Gas engineering construction quality evaluation system based on big data

Abstract

The invention belongs to the technical field of gas pipelines, and particularly relates to a gas engineering construction quality evaluation system based on big data, which comprises the following steps of: firstly, constructing a defect feature tensor, introducing a space-time grid analysis technology to analyze defect disposal aging interval data and location defect reproduction frequency data, and constructing a defect feature tensor on the basis of a preset quality-efficiency reliability quantification rule; defect features are converted into quantifiable quality-effect scores, then deep fusion of multi-source heterogeneous data is achieved by establishing a space-time incidence matrix of welding defects, welding process parameters and environment monitoring parameters, and on this basis, a fuzzy correlation analysis technology and a preset process compliance quantification rule are adopted to achieve quality-effect evaluation of the welding defects. Complex process parameters and environmental conditions are converted into quantifiable process scores, the process scores and quality-effect scores are integrated to generate a gas pipeline welding quality comprehensive evaluation value of a target contractor, and a quantifiable intelligent decision support system is provided for gas pipeline construction quality management and control of the contractor.

Description

Technical Field

[0001] The present invention belongs to the technical field of gas pipelines, and specifically relates to a gas engineering construction quality assessment system based on big data. Background Art

[0002] In the context of the rapid expansion of the current urban gas network, the construction quality of gas projects, especially the construction quality of PE pipelines, has become a key factor in ensuring public safety and system reliability. As the main body of construction, the construction quality of the contractor directly determines the safety and service life of the project. However, traditional construction quality assessment methods have problems such as strong subjectivity and low data utilization rate, making it difficult to meet the high standards of modern engineering construction. Therefore, it is particularly important to accurately assess the construction quality of the contractor's gas project based on the performance dimensions of PE pipeline construction.

[0003] In the prior art, there are also some solutions related to the analysis of PE pipeline construction quality. For example, a method and system for determining the risk of buried high-molecular polyethylene pipelines for gas use with the Chinese patent publication number CN113379280A collects and summarizes the historical accident machine detection report data of buried high-molecular polyethylene pipelines for gas use, establishes a risk database for buried high-molecular polyethylene pipelines for gas use, determines the classification of the failure risk level of buried high-molecular polyethylene pipelines for gas use, and determines the failure risk judgment indicators for buried high-molecular polyethylene pipelines for gas use. According to the characteristics of polyethylene gas pipelines and the statistical data on the accident causes of urban gas polyethylene pipeline systems at home and abroad, combined with the Kent scoring method and the analytic hierarchy process, and introducing extension theory, it is planned to design 4 main risk judgment indicators of third-party damage, design factors, misoperation factors, and inherent safety quality, establish a failure risk level assessment model for urban gas polyethylene pipelines, determine the scoring of each risk judgment indicator and the calculation method of its weight value, and develop corresponding software.

[0004] Another Chinese patent with the publication number CN119476958A, a method and Internet of Things system for intelligent monitoring of gas pipeline welding based on government supervision, is executed based on the gas company management platform of the intelligent gas pipeline welding monitoring Internet of Things system, including: obtaining at least one set of welding point information and corresponding pipeline construction information in a preset pipeline area. Based on the pipeline construction information, determine the first welding risk. Based on the first welding risk and historical welding risks, determine the first risk value. In response to the first risk value meeting the preset conditions, generate a welding adjustment instruction. Obtain the welding process information uploaded by the welder's terminal. Perform preset processing on the welding process information to obtain the key welding information and store it in the government supervision comprehensive database. Determine the monitoring parameters of the preset pipeline area. It can more comprehensively monitor the on-site welding quality of gas pipelines for different environmental factors to improve the safety and efficiency of welding operations.

[0005] Although the above solution proposes a quality analysis method for the construction performance dimension of PE pipelines, there are still limitations in its analysis logic when introducing the construction quality assessment of contractors. Specifically: 1. The existing technology has a single data dimension for the construction quality assessment of PE pipelines. It mostly relies on weld images or manual sampling records, lacks the multi-source data fusion of welding process parameters and environmental monitoring parameters, and cannot construct a complete quality assessment element system.

[0006] 2. The existing technology has a fragmented dimension for the construction quality assessment of PE pipelines. Quality assessment and process assessment are carried out separately. It not only lacks the spatio-temporal correlation analysis of defect characteristics and process parameters, but also cannot quantify the dynamic impact of environmental variables on welding quality. The determination of process compliance depends on static thresholds set in advance by humans, which in turn leads to the inability to effectively feedback the specific level of the contractor's gas construction quality. Summary of the Invention

[0007] In order to overcome the shortcomings in the background technology, the embodiments of the present invention provide a gas engineering construction quality assessment system based on big data, which can effectively solve the problems involved in the above background technology.

