Mechanical equipment state evaluation model construction and evaluation method, device and medium

Abstract

The application provides a kind of mechanical equipment state evaluation model construction and evaluation method, equipment and medium, belong to mechanical equipment state evaluation technical field, including: the parts of function correlation, having connection relationship and same contact type in mechanical equipment are divided into the same module;According to the hierarchical order of "part-module-mechanical equipment", the static characteristic measurement index of mechanical equipment is detected, and detection data is obtained;Based on the weighted quantitative evaluation method and / or fuzzy comprehensive evaluation method, a mathematical model for describing the relationship between each static characteristic measurement index and the state of the mechanical equipment is constructed;Based on the detection data, the weights of each static characteristic measurement index in the mathematical model are fitted and optimized to obtain a final mechanical equipment state evaluation model for evaluating the state of the mechanical equipment according to the measurement value of the static characteristic measurement index.The application effectively solves the technical problem that the effectiveness of the existing technology for evaluating the state of the mechanical equipment is limited.

Description

Technical Field

[0001] This invention relates to the field of mechanical equipment condition assessment technology, specifically to a method, equipment, and medium for constructing and assessing a mechanical equipment condition assessment model. Background Technology

[0002] In modern manufacturing, the performance and reliability of mechanical equipment play a crucial role in production efficiency and product quality. Mechanical equipment is a complex integrated system of "mechanical-electrical-magnetic-hydraulic-pneumatic" components. It has numerous components that may fail, and these components influence each other. The results of component-level and part-level condition analysis are correlated, making mechanical equipment condition assessment difficult.

[0003] In the field of mechanical equipment condition assessment, researchers have explored various methods to improve the accuracy and comprehensiveness of assessments. International research is relatively mature, with many scholars employing machine learning and data mining methods to analyze historical data and assess the condition of mechanical equipment. For example, algorithms such as neural networks and support vector machines are used to model equipment operating data, identifying failure modes and health status. Furthermore, the application of Six Sigma methods in manufacturing provides theoretical support for equipment condition assessment, emphasizing data-driven decision-making and process optimization, which helps improve the accuracy and reliability of assessments. Domestic research focuses more on combining maintenance and operational performance data, using methods such as analytic hierarchy process (AHP), grey relational analysis, and principal component analysis to construct comprehensive performance assessment models. Domestic scholars have also explored the functional relationships and structural connections between components to enhance the comprehensiveness of assessments. Regarding the selection of condition assessment indicators, international researchers generally emphasize the importance analysis of indicators, identifying those with the greatest impact on equipment condition assessment through weight allocation and sensitivity analysis. However, in China, although some research has begun to focus on the selection of condition indicators, a systematic indicator system has not yet been formed, limiting the effectiveness of assessment methods.

[0004] The construction of a condition assessment method and related indicator system for mechanical equipment needs to consider the combined influence of historical maintenance data, static characteristic measurement data, and other data. It also needs to consider the structural and functional relationships between the overall mechanical equipment and its components, as well as the relationships between components. Given the large number of parts and relatively complex structure of mechanical equipment, a condition assessment method based on static characteristic measurement indicators is needed to address these issues. Summary of the Invention

[0005] In view of this, it is necessary to provide a method, equipment and medium for constructing and evaluating a mechanical equipment condition assessment model, in order to solve the technical problem that the existing technology has limited effectiveness in assessing the condition of mechanical equipment.

[0006] To achieve the above-mentioned technical effects, in a first aspect, the present invention provides a method for constructing a mechanical equipment condition assessment model, comprising: Parts in mechanical equipment that are functionally related, have connection relationships, and have the same type of contact are grouped into the same module; The static characteristic measurement indicators of the mechanical equipment are tested in the hierarchical order of "part-module-mechanical equipment" to obtain test data. The test data includes the measured values ​​of the static characteristic measurement indicators and the corresponding state of the mechanical equipment. The static characteristic measurement indicators include indicators used to evaluate the characteristics of the whole machine or module under static conditions. A mathematical model is constructed based on the weighted quantitative evaluation method and / or fuzzy comprehensive evaluation method to describe the relationship between various static characteristic measurement indicators and the state of mechanical equipment; Based on the detection data, the weights of each static characteristic measurement index in the mathematical model are fitted and optimized to obtain the final mechanical equipment condition assessment model for evaluating the condition of mechanical equipment based on the measured values ​​of the static characteristic measurement indexes.

