A quality traceability method and system for the entire industry chain of thermal insulation engineering

Abstract

The present invention discloses a method and system for tracing the quality of the entire industrial chain of thermal insulation engineering, including obtaining industrial chain data and monitoring data of thermal insulation engineering, performing traceability coding based on the industrial chain data and the monitoring data and constructing a quality traceability network for the entire industrial chain of thermal insulation engineering, determining the industrial chain data corresponding to the monitoring data to be traced through traceability coding, correcting and screening the monitoring data to be traced to obtain abnormal monitoring data to be traced, and determining the source of quality hazards of the entire industrial chain of thermal insulation engineering according to the abnormal monitoring data to be traced and the quality traceability network for the entire industrial chain of thermal insulation engineering. This method can not only improve the efficiency and accuracy of the quality traceability of the entire industrial chain of thermal insulation engineering, but also has good interpretability, and can be directly applied to the quality traceability system of the entire industrial chain of thermal insulation engineering.

Description

Technical Field

[0001] The present invention relates to the field of data traceability, and in particular to a method and system for tracing the quality of the entire industrial chain of thermal insulation engineering. Background Art

[0002] With the rapid development of the construction industry, thermal insulation projects, as a key link in improving building energy efficiency, saving energy and reducing emissions, have a quality that is directly related to the energy efficiency performance, living comfort and service life of buildings. Therefore, quality traceability of the entire thermal insulation project industry chain is of great significance to the overall quality control of thermal insulation projects.

[0003] Since the traditional thermal insulation engineering quality management model focuses on the inspection and acceptance of terminal products, but the thermal insulation engineering industry chain is long and complicated, and the data of each link are independent of each other and lack effective integration and association, once a quality problem occurs, it is difficult to quickly and accurately trace the root cause of the problem, and it is impossible to take targeted rectification measures in time. In addition, the monitoring data of different links have problems such as inconsistent formats and incomplete information, which further increases the difficulty of quality tracing of the entire industrial chain of thermal insulation engineering. The present invention matches the monitoring data of terminal products with each process chain by dividing the industrial chain data into process chains and systematically processing and analyzing them. Using machine learning technology, the causal relationship between monitoring data and process chain data is clarified, and a complete quality traceability network is constructed. The present invention proposes a quality traceability method for the entire industrial chain of thermal insulation engineering, which effectively overcomes the limitations of existing quality traceability methods, realizes the effective association of data in each link and the accurate tracing of quality problems. This helps to ensure the energy-saving effect and safety of use of buildings, and meet the needs of sustainable development of modern buildings. Summary of the invention

[0004] The purpose of the present invention is to provide a method and system for tracing the quality of the entire industrial chain of thermal insulation engineering.

[0005] To achieve the above object, the present invention is implemented according to the following technical solutions:

[0006] The present invention comprises the following steps:

[0007] Obtaining industrial chain data and monitoring data of thermal insulation projects, and performing traceability coding and spatial coding based on the industrial chain data and the monitoring data;

[0008] Constructing a quality traceability network for the entire industry chain of thermal insulation engineering based on the industry chain data and the monitoring data;

[0009] Determine the industrial chain data corresponding to the monitoring data to be traced through the traceability code, correct and screen the monitoring data to be traced for abnormalities to obtain the abnormal monitoring data to be traced;

[0010] The sources of quality hazards in the entire industry chain of thermal insulation projects are determined based on the abnormal monitoring data to be traced and the quality traceability network of the entire industry chain of thermal insulation projects.

[0011] Furthermore, the method for performing traceability coding and spatial coding based on the industrial chain data and the monitoring data includes:

[0012] The monitoring data is related to the effect of the current thermal insulation project, including energy-saving effect index, living comfort index and safety index. The method for obtaining the monitoring data is:

[0013] Collect energy consumption data of metering devices in buildings based on Internet of Things technology, and determine energy-saving effect indicators based on energy consumption data in buildings; set up different types of sensors inside buildings to collect indoor environmental data, and determine living comfort indicators based on indoor environmental data; sample the anchoring force and bonding strength of thermal insulation engineering materials, use infrared thermal imaging and high-frequency radar detection methods to determine the quality factors of external insulation systems, and directly determine the toxic substances in thermal insulation engineering;

[0014] Divide the entire industry chain of thermal insulation engineering into process chains, collect corresponding parameter information according to the process chains to obtain process chain data, set timestamps according to the process chain data, and obtain industry chain data in series according to the time series;

[0015] Determine the coding format and coding length of the process chain according to the specific process and timestamp of the process chain, store the process chain information and process chain code in the RFID chip inside the thermal insulation project, partition and store the corresponding process chain data in the industrial chain database according to the thermal insulation project process chain code, connect all process chain codes in series according to the time series to obtain the traceability code of the industrial chain data, mark the corresponding traceability code when completing the last process of the thermal insulation project, and partition and store the corresponding monitoring data in the industrial chain database according to the thermal insulation project traceability code; the traceability code reflects the relationship between the industrial chain data and the monitoring data;

[0016] Draw the REVIT building model, determine the spatial coding of the thermal insulation project according to the monitoring area and spatial coordinates of the thermal insulation project monitoring data, and determine the unique correspondence between the thermal insulation project traceability code and the spatial coding.

[0017] Furthermore, a method for constructing a quality traceability network for the entire industry chain of thermal insulation engineering based on the industry chain data and the monitoring data comprises the following steps:

[0018] Determine the traceability quality characterization vector based on the industrial chain data;

[0019] Conduct process chain traceability based on traceability quality characterization vectors and monitoring data, and divide monitoring data into each process chain;

[0020] Determine the traceability performance of each monitoring data in each process chain and generate a quality traceability network for the entire industry chain of thermal insulation engineering.

