A valve quality detection method and detection system

By analyzing the corrosion and crack characteristics of the valves in sub-blocks, and calculating the valve replacement index in combination with cost evaluation, the problem of inaccurate evaluation of whether the valve needs replacement is not possible in the prior art, achieving more efficient detection and maintenance decisions, saving Cost and time.

CN118641631BInactive Publication Date: 2024-12-03QIDONG YONGAN VALVE CO LTD
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411090256.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-09
Publication Date
2024-12-03
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing valve quality detection methods cannot effectively evaluate whether the valve stem needs to be replaced with the entire valve, resulting in production shutdowns and increased labor and financial expenditure, especially when the valve body screws are too tight and the valve stem is locked.

Method used

By dividing the valve into uniform sub-blocks, using ultrasonic detection equipment to detect the degree of corrosion and calculate the average corrosion coefficient; using equidistant segmentation method to divide the valve stem surface, calculate the crack characteristic coefficient, and clustering through the K-mean clustering analysis method. Calculate the surface damage coefficient; calculate the valve replacement index based on the maintenance and replacement costs to determine whether replacement or repair is needed.

Benefits of technology

It effectively improves the project progress speed, promptly judges the valve replacement and maintenance needs, saves project costs, and avoids unnecessary production shutdowns and maintenance costs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118641631B_ABST
    Figure CN118641631B_ABST
Patent Text Reader

Abstract

The invention relates to the field of material detection, and discloses a valve quality detection method and a detection system thereof, which are used to solve the problem that when a valve stem is stuck, replacing the valve as a whole will increase the cost of manpower and financial resources. The method comprises: screening out valve sub-blocks that are prone to corrosion, detecting the degree of corrosion on the surface of the screened valve sub-blocks in real time, calculating an average corrosion coefficient, and calculating a crack characteristic coefficient of each valve stem sub-region, clustering the crack characteristic coefficients, calculating a surface damage coefficient of the valve stem according to the clustering result, collecting the maintenance cost and the valve replacement cost, and calculating a replacement cost coefficient, comprehensively evaluating to obtain a valve replacement index, and judging whether to replace the valve or repair the valve, thereby effectively improving the progress speed of the project, enabling timely judgment on valve replacement and repair according to the valve condition when the valve stem is damaged, and effectively saving project costs.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of material detection, and more specifically to a valve quality detection method and a detection system thereof. Background Art

[0002] Valve quality inspection refers to various tests and evaluations of valves. Through inspection and evaluation, it is ensured that the quality of the valves meets the design and manufacturing standards, thereby ensuring the reliability, stability and safety of the valves during use, discovering and correcting installation problems early, and avoiding production shutdowns and equipment damage caused by valve failures.

[0003] The existing valve quality inspection is to first conduct an appearance inspection of the valve, and then test the sealing of the valve through an air tightness test or a water tightness test. If there is a problem with the valve, the valve is replaced to ensure that the valve can be used normally. However, this quality inspection method will result in damage to the valve stem when the screws on the valve body are too tight and the valve stem is locked. It is impossible to assess whether the entire valve needs to be replaced under the current situation. If the entire valve is replaced, production needs to be stopped, and replacing the entire valve will increase the consumption of manpower and financial resources.

[0004] In view of the above problems, the present invention proposes a solution. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, the present invention provides a valve quality detection method and a detection system thereof to solve the problems existing in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A valve quality detection method comprises the following steps: S001: according to the shape and structure of the valve, the valve is divided into n uniform sub-blocks, recorded as valve sub-blocks, and the easily corroded valve sub-blocks are screened out. The surface of the screened valve sub-blocks is tested for corrosion degree in real time using ultrasonic detection equipment, and the calculation formula is: ,in Expressed as the degree of corrosion, represents the actual thickness of the sub-block measured by ultrasonic testing, It is expressed as the initial thickness of the sub-block, and the average corrosion coefficient is calculated by the average value algorithm. The calculation formula is: ,in Expressed as the average corrosion coefficient, It is expressed as the number of sub-blocks filtered out. It represents the corrosion degree of the i-th screened sub-block;