[0007] In some embodiments of the present invention, the step of grouping functionally related, interconnected, and contact-type parts in a mechanical device into the same module includes: The evaluation value assignment criteria for the module division criteria are formulated, and the weight coefficients are set. The module division criteria include: functional relevance criteria, contact type criteria, and connection relationship criteria. Calculate the comprehensive correlation degree between any two ordinary parts in the mechanical equipment after removing fasteners. The formula for calculating the comprehensive correlation degree includes:

[0008] in, Indicates parts , The overall correlation Indicates the strength of functional association. Indicates the strength of contact association. Indicates the degree of connection relevance. This indicates the weighting coefficient of the functional relevance criterion in the overall correlation degree, whereby the functional relevance criterion represents classifying functionally related parts into the same module. This represents the weighting coefficient of the contact type criterion in the overall correlation coefficient, whereby the contact type criterion indicates that parts within the same module have the same contact type. This represents the weighting coefficient of the connection type criterion in the overall correlation degree. The connection type criterion indicates that parts within the same module have a connection relationship and satisfy the following constraints: ; Establish a correlation matrix; Based on the degree of correlation, hierarchical clustering analysis is used to merge the two closest individuals in turn to obtain a hierarchical structure diagram. The module size and number of modules are determined based on the number and complexity of the mechanical equipment parts, and the hierarchical structure diagram is divided into several modules. After completing the modular division of ordinary parts, fasteners are added to the modules to achieve complete modular division.

[0009] In some embodiments of the present invention, the step of detecting the static characteristic measurement indicators of the mechanical equipment in a hierarchical order of "part-module-mechanical equipment" to obtain detection data includes: The various indicators of the mechanical equipment are tested according to the hierarchical order of "parts-modules-mechanical equipment" to construct a set of status assessment indicators; Based on the set of state assessment indicators, importance analysis is performed to select the static characteristic measurement indicators. The measured values ​​of the static characteristic measurement indicators are obtained by sorting them out.

[0010] In some embodiments of the present invention, the formula for constructing the state evaluation index set includes: The calculation formula for the index set of the mechanical equipment condition assessment model is as follows:

[0011]

[0012] in, Indicates the number of indicators. Indicates the number of tests. Indicates the number of samples. Indicates the first The first indicator This measurement value, Indicates the first of multiple tests The average of the indicators, Indicates the first The first sample The standard deviation of each indicator This is a set of status assessment indicators.

[0013] In some embodiments of the present invention, the step of performing importance analysis based on the set of state evaluation indicators to screen and obtain the static characteristic measurement indicators includes: The importance of the state evaluation index set is analyzed based on the random forest model, wherein the formula for calculating the importance of the state evaluation index includes: feature Importance within a single tree:

[0014] in, For a single tree, This represents the number of non-leaf nodes. For nodes The reduction in squared loss after splitting. For nodes Related features; Calculation features After determining the importance of each individual tree, the importance values ​​of all trees are summed, averaged, and then normalized to calculate the feature. Global importance value:

[0015] in, The number of trees. , Indicates the first Tree; State indicators with a global importance value greater than a preset threshold are retained as the static characteristic measurement indicators.

[0016] In some embodiments of the present invention, an initial model is constructed based on a weighted quantitative evaluation method and the detection data, including: Constructing the quality loss function As the initial model, the calculation formula is:

[0017] in, This refers to the actual value of a single state index for an individual part of a mechanical device. This refers to the standard value of a single condition index for a single part. It is a proportionality constant; The larger the value, the worse the condition of the part.

[0018] In some embodiments of the present invention, an initial model is constructed based on the fuzzy comprehensive evaluation method and the detection data, including: For those For mechanical equipment parts with various measurement indicators, an evaluation coefficient matrix is ​​established based on the importance of each indicator to the condition assessment. The calculation formula is:

[0019] In the formula, To indicate the importance of this indicator, ; The absolute value of the deviation of each measurement indicator The evaluation matrix is ​​obtained by dividing by the tolerance zone. The calculation formula is:

[0020] In the formula, = , The deviation value for each indicator; This represents the tolerance zone for this indicator; Establish a probability matrix for excellent, good, average, and poor for each indicator. The calculation formula is:

[0021]

[0022]

[0023]

[0024]

[0025] In the formula, For the first The probability that the point is optimal; For the first The probability that a point is good; For the first The probability that the point is in the middle; For the first The probability that the point is the difference; ; Establish a rating system matrix of excellent, good, average, and poor. The measurement indicators are: excellent score (a points), good score (b points), average score (c points), and poor score (d points), i.e. ; Establish matrix The score for each indicator is determined as follows:

[0026] In the formula, , Score each indicator; Combine the impact of each indicator on the condition of mechanical equipment Establish a scoring matrix for mechanical equipment parts The calculation formula is:

[0027] As a basis for measuring the condition of mechanical equipment parts.

