A Health Assessment Method for Hydraulic Gates Based on an Improved Radar Chart Approach

Abstract

A health assessment method for hydraulic gates based on an improved radar chart approach includes the following steps: Step 1, Construction of a health assessment system for hydraulic arc-shaped gates: Establishing a multi-level health assessment index system for hydraulic arc-shaped gates, including target, criterion, and indicator layers; Step 2, Indicator data acquisition: Obtaining indicator data through detection; the indicator data includes both quantitative and non-quantitative indicators; Step 3, Indicator preprocessing; Step 4, Construction of a gate health assessment model based on the improved radar chart approach. Addressing the multi-factor influence of stress, deflection, vibration, and corrosion involved in the condition assessment of hydraulic steel gates, a gate health assessment system including target, criterion, and indicator layers is proposed. A health assessment model and method for hydraulic gates combining fuzzy hierarchical analysis and an improved radar chart approach are also proposed. This method can provide intuitive evaluation results using radar graphics and quantitative evaluation values ​​using evaluation functions, achieving a comprehensive health assessment of hydraulic gates.

Description

Technical Field

[0001] This invention relates to a health assessment model for hydraulic gates and its assessment method. Background Technology

[0002] Metal equipment in hydraulic and hydropower projects is an important tool for human beings in utilizing water resources and a key facility for ensuring the safe operation of hydraulic engineering projects. Currently, a considerable portion of the existing hydraulic metal facilities in my country have reached their service life, some even far exceeding their design life. The operational status of these facilities is unclear, and some may pose safety hazards. Accidents involving hydraulic metal facilities usually have serious consequences. Conducting health status assessments of hydraulic metal structures can effectively prevent accidents and avoid blind investment.

[0003] Safety inspection and health assessment of hydraulic gates involve multiple influencing factors, including the load-bearing capacity of the steel structure, manufacturing and installation issues, natural factors such as water flow and corrosion, and human factors such as personnel and operation and maintenance management systems. Considering only one or a few of these factors can lead to biased and distorted health assessment results. Therefore, a key issue is how to comprehensively address these multiple influences, establish a steel gate health assessment system, and construct a reasonable and practical hydraulic gate health assessment model and method.

[0004] Hydraulic gates, due to their harsh working environment and complex, variable loads, are prone to structural damage and eventual failure after long-term service. Scholars have researched methods for health assessment of these gates. Yang Guangming, Zheng Shengyi, and others have elaborated on various detection and assessment methods for gates; Li Jianbin combined the analytic hierarchy process (AHP) and mathematical methods to conduct fuzzy comprehensive evaluation of gates; Wei Wenguang introduced entropy weighting and variable weighting methods to adjust weights in the comprehensive evaluation. Wang Fei, after determining the subjective and objective weights of the gate, used a game theory combination algorithm to optimize it. Existing methods mainly focus on determining the weights of various factors affecting the health status of hydraulic gates. Summary of the Invention

[0005] The technical problem to be solved by this invention is: how to effectively conduct comprehensive evaluation and assessment of the gate's status. This invention proposes to use a combination of fuzzy hierarchical analysis and radar chart method to assess the health status of the gate, and puts forward a health assessment model and method for hydraulic gates based on the improved radar chart method.

[0006] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0007] A health assessment method for hydraulic arc gates based on an improved radar chart approach includes the following steps:

[0008] Step 1: Construction of a health assessment system for hydraulic arc gates: Establish a multi-level health assessment index system for hydraulic arc gates, including target layer, criterion layer, and indicator layer;

[0009] Step 2, Indicator Data Acquisition: Obtain indicator data through detection; the indicator data includes quantitative and non-quantitative indicators;

[0010] Step 3, indicator preprocessing: (1) Indicator standardization: For quantitative indicators, the values ​​of each indicator are mapped to the (0,1) interval. The closer the measured value is to the ideal working state, the closer its value is to 1. Conversely, the closer it is to the failure state, the closer its value is to 0. For non-quantitative indicators, they are determined by professional operation and maintenance management personnel. (2) Determine indicator weights: Fuzzy hierarchical analysis method is used to determine the indicator weights of the gate.