[0008] The object of the present invention can be achieved through the following technical solutions: A gas engineering construction quality assessment system based on big data, including: a data acquisition module, a quality and efficiency assessment module, a process assessment module, and a comprehensive assessment module.

[0009] The data acquisition module is connected to the quality and efficiency assessment module, the quality and efficiency assessment module is connected to the process assessment module, and the process assessment module is connected to the comprehensive assessment module.

[0010] The data acquisition module obtains the historical welding data set of the gas pipeline of the target contractor through a multi-source data interface. The data set includes welding process parameters, environmental monitoring parameters, and a sequence of weld images arranged according to the construction time sequence.

[0011] The quality and efficiency assessment module extracts defect features from the sequence of weld images to generate a feature tensor containing defect type codes, spatial coordinates, and timestamps, analyzes the defect handling time interval data and location defect recurrence frequency data in the feature tensor based on spatio-temporal grid analysis, and generates a quality and efficiency score through a preset quality and efficiency reliability quantification rule.

[0012] The process assessment module establishes a spatio-temporal correlation matrix of welding defects, welding process parameters, and environmental monitoring parameters, analyzes the defect-process correlation index and environment-process adaptation index in the spatio-temporal correlation matrix based on fuzzy correlation analysis, and generates a process score through a preset process compliance quantification rule.

[0013] The comprehensive evaluation module is used to perform weighted fusion calculation on the process score and the quality and efficiency score, generate the comprehensive evaluation value of the welding quality of the gas pipeline of the target contractor, and output a visualization report.

[0014] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects: (1) Through the spatio-temporal alignment of multi-source heterogeneous data, the present invention realizes the three-dimensional data coupling of welding process parameters, environmental monitoring parameters, and visual features, and effectively and accurately evaluates the historical welding quality of the gas pipeline of the target contractor from the two perspectives of process compliance and quality and efficiency reliability, thereby greatly improving the reliability and practicality of the evaluation results.

[0015] (2) The present invention realizes the structured expression of defect information by constructing a defect feature tensor. On this basis, the spatio-temporal grid analysis technology is introduced to analyze the defect disposal time interval data and the location defect recurrence frequency data, realizes the dynamic tracking and visualization of the defect distribution, and further based on the preset quality and efficiency reliability quantification rules, converts the defect features into quantifiable quality and efficiency scores, which highly reflects the construction quality level of the contractor's gas pipeline.

[0016] (3) The present invention realizes the deep fusion of multi-source heterogeneous data by establishing a spatio-temporal correlation matrix of welding defects, welding process parameters, and environmental monitoring parameters, breaks through the limitations of traditional single-data-dimensional analysis. On this basis, the fuzzy correlation analysis technology and the preset process compliance quantification rules are adopted to convert complex process parameters and environmental conditions into quantifiable process scores, and realizes the dynamic evaluation of the construction process of the contractor's gas pipeline. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The present invention is further described with reference to the accompanying drawings, but the embodiments in the drawings do not constitute any limitation to the present invention. For those of ordinary skill in the art, other drawings can be obtained according to the following drawings without creative efforts.

[0018] Figure 1 It is a schematic diagram of module connection provided by the first embodiment of the present invention.

[0019] Figure 2 It is a specific logic schematic diagram of the preset quality and efficiency reliability quantification rules in the first embodiment of the present invention.

[0020] Figure 3 It is a schematic diagram of the structure of a device provided by the second embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0021] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0022] Embodiment 1

[0023] Referring to Figure 1 As shown, the first embodiment of the present invention provides a gas engineering construction quality evaluation system based on big data, including: a data acquisition module, a quality and efficiency evaluation module, a process evaluation module, and a comprehensive evaluation module.

[0024] The data acquisition module is connected to the quality and efficiency evaluation module, the quality and efficiency evaluation module is connected to the process evaluation module, and the process evaluation module is connected to the comprehensive evaluation module.

[0025] The data acquisition module obtains the historical welding data set of the gas pipeline of the target contractor through a multi-source data interface. The data set includes welding process parameters, environmental monitoring parameters, and a sequence of weld images arranged according to the construction time sequence.

[0026] It should be noted that the above-mentioned historical welding data set of the gas pipeline of the target contractor covers the welding data of various gas engineering projects constructed by the target contractor in the historical preset period according to the construction time sequence. And the welding data is real-time recorded by a welding machine or a welding torch equipped with a multi-functional sensor and a data acquisition device during the operation. For example, welding process parameters (including but not limited to current, voltage, speed, etc.) are collected in real time by the intelligent sensor array carried by the welding machine, environmental monitoring parameters (including but not limited to temperature, humidity, wind speed, air pressure, etc.) are collected in real time by the micro meteorological sensor array integrated in the welding machine, and the sequence of weld images is dynamically obtained by the linear array CCD camera integrated in the welding machine according to the welding rhythm. Since the data acquisition process is synchronized with the welding operation, and all data has a unified time stamp and spatial coordinate identifier, the multi-dimensional information in the data set can achieve accurate spatio-temporal alignment operations, providing a high-quality data basis for subsequent spatio-temporal correlation analysis.