[0021] Furthermore, the method for determining the traceability quality characterization vector according to the industrial chain data includes:

[0022] The industrial chain data is preprocessed to obtain an industrial chain traceability vector, and the industrial chain traceability vector is input into the industrial chain data decomposition framework to obtain a first time series vector, a second time series vector, a first process vector, a second process vector and a third process vector;

[0023] The industrial chain data decomposition framework includes a time series decomposition algorithm, a principal component analysis method, a subcomponent analysis method and anomaly detection algorithm; the time series decomposition algorithm decomposes the industrial chain traceability vector into a first time series vector and a second time series vector according to the time nodes of all processes; the principal component decomposition algorithm converts multiple related variables related to process quality in the industrial chain traceability vector into a small number of irrelevant principal components, and the principal components form a first process vector; the subcomponent decomposition algorithm converts multiple related variables related to process quality in the industrial chain traceability vector into a small number of irrelevant subcomponents, and the subcomponents form a second process vector; the anomaly detection algorithm obtains a third process vector by marking abnormal operations in the industrial chain traceability vector through unsupervised learning of sequence anomalies and quantitative anomalies; the first time series vector matches the time nodes of each process chain; the second time series vector contains the time nodes of all processes;

[0024] The first process vector is crossed with the first timing vector to obtain the first process vector, the second process vector is crossed with the second timing vector to obtain the second process vector, the third process vector is fused with the first process vector feature to obtain the first traceability attribute representation vector, the third process vector is fused with the second process vector feature to obtain the second traceability attribute representation vector, and the first traceability attribute representation vector and the second traceability attribute representation vector are weighted to obtain the traceability quality representation vector.

[0025] Furthermore, the method of dividing the monitoring data into each process chain includes:

[0026] Based on the time series characteristics, the traceability quality characterization vector is segmented according to the process chain, and the monitoring data is divided into each process chain using the population search algorithm. The data is the population individual, the process chain divided by the data is the optimal position of the individual, and the search point of the discoverer is determined:

[0027]

[0028] in Search point directly in front of the finder, Search point for the finder's left side, is the search point on the right side of the finder, and the search point with the best fitness is taken as the finder No. Locations searched , For the The finder individual of the search, is the target cyclic shift factor, is an arbitrary number with a normal distribution characteristic of 0 and a variance of 1. is any uniformly distributed number, is the spatial search angle, is the maximum spatial search angle, is the search direction, For the The maximum search distance for each search, is the upper bound of the design variable, is the lower bound of the design variable;

[0029] Update the spatial search angle and maximum search distance:

[0030]

[0031]

[0032] in For the The spatial search angle of the search, is the maximum search angle, For the The maximum search distance for each search, is the search distance constant;

[0033] The coupled genetic algorithm and swarm search algorithm generate new wanderer positions, update the discoverer positions and wanderers:

[0034]

[0035]

[0036] in For the discoverer The location of the search. For the Wanderer The location of the search. , is the position update factor, for dimensional vector of any number, For the The position of the individual with the best positional fitness in the population during the first search, For the Search time The best historical position of an individual, is the scaling constant, , is the probability of the positive component appearing, for dimensional uniform distribution of any number, is the individual mutation probability;

[0037] Calculate and update the individual fitness of the population, continue to search for the updated population position until the individual fitness no longer changes, stop searching, and obtain the optimal position of the population individual to determine the process chain division of the monitoring data.

[0038] Furthermore, the method of determining the traceability of each monitoring data in each process chain includes:

[0039] The monitoring data of the process chain and the traceability quality characterization vector are combined into a traceability data set, and the traceability data set is divided into a training set and a test set;

[0040] Constructing a quality traceability network model for the entire industry chain of thermal insulation engineering, wherein the quality traceability network model for the entire industry chain of thermal insulation engineering includes a calculation module, a rule module, and a machine learning module;

[0041] The calculation module calculates the correlation between the monitoring data and the traceability quality characterization vector, and determines the correlation between the monitoring data and the process chain data in each process chain. The expression is:

[0042]

[0043]

[0044]

[0045] in For the The correlation between the monitoring data in a process chain and the traceability quality characterization vector, is the positive ideal solution of the evaluation scheme, is the negative ideal solution of the evaluation scheme, is the time series stability factor, which is determined by calculating the autocorrelation function of the time series. is the time correlation weight, is the abnormal adjustment factor, determined by statistical methods, Abnormal adjustment of association weights, For the The weight of the time scale, is the time scale quantity, is the combined weight, Represent the weights of the least squares method, coefficient of variation method and entropy weight method respectively, For the The first Monitoring data, For the The optimal value in the set of traceability quality characterization vector elements within a process chain, For the The worst value in the set of traceability quality characterization vector elements within a process chain, is the resolution factor;

[0046] The rule module determines the association between monitoring data and process chain data in each process chain by setting the correlation threshold. The specific association rules are:

[0047] When the energy-saving effect correlation between the energy-saving effect monitoring data and the traceability quality characterization vector is greater than the energy-saving effect correlation threshold, the process chain monitoring data is determined to be related to the process chain data of the first category; when the residential comfort correlation between the residential comfort monitoring data and the traceability quality characterization vector is greater than the residential comfort correlation threshold, the process chain monitoring data is determined to be related to the process chain data of the second category; when the safety correlation between the safety monitoring data and the traceability quality characterization vector is greater than the safety correlation threshold, the process chain monitoring data is determined to be related to the process chain data of the third category;

[0048] When the energy-saving effect correlation, living comfort correlation, and safety correlation are all less than or equal to the corresponding correlation threshold, the comprehensive correlation of all monitoring data of the process chain and the traceability quality characterization vector is calculated. If the comprehensive correlation is greater than the comprehensive correlation threshold, it is determined that the process chain monitoring data is comprehensively correlated with the process chain data. Otherwise, it is determined that the process chain monitoring data is not correlated with the process chain data.

[0049] The machine learning module generates a quality traceability network for the entire industry chain of thermal insulation engineering based on the correlation between monitoring data and process chain data, including graph models, classification algorithms and correlation analysis; uses correlation analysis to learn the association rules and association results output by the rule module; uses a graph model to build a relationship network between monitoring data and process chain data based on the correlation analysis results, locates and links monitoring data and process chain data within the process chain, and causally links data that conforms to data association rules, while marking relevant types on data links; uses a support vector machine (SVM) classification algorithm to predict the relationship between nodes based on the correlation analysis and graph model results;

[0050] The training set data is used to train the model, and the test set data is used to evaluate the model and optimize its parameters. The quality traceability network model of the entire industrial chain of thermal insulation projects is output, and the traceability performance of each monitoring data in each process chain is determined based on the quality traceability network model of the entire industrial chain of thermal insulation projects.

[0051] Furthermore, the method for obtaining the abnormal monitoring data to be traced includes:

[0052] Obtain the monitoring data to be traced, determine its spatial code according to the monitoring area and spatial coordinates of the monitoring data to be traced in the REVIT building model, determine the traceability code of the monitoring data to be traced according to the unique correspondence of the code, extract the process chain data to be traced by region through the traceability code, and concatenate the process chain data to be traced in time series to obtain the industrial chain data to be traced;

[0053] Obtain status data of the thermal insulation project, and perform data correction on the traceability monitoring data according to the status data; the status data includes installation location, installation process and external building environment indicators;

[0054] The corrected monitoring data to be traced are screened using the hypersphere method to obtain abnormal monitoring data to be traced.

