An industrial equipment health evaluation method, system, device, medium and product

Abstract

The present application relates to the field of industrial equipment evaluation, and provides an industrial equipment health degree evaluation method, system, device, medium and product. The industrial equipment health degree evaluation method comprises: obtaining health degree score index data of an industrial equipment, and constructing a health degree score model; constructing a health degree score index data set of the industrial equipment, calculating initial weights of the health degree score index data by using an extended entropy weight method; calculating a preliminary health degree score by using the health degree score model according to the initial weights of the health degree score index data and the health degree score index data of the industrial equipment; generating an adjustment amount of the preliminary comprehensive health degree score and an optimized health degree score by using the health degree score index data set of the industrial equipment, so as to iteratively optimize the weights to obtain optimal weights; and obtaining a final health degree score according to the optimal weights and the weighted sum of the health degree score index data of the industrial equipment. The accuracy of the industrial equipment health degree evaluation is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of industrial equipment evaluation, and in particular to an industrial equipment health degree evaluation method, system, device, medium and product. BACKGROUND

[0002] The statements in this section merely provide background information related to the present application and do not necessarily constitute the prior art.

[0003] In the industrial field, the equipment health state monitoring and evaluation technology (PHM) is a core means to ensure production safety and reduce operation and maintenance costs.

[0004] The traditional industrial equipment health state evaluation method uses the experience and knowledge of experts as the main basis for generating health degree scores, uses an artificial intelligence method, and generates a regression model based on a large amount of artificial data. However, such an artificial intelligence regression model is often one-time, i.e., the generated model has no further improvement space. That is, only by transforming the data set, re-adjusting the hyperparameters, or transforming the artificial intelligence method used, a model that is more consistent with the user's cognition than the existing model can be generated. Moreover, this method does not take into account the fact that different experts may give contradictory data for the same analysis object, such as linearly unrelated data between two experts. In principle, the explainability is poor. The regression model generated by the artificial intelligence model is often greatly different from the subjective cognition of the operation and maintenance personnel, and the results obtained cannot convince the operation and maintenance personnel. SUMMARY

[0005] In order to solve the technical problems in the background art, the present application provides an industrial equipment health degree evaluation method, system, device, medium and product. The present application is particularly suitable for single-user industrial application scenarios, only a small amount of data is required, and only the individual parameters need to be adjusted to continuously optimize the health degree score model, so that the health degree score results are more consistent with the subjective cognition of the user, and the accuracy of the industrial equipment health degree evaluation is improved.

[0006] In order to achieve the above-mentioned purpose, the present application adopts the following technical solutions:

[0007] The first aspect of the present application provides an industrial equipment health degree evaluation method.

[0008] An industrial equipment health degree evaluation method comprises:

[0009] Obtaining health degree score index data of the industrial equipment, and constructing a health degree score model;

[0010] Constructing a health degree score index data set of the industrial equipment, and calculating the initial weight of the health degree score index data using an extended entropy weight method;

[0011] According to the initial weight of the health score index data and the health score index data of the industrial equipment, a health score model is used to calculate a preliminary health score;

[0012] Using the health score index data set of the industrial equipment, an adjustment amount of the preliminary comprehensive health score and an optimized health score are generated to iteratively optimize the weight and obtain an optimal weight;

[0013] According to the optimal weight and the weighted sum of the health score index data of the industrial equipment, a final health score is obtained.

[0014] Further, the industrial equipment includes manufacturing equipment, energy industry equipment, chemical industry equipment, metallurgical industry equipment, and transportation industry equipment; if it is a five-axis linkage numerical control machine tool in the manufacturing equipment, the health score index data includes spindle speed stability, spindle bearing temperature, hydraulic system oil viscosity, feed shaft vibration intensity, and positioning accuracy; if it is a blast furnace blower in the metallurgical industry equipment, the health score index data includes fan efficiency, bearing temperature, vibration intensity, lubricating oil viscosity, outlet air pressure, motor current deviation rate, impeller dynamic balance deviation, and cooling water temperature.

[0015] Further, the initial weight of the health score index data is obtained by multiplying the weights of the multiple upper nodes of the health score index data.

[0016] Further, the method for constructing the health score index data set of the industrial equipment includes:

[0017] Selecting Q pieces of health score index data of industrial equipment to construct a data matrix;

[0018] Performing elementary row transformation on the data matrix to verify whether the rank of the data matrix is R, determine which data are linearly correlated and which data are linearly uncorrelated;

[0019] If the number of uncorrelated data is equal to R, only one piece of data in each pair of linearly correlated multiple data in the data matrix is retained, and the rest is deleted; if the number of uncorrelated data is greater than R, the linearly uncorrelated data exceeding R are deleted, only one piece of data in each pair of linearly correlated multiple data in the data matrix is retained, and the rest is deleted; if the number of uncorrelated data is less than R, linearly uncorrelated data is supplemented to R pieces;

[0020] Repeat the above steps to construct multiple data matrices to obtain the health score index data set of the industrial equipment.

[0021] Further, the method for calculating the initial weight of the health score index data using the extended entropy weight method includes:

[0022] The health score index data of each industrial equipment is standardized to obtain maximum indicators, minimum indicators, intermediate indicators and interval indicators, so as to construct a data group of the health score index of each industrial equipment;

[0023] According to the data group of the health score index of each industrial equipment, the proportional coefficient of each health score index is calculated;

[0024] According to the proportional coefficient of each health score index, the information entropy of each health score index is calculated;

[0025] According to the information entropy of each health score index and the total number of health score indexes, the initial weight of each health score index data is calculated.

[0026] Further, the health score index data set of the industrial equipment is used to generate an adjustment amount of the preliminary comprehensive health score and an optimized health score, so as to iteratively optimize the weight to obtain an optimal weight; the method comprises:

[0027] According to the preliminary comprehensive health score, an adjustment amount of the preliminary comprehensive health score is generated to perform an iterative optimization process;

[0028] In the iteration process, the preliminary comprehensive health score of the current iteration number is generated according to the weight of the last iteration number and the health score index data set of the industrial equipment of the current iteration number; the adjustment amount of the health score of the current iteration number is generated according to the preliminary comprehensive health score of the current iteration number, and the optimized comprehensive health score of the current iteration number is calculated; and then the weight of the current iteration number is calculated according to the health score index data set of the industrial equipment of the current iteration number and the calculated optimized comprehensive health score;

[0029] After the iteration is completed, the optimal weight is selected from the optimized weights of all iteration numbers according to the evaluation index.

[0030] The second aspect of the present application provides an industrial equipment health evaluation system.