[0008] S002: Use the equal distance segmentation method to divide the valve stem surface into m regions, recorded as valve stem sub-regions, and calculate the crack characteristic coefficient of each valve stem sub-region;

[0009] S003: Taking the crack characteristic coefficient as a feature, the K-means clustering analysis method is used to cluster the crack characteristic coefficient, each valve stem sub-region is grouped according to the crack characteristics, and the surface damage coefficient of the valve stem is calculated according to the clustering results;

[0010] S004: Collect the maintenance cost and valve replacement cost, and calculate the replacement cost coefficient;

[0011] S005: Based on the average corrosion coefficient, surface damage coefficient and replacement cost coefficient, the valve replacement index is calculated. The calculation formula is: ,in Expressed as the valve replacement index, Expressed as the average corrosion coefficient, Expressed as the surface damage coefficient, Expressed as the replacement cost coefficient, It is expressed as a weight coefficient of the average corrosion coefficient, the surface damage coefficient and the replacement cost coefficient. By comparing the valve replacement index with the preset threshold, a judgment is made on whether to replace the valve or repair the valve.

[0012] Preferably, the step of screening out the valve sub-blocks susceptible to corrosion is:

[0013] Use an ultrasonic thickness gauge to test the material thickness of the valve and count the thickness of each valve sub-block;

[0014] Through mathematical modeling and computer simulation, the shape and size of the valve parts are accurately described and analyzed, and the surface area of ​​each valve sub-block in contact with the medium is calculated;

[0015] The corrosion resistance coefficient is calculated based on the thickness of each valve sub-block and the surface area of ​​each valve sub-block in contact with the medium. The calculation formula is: ,in Expressed as the corrosion resistance coefficient, Expressed as the thickness of each valve sub-block, It is represented by the surface area of ​​each valve sub-block contacting the medium. The corrosion resistance coefficient of each valve sub-block is compared with the preset threshold, and the valve sub-blocks with corrosion resistance coefficients less than the preset threshold are screened out and recorded as easily corroded areas.

[0016] Preferably, the crack characteristic coefficient calculation step is:

[0017] Crack detection is performed using computer vision algorithms on each valve stem sub-region;

[0018] Extract features of detected cracks, including crack length, crack width and crack number;

[0019] According to the crack length, width and number, the crack characteristic coefficient is calculated, and the calculation formula is: ,in Expressed as crack characteristic coefficient, is the crack length, is the crack width, n is the number of cracks, Expressed as the area of ​​each stem sub-region.

[0020] Preferably, the step of using a computer vision algorithm to perform crack detection is:

[0021] Use image acquisition equipment to acquire clear images of various areas of the valve stem;

[0022] Grayscale, filter and edge detect the acquired images to reduce noise and improve image quality;

[0023] Use edge feature method to extract features from preprocessed images, and extract the contour and boundary features of objects in the image by detecting the edges in the image;

[0024] Crack detection is performed using edge detection to identify crack areas in the image and mark and record them.

[0025] Preferably, the step of clustering the crack characteristic coefficients using the K-means clustering analysis method is:

[0026] The crack characteristic coefficient of each valve stem sub-region is collected, and the crack characteristic coefficient of each valve stem sub-region is recorded as the characteristic vector of the data point, and the silhouette coefficient method is used to determine the cluster number;

[0027] Randomly select the feature vector of the data point corresponding to the number of clusters from the feature vector of the data point as the initial cluster center;

[0028] The Euclidean distance method is used to calculate the distance between the feature vector of each data point and each initial cluster center, and the feature vector of the data point is assigned together with the initial cluster center with the closest distance. The calculation formula is: ,in It is expressed as the distance between the feature vector of the data point and the cluster center. Represented as the feature vector of a data point, It is represented as the feature vector of the cluster center;

[0029] After traversing the feature vectors of all data points and assigning them, we get the initial clusters;

[0030] The new cluster center of the initial cluster is calculated by the average value algorithm, and the above clustering process is repeated until the cluster center no longer changes, and the final cluster and the final cluster center are obtained.