[0055] Furthermore, the method for determining the quality risk sources of the entire industry chain of the thermal insulation project includes:

[0056] According to the quality traceability network of the entire industry chain of thermal insulation engineering, the abnormal monitoring data to be traced are divided into each process chain, and the abnormal degree of the abnormal monitoring data to be traced in each process chain is calculated. According to the abnormal degree, the abnormal monitoring data to be traced are traced to obtain the quality hidden danger source of the entire industry chain of thermal insulation engineering. The specific traceability rules are as follows:

[0057] When the energy-saving effect abnormality of the energy-saving effect monitoring data to be traced in the process chain is greater than the energy-saving effect abnormality threshold, the first type of related traceability is carried out according to the causal link of the quality traceability network of the entire industrial chain of thermal insulation engineering; when the living comfort abnormality of the living comfort abnormality of the living comfort monitoring data to be traced in the process chain is greater than the living comfort abnormality threshold, the second type of related traceability is carried out according to the causal link of the quality traceability network of the entire industrial chain of thermal insulation engineering; when the safety abnormality of the safety abnormality of the safety abnormality of the process chain is greater than the safety abnormality threshold, the third type of related traceability is carried out according to the causal link of the quality traceability network of the entire industrial chain of thermal insulation engineering;

[0058] When the energy-saving effect abnormality, living comfort abnormality, and safety abnormality are all less than or equal to the corresponding abnormality threshold, the comprehensive abnormality of the abnormal monitoring data to be traced in the process chain is calculated. If the comprehensive abnormality is greater than the comprehensive abnormality threshold, comprehensive related traceability is performed based on the causal link of the quality traceability network of the entire industry chain of thermal insulation engineering. Otherwise, quality traceability is not performed.

[0059] When conducting causal link tracing, the corresponding link annotations and corresponding process chain data are extracted, and the quality risk sources of the process chain are determined based on the link annotations and the corresponding process chain data. All the quality risk sources of the process chain are combined to determine the quality risk sources of the entire industrial chain of thermal insulation projects.

[0060] Second, a quality traceability system for the entire industry chain of thermal insulation engineering, including:

[0061] Data acquisition module: used to obtain industrial chain data and monitoring data of thermal insulation projects, and to perform traceability coding and spatial coding on the industrial chain data and the monitoring data;

[0062] Data partitioning module: used to determine the traceability quality characterization vector according to the industrial chain data, to perform process chain traceability according to the traceability quality characterization vector and the monitoring data, and to divide the monitoring data into each process chain using a population search algorithm;

[0063] Traceability network module: used to construct a quality traceability network model for the entire industry chain of thermal insulation engineering based on monitoring data and traceability quality characterization vectors, including calculating the correlation between the monitoring data and the traceability quality characterization vectors, setting a correlation threshold to determine the correlation between the monitoring data and the process chain data in each process chain, using a machine learning model to learn the correlation between the data, and generating a quality traceability network for the entire industry chain of thermal insulation engineering;

[0064] Traceability module: used to extract the industrial chain data to be traced according to the monitoring data to be traced, to correct and screen the monitoring data to be traced to obtain abnormal monitoring data to be traced, to calculate the abnormality degree of the abnormal monitoring data to be traced in each process chain, and to perform quality traceability according to the abnormality degree;

[0065] Intelligent supervision module: used to store, view and manage the industrial chain data, the monitoring data and the quality hidden danger sources of the entire industrial chain of the thermal insulation project, and to warn users based on the quality traceability of the entire industrial chain of the thermal insulation project.

[0066] The beneficial effects of the present invention are:

[0067] The present invention is a method and system for tracing the quality of the entire industry chain of thermal insulation engineering. Compared with the prior art, the present invention has the following technical effects:

[0068] The present invention significantly improves the data preprocessing capability and model adaptability in the quality traceability of the entire industrial chain of thermal insulation engineering through a series of steps, including data encoding, data matching, correlation determination, construction of a traceability network, data correction, anomaly screening and quality traceability. This not only improves the efficiency and accuracy of the quality traceability of the entire industrial chain of thermal insulation engineering, but also optimizes related technologies. By using this invention technology, resources can be greatly saved, work efficiency can be improved, and effective traceability and precise management of the quality of the entire industrial chain of thermal insulation engineering can be achieved. It is of great significance to improve the overall quality of building insulation engineering and improve the safety of exterior wall insulation engineering, and can adapt to the needs of the quality traceability system of the entire industrial chain of different insulation engineering and different product technology systems, meet the specific needs of different users in terminal traceability, and show its wide applicability. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 The present invention is a flow chart of the steps of a method for tracing the quality of the entire industrial chain of thermal insulation engineering. DETAILED DESCRIPTION

[0070] The present invention is further described below by means of specific embodiments. The illustrative embodiments and descriptions of the present invention are used to explain the present invention but are not intended to limit the present invention.

[0071] The present invention provides a method and system for tracing the quality of the entire industry chain of thermal insulation engineering, comprising the following steps:

[0072] like Figure 1 As shown, in this embodiment, the following steps are included:

[0073] Obtaining industrial chain data and monitoring data of thermal insulation projects, and performing traceability coding and spatial coding based on the industrial chain data and the monitoring data;

[0074] Constructing a quality traceability network for the entire industry chain of thermal insulation engineering based on the industry chain data and the monitoring data;

[0075] Determine the industrial chain data corresponding to the monitoring data to be traced through the traceability code, correct and screen the monitoring data to be traced for abnormalities to obtain the abnormal monitoring data to be traced;

[0076] The sources of quality hazards in the entire industry chain of thermal insulation projects are determined based on the abnormal monitoring data to be traced and the quality traceability network of the entire industry chain of thermal insulation projects.

[0077] In this embodiment, the method for performing traceability coding and spatial coding based on the industrial chain data and the monitoring data includes:

[0078] The monitoring data is related to the effect of the current thermal insulation project, including energy-saving effect index, living comfort index and safety index. The method for obtaining the monitoring data is:

[0079] The consumption of electricity, water, gas and heat in the building is collected through smart electricity meters, water meters, gas meters and other metering devices. The energy consumption data in the building is collected by connecting the metering devices in the building based on the Internet of Things technology, and the thermal conductivity, thermal insulation performance, heat insulation performance and moisture resistance of the thermal insulation materials are calculated, and the energy consumption indicators of the thermal insulation project are calculated;

[0080] An indoor distributed sensor network is constructed by installing temperature and humidity sensors, light sensors and carbon dioxide measuring devices indoors. The indoor temperature, humidity, light intensity and carbon dioxide concentration are collected and updated by constructing the indoor distributed sensor network, and the living comfort index is calculated based on the data collected by the indoor distributed sensor network.

