CN-122024974-A - Intelligent component optimization method and system for anti-condensation insulating paint

CN122024974ACN 122024974 ACN122024974 ACN 122024974ACN-122024974-A

Abstract

The invention discloses an intelligent component optimization method and system of an anti-condensation insulating coating, and relates to the technical field of data processing, wherein the method comprises the steps of establishing a component database of the anti-condensation insulating coating, determining the initial range of components and establishing a component-performance prediction model; the method comprises the steps of taking target performance parameters analyzed by an application scene as targets, optimizing an initial range of a component through a component-performance prediction model to obtain a predicted theoretical component, carrying out reliability deviation peak value analysis according to the target performance parameters and the components of the predicted theoretical component and the initial component, positioning a test verification component, carrying out test verification on the basis of the test verification component, and determining a final optimized target coating component. The invention solves the technical problems that the optimization of the anti-condensation insulating paint formula in the prior art depends on manual experience and is difficult to realize the collaborative optimization of multiple performance indexes, and achieves the technical effects of realizing the intelligent collaborative optimization of multiple target performances and improving the efficiency and accuracy of the formula optimization.

Inventors

ZHANG CHUANCHI
TENG SONG
LI JIAQI
LU SHIYONG
LI XINGCHEN
Wang rizhao

Assignees

国网江苏省电力有限公司徐州供电分公司

Dates

Publication Date: 20260512
Application Date: 20260415

Claims (10)