[0031] An industrial equipment health evaluation system comprises:

[0032] A data acquisition module configured to acquire health score index data of industrial equipment and construct a health score model;

[0033] An initial weight calculation module configured to construct a health score index data set of industrial equipment and calculate the initial weight of the health score index data by using an extended entropy weight method;

[0034] a health degree score calculation module configured to calculate a preliminary health degree score according to initial weights of health degree score index data and the health degree score index data of the industrial equipment by using a health degree score model;

[0035] a weight optimization module configured to generate an adjustment amount of the preliminary comprehensive health degree score and an optimized health degree score by using a health degree score index data set of the industrial equipment, to iteratively optimize the weights to obtain optimal weights;

[0036] an output module configured to obtain a final health degree score according to the optimal weights and a weighted sum of the health degree score index data of the industrial equipment.

[0037] A third aspect of the present application provides a computer device, the device comprising:

[0038] a processor adapted to execute a computer program;

[0039] a computer readable storage medium, the computer readable storage medium storing a computer program, the computer program being executed by the processor to implement the steps in the industrial equipment health degree evaluation method according to the first aspect.

[0040] A fourth aspect of the present application provides a computer readable storage medium storing a computer program, the computer program being adapted to be loaded and executed by a processor to implement the steps in the industrial equipment health degree evaluation method according to the first aspect.

[0041] A fifth aspect of the present application provides a computer program product or a computer program.

[0042] The present application provides a computer program product or a computer program, the computer program product or the computer program comprising computer instructions stored in a computer readable storage medium. The processor of the computer device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions to make the computer device execute the steps in the industrial equipment health degree evaluation method according to the first aspect.

[0043] Compared with the prior art, the present application has the following beneficial effects:

[0044] The application provides an industrial equipment health degree evaluation method, system, device, medium and product, health degree score index data of an industrial equipment is acquired, and a health degree score model is constructed; a health degree score index data set of the industrial equipment is constructed, an initial weight of the health degree score index data is calculated by using an extended entropy weight method; a preliminary health degree score is calculated by using the health degree score model according to the initial weight of the health degree score index data and the health degree score index data of the industrial equipment; an adjustment amount of the preliminary comprehensive health degree score and an optimized health degree score are generated by using the health degree score index data set of the industrial equipment, so as to iteratively optimize the weight and obtain an optimal weight; and a final health degree score is obtained according to the optimal weight and a weighted sum of the health degree score index data of the industrial equipment. The application is used for guiding a user to establish a health degree score model in an industrial scene, can provide an objective initial weight, if the user is not satisfied with a health degree evaluation result, can interact with the user through a machine based on a subjective evaluation method, and continuously optimizes the health degree score model until the health degree score result meets the requirement of the user, thereby improving the accuracy of the health degree evaluation of the industrial equipment. And a specific index is needed to quantify the satisfaction degree of the user to the optimized health degree evaluation model.

[0045] The application can greatly reduce the amount of training data, only needs dozens of data to optimize the result that is very satisfactory to an operation and maintenance personnel, reduces the data calculation amount, and improves the accuracy of the evaluation. BRIEF DESCRIPTION OF DRAWINGS

[0046] The drawings accompanying the specification of the application form a part of the application and serve to further understand the application. The schematic embodiments of the application and the description thereof are used to explain the application, and do not constitute an improper limitation on the application.

[0047] Figure 1 is a flowchart of the health degree evaluation method of the industrial equipment according to the embodiment of the application;

[0048] Figure 2 is a flowchart of the health degree weight optimization method according to the embodiment of the application;

[0049] Figure 3 is an example diagram of a tree model according to the embodiment of the application;

[0050] Figure 4 is an example diagram of a tree model and each index weight according to the embodiment of the application;

[0051] Figure 5 is an example diagram of a tree model of the health degree of a five-axis linkage numerical control machine tool of a certain factory according to the embodiment of the application;

[0052] Figure 6 is a flowchart of how to make a qualified data set according to the embodiment of the application;

[0053] Figure 7 is a flow chart of the weight optimization method shown in the embodiments of the present application;

[0054] Figure 8 is a structural diagram of the industrial equipment health degree evaluation system shown in the embodiments of the present application;

[0055] Figure 9 is a structural diagram of the computer device shown in the embodiments of the present application. DETAILED DESCRIPTION

[0056] The present application will be further described below in conjunction with the accompanying drawings and embodiments.

[0057] It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the present application. Unless otherwise indicated, all technical and scientific terms used herein have the same meaning as would be understood by one of ordinary skill in the art to which the present application pertains.

[0058] It should be noted that the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, the singular form is intended to include the plural form unless the context clearly indicates otherwise, and it should also be understood that when the terms "comprise" and / or "include" are used in the specification, there is a presence of a feature, step, operation, device, component and / or combinations thereof.

[0059] Figure 1 is a flow chart of the industrial equipment health degree evaluation method shown in the embodiments of the present application, with reference to Figure 1 , the method comprises:

[0060] obtaining health degree score index data of the industrial equipment, and constructing a health degree score model;

[0061] constructing a health degree score index data set of the industrial equipment, and calculating initial weights of the health degree score index data using an extended entropy weight method;

[0062] calculating a preliminary health degree score according to the initial weights of the health degree score index data and the health degree score index data of the industrial equipment using the health degree score model;

[0063] generating an adjustment amount of the preliminary comprehensive health degree score and an optimized health degree score using the health degree score index data set of the industrial equipment, to iteratively optimize the weights and obtain optimal weights;

[0064] obtaining a final health degree score according to the optimal weights and a weighted sum of the health degree score index data of the industrial equipment.

[0065] The application is particularly suitable for single-user industrial application scenarios, only a small amount of data is required, and only individual parameters need to be adjusted to continuously optimize a health score model, so that the resulting health score result is more in line with the subjective cognition of the user, and the accuracy of the health evaluation of the industrial equipment is improved.

[0066] The industrial equipment health evaluation method provided by the application is suitable for scenarios in various industrial fields that need to evaluate the health status of equipment or systems, and mainly includes the following fields:

[0067] Manufacturing industry: health evaluation of key equipment such as numerical control machine tools, industrial robots, machining centers, etc.

[0068] Energy industry: condition monitoring and health management of generator sets, fans, compressors, power transmission and transformation equipment, etc.

[0069] Chemical industry: health degree analysis of equipment such as reaction kettles, pressure vessels, and pipeline systems to prevent accidents such as leaks and explosions.

[0070] Metallurgical industry: health status evaluation of blast furnaces, rolling mills, smelting equipment, etc. to ensure production continuity.

[0071] Transportation: health monitoring of key components such as traction motors and gearboxes in rail transit vehicles, and aeroengines.