[0081] Supervise and randomly test the anchoring force of anchors, the bonding strength between insulation boards and the base layer, use infrared thermal imaging and high-frequency radar detection to determine cracking, hollowing, shedding and leakage, use gas chromatography to determine the VOC content of spare thermal insulation engineering materials, use polarizing microscope to determine the asbestos content of spare thermal insulation engineering materials, and conduct combustion tests on spare thermal insulation engineering materials to determine their flame retardant properties;

[0082] The entire industry chain of thermal insulation engineering is divided into process chains such as raw material production, material processing, material transportation, material sales, engineering installation and engineering acceptance. The corresponding parameter information is collected according to the process chain to obtain process chain data, and the timestamp is set according to the process chain data. The industry chain data is obtained by connecting in series according to the time series.

[0083] Determine the coding format and coding length of the process chain according to the specific process and timestamp of the process chain, store the process chain information and process chain code in the RFID chip inside the thermal insulation project, partition and store the corresponding process chain data in the industrial chain database according to the thermal insulation project process chain code, connect all process chain codes in series according to the time series to obtain the traceability code of the industrial chain data, mark the corresponding traceability code when completing the last process of the thermal insulation project, and partition and store the corresponding monitoring data in the industrial chain database according to the thermal insulation project traceability code; the traceability code reflects the relationship between the industrial chain data and the monitoring data;

[0084] Draw the REVIT building model, determine the spatial code of the thermal insulation project according to the monitoring area and spatial coordinates of the thermal insulation project monitoring data, and determine the unique correspondence between the thermal insulation project traceability code and the spatial code;

[0085] In the actual evaluation, taking the whole industry chain quality traceability of a certain building thermal insulation project as an example, the specific rules for determining the thermal insulation project space coding according to the monitoring area and spatial coordinates of the thermal insulation project monitoring data are as follows: one-stage coding is performed according to the building, such as the second civil building uses the English abbreviation RB of Residential Building and the digital number 2; two-stage coding is performed according to the floor, such as the thermal insulation project corresponding to the monitoring data is located on the 5th floor, it is F05; three-stage coding is performed according to the direction of the thermal insulation project, such as the thermal insulation project is located at 45 degrees east of south, the direction abbreviation EN and the digital number 45 are used; four-stage coding is performed according to the regional division and monitoring area of ​​the corresponding thermal insulation project, such as the thermal insulation project is divided into the second area of ​​the floor and the monitoring area is 10 square meters, it is 02-10; the final space coding is RB2-F05-EN45-02-10;

[0086] The coding rules for traceability coding are as follows: the coding format of the raw material production process chain is [material abbreviation] [category number] [batch number], for example, PLC01B01 represents the category number 01 and batch number B01 of polystyrene foam PLC (Polystyrene Foam); the coding format of the material processing process chain is [processing category] [number of processing steps], for example, RM3FM5 represents that the material has undergone 3 rough machining RM (Rough Machining) and 5 fine machining FM (Fine Machining). Machining); the coding format of the material transportation process chain is [transfer point] [transportation tool] [transportation time] [transfer point], for example, BJTRK48hZZ means that the material was transported from Beijing to Zhengzhou by truck TRK for 48 hours; the coding format of the material sales process chain is [storage method] [storage time], for example, WHS33DAY means that the material was stored in the warehouse WHS (Warehouse) for 33 days; the coding format of the material installation process chain is [installation process] [process time] [installer], for example, PLS04hZQ means that the construction worker Zhang Qiang spent 4 hours on the plastering PLS (Plastering); the coding format of the project acceptance process chain is [acceptance type] [acceptance stage, serial number] [key indicator / time / batch], for example, EEA02D02B01 means that the second test (D02) in the second generation implementation stage of energy-saving acceptance EEA (Energy Efficiency Acceptance) is aimed at the B01 indicator.

[0087] In this embodiment, the method for constructing a quality traceability network for the entire industrial chain of thermal insulation engineering according to the industrial chain data and the monitoring data includes the following steps:

[0088] Determine the traceability quality characterization vector based on the industrial chain data;

[0089] Conduct process chain traceability based on traceability quality characterization vectors and monitoring data, and divide monitoring data into corresponding process chains;

[0090] Determine the traceability performance of monitoring data in the corresponding process chain and generate a quality traceability network for the entire industry chain of thermal insulation engineering.

[0091] In this embodiment, the method for determining the traceability quality characterization vector according to the industrial chain data includes:

[0092] The industrial chain data is preprocessed to obtain an industrial chain traceability vector, and the industrial chain traceability vector is input into the industrial chain data decomposition framework to obtain a first time series vector, a second time series vector, a first process vector, a second process vector and a third process vector;

[0093] The industrial chain data decomposition framework includes a time series decomposition algorithm, a principal component analysis method, a subcomponent analysis method and anomaly detection algorithm; the time series decomposition algorithm decomposes the industrial chain traceability vector into a first time series vector and a second time series vector according to the time nodes of all processes; the principal component decomposition algorithm converts multiple related variables related to process quality in the industrial chain traceability vector into a small number of irrelevant principal components, and the principal components form a first process vector; the subcomponent decomposition algorithm converts multiple related variables related to process quality in the industrial chain traceability vector into a small number of irrelevant subcomponents, and the subcomponents form a second process vector; the anomaly detection algorithm obtains a third process vector by marking abnormal operations in the industrial chain traceability vector through unsupervised learning of sequence anomalies and quantitative anomalies; the first time series vector matches the time nodes of each process chain; the second time series vector contains the time nodes of all processes;

[0094] The first process vector is crossed with the first timing vector to obtain the first process vector, the second process vector is crossed with the second timing vector to obtain the second process vector, the third process vector is fused with the first process vector feature to obtain the first traceability attribute representation vector, the third process vector is fused with the second process vector feature to obtain the second traceability attribute representation vector, and the first traceability attribute representation vector and the second traceability attribute representation vector are weighted to obtain the traceability quality representation vector.