1. The intelligent component optimization method of the anti-condensation insulating paint is characterized by comprising the following steps of: Collecting physical and chemical parameter data of matrix resin, modified filler, functional auxiliary agent and curing system and historical formula-performance associated data, and establishing a component database of the anti-condensation insulating coating; determining an initial range of components based on the component database and establishing a component-performance prediction model; optimizing the initial range of the component by using the component-performance prediction model with the target performance parameter analyzed by the application scene as a target to obtain a predicted theoretical component; Performing reliability deviation peak analysis according to the target performance parameters and the components of the predicted theoretical components and the initial components, and positioning and testing the verified components; And performing test verification based on the test verification component to determine a final optimized target paint component.
2. The method for intelligently optimizing components of the anti-condensation insulating paint according to claim 1, wherein the step of establishing a component database of the anti-condensation insulating paint comprises the following steps: According to the existing research test data set, analyzing the influence degree of each component on the performance of the coating and the performance fluctuation condition of the components in actual production, and classifying the components, wherein the component types comprise stable key components with large performance influence and small fluctuation, suspicious key components with large performance influence and large fluctuation, stable auxiliary components with small performance influence and small fluctuation and suspicious auxiliary components with small performance influence and large fluctuation; And mapping and correlating the component types with target performance parameters according to the influence relation between the component types and the performance parameters, and establishing a component database of the anti-condensation insulating coating.
3. The method for intelligently optimizing components of an anti-condensation insulating paint according to claim 2, wherein determining initial ranges of components based on the component database comprises: Acquiring statistical distribution of the content of each component in the existing formula stored in the component database, and determining a feasible interval of each component by combining physical and chemical constraint conditions and expert experience; based on the feasible interval of each component, screening the performance fixed value or interval of each component by taking the basic performance of the anti-condensation insulating paint as a reference to obtain the fixed value or interval of each component; The initial ranges of the components are determined according to fixed values or intervals of the components.
4. The method for intelligently optimizing components of the anti-condensation insulating paint according to claim 2, wherein the step of establishing a component-performance prediction model comprises the following steps: taking historical formula data and historical formula-performance associated data in the component database as training samples, taking component content as input characteristics, taking target performance parameters as output labels, and training an initial prediction model by adopting a machine learning algorithm; and evaluating the prediction accuracy of the initial prediction model by using a cross verification method, analyzing a data sparse area, supplementing experimental points in the data sparse area, expanding a training sample by using the supplemented acquisition experimental data, and retraining until the prediction accuracy meets the preset standard, so as to obtain the component-performance prediction model.
5. The method for intelligently optimizing components of the anti-condensation insulating paint according to claim 4, wherein the component initial range is optimized by the component-performance prediction model with the objective performance parameter of application scene analysis as an objective, and a predicted theoretical component is obtained, comprising: Resolving condensation-preventing insulation requirements according to application scenes, and determining scene requirement performance; performing demand differential analysis on the scene demand performance by taking the basic performance of the anti-condensation insulating paint as a starting point, and determining the target performance parameter; And taking the target performance parameter as input, and carrying out component performance response analysis through the component-performance prediction model to obtain a predicted theoretical component.
6. The method for intelligently optimizing components of an anti-condensation insulating coating according to claim 5, wherein the target performance parameter is used as input, the component performance response analysis is performed through the component-performance prediction model, and a predicted theoretical component is obtained, and the method comprises the following steps: Generating a candidate recipe set based on the initial range of components, the candidate recipe set covering a fixed value of the stabilizing key component, a full range variation of the suspected key component, a fixed value of the stabilizing auxiliary component, and a narrow range variation of the suspected auxiliary component; Inputting the candidate formula set into the component-performance prediction model, and calculating multi-target performance predicted values of each candidate formula, wherein the multi-target performance predicted values comprise an anti-condensation performance index, an electrical insulation performance index, a durability performance index and a process environmental protection performance index; according to the priority weight in the target performance parameter set, carrying out weighted calculation on the multi-target performance predicted value to obtain the comprehensive performance score of each candidate formula; Screening candidate formulas with comprehensive performance scores higher than a preset threshold as a preferred candidate group, performing pareto front analysis on each candidate formula in the preferred candidate group, and identifying a non-dominant solution set without disadvantages on multiple performance indexes; Selecting a candidate formula with highest matching degree with the target performance parameter set from the non-dominant solution set as a first prediction theoretical component, selecting a candidate formula with next highest matching degree and remarkable component constitution difference as a second prediction theoretical component, and selecting a candidate formula with prominent single performance at the pareto front boundary as a third prediction theoretical component to jointly form a prediction theoretical component set; and outputting the prediction theory component set, wherein the prediction theory component set comprises the detailed proportion of each prediction theory component, a prediction performance index, a confidence level evaluation and a technological condition suggestion required for realizing the prediction performance.
7. The intelligent component optimization method of the anti-condensation insulating paint according to claim 6, wherein the reliability deviation peak analysis is performed according to the target performance parameter in combination with the components of the predicted theoretical component and the initial component, and the positioning test verification component comprises: obtaining a deviation component based on the predicted theoretical component and the initial component, and positioning a search key evolution link of a suspicious component type for the deviation component according to a search determination process of the predicted theoretical component; analyzing the test response relation according to the search key evolution link, and determining a test verification component interval; And positioning the test verification components according to the test verification component intervals and the suspicious components.
8. The method for intelligently optimizing components of an anti-condensation insulating paint according to claim 7, wherein locating a search key evolution link of a suspected component type for the deviation component according to a search determination process of a predicted theoretical component comprises: extracting iteration track data of an optimization algorithm when searching an optimal solution according to the searching and determining process of the predicted theoretical components, wherein the iteration track data comprises component content changes, comprehensive performance score changes and searching direction adjustment records of candidate solutions in each round of iteration; And for each deviation component, tracing the content evolution path in the iterative track, marking the content change rate mutation point, the search direction reverse point and the multi-round iterative stagnation point, and obtaining the search key evolution link of the deviation component.
9. The intelligent component optimization method of the anti-condensation insulating paint according to claim 8, wherein the analyzing the test response relation according to the search key evolution link to determine the test verification component interval comprises the following steps: identifying speed mutation points and direction reversal points which are positioned in a high-sensitivity core area in a search key evolution link, expanding the speed mutation points and the direction reversal points to two sides by taking the key points as centers, and establishing a complete high-sensitivity section; obtaining strong causal relation between the component content in the high-sensitivity section and the target performance parameter, and identifying nonlinear characteristics of performance response in the section, wherein the nonlinear characteristics comprise a mutation threshold value, an extreme point or a sensitive window; And taking a speed mutation point and a direction reversal point in the high-sensitivity section as anchor points, and combining the nonlinear characteristics of the performance response in the section to locate and test the verification component section.
10. A component intelligent optimization system of an anti-condensation insulating paint, characterized in that the system is used for executing a component intelligent optimization method of an anti-condensation insulating paint according to any one of claims 1 to 9, the system comprising: The database building module is used for collecting physical and chemical parameter data of matrix resin, modified filler, functional auxiliary agent and curing system and historical formula-performance associated data and building a component database of the anti-condensation insulating coating; the model building module is used for determining the initial range of the components based on the component database and building a component-performance prediction model; the optimizing module is used for optimizing the initial range of the component by taking the target performance parameter analyzed by the application scene as a target and obtaining a predicted theoretical component through the component-performance prediction model; The analysis module is used for carrying out reliability deviation peak analysis by combining the components of the predicted theoretical component and the initial component according to the target performance parameter, and positioning a test verification component; and the test verification module is used for carrying out test verification on the basis of the test verification component and determining a final optimized target paint component.