[0072] If it is a five-axis linkage numerical control machine tool in the manufacturing industry equipment, the health score index data includes: spindle speed stability, spindle bearing temperature, hydraulic system oil viscosity, feed shaft vibration intensity, and positioning accuracy; if it is a blast furnace blower in the metallurgical industry equipment, the health score index data includes: fan efficiency, bearing temperature, vibration intensity, lubricating oil viscosity, outlet air pressure, motor current deviation rate, impeller dynamic balance deviation, and cooling water temperature.

[0073] The following will be illustrated with the specific application and data of a five-axis linkage numerical control machine tool:

[0074] Figure 2 is a flowchart of the health degree weight optimization method illustrated by the embodiment of the application, with reference to Figure 2 , the method comprises:

[0075] First, how to establish a health degree analysis model based on the mechanism of health degree analysis of equipment or systems in the industrial field is introduced.

[0076] Based on mechanism modeling refers to the user explicitly knows the health score of the evaluation index and its hierarchy, and can be established according to the different level of index membership health score model. In this case, the user should be defined and operated by the user operation module to complete the construction of health score model, health score model can be used in the form of tree to intuitively display to the user, tree model example as shown in Figure 3

[0077] In the calculation module, the user-defined health score model is converted into a form that can be calculated by a computer for use in the weight optimization method described later. Next, how to solve this model is introduced, and the corresponding mathematical derivation is carried out:

[0078] In order to determine the weight of each index, several proper nouns are defined as follows, and a five-axis linkage CNC machine tool in a factory is taken as an example to illustrate:

[0079] (1) Terminal index, refers to the health score index that cannot be further decomposed into sub-indexes, which is represented in the tree structure of the model as the index at the lowest level without branches. In modeling, the health score value of the terminal index is marked as , the total number of terminal indexes is R, and the ranking of each terminal index is r, . For example Figure 4 In the model, there are 5 terminal indexes, and the health scores of these terminal indexes are , , , and . The way to obtain the health score of the terminal index is various, for example, the vibration intensity is scored according to the international standard, the temperature is referred to the normal range and the fluctuation, the performance is compared with the design parameters to see the decline, and the failure rate is counted by the number of faults and the maintenance time. Without considering how to obtain a terminal index health score, it is assumed that all terminal index health scores have been obtained through the existing calculation method and the collected data, and the invention only needs to use these scores for subsequent calculation.

[0080] (2) Node, is a health score index that can be further decomposed into sub-indexes, which is represented in the tree structure diagram of the model as a health score index with branches below, for example Figure 4 The index a in can be considered as a node, because it can be further decomposed into sub-indexes and .

[0081] (3) Branch, is a sub-index that is decomposed from a node, in the tree structure diagram of the model, if a node can be further decomposed into sub-indexes, then the sub-index can be called the branch of the node, for example Figure 4 ​Mid-terminal indicators and These are the two branches of node indicator 'a'.

[0082] (4) Sub-weights refer to the strength of the linear relationship between a node and its sub-indicators, and the sum of the weights of multiple sub-indicators under the same branch is 1. For example Figure 4 Two sub-indicators were divided at node a. and Their sub-weights are respectively and ,and and Satisfying Relationship: The subscript under the subweight indicates the number of its preceding node and the current branch, while the superscript indicates the linear relationship between which level in the tree structure.

[0083] for Figure 5 Taking the health analysis of a five-axis CNC machine tool in a factory as an example, based on the mechanism, the health status of a five-axis CNC machine tool can be evaluated from aspects such as the spindle system and feed system. Each aspect includes several specific indicators, forming a clear hierarchical structure. Users define and construct the following tree-like health scoring model in the user operation module: Spindle system (node ​​A), branches include spindle speed stability (terminal indicator a1), spindle bearing temperature (terminal indicator a2), and hydraulic system oil viscosity (terminal indicator a3), with their sub-weights being... 0.2 0.3 0.5; Feed system (node ​​B), branches are feed axis vibration intensity (terminal index a4) and positioning accuracy (terminal index a5), their weights are respectively 0.5 0.5. The sub-weight of node A is Sub-weight of node B The above sub-weights satisfy the following relationship: , and In this model, the terminal indicators are the bottom-level indicators without branches in the tree structure, such as spindle speed stability and positioning accuracy of each axis. Their health scores in this example are a1, a2, a3, a4, and a5, with a total number of R=5.

[0084] (5) Weight, in the construction of the mathematical model, refers to the strength of the linear relationship between the terminal indicator or node and the overall health score. The terminal indicator is defined as follows: The corresponding weight is ,For example Figure 4 Mid-terminal indicators The weight is The weight of the node index a is .

[0085] (6) The weight path is a link formed by the sub-weight connection between each layer from the comprehensive health degree to a terminal index. For example, in the terminal index Figure 4 , its weight path is to to . L is the total number of layers of the model. For irregular health degree scoring models, the number of layers of any weight path is not necessarily L. It can only be said that the maximum is L. It is worth noting that the weight path of any terminal index conforms to the chain rule and can be described by a mathematical expression in the form of mapping: . Wherein represents the sub-weight between the -1 layer and the l layer on the weight path of l . For example, in Figure 4 , the total number of layers L is 4, and two different weights and are used to illustrate the calculation method of the weight, wherein passes through 4 layers, but passes through only 3 layers. In Figure 4 :

[0086]

[0087] The mapping relationship is: in the expression, , and correspond to the sub-weights on the weight path in Figure 4 . Similarly:

[0088]

[0089] In the expression, and correspond to the sub-weights and on the weight path in Figure 4 .

[0090] ​By the weight value of the terminal index, the weight value of the upper layer is deduced step by step until all the sub-weights are obtained. The reasoning order needs to start from the terminal index, and first needs to find the terminal index of a certain weight path. Then the upper node corresponding to the terminal index (or branch) is found, and the weights (or sub-weights) of all terminal indexes (or branches) under the same node are added, that is, the sub-weight of the upper layer node is obtained. The derivation is carried out step by step until all the sub-weights are obtained.

[0091] For example Figure 4 In the first step, the node of the third layer, index a, and its branch, index and , whose weight relationship satisfies the following formula:

[0092]

[0093]

[0094]

[0095] Therefore, the weight of the node, index a, is deduced , and the sub-weights of the fourth layer are and . Further, the node of the second layer, preliminary health A, because it has only one branch, index a, in the third layer, the sub-weights of the second and third layers are , and the weight of the node of the second layer is . Using this method, and extending to other weight paths, all the sub-weights in this model, as well as the weights of all nodes and terminal indexes, can be deduced.