[0027] The quality and efficiency evaluation module extracts defect features from the sequence of weld images, generates a feature tensor containing defect type codes, spatial coordinates, and time stamps, analyzes the defect disposal time interval data and the location defect recurrence frequency data in the feature tensor based on spatio-temporal grid analysis, and generates a quality and efficiency score through a preset quality and efficiency reliability quantification rule.

[0028] It should be added that for the extraction of defect features of the above weld image sequence, specifically refer to the preset defect feature library that has integrated various weld defect image features. The cosine similarity is used to calculate the matching degree between the feature vectors extracted from each image to be inspected in the weld image sequence and the image features of various weld defects in the preset defect feature library. If the matching degree is greater than or equal to the preset matching degree standard threshold, it is regarded as a successful match and corresponding defect feature extraction processing is carried out.

[0029] In a preferred embodiment of the present invention, the specific defect disposal time interval data is as follows: Based on the defect type coding and spatial coordinates in the feature tensor, a three-dimensional space grid is constructed.

[0030] The time interval from the welding time stamp to the defect elimination time stamp of the same defect type coding in the grid cell is used as the single defect disposal time interval. Thus, the time intervals of each defect type in each grid cell for each disposal are collected as the defect disposal time interval data.

[0031] It should be noted that the camera device integrated in the welding machine usually collects at the millisecond level, and the time stamp is accurate to the welding moment. Therefore, multiple weld images can be collected during the same process of the weld joint. The calculation of the defect disposal time interval is the defect response time within the same process of the weld joint, specifically referring to the absolute difference between the welding time stamp and the elimination time stamp when a certain type of defect appears at the same spatial position of the weld joint (for example, a porosity is found at the 10th second of the root welding, and the porosity disappears after adjusting the current at the 12th second, with an interval of 2 seconds). If the defect persists until the end of the welding in this process or is left over across processes, it means that this type of defect has not been eliminated and the defect disposal time interval is automatically set to NULL.

[0032] In a preferred embodiment of the present invention, the specific area defect recurrence frequency data is as follows: Record the time and type of each defect occurrence in the same grid cell within the preset time window, and introduce a time decay factor and a defect type weight factor for weighting to obtain the defect recurrence frequency of the same grid cell.

[0033] It should be added that the defect recurrence frequency of the above same grid cell can be calculated according to the following formula: , represents the defect type weight factor, specifically the preset weight corresponding to the defect type when the th defect occurs in the same grid cell. It can be set artificially in advance according to factors such as the harm degree and repair cost of the defect, and is used to represent the influence degree of different defect types on the recurrence frequency and distinguish the importance. For example, the weight of a serious defect (such as a crack) is higher, and the weight of a minor defect (such as a surface porosity) is lower, while represents the time decay factor, which controls the influence of the defect occurrence time on the recurrence frequency. The closer the time, the greater the contribution, and the farther the time, the smaller the contribution. Among them is the preset attenuation coefficient, which is used to control the rate of time decay and can be set manually at the initial stage of system operation. is the time span corresponding to the preset time window. is the time interval between the time of the th defect occurrence in the same grid cell and the start time of the preset time window. is the number of defect occurrences in each time in the same grid cell. .

[0034] It should also be added that the weighted logic adopted in the example calculation formula for the defect recurrence frequency of the above-mentioned same grid cell is as follows: by assigning different weights to different types of defects, and by assigning decay weights to defects at different occurrence times, multiplying the double weights and summing them up. In essence, it is also a weighted sum of the impacts of each defect. This is explained here to avoid ambiguity.

[0035] As an example, the preset time window is 10 days, the time decay coefficient is 0.1, and the defect occurrence records of a certain grid cell within the preset time window are shown in Table 1 below.

[0036] Table 1 Defect occurrence record data within the preset time window

[0037]

[0038] Therefore, after calculation, the defect recurrence frequency of this grid cell is 2.7585.

[0039] Statistically count the number of grid cells covering the same defect type within the preset time window, and take the ratio of it to the total number of three-dimensional space grids as the location recurrence frequency of the same defect type.

[0040] Collect the defect recurrence frequencies of each grid cell and the location recurrence frequencies of each defect type within the preset time window, and jointly use them as the location defect recurrence frequency data.

[0041] It should be noted that the welding of a gas pipeline diameter interface (weld) is divided into multiple processes (root welding, hot welding, filling welding, capping welding), and each process corresponds to a different spatial coordinate interval (not an absolute physical point). For example, the root welding covers a depth of 0 - 3 mm (axial) of the weld, the filling welding covers a depth of 3 - 8 mm, and the capping welding is 0 - 1 mm on the surface. Defects in different processes of the same weld (such as lack of fusion in root welding vs. porosity in filling welding) belong to different spatial coordinates, not the same position. Therefore, the theoretical basis for the above two types of location defect recurrence frequency data can be explained by Table 2 below.