[0011] Step 4: Improve the construction of the gate health assessment model of radar chart: (1) Comprehensive evaluation of radar chart; (2) Improved method of radar chart method.

[0012] In step 3, the mapping function P(x) for the quantitative indicators i )for

[0013]

[0014] P(x i () represents the value of the original data after being mapped by the mapping function, x i x op x max These represent the original data input, the best value, and the maximum value, respectively.

[0015] Step 3 of the fuzzy hierarchical analysis method includes the following steps:

[0016] The established fuzzy judgment matrix R:

[0017]

[0018] The consistency processing procedure for the judgment matrix is ​​as follows:

[0019] 1) Calculate the row sum based on the judgment matrix

[0020]

[0021] 2) Establish a priority relation matrix:

[0022]

[0023] In the formula: p ij =(p i -p j ) / 2n+0.5(i,j=1,2,...,n);

[0024] 3) Determine the relative weights

[0025] The eigenvectors of the fuzzy consistency judgment matrix are calculated, and then normalized using a single-layer weighting formula. This allows us to determine the relative weights of each evaluation index factor. The calculation formula is as follows:

[0026]

[0027] In the formula, w i ' represents the weight value of the i-th indicator factor; n is the order of r; a = (n-1) / 2; r ij To determine the priority of factor i compared to factor j in the matrix;

[0028] The normalization formula is:

[0029]

[0030] W i That is, the weight of the i-th indicator factor.

[0031] In step 4, the comprehensive evaluation of the radar chart includes:

[0032] a) Feature extraction: Take the area of ​​the radar map of the i-th evaluation object as S i and perimeter L i As a feature vector, the number of indicators is n, and the standard value of the j-th indicator is z. ij The perimeter and area of ​​the radar image can be calculated using the sine and cosine laws of trigonometric functions:

[0033]

[0034]

[0035] b) Construct the evaluation vector V i

[0036] V i =[v i1 ,v i2 ]

[0037] in,

[0038]

[0039] In the formula S m =max{s i}; Vector component v il It is the area evaluation value, the size of which reflects the overall level of the evaluated object; v i2It is the perimeter evaluation value, which is the area ratio for the same perimeter, reflecting the balance of the evaluation indicators;

[0040] c) Constructing the evaluation function: The evaluation function takes the geometric mean of the evaluation vectors.

[0041]

[0042] Calculate the evaluation value and rank the evaluation objects according to the evaluation value.

[0043] Improvements to the radar chart method include:

[0044] a) Introduce a standards system;

[0045] b) Improved drawing method: The polygon radar chart drawing method is adopted. The weight of each indicator is converted into angle values ​​to divide the area size. The angle bisector of the corresponding sector area of ​​each indicator is used as the indicator data axis. The marker point of the indicator data axis, the endpoint of the area boundary line and the center of the unit circle are connected in sequence to form a quadrilateral area to represent the radar chart area of ​​an indicator.

[0046] c) Introducing entropy as a feature vector: Assume there are n objects being evaluated using radar charts, with a total of m evaluation indicators, x ij p represents the magnitude of the j-th indicator value of the i-th evaluation object. ij For x ij The values ​​obtained after normalization are:

[0047]

[0048] The entropy value of the i-th evaluation object can be expressed as:

[0049]

[0050] d) A two-layer radar chart structure is adopted: the first layer radar chart takes the criteria layer as the evaluation target, plots all the indicator data contained in this layer on the same radar chart, and obtains the comprehensive evaluation result of this layer through quantitative analysis; the second layer radar chart takes the gate health assessment as the target, and evaluates it using the evaluation results of the three first layer radar charts as indicator data.

[0051] e) Draw a radar chart of the criteria layer.