[0096] For example, Figure 5 Taking the health degree analysis of a five-axis linkage numerical control machine tool in a factory as an example, the weight path of index a1 relative to the comprehensive health degree of the five-axis linkage numerical control machine tool is , 0.2, the weight of index a1 relative to the comprehensive health degree of the five-axis linkage numerical control machine tool can be calculated as . The weights of other indexes a2, a3, a4 and a5 can also be calculated by this method:

[0097]

[0098]

[0099]

[0100]

[0101] Finally, after the model is built, all that is needed is the total number R of terminal indicators to construct a health assessment model and obtain a comprehensive health score. :

[0102]

[0103] For example Figure 5 In this context, the weight model based on mechanism modeling can be expressed as:

[0104]

[0105] For example, Figure 5 The overall health score of a five-axis CNC machine tool in a certain factory in China is:

[0106]

[0107] This mechanism-based modeling approach, through a clear hierarchical structure of indicators and mathematical derivation, can accurately reflect the degree of influence of each part of the equipment on the overall health, providing a scientific basis for equipment condition monitoring and maintenance decisions. In practical industrial applications, corresponding health scoring models can be flexibly defined and constructed according to different types of equipment and specific analytical needs, and the weights can be solved using the above method to achieve accurate assessment of equipment health.

[0108] Based on the above mathematical derivation, the following principle is established: knowing the weights of all terminal indicators allows us to determine the sub-weights of all weight paths and the weights of each indicator in the entire model. Therefore, it is necessary to use a dataset collected under real-world conditions to solve for the weights of all terminal indicators. Given a model as follows:

[0109]

[0110] Rewritten in matrix form:

[0111]

[0112] In this model, there are a total of R terminal indicators, so the weights will be... If we consider each variable as a separate variable, then there are R variables. Therefore, to calculate this weight matrix, we need R linearly uncorrelated data points, each containing health scores for R terminal indicators. These data points are combined into a data set matrix A of rank R, in its complete form:

[0113]

[0114] in, This represents the health score of the r-th terminal indicator in the j-th data set. ; The comprehensive health degree score is obtained according to the jth data and the weight The left and right sides of the equation are multiplied by an inverse matrix of A

[0115]

[0116]

[0117]

[0118] According to the above derivation, to determine a model weight, a set of data sets are required, wherein each data contains the score value of each terminal index, and the rank of the matrix composed of the data set is R. Therefore, when making the data set, the collected data is grouped, and the Gaussian elimination method is used in each group to make N linearly independent data sets with a rank of R for standby. The N data sets are uniformly represented in the following format, wherein the superscript n represents the nth data set in the N data sets, and the mathematical form is as follows:

[0119]

[0120] In the formula, represents the nth data set, ; represents the health degree score value of the rth terminal index in the jth data in the nth data set, .

[0121] Taking a five-axis linkage numerical control machine tool of an automobile parts factory in Figure 5 , a health degree model needs to be made for it, which is used for subsequent optimization of terminal index weight. Ten data are collected from the machine tool every day for 30 days, and a total of 10x30=300 data are obtained. In order to facilitate understanding, the health degree score of each index is directly shown in Table 1, and the health degree score based on the collected time series data is shown, as described above, and the process of converting time series data into specific scores is omitted.

[0122] Table 1: Example of terminal index health degree score data (part)

[0123]

[0124] Figure 6 is a flowchart of how to make a qualified data set according to the embodiment of the present application, and the specific data set making method is as follows: Figure 6

[0125] Step A: From the existing data, select data and generate a data set A containing Q (and Q ) data. ​​​

[0126] For example, randomly select Q=10 from 300 data (Q≥R=5) to form the initial version of the first data set , The health score value of the rth terminal index in the jth data is represented, and the matrix form is (in order to facilitate the reader to understand, it is presented in the form of each data, not directly given in the form of matrix, the following is the same):

[0127]

[0128] Step B: Verify whether the rank of the Q data is R by elementary row transformation (row exchange, multiplication, and addition). At the same time, it can be determined which data are linearly related and which data are linearly unrelated. If there are NLC linearly unrelated data.

[0129] For example, the data set is as follows:

[0130]

[0131] For each data in , elimination is performed by elementary row transformation (row exchange, multiplication, and addition). The following are the specific steps of the elementary row transformation process (column-by-column elimination) of the matrix:

[0132] (1) Perform the selection and elimination of the first column as the main element row, select row 1 (the first row with the largest non-zero element) as the main element row, and perform row transformation on rows 2 to 10: rowj = rowj - (rowj[column1] / row1[column1]) x row1. After transformation, the column 1 of rows 2 to 10 becomes 0 or approximately 0, for example, row 2 after transformation:

[0133] (2) Continue to select and eliminate the second column as the main element row. The value of column 2 of row 3 (90) is the largest, which is selected as the new main element row, and rows 2, 4 to 10 are transformed: rowj = rowj - (rowj[column2] / row3[column2]) x row3. After transformation, the column 2 of row 2 becomes 0, and the column 2 of row 4 becomes: 70 - (70 / 90) x 90 = 0

[0134] (3) Continue to select and eliminate the third column as the main element row. The value of column 3 of row 4 (92) is the largest, and the elimination is performed: rowj = rowj - (rowj[column3] / row4[column3]) x row4, and the column 3 of row 5 becomes: 90 - (90 / 92) x 92 = 0

[0135] (4) Continue to select and eliminate the fourth column as the main element row. The value of column 4 of row 5 (73) is non-zero, and after elimination, the column 4 of row 8 becomes: 67 - (67 / 73) x 73 = 0

[0136] (5) Continue to perform the main element row selection and elimination on the fifth column. Row 8 column 5 value (89) is nonzero, and after elimination, row 9 column 5 becomes: 92 - (92 / 89) x 89 = 0

[0137] (6) Finally, after row transformation, the nonzero rows (linearly independent rows) are: row 1 (data group 1), row 3 (data group 3), row 4 (data group 4), row 5 (data group 5), row 8 (data group 8), row 9 (data group 9), = 6; the zero rows (linearly dependent rows) are row 2, 6, 7, and 10.

[0138] Step C: According to the number of linearly independent data, the following operations are performed:

[0139] (1) If , it means that the current data group meets the requirements. In the data group, only one of each pair of linearly dependent multiple data is retained, and the rest are deleted. After completion, a qualified data group is obtained, which contains R data.

[0140] (2) If , it means that some linearly independent data need to be removed, and the number of linearly independent data to be deleted is . At the same time, in the data group, only one of each pair of linearly dependent multiple data is retained, and the rest are deleted. After completing these two deletion operations, a qualified data group is obtained, which contains R data.