[0031] Preferably, the step of calculating the surface damage coefficient of the valve stem according to the clustering result is:

[0032] The weight of each final cluster is calculated based on the total number of feature vectors of the data points and the number of samples in the final cluster. The calculation formula is: ,in Represented as the weight of the i-th final cluster, Represents the number of samples in each final cluster, It is represented as the total number of feature vectors of data points, and the sum of the number of samples of all final clusters is ;

[0033] According to the center of the final cluster and the weight of the final cluster, the surface damage coefficient is calculated, and its calculation formula is: ,in is represented by the surface damage coefficient, k is represented by the number of clusters, Represented as the weight of the i-th final cluster, It is represented as the cluster center of the i-th cluster, and the cluster center of the cluster is the final cluster center.

[0034] Preferably, the replacement cost coefficient calculation steps are:

[0035] Collect historical valve maintenance costs and calculate based on the mean to obtain the average maintenance cost, which includes labor costs and material costs;

[0036] Collect the cost of replacing the valve according to the valve model, which includes labor costs, material costs and losses caused by downtime;

[0037] The replacement cost coefficient is calculated based on the average maintenance cost and the cost of replacing the valve. The calculation formula is: ,in Expressed as the replacement cost coefficient, Expressed as the average maintenance cost, Expressed as the cost of replacing the valve.

[0038] Preferably, a valve quality detection system comprises:

[0039] Testing equipment preparation module, testing equipment used to collect valve data, including ultrasonic testing equipment, ultrasonic thickness gauge and image acquisition equipment;

[0040] A valve data acquisition module, used to collect various valve data through the equipment prepared by the detection equipment preparation module;

[0041] Valve data evaluation module, used to conduct comprehensive analysis on various valve data collected by the valve data collection module;

[0042] The valve quality assessment module is used to make valve replacement decisions based on the comprehensive analysis results of the valve data assessment module.

[0043] Technical effects and advantages of the present invention:

[0044] Screen out the valve sub-blocks that are prone to corrosion, detect the degree of corrosion on the surface of the screened valve sub-blocks in real time, calculate the average corrosion coefficient, and calculate the crack characteristic coefficient of each valve stem sub-area. Cluster the crack characteristic coefficients, and calculate the surface damage coefficient of the valve stem based on the clustering results. Collect the maintenance cost and valve replacement cost, and calculate the replacement cost coefficient. Comprehensively evaluate the valve replacement index to determine whether to replace or repair the valve, effectively improving the progress of the project. When the valve stem is damaged, the valve replacement and repair can be determined in time according to the valve condition, effectively saving project costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is the overall flow chart of the present invention.

[0046] Figure 2 The figure is a flow chart of crack characteristic coefficient calculation of the present invention. DETAILED DESCRIPTION

[0047] The technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the present invention. In addition, the forms of the various structures recorded in the following embodiments are only examples. The valve quality detection method and the detection system thereof involved in the present invention are not limited to the various structures recorded in the following embodiments. All other implementations obtained by ordinary technicians in this field without making creative work belong to the scope of protection of the present invention.

[0048] The present invention provides a valve quality detection method, comprising the following steps:

[0049] S001: According to the shape and structure of the valve, the valve is divided into n uniform sub-blocks, recorded as valve sub-blocks, and the valve sub-areas prone to corrosion are screened out. The ultrasonic detection equipment is used to detect the corrosion degree of the surface of the screened valve sub-blocks in real time. The calculation formula is: ,in Expressed as the degree of corrosion, represents the actual thickness of the sub-block measured by ultrasonic testing, It is expressed as the initial thickness of the sub-block, which is obtained from the valve manufacturer or the manual. It is expressed as the minimum safe thickness of the material. The minimum safe thickness of the material is obtained from the valve manufacturer or the manual, and the average corrosion coefficient is calculated by the average value algorithm. The calculation formula is: ,in Expressed as the average corrosion coefficient, It is expressed as the number of sub-blocks filtered out. It represents the corrosion degree of the i-th screened sub-block;

[0050] S002: Use the equal distance segmentation method to divide the valve stem surface into m regions, recorded as valve stem sub-regions, and calculate the crack characteristic coefficient of each valve stem sub-region;