[0095] In this embodiment, the method of dividing the monitoring data into each process chain includes:

[0096] Based on the time series characteristics, the traceability quality characterization vector is segmented according to the process chain, and the population search algorithm is used to divide the monitoring data into each process chain. A certain data in the monitoring data is defined as a population individual, and the process chain divided by the data is the optimal position of the individual, and the search point of the discoverer is determined:

[0097]

[0098] in Search point directly in front of the finder, Search point for the finder's left side, is the search point on the right side of the finder, and the search point with the best fitness is taken as the finder No. Locations searched , For the The finder individual of the search, is the target cyclic shift factor, is an arbitrary number with a normal distribution characteristic of 0 and a variance of 1. is any uniformly distributed number, is the spatial search angle, is the maximum spatial search angle, is the search direction, For the The maximum search distance for each search, is the upper bound of the design variable, is the lower bound of the design variable;

[0099] Update the spatial search angle and maximum search distance:

[0100]

[0101]

[0102] in For the The spatial search angle of the search, is the maximum search angle, For the The maximum search distance for each search, is the search distance constant;

[0103] The coupled genetic algorithm and swarm search algorithm generate new wanderer positions, update the discoverer positions and wanderers:

[0104]

[0105]

[0106] in For the discoverer The location of the search. For the Wanderer The location of the search. , is the position update factor, for dimensional vector of any number, For the The position of the individual with the best positional fitness in the population during the first search, For the Search time The best historical position of an individual, is the scaling constant, , is the probability of the positive component appearing, for dimensional uniform distribution of any number, is the individual mutation probability;

[0107] Calculate and update the individual fitness of the population, continue to search for the updated population position until the individual fitness no longer changes, stop searching, and obtain the optimal position of the population individual to determine the process chain division of the monitoring data;

[0108] In the actual assessment, the process chain division of monitoring data was obtained through data search: 1. Raw material production process chain: raw material type and composition, thermal conductivity, thermal insulation performance, VOC content, asbestos content; 2. Material processing process chain: thermal insulation performance, heat insulation performance, moisture resistance, flame retardant performance; 3. Material transportation process chain: thermal conductivity, moisture resistance, and the bonding strength between the insulation board and the base layer; 4. Material sales process chain: VOC content, asbestos content, moisture resistance, and anchoring force of anchors; 5. Material installation process chain: thermal insulation performance, heat insulation performance, moisture resistance, anchoring force of anchors, bonding strength between the insulation board and the base layer, cracking, hollowing, falling off, and leakage.

[0109] In this embodiment, the method for determining the traceability performance of each monitoring data in each process chain includes:

[0110] The monitoring data of the process chain and the traceability quality characterization vector are combined into a traceability data set, and the traceability data set is divided into a training set and a test set;

[0111] Constructing a quality traceability network model for the entire industry chain of thermal insulation engineering, wherein the quality traceability network model for the entire industry chain of thermal insulation engineering includes a calculation module, a rule module, and a machine learning module;

[0112] The calculation module calculates the correlation between the monitoring data and the traceability quality characterization vector, and determines the correlation between the monitoring data and the process chain data in each process chain. The expression is:

[0113]

[0114]

[0115]

[0116] in For the The correlation between the monitoring data in a process chain and the traceability quality characterization vector, is the positive ideal solution of the evaluation scheme, is the negative ideal solution of the evaluation scheme, is the time series stability factor, which is determined by calculating the autocorrelation function of the time series. is the time correlation weight, is the abnormal adjustment factor, determined by statistical methods, Abnormal adjustment of association weights, For the The weight of the time scale, is the time scale quantity, is the combined weight, Represent the weights of the least squares method, coefficient of variation method and entropy weight method respectively, For the The first Monitoring data, For the The optimal value in the set of traceability quality characterization vector elements within a process chain, For the The worst value in the set of traceability quality characterization vector elements within a process chain, is the resolution factor;

[0117] The rule module determines the association between monitoring data and process chain data in each process chain by setting the correlation threshold. The specific association rules are:

[0118] When the energy-saving effect correlation between the energy-saving effect monitoring data and the traceability quality characterization vector is greater than the energy-saving effect correlation threshold, the process chain monitoring data is determined to be related to the process chain data of the first category; when the residential comfort correlation between the residential comfort monitoring data and the traceability quality characterization vector is greater than the residential comfort correlation threshold, the process chain monitoring data is determined to be related to the process chain data of the second category; when the safety correlation between the safety monitoring data and the traceability quality characterization vector is greater than the safety correlation threshold, the process chain monitoring data is determined to be related to the process chain data of the third category;

[0119] When the energy-saving effect correlation, living comfort correlation, and safety correlation are all less than or equal to the corresponding correlation threshold, the comprehensive correlation of all monitoring data of the process chain and the traceability quality characterization vector is calculated. If the comprehensive correlation is greater than the comprehensive correlation threshold, it is determined that the process chain monitoring data is comprehensively correlated with the process chain data. Otherwise, it is determined that the process chain monitoring data is not correlated with the process chain data.

[0120] The machine learning module generates a quality traceability network for the entire industry chain of thermal insulation engineering based on the correlation between monitoring data and process chain data, including graph models, classification algorithms and correlation analysis; uses correlation analysis to learn the association rules and association results output by the rule module; uses a graph model to build a relationship network between monitoring data and process chain data based on the correlation analysis results, locates and links monitoring data and process chain data within the process chain, and causally links data that conforms to data association rules, while marking relevant types on data links; uses a support vector machine (SVM) classification algorithm to predict the relationship between nodes based on the correlation analysis and graph model results;

[0121] The training set data is used to train the model, and the test set data is used to evaluate the model and optimize its parameters. The quality traceability network model of the entire industrial chain of thermal insulation projects is output, and the traceability performance of each monitoring data in each process chain is determined based on the quality traceability network model of the entire industrial chain of thermal insulation projects.

[0122] In this embodiment, the method for obtaining the abnormal monitoring data to be traced includes:

[0123] Obtain the monitoring data to be traced, determine its spatial code according to the monitoring area and spatial coordinates of the monitoring data to be traced in the REVIT building model, determine the traceability code of the monitoring data to be traced according to the unique correspondence of the code, extract the process chain data to be traced by region through the traceability code, and concatenate the process chain data to be traced in time series to obtain the industrial chain data to be traced;

[0124] Obtain status data of the thermal insulation project, and perform data correction on the traceability monitoring data according to the status data; the status data includes installation location, installation process and external building environment indicators;

[0125] The corrected monitoring data to be traced are screened by the hypersphere method to obtain abnormal monitoring data to be traced;

[0126] In the actual evaluation, the traceability monitoring data of a certain building thermal insulation project were obtained: 1. Energy saving effect index (thermal conductivity 0.045W / (m·K), thermal insulation performance 75%, heat insulation performance 70%, moisture resistance 85%, energy consumption index 120kWh / m 2 every year); 2. Living comfort index (indoor temperature 20℃, humidity 60%, light intensity 380lux, carbon dioxide concentration 780ppm); 3. Safety performance index (anchoring force 20kN, insulation layer and base bonding strength 0.38MPa, local cracking, no hollowing phenomenon, VOC content 20mg / m 3 , asbestos content 0%, flame retardant performance V2 grade);

[0127] After correcting and screening the traceability monitoring data according to the status data, the traceability monitoring data are determined to be: thermal insulation performance / too low, heat insulation performance / too low, energy consumption index / too high, indoor temperature / too low, anchoring force of anchors / too low, bonding strength between insulation board and base layer / too low, and local cracks.