[0141] (3) If , linearly independent data need to be supplemented, and the number of linearly independent data to be supplemented is at least . After supplementation, return to step B.

[0142] Continue the above list, because the , row 9 (data group 9) should be deleted, and rows 1 (data group 1), 3 (data group 3), 4 (data group 4), 5 (data group 5), and 8 (data group 8) should be retained to form the final first group of data groups . Here, it is assumed that row 9 (data group 9) is deleted, and rows 1 (data group 1), 3 (data group 3), 4 (data group 4), 5 (data group 5), and 8 (data group 8) are retained, and finally as follows:

[0143]

[0144] Step D: Repeat the above steps A to C to continue to make the second group, the third group, etc. until a sufficient number of data groups are generated. At this time, the number of data groups is N, and these N data groups constitute the required data training set.

[0145] In the example, the above steps are repeated to continue creating the second and third datasets from the remaining 290 data points, until the user obtains a satisfactory number of datasets.

[0146] Traditional entropy weighting methods can only normalize and determine the weights of extremely large and extremely small indicators. This invention proposes an extended entropy weighting method, addressing the situation where four main indicator types may coexist in a health scoring model under real-world conditions. The four indicators are extremely large, extremely small, intermediate, and interval indicators, and their weights are determined objectively. Descriptions and examples are shown in Table 2.

[0147] Table 2 Indicator Types

[0148]

[0149] against Figure 5 The five-axis CNC machine tool in a certain automotive parts factory needs to have its initial weights determined using the extended entropy weight method to provide an objective basis for subsequent optimization of the health index weights; Table 3 shows the index types for the five-axis CNC machine tool.

[0150] Table 3. Types of Indicators for Five-Axis CNC Machine Tools

[0151]

[0152] The specific steps for determining the initial weights using the extended entropy weight method are as follows:

[0153] Assuming the first set of data There are R indicators and R data points, among which For the value of the j-th data point under the r-th indicator, that is, the data in the j-th row and r-th column, :

[0154]

[0155] Step a: Standardize each indicator sequentially. Let the r-th terminal indicator be... hour, The original data for data j under index r. It is the minimum value of data j. It is the maximum value of data j. For standardized data, then:

[0156] Extremely large indicators:

[0157] Minimal Indicators:

[0158] Intermediate indicators:

[0159] wherein, is the best value of the jth data, the maximum gap value of the data j .

[0160] Interval type index: wherein, the best interval is , the maximum distance from the best interval to the boundary of the best interval .

[0161] Take the first set of data made , for example, take the original data value of the first set of data (the first row) as an example to illustrate the process of standardizing each terminal index, and Table 4 shows the first row standardization process of the first set of data of the five-axis linkage CNC machine tool:

[0162] Table 4 First row standardization process of the first set of data of the five-axis linkage CNC machine tool

[0163]

[0164] According to the steps described in the above table, the original data standardization of the second, third, fourth and fifth sets of data in is completed in turn. After standardization, the data of each terminal index in the five sets of data is composed of the following arrays (columns of several data):

[0165] = [0.92, 0.85, 0.90, 0.88, 0.95]

[0166] = [0.4167, 0.5333, 0.3, 0.5, 0.5833]

[0167] = [0.3, 0.7, 0.5, 1.0, 0.4]

[0168] = [0.9077, 0.9692, 0.9846, 0.8923, 0.9846]

[0169] = [0.7, 0.75, 0.6, 0.8, 0.82]

[0170] Step b: Calculate the proportional coefficient of the sample under each index. is the proportional coefficient of the jth data under the rth terminal index, For the standardized data:

[0171]

[0172] The terminal index standardization value Take the hydraulic oil viscosity as an example, the sum of the standardization values is: 0.3 + 0.7 + 0.5 + 1.0 + 0.4 = 2.9, and the proportion coefficient in each group of data is , , , , .

[0173] Step c: Calculate the information entropy of each index in turn. The information entropy of the rth index is:

[0174]

[0175] wherein is a constant for standardizing the entropy value. When , define .

[0176] Take the terminal index standardization value x3 (hydraulic oil viscosity) as an example, R = 5, r = 3, , and calculate . Similarly, the information entropy of the five terminal indexes is calculated in turn as: , , , , .

[0177] Step d: Calculate the weight of each index in turn. According to the information entropy, the weight of the jth index is calculated as:

[0178]

[0179] Then the initial weight is obtained as .

[0180] Taking the above example, the total entropy value is: = 0.98 + 0.85 + 0.918 + 0.89 + 0.87 = 4.508

[0181] = (1-0.98) / (5-4.508) ≈ 0.0407

[0182] = (1-0.85) / (5-4.508) ≈ 0.305

[0183] = (1-0.918) / (5-4.508) ≈ 0.167

[0184] = (1-0.89) / (5-4.508) ≈ 0.224

[0185] = (1-0.87) / (5-4.508) ≈ 0.264

[0186] Initial weight .

[0187] The mathematical principle of weight optimization includes:

[0188] (1) Using the prepared data set to optimize the weight, set:

[0189]

[0190]

[0191] In the formula, represents the weight coefficient generated after the nth iteration calculation, when n is 0, it means the initial weight. Record the initial weight as . The initial weight is determined by the objective method, and the detailed method will be introduced later. represents the health degree score value of the rth terminal index in the nth data group, ; represents the comprehensive health degree score value of the jth data in the nth data group, .

[0192] After the data set is prepared, the first data group is selected, and the preliminary health degree score of the first group of data is calculated according to the initial weight :

[0193]

[0194] Take the result calculated by the five-axis linkage numerical control machine tool , as an example, get .

[0195] (2) Adjust the preliminary comprehensive health degree score obtained from the first data group . For a certain preliminary comprehensive health degree score , the adjustment amount is , and the optimized comprehensive health degree score is :

[0196]

[0197] The optimized comprehensive health degree score is , and the adjustment amount is Therefore, the derivation can be obtained:

[0198]

[0199] For example, the 4th data (main shaft bearing temperature 62 points, vibration intensity 73 points) is abnormal, but the preliminary comprehensive health degree score 79.00 points does not fully reflect the risk, and is adjusted to 70.00 points, the adjustment amount , and other data is not adjusted, so the adjustment amount , the optimized comprehensive health degree score is .

[0200] (3) Using the subjective adjustment method, the optimized comprehensive health degree score and the data group can derive the new weight after the first iteration .

[0201] Because: Therefore, the inverse matrix of is can be obtained:

[0202]

[0203]

[0204]

[0205]

[0206] In this way, the optimization from the initial weight to the weight after the first iteration is completed.