[0051] S003: Taking the crack characteristic coefficient as a feature, the K-means clustering analysis method is used to cluster the crack characteristic coefficient, each valve stem sub-region is grouped according to the crack characteristics, and the surface damage coefficient of the valve stem is calculated according to the clustering results;

[0052] S004: Collect the maintenance cost and valve replacement cost, and calculate the replacement cost coefficient;

[0053] S005: Based on the average corrosion coefficient, surface damage coefficient and replacement cost coefficient, the valve replacement index is calculated. The calculation formula is: ,in Expressed as the valve replacement index, It is expressed as the average corrosion coefficient. The average corrosion coefficient reflects the degree of corrosion of the valve. The more serious the corrosion of the valve, the greater the impact on its performance and structure, resulting in a shortened life of the valve and the need for earlier replacement. Therefore, the higher the average corrosion coefficient, the higher the valve replacement index. Expressed as the surface damage coefficient, the surface damage degree of the valve stem is usually regarded as one of the important indicators for valve maintenance and replacement decisions. The higher the surface damage coefficient of the valve stem, the more serious the surface damage of the valve stem, and its performance and service life may be more affected. It is necessary to consider replacing the valve sooner. Therefore, the higher the surface damage coefficient, the higher the valve replacement index. It is expressed as the replacement cost coefficient. The higher the replacement cost coefficient, the higher the cost of repair. The more serious the damage, the higher the repair cost. It also increases the possibility of valve replacement. Therefore, the higher the replacement cost coefficient, the higher the valve replacement index. It is expressed as the weight coefficient of the average corrosion coefficient, surface damage coefficient and replacement cost coefficient, and this embodiment is not The specific value is calculated specifically, and the valve replacement index is compared with the preset threshold. If the valve replacement index is less than the preset threshold, the valve is not replaced and only repaired. If the valve replacement index is greater than the preset threshold, the valve is replaced.

[0054] Valves play a key role in industrial production, building facilities and other fields. They control the flow and pressure of fluids such as liquids, gases, and steam, and are directly related to the safety of equipment and personnel. Valves with unqualified quality may cause safety accidents such as leakage, explosion, and fire. Therefore, ensuring the quality of valves through testing can effectively ensure the safety of workplaces and equipment. Qualified valves should have stable performance, including flexibility in opening and closing, sealing, corrosion resistance, etc. Testing can evaluate the performance of valves under various working conditions to ensure that they meet design requirements and can operate stably and reliably.

[0055] By testing qualified valves, we can ensure their good sealing, reduce fluid leakage and energy waste, improve the energy efficiency of the process system, and reduce energy consumption in the production process, so as to achieve the purpose of energy conservation and emission reduction. Valves are usually installed in pipeline systems. If the valve quality is unqualified, it is easy to cause frequent failures and repairs, increase maintenance costs and downtime. By testing to ensure the quality of valves, we can reduce maintenance costs, extend equipment life, and improve production efficiency. In some industries and fields, valves may need to comply with specific laws, regulations, standards and industry standards, such as pressure vessel safety technical specifications, ISO standards, etc., to ensure that the valves meet relevant compliance requirements and avoid legal liability and economic losses caused by violations.

[0056] Use ultrasonic testing equipment to inspect the surface of the valve sub-block. Ultrasonic waves will propagate in the material and reflect when encountering unevenness or defects. By measuring the propagation time and intensity changes of ultrasonic waves, the degree of corrosion on the material surface can be detected. Generally speaking, when the degree of corrosion reaches a certain level, it may affect the performance and safety of the valve, and corresponding maintenance or replacement measures need to be taken.

[0057] Mathematical modeling and computer simulation can provide highly accurate descriptions of valve shape and size, which is much more accurate than traditional manual measurement methods. This accuracy helps to more accurately predict the contact surface area between the valve and the medium. Mathematical modeling and computer simulation can generate intuitive three-dimensional models and visual results, allowing users to clearly understand the shape and size characteristics of the valve, helping engineers and designers to better understand the structure of the valve and conduct more accurate analysis and evaluation. It can easily analyze multiple parameters, such as different shapes, sizes and materials of valves. By adjusting these parameters, their impact on the contact surface area of ​​the medium can be evaluated.