[0052] Improvements to the radar chart method also include:

[0053] f) Quantitative evaluation of the criterion-layer radar chart; the formula for calculating the area of ​​each criterion-layer radar chart is:

[0054]

[0055] Where, θi P(x) represents the angle value of the sector region corresponding to the i-th index; i ) represents the preprocessed value of the i-th indicator data; m represents the number of indicators contained in this criterion layer;

[0056] Let S be the area of ​​the radar chart of the i-th evaluation object. i and entropy value H j As a feature vector, the comprehensive evaluation result S of the object I Represented as

[0057]

[0058] Where S op This represents the area of ​​the best sample in the radar graph in this evaluation, that is, the area when all index data in the radar graph is 1.

[0059] g) Draw a target-layer radar chart and perform quantitative evaluation: Following the analysis steps described above, draw a target-layer radar chart based on the comprehensive evaluation results of the radar charts of the above criteria layers, where the angles of the corresponding fan-shaped areas on the target-layer radar chart for each criterion are 180 degrees, 72 degrees, and 108 degrees, respectively.

[0060] Methods for drawing criterion-layer radar charts include:

[0061] 1) Draw a unit circle with center 0 and radius 1;

[0062] 2) Based on the weights w of each indicator i Determine the corresponding sector angle θ i The sector areas corresponding to each indicator are divided by dashed lines according to the order of the indicator system, where θ i =2πw i ;

[0063] 3) Draw the diagonal of the area of ​​each sector as the index axis x using a solid line. i Determine the dividing points of the intervals corresponding to each level on the indicator axis;

[0064] 4) Connect the dividing points on each indicator axis with lines of different markings to form a polygon.

[0065] This invention, employing the above technical solution, addresses the multi-factor impact of stress, deflection, vibration, and corrosion on the condition assessment of hydraulic steel gates. It proposes a gate health assessment system comprising a target layer, a criterion layer, and an indicator layer. Furthermore, it presents a hydraulic gate health assessment model and method combining fuzzy hierarchical analysis (AHP) and an improved radar chart method. This method first uses AHP to determine the weights of each indicator, then uses the improved radar chart method to evaluate the criterion and target layers separately, and finally standardizes the gate evaluation level and plots it in the radar chart. This approach allows for both intuitive evaluation results via radar graphics and quantitative evaluation values ​​via evaluation functions, achieving a comprehensive health assessment of hydraulic gates.

[0066] Radar charts, as an intuitive and comprehensive evaluation and analysis method, can scientifically evaluate research objects with multiple indicators. They are intuitive and efficient for assessing the merits and demerits of objects with multiple variables, and can yield both qualitative and quantitative results. When using radar charts for evaluation, both the overall level of the evaluated object and the interrelationships between its various indicators are considered, making them widely applicable in the health assessment of hydraulic gates. Attached Figure Description

[0067] Figure 1 A radar chart based on safety criteria;

[0068] Figure 2 A radar chart based on applicability criteria;

[0069] Figure 3 A radar chart based on durability criteria;

[0070] Figure 4 This is a radar image based on the target layer;

[0071] Figure 5 This is a radar chart based on security criteria in an example.

[0072] Figure 6 The radar chart is based on the applicability criterion in the example.

[0073] Figure 7 The radar chart is based on the durability criterion in the example.

[0074] Figure 8 This is a radar chart based on the target layer in the example. Detailed Implementation

[0075] This invention relates to a method for health assessment of hydraulic gates based on an improved radar chart method, comprising the following steps:

[0076] 1. Construction of a Health Assessment System for Hydraulic Gates: This study analyzes the influence of multiple factors, including stress, deflection, vibration, and corrosion, on the condition assessment of hydraulic steel gates. A multi-level health assessment index system for hydraulic gates, comprising target, criterion, and indicator layers, is established, as shown in Table 1 below.

[0077] Table 1 Quality Index System for Health Assessment of Hydraulic Gates

[0078]

[0079] 2. Data Acquisition: This includes quantitative indicators (such as corrosion detection, non-destructive testing of welds, structural stress verification, and structural vibration verification, which can be quantitatively detected and characterized by instruments and equipment) and non-quantitative indicators (such as inspection and gate appearance inspection, which rely on manual experience for qualitative description). Based primarily on the requirements of the Technical Specification for Safety Inspection of Hydraulic Steel Gates and Hoists (SL 101-2014), the inspection content for the operational safety of gate equipment generally includes: inspection, gate appearance inspection, corrosion detection, non-destructive testing of welds, structural stress verification, and structural vibration verification.