[0042] Table 2 Connotation of location defect recurrence frequency

[0043]

[0044] Referring to Figure 2 As shown, in a preferred embodiment of the present invention, the preset quality and reliability quantification rules include the following: By referring to the preset scoring mapping table, the corresponding score values of each defect type at each disposal time interval for each grid unit are obtained, and the timeliness score is obtained through hierarchical averaging and normalization processing.

[0045] It should be noted that the above hierarchical averaging process includes the first, second, and third layer averaging processes. Specifically, the first layer performs arithmetic averaging on the corresponding scores of multiple disposal time intervals of the same defect to eliminate the contingency of single disposal. The second layer performs arithmetic averaging on the average scores of all defect types under the same grid unit. The third layer performs global averaging on the final scores of all grid units.

[0046] The average defect recurrence frequency of the grid unit and the average location recurrence frequency of the defect type within the preset time window are respectively substituted into the preset penalty function to quantify the grid recurrence degradation value and the type diffusion degradation value. The higher value of the two-dimensional degradation value is taken, and the initial stability full score is deducted according to the higher value to obtain the stability score.

[0047] It should be added that the above preset penalty function can refer to the following calculation formula: , is the average defect recurrence frequency of the grid unit or the average location recurrence frequency of the defect type within the preset time window, is the preset allowable reference frequency for grid unit defect recurrence or the allowable reference frequency for defect type location recurrence, which is set based on historical data or industry standards, is the preset penalty coefficient corresponding to grid unit defect recurrence and defect type location recurrence, used to control the growth rate of the degradation value, and needs to be determined through sensitivity experiment analysis, is the preset control factor, used to control the numerical range of the output result of the preset penalty function. In this embodiment takes the value of 0.5, takes the value of 2, takes the value of 1.

[0048] This formula selects an exponential function, aiming to non-linearly amplify the risk of exceeding the reference value, which conforms to the characteristic of "small problems accumulating to cause qualitative changes" in actual engineering.

[0049] As an example, the average defect recurrence frequency of the grid unit within the preset time window is 2.5, the average location recurrence frequency of the defect type is 0.12, the allowable reference frequency for grid unit defect recurrence is 2, and the allowable reference frequency for defect type location recurrence is 0.1. After calculation, the grid recurrence degradation value is 0.133, and the type diffusion degradation value is 0.491.

[0050] It should also be supplemented that the above high-value deduction processing specifically refers to multiplying the initial stability full score (100 points) by the absolute difference between the higher value of the two-dimensional degradation value and 1. Based on the above example, further calculation is carried out, where the higher value of the two-dimensional degradation value is 0.491, and the high-value deduction processing process is , and the resulting stability score is 50.9 points.

[0051] The sum of the timeliness score and the stability score is used as the quality and efficiency score.

[0052] In a preferred embodiment of the present invention, the preset score mapping table includes a non-linear mapping relationship between the disposal time interval of each defect type and the score value.

[0053] In the embodiment of the present invention, the defect information is structurally expressed by constructing a defect feature tensor. On this basis, the spatio-temporal grid analysis technology is introduced to analyze the defect disposal time interval data and the location defect recurrence frequency data, realize the dynamic tracking and visualization of the defect distribution, and further based on the preset quality and efficiency reliability quantification rules, convert the defect features into quantifiable quality and efficiency scores, which highly reflects the construction quality level of the contractor's gas pipeline.

[0054] The process evaluation module establishes a spatio-temporal correlation matrix of welding defects, welding process parameters and environmental monitoring parameters, analyzes the defect-process correlation degree index and the environment-process adaptation degree index in the spatio-temporal correlation matrix based on fuzzy correlation analysis, and generates a process score through preset process compliance quantification rules.

[0055] In a preferred embodiment of the present invention, the spatio-temporal correlation matrix establishes a mapping relationship between each welding defect and the corresponding process environment conditions through the spatio-temporal alignment of welding process parameters and environmental monitoring parameters.

[0056] In a preferred embodiment of the present invention, the specific process of the defect-process correlation degree index is as follows: retrieve the synchronous welding process parameters corresponding to the time sequence interval of each welding defect occurrence, construct a fuzzy grade division standard for process parameters using a trapezoidal membership function, and map the welding process parameters to the predefined fuzzy grade through fuzzy processing to generate the process parameter membership matrix corresponding to each welding defect moment.