[0058] In this embodiment, it should be specifically explained that the step of screening out the easily corrosive areas is:

[0059] The material thickness of the valve is tested using an ultrasonic thickness gauge, which is a portable device used to measure the thickness of materials. The ultrasonic thickness gauge determines the thickness of the material by sending ultrasonic pulses to the surface of the material and measuring the time it takes for the ultrasonic signal to be reflected from the surface. The speed at which ultrasonic waves propagate in the material is usually known, so the thickness of the material can be calculated based on the propagation time of the ultrasonic wave.

[0060] Through mathematical modeling and computer simulation, the shape and size of the valve parts are accurately described and analyzed, and the surface area of ​​contact with the medium is calculated;

[0061] The corrosion resistance coefficient is calculated based on the valve material thickness and the surface area of ​​the contact medium. The calculation formula is: ,in Expressed as the corrosion resistance coefficient, Expressed as the thickness of each valve sub-block, It is represented by the surface area of ​​each valve sub-block contacting the medium. The corrosion resistance coefficient of each valve sub-block is compared with a preset threshold value, and the valve sub-blocks with corrosion resistance coefficients less than the preset threshold value are screened out.

[0062] In this embodiment, it should be specifically explained that the steps of using an ultrasonic thickness gauge to detect the material thickness of the valve are as follows:

[0063] According to the actual structure and design drawings of the valve, use computer-aided design software to establish a three-dimensional geometric model of the valve part;

[0064] According to the shape and structure of the valve, the valve surface is divided into n uniform sub-blocks, which are recorded as valve sub-blocks;

[0065] Use an ultrasonic thickness gauge to place it on the valve surface according to the pre-divided sub-blocks and perform measurements. In each sub-block, the gauge will send an ultrasonic signal and receive an echo signal, and the thickness of the material is determined by calculating the time difference of the echo signal;

[0066] Count the material thickness data of each sub-block.

[0067] In this embodiment, it should be specifically explained that the step of calculating the surface area of ​​the contact medium is:

[0068] According to the actual structure and design drawings of the valve, use computer-aided design software to establish a three-dimensional geometric model of the valve part to accurately describe the shape and size of the valve part;

[0069] The established three-dimensional geometric model is meshed and the valve surface is divided into small areas using finite element analysis software for subsequent calculation and analysis;

[0070] For each small area, calculate its surface area, and add up the surface areas of all small areas to obtain the total surface area of ​​the valve part in contact with the medium.

[0071] In this embodiment, it should be specifically explained that Figure 2 As shown, the crack characteristic coefficient calculation steps are:

[0072] Perform crack detection on each valve stem sub-area using a computer vision algorithm, which is a technology that uses computers to process and analyze images or videos. It involves a series of processes such as image acquisition, preprocessing, feature extraction, and object recognition to achieve understanding, analysis, and application of image content;

[0073] Extract features of detected cracks, including crack length, crack width and crack number;

[0074] According to the crack length, width and number, the crack characteristic coefficient is calculated, and the calculation formula is: ,in Expressed as crack characteristic coefficient, is the crack length, is the crack width, n is the number of cracks, Expressed as the area of ​​each stem sub-region.

[0075] Compared with traditional manual detection methods, computer vision algorithms can realize automated and efficient crack detection, greatly improving detection efficiency. It can quickly process large amounts of image data and complete detection tasks in a short time. Computer vision algorithms can accurately identify cracks in images and provide accurate detection results. They can also accurately analyze and judge the shape, size and position of cracks based on pre-defined rules and models. Since the working method of computer vision algorithms is based on automated processing based on pre-defined algorithms and models, it can ensure consistent detection results in different situations and avoid differences introduced by human factors. Computer vision algorithms can detect different types of cracks, and can respond flexibly whether on different material surfaces or under different environmental conditions, and have certain versatility and scalability.