[0207] Continue with the example,

[0208] The weight after the first iteration is:

[0209] =

[0210] Such iteration also has significance in industry. After adjustment, the comprehensive health degree score model significantly improves the attention to the main shaft speed stability and the hydraulic oil viscosity, and reduces the dependence on vibration intensity, which is more in line with the fault characteristics of the main shaft bearing temperature abnormality.

[0211] (4) Continue with the second set of data with the weights after the first iteration , repeat the above steps for the second iteration of weights as follows:

[0212] First calculate the preliminary overall health score for the second set of data :

[0213]

[0214] Continue with the weights after the first iteration , adjust the preliminary overall health score for the second set of data . For one of the preliminary overall health scores , the adjustment is , and the optimized overall health score is :

[0215]

[0216] Let the optimized overall health score be , and the adjustment be , so we have:

[0217]

[0218] Finally, using the optimized overall health score and the second set of data , we can derive the new set of weights after the second iteration . This completes the optimization from the weights after the first iteration to the weights after the second iteration .

[0219] (5) Using induction, generalize the above process to the nth set of data :

[0220]

[0221]

[0222] where represents the preliminary overall health score for the nth set of data using the weights after the (n-1)th iteration . When n is 1, it refers to the score generated for the first set of data using the initial weights . . ​represents the n th iteration, based on the n th data set generated and adjustment amount , the generated optimized comprehensive health score.

[0223] Figure 7 is a flowchart of the weight optimization method shown in the embodiments of the present application, with reference to Figure 7 , the algorithm for continuously iterating the comprehensive health score model weight W, including:

[0224] Step 1: the user sets a maximum number of iterations e, and .

[0225] Step 2: in the n th iteration, multiply the n th data set and the weight generated in the n-1 th iteration to obtain the preliminary comprehensive health score . When n = 1, the initial weight is used .

[0226] Step 3: using the subjective evaluation method, generate the adjustment amount of the preliminary comprehensive health score , and generate the optimized comprehensive health score .

[0227] Step 4: using the n th data set and the optimized comprehensive health score , using the formula , calculate the weight after the n th iteration.

[0228] For , , , and five parameters:

[0229]

[0230]

[0231]

[0232]

[0233] Therefore, the formula is simplified as: Therefore, the core of using this method is to find the adjustment amount of the n th data set . For this purpose, the present application proposes an evaluation adjustment method to generate the adjustment amount in step 2 This method is suitable for the case that "the experience of the expert is not enough to directly give the accurate health degree score, but can give the evaluation of the optimization direction of the health degree score". This method needs to be set by the user, and the setting items include "health degree score optimization direction" and "health degree score optimization ratio" corresponding to "health degree score".

[0234] For example, taking the above five-axis linkage numerical control machine tool comprehensive health degree score as an example, the "health degree score optimization ratio" is set to "conform", "score too large" and "score too small" three gears. For the "conform" option, it means that the current score meets the requirements and does not need to be changed. For the "score too large" option, it means that the score should be reduced in the reviewer's mind, so the "health degree score optimization direction" is reduced, and the "health degree score optimization ratio" is set to 95% of the preliminary comprehensive health degree score. Similarly, for the "score too small" option, it means that the score should be increased in the reviewer's mind, so the "health degree score optimization direction" is increased, and the "health degree score optimization ratio" is set to 105% of the preliminary comprehensive health degree score. For the above calculated preliminary health degree score The score personnel selects "conform", "conform", "score too large", "score too small", and "conform" for the five health degree comprehensive scores, respectively, and the adjusted comprehensive health degree score is The adjustment amount is .

[0235] It is worth noting that the "health degree score result optimization ratio" set by the user is similar to the concept of learning rate in machine learning, and should be reasonably designed. If not properly set, it will adversely affect the optimization effect of the weight. When the ratio is set too large, it may always fail to get a result that the user is satisfied with, and in extreme cases it may even cause the weight parameter to update too much, eventually causing the model training process to diverge. When the ratio is set too small, the step of model parameter update will be very small, which will cause the model training process to be very slow and require more iterations to get a result that the user is satisfied with. In order to prevent the above two cases, an iteration number needs to be set, and at the same time, an evaluation index needs to be set to evaluate the optimization result.

[0236] Step 5: Determine whether the iteration is completed.

[0237] If , the iteration is not completed, n = n + 1, and return to step 2 to continue iteration.

[0238] At the same time, if But the user has reached a more satisfactory degree of optimization results, think that can stop iteration, can also directly complete the iteration.

[0239] If , complete iteration.

[0240] After completing the iteration, the optimization weight obtained in the iteration process is output

[0241] Step 6: Select the optimal weight. Set an evaluation index To judge the pros and cons of the optimization weight:

[0242]

[0243] In the formula, The n-th iteration, The number of The number of The index describes the degree of satisfaction of the generated comprehensive health score of each data set. The larger the index, the closer the estimated value to the actual value, and the better the result. According to the index result, the optimal weight is set as

[0244]

[0245] For example: e=3 iterations, 12 data, the number of satisfaction is 4, 3, 5 respectively, then , , Because is the largest, so = .

[0246] Step 7: export to the health score model and deploy. The model can be deployed to specific projects through the calculation module. That is, the collected data can be used to score the health degree.

[0247] The above combines Figure 1 , Figure 2 to introduce the industrial equipment health evaluation method provided by the embodiment of the application in detail. Next, the industrial equipment health evaluation system provided by the embodiment of the application will be introduced with reference to the accompanying drawings.

[0248] Figure 8 It is the structure diagram of the industrial equipment health evaluation system shown in the embodiment of the application. With reference to Figure 8 , the system described in the application comprises:

[0249] The data acquisition module is configured to acquire health score index data of the industrial equipment and construct a health score model.

[0250] The initial weight calculation module is configured to construct a health score index data set of the industrial equipment and calculate initial weights of the health score index data by using an extended entropy weight method.

[0251] The health score calculation module is configured to calculate a preliminary health score according to the initial weights of the health score index data and the health score index data of the industrial equipment by using the health score model.

[0252] The weight optimization module is configured to generate an adjustment amount of the preliminary comprehensive health score and an optimized health score by using the health score index data set of the industrial equipment, to iteratively optimize the weights, and to obtain optimal weights.

[0253] The output module is configured to obtain a final health score according to the optimal weights and a weighted sum of the health score index data of the industrial equipment.