[0080] 3. Indicator preprocessing:

[0081] The preprocessing method for hydraulic gate indicator data involves considering both quantitative and non-quantitative indicators in the health evaluation of these gates. Quantitative indicators are further divided into absolute and relative quantitative indicators. The final effect of standardizing these indicators is to map their values ​​to the (0,1) interval using a specific correspondence rule. The closer the measured value is to the ideal working state, the closer its value is to 1; conversely, the closer it is to the failure state, the closer its value is to 0. The correspondence between the indicator level standard and the standardized interval is as follows:

[0082] Grade A: (0.75, 1.0]; Grade B: (0.5, 0.75]; Grade C: (0.25, 0.5]; Grade D: [0, 0.25].

[0083] (1) Indicator standardization: Specifically, the values ​​of each indicator are mapped to the (0,1) interval through a mapping function. The closer the measured value is to the ideal working condition, the closer its value is to 1; conversely, the closer it is to the failure state, the closer its value is to 0. Quantitative indicators are all of the type where smaller is better, and can be determined according to the mapping function P(xi).

[0084]

[0085] P(x i () represents the value of the original data after being mapped by the mapping function, x i x op x maxThese represent the original data input, the best data value, and the maximum data value, respectively.

[0086] Non-quantifiable indicators are determined by scores from professional operations and maintenance management personnel.

[0087] (2) Determine the weights of the indicators:

[0088] Establishing the fuzzy judgment matrix for the evaluation indicators of hydraulic gates is the first step in determining the weights of the gate indicators using the fuzzy hierarchical analysis method. The values ​​at each position in the matrix are determined using the method shown in Table 2.

[0089] Table 2. Scaling Judgment Criteria and Their Meaning

[0090]

[0091] The established fuzzy judgment matrix R can be represented as:

[0092]

[0093] The consistency processing procedure for the judgment matrix is ​​as follows:

[0094] 1) Calculate the row sum based on the judgment matrix

[0095]

[0096] 2) Establish the priority relation matrix P:

[0097]

[0098] In the formula: p ij =(p i -p j ) / 2n+0.5(i,j=1,2,...,n).

[0099] 3) Determine the relative weights

[0100] The eigenvectors of the fuzzy consistency judgment matrix are calculated, and then normalized using a single-layer weighting formula to determine the relative weights of each evaluation index factor. The calculation formula is as follows:

[0101]

[0102] In the formula, w i ' represents the weight value of the i-th indicator factor; n is the order of r; a = (n-1) / 2; r ij To determine the element values ​​in the matrix, we need to indicate the priority of factor i compared to factor j.

[0103] The normalization formula is:

[0104]

[0105] W i This represents the weight of the i-th indicator factor. Based on the above calculation method, the weights of each indicator in the hydraulic gate health assessment system are shown in Table 3.

[0106] Table 3 Weights of Health Assessment Indicators

[0107]

[0108] 4. Improve the construction of the radar chart gate health assessment model:

[0109] (1) Comprehensive evaluation of radar charts includes the following steps:

[0110] a) Feature extraction

[0111] Let S be the area of ​​the radar chart of the i-th evaluation object. i and perimeter L i As a feature vector, the number of indicators is n, and the standard value of the j-th indicator is z. ij According to the sine and cosine theorems of trigonometric functions, the perimeter and area of ​​the radar image can be calculated as follows:

[0112]

[0113]

[0114] b) Construct the evaluation vector V i

[0115] V i =[v i1 ,v i2 (4.3)

[0116] in,

[0117]

[0118] In the formula S m =max{s i}. Vector component v il It is the area evaluation value, the size of which reflects the overall level of the evaluated object; v i2 This is the perimeter evaluation value, which is the area ratio for the same perimeter, reflecting the degree of balance of the evaluation indicators. It can be seen that the evaluation vector comprehensively considers both the overall level and the degree of balance of the indicators.