[0128] In this embodiment, the method for determining the quality risk sources of the entire industry chain of the thermal insulation project includes:

[0129] According to the quality traceability network of the entire industry chain of thermal insulation engineering, the abnormal monitoring data to be traced are divided into each process chain, and the abnormal degree of the abnormal monitoring data to be traced in each process chain is calculated. According to the abnormal degree, the abnormal monitoring data to be traced are traced to obtain the quality hidden danger source of the entire industry chain of thermal insulation engineering. The specific traceability rules are as follows:

[0130] When the energy-saving effect abnormality of the energy-saving effect monitoring data to be traced in the process chain is greater than the energy-saving effect abnormality threshold, the first type of related traceability is carried out according to the causal link of the quality traceability network of the entire industrial chain of thermal insulation engineering; when the living comfort abnormality of the living comfort abnormality of the living comfort monitoring data to be traced in the process chain is greater than the living comfort abnormality threshold, the second type of related traceability is carried out according to the causal link of the quality traceability network of the entire industrial chain of thermal insulation engineering; when the safety abnormality of the safety abnormality of the safety abnormality of the process chain is greater than the safety abnormality threshold, the third type of related traceability is carried out according to the causal link of the quality traceability network of the entire industrial chain of thermal insulation engineering;

[0131] When the energy-saving effect abnormality, living comfort abnormality, and safety abnormality are all less than or equal to the corresponding abnormality threshold, the comprehensive abnormality of the abnormal monitoring data to be traced in the process chain is calculated. If the comprehensive abnormality is greater than the comprehensive abnormality threshold, comprehensive related traceability is performed based on the causal link of the quality traceability network of the entire industry chain of thermal insulation engineering. Otherwise, quality traceability is not performed.

[0132] When tracing the cause-effect links, extract the corresponding link annotations and corresponding process chain data, determine the quality risk sources of the process chain based on the link annotations and corresponding process chain data, and combine all the quality risk sources of the process chain to determine the quality risk sources of the entire industry chain of thermal insulation engineering;

[0133] In the actual evaluation, the process chain division results of the above-mentioned monitoring data to be traced are determined based on the quality traceability network of the entire industry chain of thermal insulation engineering: 1. Raw material production process chain (thermal insulation performance / low); 2. Material processing process chain (thermal insulation performance / low, thermal insulation performance / low); 3. Material transportation process chain (insulation board and base layer bonding strength / low); 4. Material sales process chain (anchoring force / low); 5. Material installation process chain (thermal insulation performance / low, thermal insulation performance / low, anchoring force / low, insulation board and base layer bonding strength / low, local cracking); Energy consumption indicators, indoor temperature and thermal insulation / thermal insulation performance are directly bound;

[0134] Calculate the abnormal degree of the abnormal monitoring data to be traced in each process chain, trace the abnormal detection data to be traced that meets the traceability rules according to the quality traceability network of the entire industry chain of thermal insulation engineering to obtain the corresponding causal link, and extract the abnormal operation records of the third process vector of the corresponding process chain to determine the source of quality hazards in the process chain: 1. The low bonding strength between the insulation board and the base layer is not related to the material transportation process chain, but is related to the material installation process chain. The specific influencing factor is the insufficient bonding strength of the bonding material during the installation process; 2. The anchoring force of the anchor is low and is related to the material sales process chain. The specific influencing factor is the decrease in the anchoring force due to the extrusion structure deformation during material storage; 3. The thermal insulation performance is low and the heat insulation performance is low and is related to the material installation process chain. The specific influencing factors are the insufficient thickness of the insulation layer and the existence of gaps; 4. The existence of local cracks is related to the material installation process chain. The specific influencing factors are the low surface flatness requirements of the base layer during installation and the insufficient bonding strength of the bonding material.

[0135] Second, a quality traceability system for the entire industry chain of thermal insulation engineering, including:

[0136] Data acquisition module: used to obtain industrial chain data and monitoring data of thermal insulation projects, and to perform traceability coding and spatial coding on the industrial chain data and the monitoring data;

[0137] Data partitioning module: used to determine the traceability quality characterization vector according to the industrial chain data, to perform process chain traceability according to the traceability quality characterization vector and the monitoring data, and to divide the monitoring data into each process chain using a population search algorithm;

[0138] Traceability network module: used to construct a quality traceability network model for the entire industry chain of thermal insulation engineering based on monitoring data and traceability quality characterization vectors, including calculating the correlation between the monitoring data and the traceability quality characterization vectors, setting a correlation threshold to determine the correlation between the monitoring data and the process chain data in each process chain, using a machine learning model to learn the correlation between the data, and generating a quality traceability network for the entire industry chain of thermal insulation engineering;

[0139] Traceability module: used to extract the industrial chain data to be traced according to the monitoring data to be traced, to correct and screen the monitoring data to be traced to obtain abnormal monitoring data to be traced, to calculate the abnormality degree of the abnormal monitoring data to be traced in each process chain, and to perform quality traceability according to the abnormality degree;

[0140] Intelligent supervision module: used to store, view and manage the industrial chain data, the monitoring data and the quality hidden danger sources of the entire industrial chain of the thermal insulation project, and to warn users based on the quality traceability of the entire industrial chain of the thermal insulation project.