[0076] In this embodiment, it should be specifically explained that the steps of using a computer vision algorithm to perform crack detection are:

[0077] Use an image acquisition device to acquire clear images of various regions of the valve stem, wherein the acquisition device may be a camera;

[0078] Grayscale, filter and edge detect the acquired images to reduce noise and improve image quality;

[0079] The edge feature method is used to extract features from the preprocessed image to facilitate the recognition and detection of cracks. Common features include texture features, shape features, and edge features. The edge feature method is a commonly used computer vision algorithm for extracting features from images. This method mainly focuses on the edge information in the image. By detecting the edges in the image, the contour and boundary features of the object in the image are extracted.

[0080] Crack detection is performed using edge detection to detect crack areas in the image and mark and record them.

[0081] The steps of using edge detection method to perform crack detection are:

[0082] The Sobel operator is used to detect the edge of the preprocessed image, and the complete edge line is formed by connecting the pixels of the edge points;

[0083] Analyze and process the edge image to identify the edge lines associated with the cracks;

[0084] Cracks are identified based on the detected crack edge lines.

[0085] In image processing, edge detection is a common technique used to detect the position of the boundary of objects in an image. Among edge detection algorithms, the Sobel operator is one of the classic edge detection operators. The Sobel operator is a discrete spatial filter used to detect edges in an image. It utilizes the differences between pixels in an image, especially where pixel values ​​change dramatically, such as object boundaries.

[0086] In this embodiment, it should be specifically explained that the steps of clustering the crack characteristic coefficients using the K-means clustering analysis method are as follows:

[0087] The crack characteristic coefficient of each valve stem sub-region is collected, and the crack characteristic coefficient of each valve stem sub-region is recorded as the characteristic vector of the data point. The number of clusters is determined using the silhouette coefficient method. For example, the characteristic vector of the collected data point is , the determined number of clusters is 3, the feature vector is a vector used to describe the characteristics of each sample, a vector composed of the characteristics of the sample, and each feature corresponds to a dimension in the vector;

[0088] Randomly select the feature vector of the data point corresponding to the number of clusters from the feature vector of the data point as the initial cluster center. For example, the selected initial cluster center is {x1, x2, x3}. The cluster center refers to the center point of each cluster, which represents the average position of all samples in the cluster.

[0089] The Euclidean distance method is used to calculate the distance between the feature vector of each data point and each initial cluster center, and the feature vector of the data point is assigned together with the initial cluster center with the closest distance. The calculation formula is: ,in It is expressed as the distance between the feature vector of the data point and the cluster center. Represented as the feature vector of a data point, The Euclidean distance method is a commonly used distance measurement method used to measure the straight-line distance between two points in Euclidean space. In cluster analysis, the Euclidean distance is often used to measure the similarity or distance between samples.

[0090] After traversing the feature vectors of all data points and assigning them, we get the initial clusters;

[0091] The new cluster center of the initial cluster is calculated by the average value algorithm. For example, the initial cluster 1 is , then its new cluster center is , repeat the above clustering process until the cluster center no longer changes and obtain the final cluster.

[0092] In this embodiment, it should be specifically explained that the step of calculating the surface damage coefficient of the valve stem according to the clustering result is:

[0093] The weight of each final cluster is calculated based on the total number of feature vectors of the data points and the number of samples in the final cluster. For example, the total number of feature vectors of the data points is , the number of samples in the final cluster 1 is , the number of samples in the final cluster 2 is , the number of samples in the final cluster 3 is ,and , then the final weight of cluster 1 is , the final weight of cluster 2 is , the final weight of cluster 3 is ;

[0094] According to the center of the final cluster and the weight of the final cluster, the surface damage coefficient is calculated, and its calculation formula is: ,in is represented by the surface damage coefficient, k is represented by the number of clusters, Represented as the weight of the i-th cluster, Represented as the cluster center of the i-th cluster.

[0095] Valve replacement is a comprehensive project. The valve replacement process requires technical workers to operate, including dismantling old valves, installing new valves, connecting pipes, debugging, etc., and also requires the use of new valves and their accessories. These material costs include the purchase cost of new valves, the cost of accessories such as sealing gaskets, bolts, and pipe fittings. Valve replacement may cause shutdowns or production interruptions in industrial production, construction facilities, etc., which will bring production losses and shutdown costs. Therefore, the shutdown costs include capacity losses, delay losses, personnel costs, equipment depreciation, etc. caused by shutdowns.