[0119] c) Constructing the evaluation function

[0120] Evaluation function f i Take the geometric mean of the evaluation vector.

[0121]

[0122] Calculate the evaluation value and rank the evaluation objects according to the evaluation value.

[0123] (2) Improved methods for radar chart method:

[0124] a) Introducing a standard system: The current radar chart analysis method is mainly used to sort different evaluation objects or different stages of the same object. Its indicator data preprocessing is often to compare and process the indicator data of each evaluation object or the indicator data of each stage, resulting in the final evaluation value changing with the change of the indicator data of the evaluation object.

[0125] To ensure the evaluation values ​​are deterministic and comparable, and to allow for evaluation of any object at any time, the radar chart analysis method needs to incorporate an evaluation standard system. Therefore, the classification of health assessment levels for hydraulic gates and the standard classification of each indicator level are introduced into the radar chart analysis method, making the evaluation more uniform and expanding the applicability of the evaluation method.

[0126] The specific approach involves first preprocessing the data that defines the A, B, C, and D levels of each indicator included in safety, suitability, and durability. Then, different colors are used to represent these levels on a radar chart, thus dividing the area into four levels. Based on this, a health assessment is then performed on a specific gate at a given moment, which ensures the certainty and uniqueness of the assessment.

[0127] b) Improve the drawing method:

[0128] In radar chart creation, a polygon radar chart method is adopted. The weights of each indicator are converted into angle values ​​to divide the region size. The angle bisectors of the corresponding sector regions for each indicator are used as the indicator data axes. Connecting the marker points of the indicator data axes, the endpoints of the region boundaries, and the center of the unit circle sequentially forms a quadrilateral region to represent the radar chart area of ​​one indicator. This method avoids discrepancies in overall evaluation caused by the correlation between indicators.

[0129] c) Introduce entropy values ​​as feature vectors:

[0130] In using radar charts for quantitative assessment, in addition to using the area of ​​the radar chart as a feature vector, the entropy value of the indicators is introduced instead of the perimeter to measure the balance of development of the evaluated object, that is, the difference between the data of various indicators. The greater the difference between the data of various indicators, the smaller the entropy value, indicating that these indicators provide less reliable information; conversely, the smaller the difference, the larger the entropy value, which can provide more information. The following is the calculation process of the entropy value.

[0131] Suppose that in a certain comprehensive evaluation using radar charts, there are n objects and a total of m evaluation indicators, x ij p represents the magnitude of the j-th indicator value of the i-th evaluation object. ij For x ij The value obtained after normalization.

[0132]

[0133] Then the entropy value H of the i-th evaluation object j It can be expressed as

[0134]

[0135] d) A two-layer radar chart structure is adopted:

[0136] Based on the constructed three-layer health assessment system for hydraulic gates (target layer, criterion layer, and indicator layer), the target layer represents the gate health assessment results. The criterion layer is divided into three criteria: safety, suitability, and durability. Each criterion layer contains 5-6 specific indicators, totaling 16. Plotting all these indicators on a single radar chart would make the chart overly complex, information-disorganized, and difficult to discern patterns. Therefore, a two-layer radar chart approach is proposed. The first-layer radar chart uses the criterion layer as the assessment target, plotting all indicator data within that layer on a single radar chart, and deriving the comprehensive evaluation result of that layer through quantitative analysis. The second-layer radar chart focuses on the gate health assessment itself, using the evaluation results from the three first-layer radar charts as indicator data for evaluation. This two-layer radar chart method aligns with the gate health assessment system's structure, is rational and orderly, and facilitates the analysis of the gate assessment criterion layer.

[0137] e) Draw the radar chart of the criterion layer.

