House quality problem detection method and system based on graph theory correlation analysis

Abstract

The invention discloses a house quality problem detection method and system based on graph theory correlation analysis, and the method and system have the advantages that a correlation graph model is established, a correlation graph between factors is provided by integrating construction process, material and maintenance data and dynamically recognizing high-risk nodes, the recognition of potential risks is improved, and the detection efficiency is improved. The node weight and the edge association strength can be adjusted in real time, and the quality can be effectively supervised; a multi-stage evolutionary game model is constructed, analysis is carried out in the early stage, the middle stage and the later stage of construction in combination with an association graph model, and a more accurate risk assessment model is formed; a sub-shortest path algorithm is introduced, a quality problem extension path is accurately predicted, a potential problem source can be quickly positioned, and the early warning efficiency is improved; historical data and game results are utilized to form a quality risk assessment model, high-risk nodes are deeply analyzed, potential quality problem extension trends are identified, prevention measures are taken in advance, later rework cost is reduced, and delivery safety and reliability are improved.

Description

Technical Field

[0001] The present invention mainly relates to the technical field of house quality, and in particular to a house quality problem detection method and system based on graph theory association analysis. Background Art

[0002] In the field of modern construction, the prediction and detection of housing quality problems have gradually received attention. Especially in the construction process, due to the influence of various factors, housing quality problems frequently occur. Existing technologies mainly rely on traditional quality inspection methods including on-site visual inspection, material testing and review of construction records. Although traditional methods can detect problems in construction to a certain extent, they often have the defects of low efficiency and insufficient data analysis capabilities.

[0003] Current quality inspection technologies are mostly static inspections, lacking the ability to monitor dynamic changes during the construction process in real time. In many cases, potential quality problems that arise during the construction process are often not discovered until the final acceptance stage, which not only delays the progress of the project, but may also lead to high rework costs. In addition, traditional inspection methods often rely on manual experience and are limited by the professional level and subjective judgment of inspectors, which can easily lead to missed inspections or misjudgments, posing a threat to the safety and durability of the house. Another significant problem is that existing technologies generally lack systematic analysis of construction data. Although some companies have begun to try to use data management systems, the systems often cannot effectively integrate multidimensional data during the construction process. This information island phenomenon makes it difficult for various types of data to work together and form a comprehensive quality assessment model.

[0004] In addition, with the continuous development of construction technology, the complexity and uncertainty of the construction process are also increasing. Existing technologies have failed to effectively cope with the changing environment during construction and lack sufficient predictive capabilities, resulting in insufficient assessment of construction risks. As a result, many potential quality problems have not been discovered and resolved in a timely manner, thus posing a hidden danger to building safety.

[0005] In summary, the existing housing quality prediction and detection technologies have significant deficiencies in dynamic monitoring capabilities, data integration and analysis, and risk assessment. An innovative method is urgently needed to improve the ability to identify and predict quality problems during the construction process to ensure the safety and durability of the house. Summary of the invention

[0006] Based on this, it is necessary to provide a housing quality problem detection method and system based on graph theory association analysis to address the existing problems.

[0007] In a first aspect, an embodiment of the present application provides a housing quality problem detection method based on graph theory association analysis, comprising the following steps:

[0008] S1: Obtain the construction process data, construction material data and construction maintenance data of the house construction to build the basic data set of the house;

[0009] S2: extracting multi-dimensional features from the basic house data set to obtain a house key factor set consisting of multiple key factors related to construction process, construction materials and construction maintenance;

[0010] S3: Performing graph theory modeling on the set of key factors of the house to obtain a correlation graph model of factors affecting house quality;

[0011] S4: dynamically adjusting the association graph model and performing graph theory analysis to obtain a set of potential high-risk nodes;

[0012] S5: using a multi-stage evolutionary game model to process the potential high-risk node set to obtain a high-risk node set;

[0013] S6: using a second shortest path algorithm to process the high-risk nodes of the high-risk node set and their impact chains, to obtain an impact chain set of housing quality issues and a comprehensive risk value of each impact chain;

[0014] S7: Based on the set of impact chains of the housing quality problem and the comprehensive risk value of each impact chain, a housing quality prediction model is constructed and housing construction is monitored in real time. If an abnormal state of a risk node is detected, an early warning is automatically issued and a housing quality problem report is generated.