[0096] In this embodiment, it should be specifically explained that the replacement cost coefficient calculation steps are:

[0097] Collect historical valve maintenance costs and calculate based on the mean to obtain the average maintenance cost, which includes labor costs and material costs;

[0098] Collect the cost of replacing the valve according to the valve model, which includes labor costs, material costs and losses caused by downtime;

[0099] The replacement cost coefficient is calculated based on the average maintenance cost and the cost of replacing the valve. The calculation formula is: ,in Expressed as the replacement cost coefficient, Expressed as the average maintenance cost, Expressed as the cost of replacing the valve.

[0100] In this embodiment, it is necessary to specifically explain a valve quality detection system, the system comprising:

[0101] Testing equipment preparation module, testing equipment used to collect valve data, including ultrasonic testing equipment, ultrasonic thickness gauge and image acquisition equipment;

[0102] A valve data acquisition module, used to collect various valve data through the equipment prepared by the detection equipment preparation module;

[0103] Valve data evaluation module, used to conduct comprehensive analysis on various valve data collected by the valve data collection module;

[0104] The valve quality assessment module is used to make valve replacement decisions based on the comprehensive analysis results of the valve data assessment module.

[0105] Finally: 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 spirit and principles of the present invention should be included in the protection scope of the present invention.

[0106] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A valve quality inspection method, It is characterized in that The following steps are involved: S001: According to the shape and structure of the valve, the valve is divided into n uniform sub-blocks, recorded as valve sub-blocks, and the easily corroded valve sub-blocks are screened out. The ultrasonic detection equipment is used to detect the corrosion degree of the surface of the screened valve sub-blocks in real time. The calculation formula is: ,in Expressed as the degree of corrosion, represents the actual thickness of the sub-block measured by ultrasonic testing, is the initial thickness of the sub-block, It is expressed as the minimum safe thickness of the material, and the average corrosion coefficient is calculated by the average value algorithm. The calculation formula is: ,in Expressed as the average corrosion coefficient, It is expressed as the number of sub-blocks filtered out. It represents the corrosion degree of the i-th screened sub-block; S002: Use the equal distance segmentation method to divide the valve stem surface into m regions, recorded as valve stem sub-regions, and calculate the crack characteristic coefficient of each valve stem sub-region; S003: Taking the crack characteristic coefficient as a feature, the K-means clustering analysis method is used to cluster the crack characteristic coefficient, each valve stem sub-region is grouped according to the crack characteristics, and the surface damage coefficient of the valve stem is calculated according to the clustering results; S004: Collect the maintenance cost and valve replacement cost, and calculate the replacement cost coefficient; S005: Based on the average corrosion coefficient, surface damage coefficient and replacement cost coefficient, the valve replacement index is calculated. The calculation formula is: ,in Expressed as the valve replacement index, Expressed as the average corrosion coefficient, Expressed as the surface damage coefficient, Expressed as the replacement cost coefficient, The weight coefficients are expressed as average corrosion coefficient, surface damage coefficient and replacement cost coefficient. By comparing the valve replacement index with the preset threshold, it is judged whether to replace the valve or repair the valve; The steps of screening out the valve sub-blocks susceptible to corrosion are: Use an ultrasonic thickness gauge to test the material thickness of the valve and count the thickness of each valve sub-block; Through mathematical modeling and computer simulation, the shape and size of the valve parts are accurately described and analyzed, and the surface area of ​​each valve sub-block in contact with the medium is calculated; The corrosion resistance coefficient is calculated based on the thickness of each valve sub-block and the surface area of ​​each valve sub-block in contact with the medium. The calculation formula is: ,in Expressed as the corrosion resistance coefficient, Expressed as the thickness of each valve sub-block, It is represented by the surface area of ​​each valve sub-block contacting the medium. The corrosion resistance coefficient of each valve sub-block is compared with the preset threshold value, and the valve sub-blocks with corrosion resistance coefficients less than the preset threshold value are screened out and recorded as the easily corrosive areas; The crack characteristic coefficient calculation steps are: Crack detection is performed using computer vision algorithms on each valve stem sub-region; Extract features of detected cracks, including crack length, crack width and crack number; According to the crack length, width and number, the crack characteristic coefficient is calculated, and the calculation formula is: in Expressed as crack characteristic coefficient, is the crack length, is the crack width, n is the number of cracks, It is expressed as the area of ​​each stem sub-region; The steps of calculating the surface damage coefficient of the valve stem according to the clustering results are as follows: The weight of each final cluster is calculated based on the total number of feature vectors of the data points and the number of samples in the final cluster. The calculation formula is: ,in Represented as the weight of the i-th final cluster, Represents the number of samples in each final cluster, It is represented as the total number of feature vectors of data points, and the sum of the number of samples of all final clusters is ; According to the center of the final cluster and the weight of the final cluster, the surface damage coefficient is calculated, and its calculation formula is: ,in is represented by the surface damage coefficient, k is represented by the number of clusters, Represented as the weight of the i-th final cluster, It is represented as the cluster center of the i-th cluster, and the cluster center of the cluster is the final cluster center.