[0138] First, draw a radar chart corresponding to the standard levels based on security criteria, as follows:

[0139] 3) Draw a unit circle with center 0 and radius 1;

[0140] 4) Based on the weights w of each indicator obtained in "3. Indicator Preprocessing" i Determine the corresponding sector angle θ i The sector areas corresponding to each indicator are divided by dashed lines according to the order of the indicator system, where θ i =2πw i ;

[0141] 3) Draw the diagonal of the area of ​​each sector as the index axis x using a solid line. i On the indicator axis, determine the dividing points of the intervals corresponding to each level A, B, C, and D;

[0142] 4) Connect the boundary points on each indicator axis sequentially using "double long dash-short dash", "double dash-dot-dot", "double dash-dot", and "half-long dash-short dash" to form polygons. The closed area formed between the "double long dash-short dash" and "double dash-dot-dot" polygons represents the radar chart shape corresponding to level A; the closed area formed between the "double dash-dot-dot" and "double dash-dot" polygons represents the radar chart shape corresponding to level B; the closed area formed between the "double dash-dot" and "half-long dash-short dash" polygons represents the radar chart shape corresponding to level C; and the closed area formed within the "half-long dash-short dash" polygon represents the radar chart shape corresponding to level D. For example... Figure 1 As shown. Using the same method, radar charts based on the applicability criterion and the durability criterion are drawn sequentially. For example... Figure 2 , Figure 3 As shown.

[0143] Table 4.1 Radar chart area angles corresponding to each indicator

[0144]

[0145] f) Quantitative evaluation of the criterion-level radar chart

[0146] The formula for calculating the area of ​​each criterion layer radar image is:

[0147]

[0148] Where, θ i P(x) represents the angle value of the sector region corresponding to the i-th index; i ) represents the preprocessed value of the i-th indicator data; m represents the number of indicators contained in this criterion layer.

[0149] Let S be the area of ​​the radar chart of the i-th evaluation object. i and entropy value H j As a feature vector, the comprehensive evaluation result S of the object I Represented as

[0150]

[0151] Where S op The area represents the best sample in the radar graph in this evaluation, that is, the area where all index data in the radar graph is 1. The comprehensive evaluation results of the steel gate based on the criteria layer radar graphs A, B, C, and D are shown in Tables 4.2 to 4.4.

[0152] Table 4.2 Safety Criteria Radar Chart Quantitative Assessment

[0153]

[0154] Table 4.3 Applicability Criteria Radar Chart Quantitative Assessment

[0155]

[0156] Table 4.4 Durability Criteria Quantitative Assessment via Radar Chart

[0157]

[0158] g) Draw a radar map of the target layer and conduct a quantitative assessment.

[0159] Following the analytical steps described above, a radar chart based on the target layer is drawn using the comprehensive evaluation results of the radar charts at the criterion layer as indicators. The corresponding sector angles for each criterion on the target layer radar chart are 180 degrees, 72 degrees, and 108 degrees, respectively. For example... Figure 4 As shown.

[0160] The evaluation results were calculated and are shown in Table 4.5.

[0161] Table 4.5 Quantitative Assessment of Target Layer Radar Image

[0162]

[0163] h) Example Analysis:

[0164] A reservoir in Henan Province, located in the Yellow River basin, is a large (I) water conservancy project primarily for flood control, but also providing power generation and water supply. The dam is 55m high, and the total reservoir capacity is 1.32 billion m³. 3 The reservoir was started in December 1959 and completed at the end of August 1965. In February 1970, an irrigation and power generation tunnel was added. Dam reinforcement work began in 1986, and the government invested further funds in 2001 and 2003 for repairs and reinforcement. The reservoir's spillway gates are exposed arched gates, primarily made of Q345B steel. The gates are 12m high, 10.5m wide, and have a maximum water level of 323m. There are three gates with a maximum spillway capacity of 3810m³. 3 The spillway is 435m long. Based on reservoir operation and management, and on-site inspection, the gate opening and closing operations are normal, the steel gate's appearance is relatively good, and the overall structure has slight deformation. There is some leakage at the top stop of the gate, and the gate body is vibrating. The outer material of the gate is basically intact, with some overall rust and some areas severely corroded. The on-site operation and maintenance management department is generally able to carry out inspections and maintenance as required.

[0165] First, based on the on-site monitoring of the gate, analysis of historical documents, numerical calculations, expert scoring based on experience, and internal data from relevant departments, the gate's various indicator data are shown in Table 9.

[0166] Table 9. Indicator data for the arc-shaped gate of a reservoir's spillway.