[0057] Scan the preset association rule library through a preset association rule algorithm (which can be exemplified as the Apriori association rule algorithm), extract the strong association rule set that matches the process parameter membership matrix corresponding to the welding defect moment, record the confidence weights of each strong association rule, and use a linear weighted calculation method to obtain the association degree between each welding defect and its corresponding welding process parameters at that moment, and output the association degree index between the welding defect and the welding process parameters through mean calculation.

[0058] It should be added that the above-mentioned preset association rule library is a set of rules established in advance through historical data mining or expert experience, which is used to describe the association relationship between welding process parameters and defect types, as well as the adaptation relationship between welding variables and environmental variables. Each rule in the rule library contains the following core information: a) Defect association rule: low current low voltage defect , this example is interpreted as the combination of low current level and low voltage level results in defect occurring.

[0059] b) Environmental adaptation rule: high wind speed high welding speed. This example is interpreted as if the wind speed in the welding environment is at a high level, the welding speed needs to be at a corresponding high level. That is, the strongly associated environmental variable corresponding to the welding speed is the wind speed. Synchronously, the preset association rule library stores the benchmark values predefined for each environmental variable and the numerical intervals for dividing low, medium, and high levels.

[0060] c) Support degree: The frequency of the rule appearing in historical data (such as the number of times the rule appears / the total number of samples).

[0061] d) Confidence degree: The probability that the consequent occurs when the antecedent of the rule occurs (such as the number of times the antecedent and the consequent appear together / the number of times the antecedent appears).

[0062] Special note: Each environmental adaptation rule stored in the preset association rule library points to its influencing welding variable with a single strongly associated environmental variable. Although the welding variable may be jointly affected by multiple environmental variables, at the initial stage of establishing the preset association rule library, the influence degree of each environmental variable on the same welding variable has been analyzed by manual comparison, and the environmental variable with the greatest influence degree has been selected as the single strongly associated environmental variable. Therefore, the single strongly associated environmental variable in the preset association rule library does not mean that the welding variable is only affected by one environmental variable, but is to simplify the rule library structure and highlight the main influencing factors. This special note is hereby made to avoid ambiguity.

[0063] It also needs to be added that the above-mentioned set of strongly associated rules matching the membership degree matrix of process parameters corresponding to the moment of welding defects needs to meet the following conditions: (1) The antecedent of the rule needs to be completely included in the membership degree matrix of process parameters corresponding to the moment of welding defects. Among them, the numerical value of the membership degree matrix element represents the fuzzy level of the process parameter. The fuzzy level of each process parameter in the antecedent of the rule needs to be exactly the same as the numerical meaning of the corresponding element in the membership degree matrix. Only when this condition is met is it regarded as a matching association rule. On this basis, the rules that meet the condition are only regarded as matching association rules.

[0064] (2) To match the association rules, it is necessary to simultaneously satisfy that the support degree is greater than or equal to the preset critical support degree, and the confidence degree of the matching association rules is greater than or equal to the preset critical confidence degree, so as to be further regarded as strong association rules. Among them, the preset critical support degree and the preset critical confidence degree are determined by mean analysis or quartile method based on the support degree and confidence degree data of all rules in the preset association rule library at the initial stage of establishing the library.

[0065] The following are welding defects Specific calculation example of the correlation degree between the welding process parameters at the corresponding moment: The fuzzification results of the welding process parameters at this moment are shown in Table 3 below (i.e., the tabular form of the membership matrix)

[0066] Table 3 Fuzzification results of welding process parameters

[0067]

[0068] The matching association rules in the preset rule library involved in the welding process parameters at this moment are shown in Table 4 below.

[0069] Table 4 Matching association rules corresponding to the fuzzification results of welding process parameters

[0070]

[0071] Record the confidence degrees of each strong association rule in Table 4 (i.e., 0.8, 0.85), from Calculate to obtain the welding defects The correlation degree between the welding process parameters at the corresponding moment.

[0072] In a preferred embodiment of the present invention, the specific analysis process of the environment-process adaptability index is as follows: Randomly extract a certain welding variable in the welding process parameters at a certain moment at a certain spatial coordinate, mark the welding variable as the target welding variable, retrieve the strongly associated environmental variables matching the target welding variable through the preset association rule library, and retrieve the monitoring values of the strongly associated environmental variables at the same spatial coordinate and the same moment.

[0073] Define the standard value of the target welding variable according to the material characteristics of the gas pipeline at this spatial coordinate, and correct the standard value of the target welding variable through the monitoring values of the strongly associated environmental variables to generate the process adaptation interval of the target welding variable.

[0074] It should be added that when defining the standard values of welding variables according to the material characteristics of the gas pipeline, the welding variable specification range stipulated in the gas pipeline material manual is mainly referred to, and the median of this range is selected as the standard value. At the same time, the fluctuation range of the welding variable specification range stipulated in the material manual is recorded and used as the basis for the fluctuation range of the subsequent generated process adaptation range. Specifically, the process adaptation range of the target welding variable is a dynamic numerical range generated with the adapted value after correcting the target welding variable as the median of the range and the fluctuation range stipulated in the material manual as the range width.