[0028] Secondly, the present invention also provides a method for assessing the condition of mechanical equipment, comprising: Construct a mechanical equipment condition assessment model based on any one of the above methods; The measured values ​​of the static characteristic indicators are input into the mechanical equipment condition assessment model, and the assessment result of the mechanical equipment condition is determined based on the output value of the mechanical equipment condition assessment model.

[0029] Thirdly, the present invention also provides an apparatus comprising: Memory, used to store programs; A processor, coupled to the memory, is used to execute the program stored in the memory to implement the steps in the method for constructing a mechanical equipment condition assessment model or the mechanical equipment condition assessment method described in any of the above method items.

[0030] Fourthly, the present invention also provides a medium comprising: Used to store computer-readable programs or instructions, which, when executed by a processor, can implement the steps in the method for constructing a mechanical equipment condition assessment model or the mechanical equipment condition assessment method described in any of the above methods.

[0031] The beneficial effects of this invention are as follows: This invention provides a method for constructing a mechanical equipment condition assessment model. First, the parts of the mechanical equipment are divided into separate modules, which eliminates the curse of dimensionality while preserving coupling relationships and avoiding distortion of the index weights in the subsequent assessment model. Then, equipment inspection is carried out at three levels: "parts-modules-mechanical equipment," and the inspection data is acquired. Combined with the condition assessment method, a mathematical model and parameter fitting are established to obtain the final condition assessment model. While incorporating static characteristic measurement indicators into the assessment system, this invention also effectively avoids the problem of chaotic granularity of measurement objects, thereby achieving accurate condition assessment of mechanical equipment, effectively reducing assessment errors, improving equipment maintenance efficiency, reducing failure rates, and ultimately optimizing resource allocation and reducing operation and maintenance costs. This solves the technical problem of limited effectiveness of existing technologies in assessing the condition of mechanical equipment. Attached Figure Description

[0032] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0033] Figure 1 A flowchart illustrating an embodiment of the method for constructing a mechanical equipment condition assessment model provided by the present invention; Figure 2 for Figure 1A flowchart illustrating an embodiment of step S101; Figure 3 The hierarchical structure diagram of the mechanical equipment provided by this invention; Figure 4 for Figure 1 A flowchart illustrating an embodiment of step S102; Figure 5 for Figure 4 A flowchart illustrating an embodiment of step S402; Figure 6 A typical part drawing of a certain mechanical equipment provided by the present invention; Figure 7 A flowchart illustrating an embodiment of the mechanical equipment condition assessment method provided by the present invention; Figure 8 A general framework diagram of the mechanical equipment condition assessment method provided by the present invention; Figure 9 This is a schematic diagram of the structure of an embodiment of the device provided by the present invention. Detailed Implementation

[0034] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0035] In the description of the embodiments of the present invention, unless otherwise stated, "multiple" means two or more. "And / or" describes the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone.

[0036] The terms "first," "second," etc., used in the embodiments of this invention are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a technical feature defined with "first" or "second" may explicitly or implicitly include at least one of that feature.

[0037] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0038] This invention provides a method, equipment, and medium for constructing and evaluating a mechanical equipment condition assessment model, which will be described below.

[0039] Before proceeding, it should be noted that the implementing entity of this invention can be an automated information processing system. This system integrates data acquisition, correlation calculation, indicator weight analysis, and status assessment functions, completing the quantitative assessment of the static characteristics of mechanical equipment without human intervention. This invention can be applied to final inspection before leaving the factory, as well as to preventive maintenance and other scenarios.

[0040] like Figure 1 As shown, in a first aspect, the present invention provides a method for constructing a mechanical equipment condition assessment model, comprising: S101. Group functionally related, connected, and contact-type parts in mechanical equipment into the same module.

[0041] It should be noted that when formulating module division criteria, the differences between different modules should be as obvious as possible, while the differences between parts within the same module should be as small as possible.