Description

Intelligent component optimization method and system for anti-condensation insulating paint Technical Field The invention relates to the technical field of data processing, in particular to an intelligent component optimization method and system of an anti-condensation insulating coating. Background In the research and development process of the anti-condensation insulating coating, the formula design generally depends on technicians to adjust component proportion and repeatedly test and prepare according to experience, and complex coupling relation exists among different components, so that electrical insulation performance, durability or construction performance are often influenced when the anti-condensation performance is improved, and multiple performance indexes are difficult to consider simultaneously. Due to the lack of systematic data support and quantitative analysis means, the formula optimization process has long period, high test cost and insufficient stability, and the cooperative promotion and the accurate control of the multi-performance targets are difficult to realize. Disclosure of Invention The application provides an intelligent component optimization method and system for anti-condensation insulating paint, which are used for solving the technical problems that the formula optimization of the anti-condensation insulating paint in the prior art depends on manual experience and is difficult to realize multi-performance index collaborative optimization. In view of the above problems, the application provides an intelligent optimization method and system for components of an anti-condensation insulating coating. In a first aspect of the application, there is provided a method for intelligently optimizing components of an anti-condensation insulating coating, the method comprising: Collecting physical and chemical parameter data of matrix resin, modified filler, functional auxiliary agent and curing system, and historical formula-performance related data, establishing a component database of the anti-condensation insulating coating, determining the initial range of components based on the component database, establishing a component-performance prediction model, optimizing the initial range of the components by taking the target performance parameter analyzed by an application scene as a target, obtaining a predicted theoretical component, carrying out reliability deviation peak value analysis according to the target performance parameter in combination with the components of the predicted theoretical component and the initial component, positioning a test verification component, and carrying out test verification based on the test verification component to determine the final optimized target coating component. In a second aspect of the present application, there is provided a component intelligent optimization system for an anti-condensation insulating paint, the system comprising: The anti-condensation insulating coating comprises a database establishing module, a model establishing module, an optimizing module, an analyzing module and a test verifying module, wherein the database establishing module is used for acquiring physical and chemical parameter data of matrix resin, modified filler, functional auxiliary agent and a curing system and historical formula-performance related data to establish a component database of the anti-condensation insulating coating, the model establishing module is used for determining an initial range of a component based on the component database and establishing a component-performance prediction model, the optimizing module is used for optimizing the initial range of the component by taking a target performance parameter analyzed by an application scene as a target through the component-performance prediction model to obtain a predicted theoretical component, the analyzing module is used for carrying out reliability deviation peak value analysis according to the target performance parameter combined with the components of the predicted theoretical component and the initial component to locate the test verifying component, and the test verifying module is used for carrying out test verification based on the test verifying component to determine the final optimized target coating component. One or more technical schemes provided by the application have at least the following technical effects or advantages: The method comprises the steps of collecting physical and chemical parameter data of matrix resin, modified filler, functional auxiliary agent and curing system, establishing a component database of the anti-condensation insulating coating, determining an initial range of components based on the component database, establishing a component-performance prediction model, optimizing the initial range of the components by taking target performance parameters analyzed by an application scene as targets through the component-performance prediction model to ob