[0075] It also needs to be added that the specific process of correcting the standard value of the target welding variable through the monitored values of the strongly correlated environmental variables includes: extracting the predefined reference value of the strongly correlated environmental variable and its correction direction for the target welding variable from the preset correlation rule library (depending on the environmental adaptation rule between the target welding variable and the strongly correlated environmental variable, for example, when the wind speed is high and the welding speed is high, the correction direction for the welding speed is upward).

[0076] Calculate the deviation ratio between the monitored value of the strongly correlated environmental variable and its reference value. The specific calculation logic is the ratio of the absolute difference between the two to the reference value.

[0077] Multiply the deviation ratio by the confidence level of the environmental adaptation rule of the strongly correlated environmental variable for the target welding variable to obtain a correction factor, and determine the sign of the correction factor according to the correction direction (upward is positive, downward is negative).

[0078] Define the correction coefficient as the cumulative value of 1 and the correction factor, and take the product of the correction coefficient and the standard value of the target welding variable as the adapted value after correcting the target welding variable.

[0079] As an example, assume that the target welding variable is the welding current, its standard value is 180A, the strongly correlated environmental variable matching the target welding variable is the temperature, its predefined reference value is 25°C, and the monitored value retrieved at the same spatial coordinate and the same moment is 30°C. Search the preset correlation rule library for the environmental adaptation rule of the strongly correlated environmental variable for the target welding variable as high temperature and high welding current, and the confidence level is 0.8. It can be judged that the correction direction is positive. Then, through calculation, it can be known that: deviation ratio = , correction factor = 0.2×0.8 = 0.16, correction coefficient = 1 + 0.16 = 1.16, target adapted value = 180×1.16 = 208.8A. Therefore, the adapted value of the welding current after correction is 208.8A.

[0080] Compare the actual value of the target welding variable with its process adaptation interval, and output the adaptation degree between the target welding variable and its strongly associated environmental variable. Similarly, collect the adaptation degrees between each welding variable and its strongly associated environmental variable in the welding process parameters at each moment of each spatial coordinate, and obtain the environment-process adaptation degree index through hierarchical averaging.

[0081] In a preferred embodiment of the present invention, the specific process of outputting the adaptation degree between the target welding variable and its strongly associated environmental variable is as follows: If the actual value of the target welding variable is within its process adaptation interval, the output adaptation degree between the target welding variable and its strongly associated environmental variable is 1.

[0082] If the actual value of the target welding variable is outside its process adaptation interval, quantify the deviation amplitude between the actual value of the target welding variable and its process adaptation interval, and introduce the negative form of the deviation amplitude into the natural exponential function to solve the adaptation degree between the target welding variable and its strongly associated environmental variable.

[0083] It should be added that the specific quantification process of the deviation amplitude between the actual value of the target welding variable and its process adaptation interval is as follows: If the actual value of the target welding variable is greater than the upper limit value of its process adaptation interval, obtain the over-limit value of the actual value of the target welding variable relative to the upper limit value of its process adaptation interval, and perform ratio analysis (normalization processing) with the upper limit value of its process adaptation interval to obtain the deviation amplitude.

[0084] If the actual value of the target welding variable is less than the lower limit value of its process adaptation interval, obtain the absolute difference between the actual value of the target welding variable and the lower limit value of its process adaptation interval, and perform ratio analysis with the lower limit value of its process adaptation interval to obtain the deviation amplitude.

[0085] In a preferred embodiment of the present invention, the preset process compliance quantification rule includes the following content: According to the preset defect-process correlation degree index interval, preset environment-process adaptation degree index interval, and preset process score value corresponding to each process level ladder, respectively determine the process level ladder corresponding to the defect-process correlation degree index and environment-process adaptation degree index analyzed in the spatio-temporal correlation matrix, and calculate the preset process score values of the process level ladders corresponding to the intervals of the two indexes according to the preset weight ratio to obtain the process score.

[0086] It should be noted that the preset weight ratio setting logic of the preset process score values corresponding to the process level ladder in the intervals where the above two indicators are located is as follows: the defect-process correlation index is greater than the environment-process adaptability index. The basis is that the defect-process correlation directly reflects the contractor's process execution ability (such as welding time control, groove treatment), which belongs to 100% controllable factors. The accident statistics of the Ministry of Housing and Urban-Rural Development show that 68% of the PE pipeline defects (such as excessive misalignment, insufficient welding temperature) can be attributed to substandard process operations, and the process stability difference of the same contractor in different environments reaches 52% (for example, the welding qualification rate fluctuation of an enterprise in dry / humid environments <5%, reflecting the process anti-interference ability). The environment-process adaptability reflects the passive adaptation ability of the process to the environment (such as whether to adjust the cooling time when the humidity > 80%), which belongs to partially controllable factors. Therefore, a relatively large preset weight ratio of the preset process score values corresponding to the interval where the defect-process correlation index is located is set for the process level ladder.