[0141] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for tracing the quality of the entire industry chain of thermal insulation engineering, characterized in that: The following steps are involved: S1. Obtaining industrial chain data and monitoring data of thermal insulation projects, and performing traceability coding and spatial coding based on the industrial chain data and the monitoring data; S2. Constructing a quality traceability network for the entire industry chain of thermal insulation engineering based on the industry chain data and the monitoring data; S3. Determine the industrial chain data corresponding to the monitoring data to be traced through the traceability code, and correct and screen the monitoring data to be traced to obtain abnormal monitoring data to be traced; S4. Determine the source of quality hazards of the entire industry chain of thermal insulation engineering according to the abnormal monitoring data to be traced and the quality traceability network of the entire industry chain of thermal insulation engineering; The method for performing traceability coding and spatial coding based on the industrial chain data and the monitoring data includes: The monitoring data is related to the effect of the current thermal insulation project, including energy-saving effect index, living comfort index and safety index. The method for obtaining the monitoring data is: Collect energy consumption data of metering devices in buildings based on Internet of Things technology, and determine energy-saving effect indicators based on energy consumption data in buildings; Different types of sensors are installed inside the building to collect indoor environmental data, and the living comfort index is determined based on the indoor environmental data; Randomly measure the anchoring force and bonding strength of thermal insulation engineering materials, use infrared thermal imaging and high-frequency radar detection methods to determine the quality factors of external insulation systems, and directly measure the toxic substances in thermal insulation engineering; Divide the entire industry chain of thermal insulation engineering into process chains, collect corresponding parameter information according to the process chains to obtain process chain data, set timestamps according to the process chain data, and obtain industry chain data in series according to the time series; Determine the coding format and coding length of the process chain according to the specific process and timestamp of the process chain, store the process chain information and process chain code in the RFID chip inside the thermal insulation project, partition and store the corresponding process chain data in the industrial chain database according to the thermal insulation project process chain code, connect all process chain codes in series according to the time series to obtain the traceability code of the industrial chain data, mark the corresponding traceability code when completing the last process of the thermal insulation project, and partition and store the corresponding monitoring data in the industrial chain database according to the thermal insulation project traceability code; the traceability code reflects the relationship between the industrial chain data and the monitoring data; Draw the REVIT building model, determine the spatial code of the thermal insulation project according to the monitoring area and spatial coordinates of the thermal insulation project monitoring data, and determine the unique correspondence between the thermal insulation project traceability code and the spatial code; The method for constructing a quality traceability network for the entire industrial chain of thermal insulation engineering according to the industrial chain data and the monitoring data comprises the following steps: Determine the traceability quality characterization vector based on the industrial chain data; Conduct process chain traceability based on traceability quality characterization vectors and monitoring data, and divide monitoring data into corresponding process chains; Determine the traceability performance of monitoring data in the corresponding process chain and generate a quality traceability network for the entire industry chain of thermal insulation engineering; The method for determining the traceability quality characterization vector according to the industrial chain data includes: The industrial chain data is preprocessed to obtain an industrial chain traceability vector, and the industrial chain traceability vector is input into the industrial chain data decomposition framework to obtain a first time series vector, a second time series vector, a first process vector, a second process vector and a third process vector; The industrial chain data decomposition framework includes a time series decomposition algorithm, a principal component analysis method, a subcomponent analysis method and anomaly detection algorithm; the time series decomposition algorithm decomposes the industrial chain traceability vector into a first time series vector and a second time series vector according to the time nodes of all processes; the principal component decomposition algorithm converts multiple related variables related to process quality in the industrial chain traceability vector into a small number of irrelevant principal components, and the principal components form a first process vector; the subcomponent decomposition algorithm converts multiple related variables related to process quality in the industrial chain traceability vector into a small number of irrelevant subcomponents, and the subcomponents form a second process vector; the anomaly detection algorithm obtains a third process vector by marking abnormal operations in the industrial chain traceability vector through unsupervised learning of sequence anomalies and quantitative anomalies; the first time series vector matches the time nodes of each process chain; the second time series vector contains the time nodes of all processes; The first process vector is crossed with the first timing vector to obtain the first process vector, the second process vector is crossed with the second timing vector to obtain the second process vector, the third process vector is fused with the first process vector feature to obtain the first traceability attribute representation vector, the third process vector is fused with the second process vector feature to obtain the second traceability attribute representation vector, and the first traceability attribute representation vector and the second traceability attribute representation vector are weighted to obtain the traceability quality representation vector; The method of dividing the monitoring data into each process chain includes: Based on the time series characteristics, the traceability quality characterization vector is segmented according to the process chain, and the population search algorithm is used to divide the monitoring data into each process chain. A certain data in the monitoring data is defined as a population individual, and the process chain divided by the data is the optimal position of the individual, and the search point of the discoverer is determined: in Search point directly in front of the finder, Search point for the finder's left side, is the search point on the right side of the finder, and the search point with the best fitness is taken as the finder No. Locations searched , For the The finder individual of the search, is the target cyclic shift factor, is an arbitrary number with a normal distribution characteristic of 0 and a variance of 1. is any uniformly distributed number, is the spatial search angle, is the maximum spatial search angle, is the search direction, For the The maximum search distance for each search, is the upper bound of the design variable, is the lower bound of the design variable; Update the spatial search angle and maximum search distance: in For the The spatial search angle of the search, is the maximum search angle, For the The maximum search distance for each search, is the search distance constant; The coupled genetic algorithm and swarm search algorithm generate new wanderer positions, update the discoverer positions and wanderers: in For the discoverer The location of the search. For the Wanderer The location of the search. , is the position update factor, for dimensional vector of any number, For the The position of the individual with the best positional fitness in the population during the first search, For the Search time The best historical position of an individual, is the scaling constant, , is the probability of the positive component appearing, for dimensional uniform distribution of any number, is the individual mutation probability; Calculate and update the individual fitness of the population, continue to search for the updated population position until the individual fitness no longer changes, stop searching, and obtain the optimal position of the population individual to determine the process chain division of the monitoring data; Methods for determining the traceability of each monitoring data in each process chain, including: The monitoring data of the process chain and the traceability quality characterization vector are combined into a traceability data set, and the traceability data set is divided into a training set and a test set; Constructing a quality traceability network model for the entire industry chain of thermal insulation engineering, wherein the quality traceability network model for the entire industry chain of thermal insulation engineering includes a calculation module, a rule module, and a machine learning module; The calculation module calculates the correlation between the monitoring data and the traceability quality characterization vector, and determines the correlation between the monitoring data and the process chain data in each process chain. The expression is: in For the The correlation between the monitoring data in a process chain and the traceability quality characterization vector, is the positive ideal solution of the evaluation scheme, is the negative ideal solution of the evaluation scheme, is the time series stability factor, which is determined by calculating the autocorrelation function of the time series. is the time correlation weight, is the abnormal adjustment factor, determined by statistical methods, Abnormal adjustment of association weights, For the The weight of the time scale, is the time scale quantity, is the combined weight, Represent the weights of the least squares method, coefficient of variation method and entropy weight method respectively, For the The first Monitoring data, For the The optimal value in the set of traceability quality characterization vector elements within a process chain, For the The worst value in the set of traceability quality characterization vector elements within a process chain, is the resolution factor; The rule module determines the association between monitoring data and process chain data in each process chain by setting the correlation threshold. The specific association rules are: When the energy-saving effect correlation between the energy-saving effect monitoring data and the traceability quality characterization vector is greater than the energy-saving effect correlation threshold, the process chain monitoring data is determined to be related to the process chain data of the first category; when the residential comfort correlation between the residential comfort monitoring data and the traceability quality characterization vector is greater than the residential comfort correlation threshold, the process chain monitoring data is determined to be related to the process chain data of the second category; when the safety correlation between the safety monitoring data and the traceability quality characterization vector is greater than the safety correlation threshold, the process chain monitoring data is determined to be related to the process chain data of the third category; When the energy-saving effect correlation, living comfort correlation, and safety correlation are all less than or equal to the corresponding correlation threshold, the comprehensive correlation of all monitoring data of the process chain and the traceability quality characterization vector is calculated. If the comprehensive correlation is greater than the comprehensive correlation threshold, it is determined that the process chain monitoring data is comprehensively correlated with the process chain data. Otherwise, it is determined that the process chain monitoring data is not correlated with the process chain data. The machine learning module generates a quality traceability network for the entire industry chain of thermal insulation engineering based on the correlation between monitoring data and process chain data, including graph models, classification algorithms and correlation analysis; uses correlation analysis to learn the association rules and association results output by the rule module; uses a graph model to build a relationship network between monitoring data and process chain data based on the correlation analysis results, locates and links monitoring data and process chain data within the process chain, and causally links data that conforms to data association rules, while marking relevant types on data links; uses a support vector machine (SVM) classification algorithm to predict the relationship between nodes based on the correlation analysis and graph model results; The training set data is used to train the model, and the test set data is used to evaluate the model and optimize its parameters. The quality traceability network model of the entire industrial chain of thermal insulation projects is output, and the traceability performance of each monitoring data in each process chain is determined based on the quality traceability network model of the entire industrial chain of thermal insulation projects.