2. A valve quality detection method according to claim 1, Features: The steps of using computer vision algorithm to detect cracks are: Use image acquisition equipment to acquire clear images of various areas of the valve stem; Grayscale, filter and edge detect the acquired images to reduce noise and improve image quality; Use edge feature method to extract features from preprocessed images, and extract the contour and boundary features of objects in the image by detecting the edges in the image; Crack detection is performed using edge detection to identify crack areas in the image and mark and record them.

3. A valve quality detection method according to claim 2, Features: The steps of clustering the crack characteristic coefficients using the K-means clustering analysis method are as follows: The crack characteristic coefficient of each valve stem sub-region is collected, and the crack characteristic coefficient of each valve stem sub-region is recorded as the characteristic vector of the data point, and the silhouette coefficient method is used to determine the cluster number; Randomly select the feature vector of the data point corresponding to the number of clusters from the feature vector of the data point as the initial cluster center; The Euclidean distance method is used to calculate the distance between the feature vector of each data point and each initial cluster center, and the feature vector of the data point is assigned together with the initial cluster center with the closest distance. The calculation formula is: ,in It is expressed as the distance between the feature vector of the data point and the cluster center. Represented as the feature vector of a data point, is represented as the eigenvector of the cluster center, is the feature vector of the mth data point, is the characteristic vector of the mth cluster center; After traversing the feature vectors of all data points and assigning them, we get the initial clusters; The new cluster center of the initial cluster is calculated by the average value algorithm, and the above clustering process is repeated until the cluster center no longer changes, and the final cluster and the final cluster center are obtained.

4. A valve quality detection method according to claim 3, Features: The steps for calculating the replacement cost coefficient are: Collect historical valve maintenance costs and calculate based on the mean to obtain the average maintenance cost, which includes labor costs and material costs; Collect the cost of replacing the valve according to the valve model, which includes labor costs, material costs and losses caused by downtime; The replacement cost coefficient is calculated based on the average maintenance cost and the cost of replacing the valve. The calculation formula is: ,in Expressed as the replacement cost coefficient, Expressed as the average maintenance cost, Expressed as the cost of replacing the valve.

5. A valve quality detection system, used to implement a valve quality detection method as described in any one of claims 2 to 4 above, It is characterized in that The system comprises: Testing equipment preparation module, testing equipment for collecting valve data, including ultrasonic testing equipment, ultrasonic thickness gauge and image acquisition equipment; A valve data acquisition module, used to collect various valve data through the equipment prepared by the detection equipment preparation module; Valve data evaluation module, used to conduct comprehensive analysis on various valve data collected by the valve data collection module; The valve quality assessment module is used to make valve replacement decisions based on the comprehensive analysis results of the valve data assessment module.

Citation Information

Cited By

  • Valve quality detection method and detection system thereof

    CN121027109A