[0087] Exemplarily, the preset weight ratio of the preset process score values corresponding to the intervals where the above two indicators are located can be set to 0.72 for the defect-process correlation index and 0.28 for the environment-process adaptability index.

[0088] In the embodiment of the present invention, by establishing a spatio-temporal correlation matrix of welding defects, welding process parameters, and environmental monitoring parameters, the deep fusion of multi-source heterogeneous data is realized, breaking through the limitations of traditional single-data-dimensional analysis. On this basis, using fuzzy correlation analysis technology and preset process compliance quantification rules, complex process parameters and environmental conditions are converted into quantifiable process scores, realizing the dynamic assessment of the contractor's gas pipeline construction process.

[0089] The comprehensive evaluation module is used to perform weighted fusion calculation on the process score and the quality and efficiency score, generate the comprehensive evaluation value of the gas pipeline welding quality of the target contractor, and output a visual report.

[0090] In the embodiment of the present invention, through the spatio-temporal alignment of multi-source heterogeneous data, the three-dimensional data coupling of welding process parameters, environmental monitoring parameters, and visual features is realized, and the effective and accurate evaluation of the historical welding quality of the target contractor's gas pipeline is carried out from two perspectives of process compliance and quality and efficiency reliability, thereby greatly improving the reliability and practicality of the evaluation results.

[0091] Embodiment 2

[0092] Refer to Figure 3As shown in the figure, a device is provided in the second embodiment of the present invention, including: a processor, a memory, and a communication bus. A computer-readable program executable by the processor is stored on the memory. The communication bus realizes the connection and communication between the processor and the memory. When the processor executes the computer-readable program, the gas engineering construction quality assessment system based on big data can be realized.

[0093] Specifically, the above-mentioned memory and processor can be general memory and processor, and no specific limitation is made here. When the processor runs the computer-readable program stored in the memory, it can execute the relevant steps of the above system.

[0094] The processor may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above system can be completed by the integrated logic circuit in the hardware of the processor or the instructions in the form of software. The above-mentioned processor can be a general processor, including a central processing unit (CPU for short), a network processor (NP for short), etc. It can also be a digital signal processor (DSP for short), an application specific integrated circuit (ASIC for short), a field-programmable gate array (FPGA for short), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. It can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present application. The general processor can be a microprocessor or the processor can also be any conventional processor, etc. The relevant method steps of the system disclosed in combination with the embodiments of the present application can be directly embodied as being executed and completed by the hardware decoding processor, or by a combination of the hardware and software modules in the decoding processor. The software module can be located in a mature storage medium in the art such as random access memory, flash memory, read-only memory, programmable read-only memory, or electrically erasable programmable memory, registers, etc. This storage medium is located in the memory, and the processor reads the information in the memory and combines its hardware to complete the steps of the above system.

[0095] The above content is only an example and explanation of the structure of the present invention. Those skilled in the art of the present technology can make various modifications or supplements to the described specific embodiments or use similar methods for substitution, as long as they do not deviate from the structure of the invention or exceed the scope defined by the present invention, they should all fall within the protection scope of the present invention.

Claims

1. A gas engineering construction quality assessment system based on big data, characterized in that, The system includes: A data acquisition module that obtains a historical welding data set of gas pipelines of a target contractor through a multi-source data interface. The data set includes welding process parameters, environmental monitoring parameters, and a sequence of weld images arranged according to the construction time sequence. A quality and efficiency evaluation module that extracts defect features from the sequence of weld images, generates a feature tensor containing defect type codes, spatial coordinates, and timestamps, analyzes the defect disposal time interval data and location defect recurrence frequency data in the feature tensor based on spatio-temporal grid analysis, and generates a quality and efficiency score through a preset quality and efficiency reliability quantification rule. A process evaluation module that establishes a spatio-temporal correlation matrix of welding defects, welding process parameters, and environmental monitoring parameters, analyzes the defect-process correlation index and environment-process adaptability index in the spatio-temporal correlation matrix based on fuzzy correlation analysis, and generates a process score through a preset process compliance quantification rule. A comprehensive evaluation module that is used to perform weighted fusion calculation on the process score and the quality and efficiency score, generate a comprehensive evaluation value of the welding quality of the gas pipelines of the target contractor, and output a visual report.

2. The gas engineering construction quality assessment system based on big data according to claim 1, characterized in that: The specific reference for the defect disposal time interval data is as follows: a three-dimensional space grid is constructed based on the defect type code and spatial coordinates in the feature tensor. The time interval from the welding timestamp to the defect elimination timestamp of the same defect type code in the grid cell is used as the single-defect disposal time interval. Thus, the disposal time intervals of each defect type in each grid cell are collected as the defect disposal time interval data.