2. According to claim 1, a method for tracing the quality of the entire industry chain of thermal insulation engineering is characterized in that: The method for obtaining the abnormal monitoring data to be traced includes: Obtain the monitoring data to be traced, determine its spatial code according to the monitoring area and spatial coordinates of the monitoring data to be traced in the REVIT building model, determine the traceability code of the monitoring data to be traced according to the unique correspondence of the code, extract the process chain data to be traced by region through the traceability code, and concatenate the process chain data to be traced in time series to obtain the industrial chain data to be traced; Obtain status data of the thermal insulation project, and perform data correction on the traceability monitoring data according to the status data; the status data includes installation location, installation process and external building environment indicators; The corrected monitoring data to be traced are screened using the hypersphere method to obtain abnormal monitoring data to be traced.

3. According to claim 1, a method for tracing the quality of the entire industry chain of thermal insulation engineering is characterized in that: The method for determining the quality hidden danger sources of the entire industry chain of the thermal insulation project includes: According to the quality traceability network of the entire industry chain of thermal insulation engineering, the abnormal monitoring data to be traced are divided into each process chain, and the abnormal degree of the abnormal monitoring data to be traced in each process chain is calculated. According to the abnormal degree, the abnormal monitoring data to be traced are traced to obtain the quality hidden danger source of the entire industry chain of thermal insulation engineering. The specific traceability rules are as follows: When the energy-saving effect abnormality of the energy-saving effect monitoring data to be traced in the process chain is greater than the energy-saving effect abnormality threshold, the first type of related traceability is carried out according to the causal link of the quality traceability network of the entire industrial chain of thermal insulation engineering; when the living comfort abnormality of the living comfort abnormality of the living comfort monitoring data to be traced in the process chain is greater than the living comfort abnormality threshold, the second type of related traceability is carried out according to the causal link of the quality traceability network of the entire industrial chain of thermal insulation engineering; when the safety abnormality of the safety abnormality of the safety abnormality of the process chain is greater than the safety abnormality threshold, the third type of related traceability is carried out according to the causal link of the quality traceability network of the entire industrial chain of thermal insulation engineering; When the energy-saving effect abnormality, living comfort abnormality, and safety abnormality are all less than or equal to the corresponding abnormality threshold, the comprehensive abnormality of the abnormal monitoring data to be traced in the process chain is calculated. If the comprehensive abnormality is greater than the comprehensive abnormality threshold, comprehensive related traceability is performed based on the causal link of the quality traceability network of the entire industry chain of thermal insulation engineering. Otherwise, quality traceability is not performed. When conducting causal link tracing, the corresponding link annotations and corresponding process chain data are extracted, and the quality risk sources of the process chain are determined based on the link annotations and the corresponding process chain data. All the quality risk sources of the process chain are combined to determine the quality risk sources of the entire industrial chain of thermal insulation projects.

4. A quality traceability system for the entire industry chain of thermal insulation engineering, used to implement the method described in any one of claims 1 to 3, characterized in that: include: Data acquisition module: used to obtain industrial chain data and monitoring data of thermal insulation projects, and to perform traceability coding and spatial coding on the industrial chain data and the monitoring data; Data partitioning module: used to determine the traceability quality characterization vector according to the industrial chain data, to perform process chain traceability according to the traceability quality characterization vector and the monitoring data, and to divide the monitoring data into each process chain using a population search algorithm; Traceability network module: used to construct a quality traceability network model for the entire industry chain of thermal insulation engineering based on monitoring data and traceability quality characterization vectors, including calculating the correlation between the monitoring data and the traceability quality characterization vectors, setting a correlation threshold to determine the correlation between the monitoring data and the process chain data in each process chain, using a machine learning model to learn the correlation between the data, and generating a quality traceability network for the entire industry chain of thermal insulation engineering; Traceability module: used to extract the industrial chain data to be traced according to the monitoring data to be traced, to correct and screen the monitoring data to be traced to obtain abnormal monitoring data to be traced, to calculate the abnormality degree of the abnormal monitoring data to be traced in each process chain, and to perform quality traceability according to the abnormality degree; Intelligent supervision module: used to store, view and manage the industrial chain data, the monitoring data and the quality hidden danger sources of the entire industrial chain of the thermal insulation project, and to warn users based on the quality traceability of the entire industrial chain of the thermal insulation project.

Images