[0254] The industrial equipment health evaluation system according to the embodiment of the present application can correspond to the method described in the embodiment of the present application, and the above and other operations and / or functions of each module of the industrial equipment health evaluation system are respectively to realize the corresponding process of each method in the embodiment of the present application, and for the sake of brevity, will not be repeated here. Figure 1

[0255] Referring to the structural diagram of the computer device shown in Figure 9 The computer device includes a processor, a communication interface and a computer readable storage medium. The processor, the communication interface and the computer readable storage medium can be connected through a bus or other means. The communication interface is used to receive and send data. The computer readable storage medium can be stored in the memory of the computer device, and the computer readable storage medium is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer readable storage medium. The processor (or CPU (Central Processing Unit, central processor)) is the computing core and control core of the computer device, which is suitable for implementing one or more instructions, and is particularly suitable for loading and executing one or more instructions to realize the corresponding steps in the embodiment of the industrial equipment health evaluation method.

[0256] ​The embodiment of the present application provides a computer readable storage medium (Memory), which is a memory device in a computer device, and is used for storing programs and data. It can be understood that the computer readable storage medium herein can include a built-in storage medium in the computer device, and of course can include an extended storage medium supported by the computer device. The computer readable storage medium provides a storage space, and the storage space stores a processing system of the computer device.

[0257] In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and the instructions can be one or more computer programs (including program codes). It should be noted that the computer readable storage medium herein can be a high-speed RAM memory, or a non-volatile memory such as at least one disk memory; optionally, the computer readable storage medium can be at least one computer readable storage medium located away from the aforementioned processor.

[0258] In one embodiment, the computer readable storage medium stores one or more instructions; the processor loads and executes the one or more instructions stored in the computer readable storage medium, so as to realize the corresponding steps in the industrial equipment health degree evaluation method embodiment.

[0259] The embodiment of the present application provides a computer program product or a computer program, and the computer program product or the computer program includes computer instructions stored in a computer readable storage medium. The processor of the computer device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions, so that the computer device executes the corresponding steps in the industrial equipment health degree evaluation method embodiment.

[0260] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system or a computer program product. Therefore, the present application can be in the form of a hardware embodiment, a software embodiment or an embodiment combining software and hardware aspects. Moreover, the present application can be in the form of a computer program product implemented on one or more computer readable storage media (including but not limited to disk storage and optical storage) containing computer usable program codes.

[0261] The above only describes the preferred embodiments of the present application and is not used to limit the present application. For those skilled in the art, the present application can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. An industrial equipment health evaluation method, characterized by, The method comprises the following steps: obtaining health score index data of industrial equipment, and constructing a health score model; the industrial equipment comprises manufacturing equipment, energy industry equipment, chemical industry equipment, metallurgical industry equipment and transportation industry equipment; if the industrial equipment is a five-axis linkage numerical control machine tool in the manufacturing equipment, the health score index data comprises spindle speed stability, spindle bearing temperature, hydraulic system oil viscosity, feed shaft vibration intensity and positioning accuracy; if the industrial equipment is a blast furnace blower in the metallurgical industry equipment, the health score index data comprises fan efficiency, bearing temperature, vibration intensity, lubricating oil viscosity, outlet air pressure, motor current deviation rate, impeller dynamic balance deviation and cooling water temperature; constructing a health score index data set of the industrial equipment, and calculating initial weights of the health score index data by using an extended entropy weight method; the method for calculating the initial weights of the health score index data by using the extended entropy weight method comprises the following steps: standardizing the health score index data of each industrial equipment to obtain maximum indicators, minimum indicators, intermediate indicators and interval indicators, so as to construct a data group of the health score index of each industrial equipment; In the first set of data There are R indicators and R pieces of data, where is the value in the jth data under the rth indicator, i.e., the data in the jth row and rth column, : Each index is standardized in turn, and let the rth terminal index be standardized as Time, is the original data of data j under the rth index, is the minimum value of data j, is the maximum value of data j, is the standardized data, then: very large scale: Miniature indicator: Intermediate indicators: wherein is the optimal value of the data j, the maximum difference value ; Interval type index: wherein the best interval for is , taking the maximum distance to the boundary of the best interval calculating proportional coefficients of each health score index according to the data group of the health score index of each industrial equipment; calculating information entropy of each health score index according to the proportional coefficients of each health score index; information entropy of the rth index: wherein is a constant used to normalize the entropy values, when is defined as ; calculating initial weights of each health score index data according to the information entropy of each health score index and the total number of health score indexes; calculating the weight of the jth index according to the information entropy: Thus, the initial weights are obtained as ; calculating a preliminary health score by using the health score model according to the initial weights of the health score index data and the health score index data of the industrial equipment; generating an adjustment amount of the preliminary health score by using the health score index data set of the industrial equipment, so as to iteratively optimize the initial weights and obtain optimal weights; obtaining a final health score according to the optimal weights and a weighted sum of the health score index data of the industrial equipment; the method for generating the adjustment amount of the preliminary comprehensive health score and the optimized health score, and iteratively optimizing the weights to obtain the optimal weights, comprises the following steps: generating the adjustment amount of the preliminary comprehensive health score according to the preliminary comprehensive health score, and performing an iterative optimization process; in the iterative process, a maximum number of iterations is set by a user, a preliminary comprehensive health score of the current iteration number is generated according to the weights of the last iteration number and the health score index data set of the industrial equipment of the current iteration number, an adjustment amount of the health score of the current iteration number is generated according to the preliminary comprehensive health score of the current iteration number, and an optimized comprehensive health score of the current iteration number is calculated; then, the weights of the current iteration number are calculated according to the health score index data set of the industrial equipment of the current iteration number and the calculated optimized comprehensive health score. The adjustment amount is generated by an evaluation adjustment method, which is set by the user, and the setting items include the health score optimization direction and the health score optimization proportion corresponding to the health score; After the iteration is completed, the optimal weight is selected from the weights optimized by all iteration times according to the evaluation index; The algorithm for continuously iterating the comprehensive health score model weight W includes: Step 1: User sets a maximum number of iterations e, and ; Step 2: In the nth iteration, use the nth set of data groups and multiply by the weight generated at the (n-1)th iteration to obtain a preliminary comprehensive health score When n = 1, use the initial weight ; Step 3: Generate preliminary overall health score using subjective evaluation method adjustment amount and generate optimized overall health score ; Step 4: Utilize the nth set of data groups and the optimized overall health score , using the formula to calculate the weights after the nth iteration ; For , , , and five parameters: The equation simplifies to: ; Step 5: judging whether the iteration is completed; If then the iteration is not complete, n = n + 1, and the process returns to step 2 to continue the iteration; Or, if the user approves the optimization result, the iteration is completed directly; If , the iteration is completed, and after the iteration is completed, the optimization weight obtained in the iteration process is output ; Step 6: Select the best weight; set an evaluation index The evaluation index used to judge the optimization weight In the formula, for the nth iteration, are evaluated as satisfactory the number of According to the index result, set the optimal weight as ; Step 7: Export to the health score model and deploy. Exported into the health score model and deployed.