3. The gas engineering construction quality assessment system based on big data according to claim 2, characterized in that: The specific reference for the location defect recurrence frequency data is as follows: record the time and type of each defect occurrence in the same grid cell within a preset time window, introduce a time decay factor and a defect type weight factor for weighting to obtain the defect recurrence frequency of the same grid cell. Count the number of grid cells covered by the same defect type within a preset time window, and take the ratio of it to the total number of three-dimensional space grids as the location recurrence frequency of the same defect type. Collect the defect recurrence frequencies of each grid cell and the location recurrence frequencies of each defect type within a preset time window, which are jointly used as the location defect recurrence frequency data.

4. The gas engineering construction quality assessment system based on big data according to claim 3, characterized in that: The preset quality and efficiency reliability quantification rule includes the following: obtain the corresponding score values of the disposal time intervals of each defect type in each grid unit through a preset score mapping table, and obtain the timeliness score through hierarchical averaging and normalization processing. Substitute the average defect recurrence frequency of the grid cell and the average location recurrence frequency of the defect type within a preset time window into a preset penalty function respectively to quantify the grid recurrence deterioration value and the type diffusion deterioration value, take the higher value of the two-dimensional deterioration values, and perform a high-deduction processing on the initial stability full score to obtain the stability score. The sum of the timeliness score and the stability score is used as the quality and efficiency score.

5. The gas engineering construction quality assessment system based on big data according to claim 4, characterized in that: The preset score mapping table contains the non-linear mapping relationship between the disposal time interval of each defect type and the score value.

6. The gas engineering construction quality assessment system based on big data according to claim 1, characterized in that: The spatio-temporal correlation matrix establishes the mapping relationship between each welding defect and the corresponding process environment conditions through the spatio-temporal alignment of welding process parameters and environmental monitoring parameters.

7. The gas engineering construction quality assessment system based on big data according to claim 1, characterized in that: The specific method for the defect-process correlation index is as follows: Retrieve the synchronous welding process parameters corresponding to the time sequence intervals of each welding defect, construct a fuzzy grade division standard for process parameters using a trapezoidal membership function, and map the welding process parameters to predefined fuzzy grades through fuzzy processing to generate a membership degree matrix of process parameters corresponding to each welding defect moment. Scan the predefined association rule library through a preset association rule algorithm, extract a set of strong association rules that match the membership degree matrix of process parameters corresponding to the welding defect moment, record the confidence weights of each strong association rule, and use a linear weighted calculation method to obtain the correlation degree between each welding defect and the welding process parameters at its corresponding moment. After average calculation, output the correlation index between the welding defect and the welding process parameters.

8. The gas engineering construction quality assessment system based on big data according to claim 1, characterized in that: The specific method for the environment-process adaptability index is as follows: Randomly extract a certain welding variable from the welding process parameters at a certain spatial coordinate and a certain moment, mark this welding variable as the target welding variable, retrieve the strongly associated environmental variables that match the target welding variable through the predefined association rule library, and retrieve the monitoring values of the strongly associated environmental variables at the same spatial coordinate and the same moment. Define the standard value of the target welding variable according to the material characteristics of the gas pipeline at this spatial coordinate, and correct the standard value of the target welding variable through the monitoring values of the strongly associated environmental variables to generate a process adaptation interval for the target welding variable. Compare the actual value of the target welding variable with its process adaptation interval, and output the adaptability between the target welding variable and its strongly associated environmental variables. Similarly, collect the adaptabilities between each welding variable in the welding process parameters at each spatial coordinate and each moment and their strongly associated environmental variables, and obtain the environment-process adaptability index through hierarchical averaging.

9. The gas engineering construction quality assessment system based on big data according to claim 8, characterized in that: The specific method for the adaptability between the target welding variable and its strongly associated environmental variables is as follows: If the actual value of the target welding variable is within its process adaptation interval, output the adaptability between the target welding variable and its strongly associated environmental variables as 1. If the actual value of the target welding variable is outside its process adaptation interval, quantify the deviation amplitude between the actual value of the target welding variable and its process adaptation interval, and introduce the negative form of the deviation amplitude into the natural exponential function to solve the adaptability between the target welding variable and its strongly associated environmental variables.

10. The gas engineering construction quality assessment system based on big data according to claim 1, characterized in that: The preset process compliance quantification rules include the following: According to the preset defect-process correlation index intervals, preset environment-process adaptability index intervals, and preset process score values corresponding to each process level ladder, respectively determine the process level ladders corresponding to the defect-process correlation index and the environment-process adaptability index analyzed in the spatio-temporal correlation matrix. Weight the preset process score values of the process level ladders corresponding to the intervals of the two indexes according to the preset weight ratio to obtain the process score.

Images