the remaining components is taken as the similarity of the target component, and the component corresponding to the minimum value in the similarity is recorded as a noise component and removed, thereby completing the data processing for the heat treatment process.

[0057] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A data processing method for a heat treatment process, characterized in that: include: Acquire various process data of the heat treatment process respectively, and use the independent component analysis method to obtain the final components of each process data to realize data processing for the heat treatment process; In each round of iteration of the independent component analysis method, in response to the change characteristics of all components being less than the preset threshold, the components of the process data are considered stable, the iteration is terminated, and the components at this time are taken as the final components. Otherwise, the next round of iteration is entered, including: Any component separated by independent component analysis is recorded as the target component; based on the Euclidean distance between any data point in the target component and two adjacent data points, the stability characteristics of any data point in the target component are calculated; clustering is performed based on the stability characteristics of each data point in the target component, and the stability of the target component is calculated based on the number of clusters and the shape and size of the clusters in the clustering results; Perform curve fitting on the data points in all components to obtain the fitting curve of each component. Calculate the curve similarity of any two components based on the area of ​​the figure enclosed by the fitting curves of the two components. Calculate the degree of separation of the two components based on the curve similarity and DTW similarity of any two components. Calculate the change characteristics of all components based on the separation degree and stationarity.

2. A data processing method for heat treatment process according to claim 1, characterized in that: The stability characteristics satisfy the following relationship: ; In the formula, For the During the iteration process The first component The stable characteristics of the data points, For the During the iteration process The first component The Euclidean distance between a data point and its two adjacent data points is For the During the iteration process The first component The value of the data point, For the During the iteration process The first component The value of the data point, For the During the iteration process The first component The value of the data point.

3. A data processing method for heat treatment process according to claim 1, characterized in that: The cluster shape and size satisfy the following relationship: ； In the formula, For the During the iteration process The clustering results of the components The shape and size of the clusters, For the During the iteration process The clustering results of the components The maximum value of the data points in the clusters, For the During the iteration process The clustering results of the components The minimum value of the data points in the clusters, For the During the iteration process The clustering results of the components The number of data points in a cluster.

4. A data processing method for heat treatment process according to claim 1, characterized in that: The stability satisfies the following relationship: ； In the formula, For the During the iteration process The stability of the components, For the During the iteration process The clustering results of the components The first-order difference value of the shape and size of the clusters, For the During the iteration process The number of clusters in the clustering results of the components, is a natural exponential function.

5. A data processing method for heat treatment process according to claim 1, characterized in that: The curve fitting is performed by the least square method.

6. A data processing method for heat treatment process according to claim 1, characterized in that: The curve similarity satisfies the following relationship: ； In the formula, For the During the iteration The components and The similarity of the curves of the components, For the During the iteration The components and The area of ​​the figure enclosed by the fitting curves of the components, For the During the iteration process The fitting curve of the components, For the During the iteration The fitting curve of the components, is the initial collection time of the process data of the component item, The last collection time of the process data of the component. is the absolute value symbol.

7. A data processing method for heat treatment process according to claim 1, characterized in that: The degree of separation satisfies the following relationship: ; In the formula, For the During the iteration The components and The degree of separation of the components For the During the iteration The components and The similarity of the curves of the components, For the During the iteration The components and DTW similarity of the components.

8. The data processing method for heat treatment process according to claim 1, characterized in that: The change characteristics satisfy the following relationship: ； In the formula, for The changing characteristics of all components during the round iteration, For the During the iteration process The components and The degree of separation of the components For the During the iteration process The stability of the components, For the During the iteration process The stability of the components, For the The number of all components during the round iteration, is a natural exponential function.

9. A data processing method for heat treatment process according to claim 1, characterized in that: The method for realizing data processing for heat treatment process comprises: The average of the DTW similarities between the target component and the remaining components is taken as the similarity of the target component, and the component corresponding to the minimum value in the similarity is recorded as a noise component and removed, thereby completing the data processing for the heat treatment process.

Images