[0042] In some embodiments of the present invention, the module partitioning criteria include functional correlation criteria and structural correlation criteria. The functional correlation criteria can ensure that the components partitioned into the same module are functionally related, thereby increasing the tightness within the module and reducing the degree of interaction between modules, achieving the purpose of low coupling and high cohesion. The structural correlation criteria refer to the spatial position and geometric relationship between components, mainly including the connection type and contact type between components. It should ensure that the connection relationship between components partitioned into the same module is as strong as possible and that the contact type is the same.

[0043] like Figure 2 Step S101 specifically includes: S201. Formulate the evaluation value assignment standard for the module division criteria and set the weight coefficient.

[0044] Preferably, the module division criteria include: functional relevance criteria, contact type criteria, and connection relationship criteria.

[0045] S202. Calculate the overall correlation between any two ordinary parts in the mechanical equipment after removing fasteners.

[0046] The formula for calculating the overall correlation degree includes: (1) In equation (1), Indicates parts , The overall correlation Indicates the strength of functional association. Indicates the strength of contact association. Indicates the degree of connection relevance. This indicates the weighting coefficient of the functional relevance criterion in the overall correlation degree, whereby the functional relevance criterion represents classifying functionally related parts into the same module. This represents the weighting coefficient of the contact type criterion in the overall correlation coefficient, whereby the contact type criterion indicates that parts within the same module have the same contact type. This represents the weighting coefficient of the connection type criterion in the overall correlation degree. The connection type criterion indicates that parts within the same module have a connection relationship and satisfy the following constraints: .

[0047] It should be noted that the method for setting the weights is mainly based on experience and subjective judgment. When there are many criteria, the weight allocation becomes more difficult. In this embodiment, the function type has the greatest impact on the difficulty of disassembly. The influence of type is greatest, followed by connection type, and then contact type.

[0048] S203. Establish the correlation matrix: From parts , The overall correlation obtained after weighted average Establish a correlation matrix : (2) In equation (2), , , correlation matrix It is a symmetric matrix, and the values ​​on the diagonal are 1, which means that the part has the greatest correlation with itself.

[0049] S204. Based on the degree of correlation, merge the two closest individuals in sequence using hierarchical clustering analysis to obtain a hierarchical structure diagram.

[0050] It should be noted that hierarchical clustering and merging aims to achieve a module division result that meets the criteria of "high functional cohesion, strong structural correlation, and reasonable module size and number".

[0051] S205. Determine the module size and number of modules based on the number and complexity of the mechanical equipment parts, and divide the hierarchical structure diagram into several modules.

[0052] S206. After completing the module division of ordinary parts, add fasteners to the modules to achieve complete module division.

[0053] When there are many parts in a mechanical device, the device can be divided into several large modules. If some large modules still contain many parts, a second division is performed until the requirements are met. Figure 3 This is a hierarchical structure diagram of mechanical equipment. The bottom layer of indivisible modules and parts are combined to form the modules or assemblies of the next layer. In this way, they are combined to form mechanical equipment. Based on this diagram, the position of each component can be found.

[0054] S102. The static characteristic measurement indicators of the mechanical equipment are tested according to the hierarchical order of "parts-modules-mechanical equipment" to obtain test data.

[0055] Preferably, the test data includes: measured values ​​of static characteristic indicators and the corresponding state of the mechanical equipment; static characteristic indicators include: indicators used to evaluate the characteristics of the whole machine or module under static conditions. These include key dimensions, surface quality, and assembly fit; key dimensions include linear dimensions, geometry, angles, etc.; surface quality includes roughness, hardness, etc.; assembly fit includes fit clearance, alignment, and assembly structural dimensions, etc.

[0056] like Figure 4 In some embodiments of the present invention, step S102 specifically includes: S401. Test various indicators of mechanical equipment in the hierarchical order of "parts-modules-complete machine" to construct a set of status assessment indicators.

[0057] Assume the sample set collected from the condition monitoring of mechanical equipment parts is as follows: In this sample set, each row represents a component, and each column represents a parameter in each testing process. Because different parameters have varying degrees of importance, some parameters are measured multiple times in different testing processes. Therefore, it is necessary to evaluate the condition of mechanical equipment components based on the stability of the mechanical equipment testing parameters. For each sample in the sample set, the standard deviation of the multiple test data for a single parameter is calculated. These calculated standard deviations are then combined into a new dataset, resulting in the fused feature dataset. .