[0167]

[0168]

[0169] Based on the indicators of safety, applicability, and durability criteria, example radar graphics were drawn on the basis of the standard radar chart (as shown below). Figure 5 (As shown by the solid black line).

[0170] Then, feature vectors are extracted and evaluation results are calculated (see Table 10 below).

[0171] Table 10 Comprehensive Evaluation of Radar Charts for the Criterion Layer in the Calculation Examples

[0172]

[0173] Based on the comprehensive evaluation results of the above criterion-level radar chart, a target-level radar chart is drawn (as follows). Figure 8 (As shown).

[0174] The comprehensive evaluation results are calculated (see Table 11 below).

[0175] Table 11 Comprehensive Evaluation of Target Layer Radar Chart in the Calculation Example

[0176]

[0177] As shown in the radar chart based on safety criteria, the overall graph is near the yellow line, with x2 and x4 positioned between the yellow and blue lines, and x1 and x5 between the yellow and red lines, with x5 being closer to the red line. From a quantitative perspective, the comprehensive evaluation result S based on the safety criteria of the radar chart is... I =0.593, falling within the C-level range (0.300, 0.597), closer to the borderline between B and C levels, thus qualifying as a C+ level. The radar chart based on applicability criteria shows a relatively even distribution of indicators, with x9 performing slightly worse. Overall, it lies within the area between the blue and yellow lines. S I =0.688, which is in the middle of the B-level range (0.500, 0.752). As can be seen from the radar chart based on durability criteria, x 10 x 12 It is near the blue line and performing well. 11 Located near the yellow line, performing poorly, S I =0.630, which is in the middle of the B-level range (0.520-0.761). As can be seen from the target-layer radar chart, the usability and durability of the gate in this example are between the blue and yellow lines, while its safety is near the yellow line. The overall evaluation result is S. I=0.627, which is in the middle of Grade B (0.546, 0.774), indicating that the main performance of the gate meets the usage requirements. However, regarding individual indicators, the gate's x5 indicator (main beam deflection) is excessive, and the x9 indicator (water-stopping performance) is relatively poor; these should be addressed with targeted improvements.

Claims

1. A health assessment method for hydraulic arc-shaped gates based on an improved radar chart method, characterized in that, It includes the following steps: Step 1: Construction of a health assessment system for hydraulic arc gates: Establish a multi-level health assessment index system for hydraulic arc gates, including target layer, criterion layer, and indicator layer; Step 2, Indicator Data Acquisition: Obtain indicator data through detection; the indicator data includes quantitative and non-quantitative indicators; Step 3, indicator preprocessing: (1) Indicator standardization: For quantitative indicators, the values ​​of each indicator are mapped to the (0,1) interval. The closer the measured value is to the ideal working state, the closer its value is to 1. Conversely, the closer it is to the failure state, the closer its value is to 0. For non-quantitative indicators, they are determined by professional operation and maintenance management personnel. (2) Determine indicator weights: Fuzzy hierarchical analysis method is used to determine the indicator weights of the gate. Step 4, Improve the construction of the gate health assessment model of radar chart: (1) Comprehensive evaluation of radar chart; (2) Improved methods for radar chart method; Improvements to the radar chart method include: a) Introduce a standards system; b) Improved drawing method: The polygon radar chart drawing method is adopted. The weight of each indicator is converted into angle values ​​to divide the area size. The angle bisector of the corresponding sector area of ​​each indicator is used as the indicator data axis. The marker point of the indicator data axis, the endpoint of the area boundary line and the center of the unit circle are connected in sequence to form a quadrilateral area to represent the radar chart area of ​​an indicator. c) Introduce entropy as an eigenvector: x ij p represents the magnitude of the j-th indicator value of the i-th evaluation object. ij For x ij The values ​​obtained after normalization, assuming there are n evaluation indicators in total, have ； Then the first i The entropy value of an evaluation object can be expressed as: ； d) A two-layer radar chart structure is adopted: the first layer radar chart takes the criteria layer as the evaluation target, and plots all the indicator data contained in this layer on the same radar chart. The comprehensive evaluation result of this layer is obtained through quantitative analysis; the second layer radar chart takes the gate health assessment as the target, and evaluates it using the evaluation results of the three first layer radar charts as indicator data. e) Draw a radar chart of the criteria layer; f) Quantitative evaluation of the criterion-layer radar chart; the formula for calculating the area of ​​each criterion-layer radar chart is as follows: ； in, θ i Representing the i Each indicator corresponds to the angle value of the sector area; P ( x i ) is the first i The preprocessed values ​​of each indicator data; m This refers to the number of indicators contained in this criterion layer; Take the first i The area of ​​the radar map of each evaluation object is S i and entropy H i As a feature vector, the comprehensive evaluation result of the object S I Represented as ； in S op This represents the area of ​​the best sample in the radar graph in this evaluation, that is, the area when all index data in the radar graph is 1. g) Draw a radar chart of the target layer and make a quantitative evaluation: Following the analysis steps described above, draw a radar chart based on the target layer using the comprehensive evaluation results of the radar chart of the above criteria layer as an indicator. The angles of the corresponding fan-shaped areas of each criterion on the target layer radar chart are 180 degrees, 72 degrees and 108 degrees respectively.