2. The industrial equipment health assessment method of claim 1, wherein The initial weight of the health score index data is obtained by multiplying the weights of the multi-level upper nodes of the health score index data.

3. The industrial equipment health assessment method of claim 1, wherein The method for constructing the health score index data set of the industrial equipment includes: Selecting Q pieces of health score index data of industrial equipment to construct a data matrix; Performing elementary row transformation on the data matrix to verify whether the rank of the data matrix is R, determine which data are linearly correlated and which data are linearly uncorrelated; If the number of uncorrelated data is equal to R, only one piece of data in each pair of linearly correlated data in the data matrix is retained, and the rest is deleted; if the number of uncorrelated data is greater than R, the linearly uncorrelated data exceeding R are deleted, only one piece of data in each pair of linearly correlated data in the data matrix is retained, and the rest is deleted; if the number of uncorrelated data is less than R, the linearly uncorrelated data is supplemented to R pieces; Repeat the above steps to construct multiple data matrices to obtain the health score index data set of the industrial equipment.

4. An industrial equipment health evaluation system characterized by comprising: It includes: A data acquisition module configured to acquire health score index data of industrial equipment and construct a health score model; The industrial equipment includes manufacturing equipment, energy industry equipment, chemical industry equipment, metallurgical industry equipment and transportation industry equipment; If the industrial equipment is a five-axis linkage numerical control machine tool in the manufacturing industry, the health score index data includes spindle speed stability, spindle bearing temperature, hydraulic system oil viscosity, feed shaft vibration intensity and positioning accuracy; if the industrial equipment is a blast furnace blower in the metallurgical industry, the health score index data includes fan efficiency, bearing temperature, vibration intensity, lubricating oil viscosity, outlet air pressure, motor current deviation rate, impeller dynamic balance deviation and cooling water temperature; An initial weight calculation module configured to construct a health score index data set of the industrial equipment and calculate the initial weight of the health score index data by using an extended entropy weight method; The method for calculating the initial weight of the health score index data by using the extended entropy weight method includes: The health score index data of each industrial equipment is standardized to obtain maximum indicators, minimum indicators, intermediate indicators and interval indicators to construct a data group of the health score index of each industrial equipment; each indicator is sequentially standardized, and it is assumed that the rth terminal indicator is , is the original data of the data j under the rth indicator, is the minimum value of the data j, is the maximum value of the data j, is the standardized data, and then very large scale: Miniature indicator: Intermediate indicators: wherein is the optimal value of the data j, the maximum difference value ; Interval type index: wherein the best interval for is , taking the maximum distance to the boundary of the best interval According to the data group of each health score index of the industrial equipment, the proportion coefficient of each health score index is calculated; According to the proportion coefficient of each health score index, the information entropy of each health score index is calculated; The information entropy of the rth index is: wherein is a constant used to normalize the entropy values, when is defined as ; According to the information entropy of each health score index and the total number of health score indexes, the initial weight of each health score index data is calculated; The weight of the jth index is calculated according to the information entropy: Thus, the initial weights are obtained as ; A health score calculation module configured to calculate a preliminary health score by using a health score model according to the initial weight of the health score index data and the health score index data of the industrial equipment; The weight optimization module is configured to: adopt the health degree score index data set of the industrial equipment, generate an adjustment amount of a preliminary comprehensive health degree score and an optimized health degree score, iteratively optimize the weight, and obtain an optimal weight; The output module is configured to: obtain a final health degree score according to the optimal weight and a weighted sum of the health degree score index data of the industrial equipment; The method comprises the following steps of: adopting the health degree score index data set of the industrial equipment, generating an adjustment amount of a preliminary comprehensive health degree score and an optimized health degree score, iteratively optimizing the weight, and obtaining an optimal weight; and generating the adjustment amount of the preliminary comprehensive health degree score according to the preliminary comprehensive health degree score, and performing an iterative optimization process. In the iterative process, a user sets a maximum number of iterations, generates a preliminary comprehensive health degree score of the current iteration number according to the weight of the last iteration number and the health degree score index data set of the industrial equipment of the current iteration number, generates an adjustment amount of the health degree score of the current iteration number according to the preliminary comprehensive health degree score of the current iteration number, and calculates an optimized comprehensive health degree score of the current iteration number; and then the weight of the current iteration number is calculated according to the health degree score index data set of the industrial equipment of the current iteration number and the calculated optimized comprehensive health degree score. The adjustment amount is generated by an evaluation adjustment method, and the evaluation adjustment method is set by the user. The setting items include a health degree score optimization direction corresponding to the health degree score and a health degree score optimization proportion; After the iteration is completed, the optimal weight is selected from the weights optimized in all iteration numbers according to the evaluation index; The algorithm for continuously iterating the weight W of the comprehensive health degree score model comprises the following steps: Step 1: User sets a maximum number of iterations e, and ; Step 2: In the nth iteration, use the nth set of data groups and multiply by the weight generated in the n-1th iteration to obtain a preliminary comprehensive health degree score When n = 1, use the initial weight as ; Step 3: Generate preliminary overall health score using subjective evaluation method adjustment amount and generate optimized overall health score ; Step 4: Utilize the nth set of data groups and the optimized overall health score , the weight after the nth iteration is calculated using the formula ;​ For , , , and five parameters: The equation simplifies to: ; Step 5: determining whether the iteration is completed; If then the iteration is not complete, n = n + 1, and the process returns to step 2 to continue the iteration; Or, if the user approves the optimization result, the iteration is completed directly; If , the iteration is completed, and after the iteration is completed, the optimization weight obtained in the iteration process is output ; Step 6: Select the best weight; set an evaluation index The evaluation index for judging the optimization weight In the formula, for the nth iteration, are evaluated as satisfactory the number of According to the index result, set the optimal weight as ; Step 7: Export to the health score model and deploy. Exported into the health score model and deployed.

5. A computer device, characterized in that, a processor adapted to execute a computer program; a computer readable storage medium, the computer readable storage medium storing a computer program, the computer program being executed by the processor to implement the steps in the industrial equipment health degree evaluation method according to any one of claims 1-3.

6. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, and the computer program is adapted to be loaded and executed by the processor to implement the steps in the industrial equipment health degree evaluation method according to any one of claims 1-3.

7. A computer program product, characterised in that, The computer program product comprises a computer program, and the computer program is executed by the processor to implement the steps in the industrial equipment health degree evaluation method according to any one of claims 1-3.

Images