[0058] Specifically, the calculation formula for the index set used in the mechanical equipment condition assessment model is as follows: (3) (4) In equations (2)-(3), Indicates the number of indicators. Indicates the number of tests. Indicates the number of samples. Indicates the first The first indicator This measurement value, Indicates the first of multiple tests The average of the indicators, Indicates the first The first sample The standard deviation of each indicator This is a set of status assessment indicators.

[0059] S402. Based on the set of state assessment indicators, perform importance analysis and select static characteristic measurement indicators.

[0060] Preferably, the present invention uses the random forest method for importance analysis. Specifically, by comparing and screening the importance values ​​of the state indicators of the random forest model, a suitable threshold is determined. Indicators with an importance value less than the threshold are considered to be poor state indicators and are deleted, while those with an importance value greater than the threshold are considered to be good state indicators and are retained.

[0061] like Figure 5 In some embodiments of the present invention, step S402 includes: S501. Based on the random forest model, perform importance analysis on the set of state evaluation indicators.

[0062] The formulas for calculating the importance of the status assessment indicators include: feature Importance within a single tree: (5) In equation (4), For a single tree, This represents the number of non-leaf nodes. For nodes The reduction in squared loss after splitting. For nodes Related features; Calculation features After determining the importance of each individual tree, the importance values ​​of all trees are summed, averaged, and then normalized to calculate the feature. Global importance value: (6) In equation (5), The number of trees. , Indicates the first A tree.

[0063] S502. Retain state indicators with a global importance value greater than a preset threshold as static characteristic measurement indicators.

[0064] S403. Organize and process the test data of the obtained static characteristic measurement indicators.

[0065] S103. Construct a mathematical model based on the weighted quantitative evaluation method and / or fuzzy comprehensive evaluation method to describe the relationship between various static characteristic measurement indicators and the state of mechanical equipment.

[0066] It should be noted that the core of the weighted quantitative assessment method is to quantify the condition of components through a quality loss function. The loss value is calculated directly based on the squared deviation between the actual value and the standard value; the larger the deviation, the worse the condition. Its characteristics include precise quantification, reliance on clearly defined numerical deviations, and suitability for scenarios where the indicator boundaries are clear and can be accurately measured.

[0067] The fuzzy comprehensive evaluation method introduces fuzzy functions into the weighted quantitative evaluation method. By establishing a probability matrix of excellent, good, average, and poor, it transforms quantitative deviations into fuzzy probabilities, and then calculates the comprehensive score by combining the scoring system and the indicator importance matrix. Its characteristics include handling fuzziness and uncertainty, and it is suitable for scenarios where the indicator boundaries are fuzzy or where hierarchical evaluation is required.

[0068] These two methods can be used selectively, or simultaneously, depending on the specific scenario. When the evaluation indicators have clear boundaries, high data accuracy, and require precise quantification of state differences, the weighted quantitative evaluation method can be used alone. When the evaluation indicators have fuzzy boundaries, or when it is necessary to transform the results into hierarchical conclusions to assist decision-making, the fuzzy comprehensive evaluation method can be used. When both accuracy and fuzziness need to be considered, the two methods can be used in combination.

[0069] In some embodiments of the present invention, an initial model is constructed based on a weighted quantitative evaluation method and combined with detection data, including: Constructing the quality loss function As the initial model, the calculation formula is: (7) In equation (7), This refers to the actual value of a single state index for an individual part of a mechanical device. This refers to the standard value of a single condition index for a single part. It is a proportionality constant; The larger the value, the worse the condition of the part.

[0070] The main working principle of weighted quantitative condition assessment is as follows: First, determine the condition assessment results of the part dimensions, then obtain the condition assessment results of each surface of the part after weighting, and finally obtain the final condition assessment result of the part.

[0071] In a specific embodiment, the steps for performing state assessment using a weighted quantitative method are as follows: Step 1: Conduct a quality evaluation of the part's dimensions, shape errors, and positional errors. Based on the actual machined dimensions of the part and the quality loss function, the actual quality score of the machined dimensions of the part is calculated; based on the actual shape error and position error of the part, the quality scores of the shape accuracy and position accuracy of the part are calculated using the same method.

[0072] Step 2: Calculate the quality score of the machined surface of the component. The processing of a surface may consist of multiple dimensional, shape, and positional errors. Once the dimensional, shape, and positional accuracy quality scores of the surface are obtained, the surface quality score of the component can be derived based on the importance of their impact on performance.