2. The method for health assessment of hydraulic arc-shaped gates based on the improved radar chart method according to claim 1, characterized in that: In step 3, the mapping function P(x i ) of the quantitative index is ； P(x i () represents the value of the original data after being mapped by the mapping function, x i x op x max These represent the original data input, the best value, and the maximum value, respectively.

3. The method for health assessment of hydraulic arc-shaped gates based on the improved radar chart method according to claim 1, characterized in that: Step 3 of the fuzzy hierarchical analysis method includes the following steps: The established fuzzy judgment matrix R: ； The consistency processing procedure for the judgment matrix is ​​as follows: 1) Calculate the row sum based on the judgment matrix ； 2) Establish a priority relation matrix: ； In the formula: p ij = ( p i - p j ) / 2 n + 0.5 ( i,j =1,2,..., n ); 3) Determine the relative weights The eigenvectors of the fuzzy consistency judgment matrix are calculated, and then normalized using a single-layer weighting formula. This allows us to determine the relative weights of each evaluation index factor. The calculation formula is as follows: ； In the formula, w i ’ Let be the weight value of the i-th indicator factor; a = ( n -1) / 2; r ij To determine the element values ​​in a matrix, indicating factors i and factors j Priority of comparison; The normalization formula is: ； w i That is, the first i The weight of each indicator factor.

4. The method for health assessment of hydraulic arc-shaped gates based on the improved radar chart method according to claim 1, characterized in that: In step 4, the comprehensive evaluation of the radar chart includes: a) Feature extraction: Take the area of ​​the radar map of the i-th evaluation object as S i and perimeter L i As a feature vector, the standard value of the j-th index is z. ij The perimeter and area of ​​the radar image can be calculated using the sine and cosine laws of trigonometric functions: ； ； b) Constructing the evaluation vector V i ； in, ， ， In the formula S m =max { s i }; Vector components v il It is the area evaluation value, the size of which reflects the overall level of the evaluated object; v i2 It is the perimeter evaluation value, which is the area ratio under the same perimeter, reflecting the balance of the evaluation indicators; c) Constructing the evaluation function: The evaluation function takes the geometric mean of the evaluation vectors. ； Calculate the evaluation value and rank the evaluation objects according to the evaluation value.

5. The method for health assessment of hydraulic arc-shaped gates based on the improved radar chart method according to claim 1, characterized in that: Methods for drawing criterion-layer radar charts include: Draw a unit circle with center 0 and radius 1; Based on the weight of each indicator w i Determine the corresponding sector angle θ i The sector areas corresponding to each indicator are divided by dashed lines according to the order of the indicator system. ; 3) Draw the diagonal of each sector's area using solid lines as the indicator axis. x i Determine the dividing points of the intervals corresponding to each level on the indicator axis; 4) Connect the dividing points on each indicator axis with lines of different markings to form a polygon.

Images