[0073]

[0074] The weighting coefficients for the importance of component dimensions are shown in Table 1.

[0075] Table 1 Weighting coefficients for the importance of component dimensions

[0076] in, .

[0077] Step 3: Calculate the surface quality score of the part. A component consists of multiple surfaces. After obtaining the quality score of each surface, the surface quality score of the entire component is obtained by multiplying the surface quality score by the corresponding weighting coefficient based on the degree of influence of these surfaces on the performance of the component.

[0078]

[0079] The weighting coefficients for the importance of part surfaces are shown in Table 2.

[0080] Table 2 Weighting coefficients for the importance of part surfaces

[0081] in, .

[0082] like Figure 6 The image shown is a typical part diagram of a certain mechanical equipment. Its data was obtained, and its condition was quantitatively assessed. The measured value was 100.2 mm. The measured value was 59.95 mm. The measured value is 45mm; The measured value was 20.02 mm. The measured value is 25.01 mm; the end face parallelism is 0.05 mm, and the measured value is 0.025 mm.

[0083] The relevant measurement data and their importance are shown in Table 3.

[0084] Table 3 Relevant Measurement Data and Importance

[0085] According to the formula and Figure 5 The quality scores for part dimensions and shape tolerances are calculated. Based on the formula and the importance weighting coefficients given in Tables 1 and 2, the surface quality scores for the parts are calculated and listed in Table 4.

[0086] Table 4 Surface Quality Scores of Parts

[0087] Importance weighting coefficients in Table 4 , , Calculate the part status score. Part status score = (98.3×1+90×5+91.7×3) / (1+5+3)=91.49, the status assessment result is excellent.

[0088] In some embodiments of the present invention, an initial model is constructed based on the fuzzy comprehensive evaluation method and combined with detection data, including: For mechanical equipment parts A set of measurement indicators are used to establish an evaluation coefficient matrix based on the importance of each indicator to the status assessment. The calculation formula is: (8) In equation (8), To indicate the importance of this indicator, ; The absolute value of the deviation of each measurement indicator The evaluation matrix is ​​obtained by dividing by the tolerance zone. The calculation formula is: (9) In equation (9), = , The deviation value for each indicator; This represents the tolerance zone for this indicator.

[0089] It should be noted that, The smaller the value, the better the state. According to... The value categorizes the measured index status into excellent, good, average, and poor. When the value is 0, the probability of the measurement indicator being excellent is 1, and the probability of the measurement indicator being good is 0; when When the value is 0.5, the probability of the measured index being "good" is 1, and the probability of the measured index being "excellent" is 0; when When the value is 1, the probability of the measured index being "medium" is 1, and the probability of the measured index being "good" is 0; when... When the value is 1.5, the probability of the measured index being poor is 1, and the probability of the measured index being good is 0; when When the value is greater than 1.5, the probability of the measured index being poor is 1.

[0090] Establish a probability matrix for excellent, good, average, and poor for each indicator. The calculation formula is: (10) (11) (12) (13) (14) In equation (10), For the first The probability that the point is optimal; For the first The probability that a point is good; For the first The probability that the point is in the middle; For the first The probability that the point is the difference; .

[0091] Establish a rating system matrix of excellent, good, average, and poor. The measurement indicators are: excellent score (a points), good score (b points), average score (c points), and poor score (d points), that is: (11) For example, the measurement index is 100 points for excellent, 70 points for good, 40 points for average, and 0 points for poor, that is... .

[0092] Establish matrix The score for each indicator is determined as follows: (12) In equation (12), , Score each indicator; Combine the impact of each indicator on the condition of mechanical equipment Establish a scoring matrix for mechanical equipment parts The calculation formula is: (13) Will As a basis for measuring the condition of mechanical equipment parts.

[0093] S104. Based on the detection data, the weights of each static characteristic measurement index in the mathematical model are fitted and optimized to obtain the final mechanical equipment condition assessment model for evaluating the condition of mechanical equipment based on the measured values ​​of the static characteristic measurement indexes.

[0094] like Figure 7 Secondly, the present invention also provides a method for assessing the condition of mechanical equipment, comprising: S701. Construct a mechanical equipment condition assessment model based on any one of the above methods. S702. Input the measured values ​​of the static characteristic measurement indexes into the mechanical equipment condition assessment model, and determine the assessment result of the mechanical equipment condition based on the output value of the mechanical equipment condition assessment model.

[0095] To better understand the specific embodiments of the present invention, Figure 8 A complete process for condition assessment of mechanical equipment is provided in a more specific embodiment.

[0096] like Figure 9 Thirdly, the present invention also provides an apparatus 90, comprising: Memory 910 is used to store programs; The processor 920, coupled to the memory 910, is used to execute the program stored in the memory 910 to implement the steps in the method for constructing a mechanical equipment condition assessment model or the mechanical equipment condition assessment method described in any of the above method items.

[0097] Fourthly, the present invention also provides a medium comprising: Used to store computer-readable programs or instructions, which, when executed by a processor, can implement the steps in the method for constructing a mechanical equipment condition assessment model or the mechanical equipment condition assessment method described in any of the above methods.

[0098] The foregoing has provided a detailed description of the construction and evaluation method, equipment, and medium for a mechanical equipment condition assessment model provided by the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for constructing a condition assessment model for mechanical equipment, characterized in that, include: The comprehensive correlation degree between any two ordinary parts in a mechanical device is calculated based on functional relevance, contact type, and connection relationship. The formula for calculating the comprehensive correlation degree includes: , Indicates parts , The overall correlation Indicates the strength of functional association. Indicates the strength of contact association. Indicates the degree of connection relevance. This indicates the weighting coefficient of the functional relevance criterion in the overall correlation degree, whereby the functional relevance criterion represents classifying functionally related parts into the same module. This represents the weighting coefficient of the contact type criterion in the overall correlation coefficient, whereby the contact type criterion indicates that parts within the same module have the same contact type. The corresponding weight coefficient of the connection type criterion in the comprehensive correlation degree is indicated, and hierarchical clustering analysis is performed on the parts based on the comprehensive correlation degree to classify the functionally related, connected and contact-type parts in the mechanical equipment into the same module. The static characteristic measurement indicators of the mechanical equipment are detected according to the hierarchical order of "part-module-mechanical equipment" to construct a state evaluation indicator set; the importance of the state evaluation indicator set is analyzed based on the random forest model to filter out the static characteristic measurement indicators, and the measured values ​​of the static characteristic measurement indicators and the corresponding state of the mechanical equipment are sorted to obtain the detection data. The detection data includes: the measured values ​​of the static characteristic measurement indicators and the corresponding state of the mechanical equipment. The static characteristic measurement indicators include: indicators used to evaluate the characteristics of the whole machine or module under static conditions. A mathematical model is constructed based on the fuzzy comprehensive evaluation method to describe the relationship between various static characteristic measurement indicators and the state of mechanical equipment. The mathematical model based on the fuzzy comprehensive evaluation method includes: establishing an evaluation coefficient matrix D according to the importance of each indicator to the state assessment; dividing the absolute value C of the deviation value of each measurement indicator by the tolerance zone to obtain the evaluation matrix V; establishing a probability matrix R that maps the values ​​of the evaluation matrix V to excellent, good, average, and poor; establishing a scoring system matrix U for excellent, good, average, and poor; and establishing a scoring matrix S for mechanical equipment parts by combining the calculation results of the evaluation coefficient matrix D, the probability matrix R, and the scoring system matrix U. Based on the detection data, the weights of each static characteristic measurement index in the mathematical model are fitted and optimized to obtain the final mechanical equipment condition assessment model for evaluating the condition of mechanical equipment based on the measured values ​​of the static characteristic measurement indexes.

2. The method for constructing a mechanical equipment condition assessment model according to claim 1, characterized in that, The method of grouping functionally related, interconnected, and contact-type parts in mechanical equipment into the same module includes: The evaluation value assignment criteria for the module division criteria are formulated, and the weight coefficients are set. The module division criteria include: functional relevance criteria, contact type criteria, and connection relationship criteria. Calculate the comprehensive correlation degree between any two ordinary parts in the mechanical equipment after removing fasteners. The formula for calculating the comprehensive correlation degree includes: in, Indicates parts , The overall correlation Indicates the strength of functional association. Indicates the strength of contact association. Indicates the degree of connection relevance. This indicates the weighting coefficient of the functional relevance criterion in the overall correlation degree, whereby the functional relevance criterion represents classifying functionally related parts into the same module. This represents the weighting coefficient of the contact type criterion in the overall correlation coefficient, whereby the contact type criterion indicates that parts within the same module have the same contact type. This represents the weighting coefficient of the connection type criterion in the overall correlation degree. The connection type criterion indicates that parts within the same module have a connection relationship and satisfy the following constraints: ; Establish a correlation matrix; Based on the degree of correlation, hierarchical clustering analysis is used to merge the two closest individuals in turn to obtain a hierarchical structure diagram. The module size and number of modules are determined based on the number and complexity of the mechanical equipment parts, and the hierarchical structure diagram is divided into several modules. After completing the modular division of ordinary parts, fasteners are added to the modules to achieve complete modular division.

3. The method for constructing a mechanical equipment condition assessment model according to claim 1, characterized in that, The formula for constructing the state assessment index set includes: The calculation formula for the index set of the mechanical equipment condition assessment model is as follows: in, Indicates the number of indicators. Indicates the number of tests. Indicates the number of samples. Indicates the first The first indicator This measurement value, Indicates the first of multiple tests The average of the indicators, Indicates the first The first sample The standard deviation of each indicator This is a set of status assessment indicators.

4. The method for constructing a mechanical equipment condition assessment model according to claim 3, characterized in that, The static characteristic measurement indicators are obtained by performing importance analysis based on the state assessment indicator set and filtering them, including: The importance of the state evaluation index set is analyzed based on the random forest model, wherein the formula for calculating the importance of the state evaluation index includes: feature Importance within a single tree: in, For a single tree, This represents the number of non-leaf nodes. For nodes The reduction in squared loss after splitting. For nodes Related features; Calculation features After determining the importance of each individual tree, the importance values ​​of all trees are summed, averaged, and then normalized to calculate the feature. Global importance value: in, The number of trees. , Indicates the first Tree; State indicators with a global importance value greater than a preset threshold are retained as the static characteristic measurement indicators.

5. The method for constructing a mechanical equipment condition assessment model according to claim 1, characterized in that, Based on the weighted quantitative evaluation method, an initial model is constructed using the aforementioned detection data, including: Constructing the quality loss function As the initial model, the calculation formula is: in, This refers to the actual value of a single state index for an individual part of a mechanical device. This refers to the standard value of a single condition index for a single part. It is a proportionality constant; The larger the value, the worse the condition of the part.

6. The method for constructing a mechanical equipment condition assessment model according to claim 1, characterized in that, Based on the fuzzy comprehensive evaluation method, an initial model is constructed using the aforementioned detection data, including: For those For mechanical equipment parts with various measurement indicators, an evaluation coefficient matrix is ​​established based on the importance of each indicator to the condition assessment. The calculation formula is: In the formula, To indicate the importance of this indicator, ; The absolute value of the deviation of each measurement indicator The evaluation matrix is ​​obtained by dividing by the tolerance zone. The calculation formula is: In the formula, = , The deviation value for each indicator; This represents the tolerance zone for this indicator; Establish a probability matrix for excellent, good, average, and poor for each indicator. The calculation formula is: In the formula, For the first The probability that the point is optimal; For the first The probability that a point is good; For the first The probability that the point is in the middle; For the first The probability that the point is the difference; ; Establish a rating system matrix of excellent, good, average, and poor. The measurement indicators are: excellent score (a points), good score (b points), average score (c points), and poor score (d points), i.e. ; Establish matrix The score for each indicator is determined as follows: In the formula, , Score each indicator; Combine the impact of each indicator on the condition of mechanical equipment Establish a scoring matrix for mechanical equipment parts The calculation formula is: As a basis for measuring the condition of mechanical equipment parts.

7. A method for assessing the condition of mechanical equipment, characterized in that, include: The mechanical equipment condition assessment model is constructed based on the construction method of the mechanical equipment condition assessment model as described in any one of claims 1-6; The measured values ​​of the static characteristic indicators are input into the mechanical equipment condition assessment model, and the assessment result of the mechanical equipment condition is determined based on the output value of the mechanical equipment condition assessment model.

8. A device, characterized in that, include: Memory, used to store programs; A processor, coupled to the memory, is configured to execute the program stored in the memory to implement the steps in the method for constructing a mechanical equipment condition assessment model according to any one of claims 1 to 6 or the method for assessing mechanical equipment condition according to any one of claims 7.

9. A medium, characterized in that, include: Used to store computer-readable programs or instructions, which, when executed by a processor, can implement the steps in the method for constructing a mechanical equipment condition assessment model as described in any one of claims 1 to 6 or the mechanical equipment condition assessment method as described in any one of claims 7.

Images