[0015] Preferably, step S1 comprises:

[0016] S11. Obtain construction process data, construction material data and construction maintenance data of the house construction; construct a construction process data set D s , Construction material dataset D m and construction maintenance dataset D n ;

[0017] S12: Construction process dataset D s , Construction material dataset D m and construction maintenance dataset D n Perform standardization to construct the basic housing data set D base ;

[0018] Among them, D s ={d s1 ,d s2 ,…,d sp}, d si is the construction time, material usage and construction process parameters of the i-th stage of the construction process data, 1≤i≤p;

[0019] Among them, Dm ={d m1 ,d m2 ,…,d mk}, d mj Material type, durability and compressive strength of construction material data, 1≤j≤k;

[0020] Among them, D n ={d n1 ,d n2 ,…,d nl}, d nq The maintenance time, maintenance content, maintenance personnel and use of repair materials and equipment involved in the construction maintenance data, 1≤q≤l;

[0021] Among them, D base ={D s ,D m ,D n}.

[0022] Preferably, step S2 comprises:

[0023] Based on the professional knowledge of experts in the field of house construction, feature extraction is performed on the house basic data set based on the three dimensions of construction process, construction materials and construction maintenance, and multiple key factors related to the three dimensions of the construction process, the construction materials and the construction maintenance are obtained. The multiple key factors constitute the house key factor set.

[0024] Preferably, step S3 comprises:

[0025] S31: Modeling each key factor in the house key factor set as a graph theory node;

[0026] S32: Based on the interaction relationship between the construction process, the construction materials and the key factors in the construction maintenance, obtain the association relationship between the graph nodes, the association relationship between the graph nodes includes the edge set E of the association relationship between the graph nodes in the construction process s , the edge set E of the association relationship between the graph nodes of the construction materials m and the edge set E of the association relationship between the construction and maintenance graph nodes n ;

[0027] S33: constructing a correlation graph model G(N, E, W) of factors affecting housing quality based on the correlation relationship between the graph theory nodes;

[0028] Where N = N s ∪N m ∪N n is the graph node set of all key factors, N sis the graph node set of key factors in the construction process, N m is the graph node set of key factors of construction materials, N s is the graph node set of key factors of construction maintenance; E = E s ∪E m ∪E n is the edge set of the association relationship between all graph theory nodes, W is the weight matrix of the association strength between key factors, and the weight matrix W is expressed by the following formula:

[0029] W=(w si,mj w mj,nq w si,nq ) (1);

[0030] Among them, w si,mj represents the association weight between the construction process graph theory node and the construction material graph theory node, w mj,nl represents the weight between the construction material graph theory node and the construction maintenance graph theory node, w si,nl Represents the weight between the construction process graph theory nodes and the construction maintenance graph theory nodes.

[0031] Preferably, step S4 comprises:

[0032] S41: According to formula (2), for each graph node n in the association graph model G(N, E, W) i The weight w i Make dynamic adjustments;

[0033] S42: Dynamically adjust the weight of each edge in the association graph model G(N, E, W) according to formula (3);

[0034] S43: Calculate the centrality C of each graph node according to formula (4) i ;

[0035] S44: Obtain the clustering coefficient of each graph node according to formula (5);

[0036] S45: Analyze the centrality C of the graph nodes according to formula (6) i and clustering coefficient γ i , obtain the comprehensive risk assessment value of each graph node;

[0037] S46: Screening graph nodes whose comprehensive risk assessment values ​​are greater than a preset risk threshold to obtain a set of potential high-risk nodes;

[0038] Wherein, formula (2) is expressed as follows:

[0039]

[0040] Wherein, formula (3) is expressed as follows:

[0041]

[0042] Wherein, formula (4) is expressed as follows:

[0043]

[0044] Wherein, formula (5) is expressed as follows:

[0045]

[0046] Wherein, formula (6) is expressed as follows:

[0047]

[0048] in, is the graph node n i The weight after the t+1th iteration, w ij is the graph node n i With graph theory node n j The weight between ij is the graph node n i With graph theory node n j The impact factor between i is the graph node n i The set of directly connected neighbor graph nodes; For edge e ij The weight after the t+1th iteration, α1 is the adjustment coefficient, Δ ij Represents edge e ij The corresponding change value of the correlation factor; a is the graph node n i is the number of actual connection edges between the neighbor graph theory nodes, b is the number of possible connection edges between all neighbor graph theory nodes; λ1, λ2, λ3 are risk weight coefficients; the preset risk threshold is associated with the construction process data, the construction material data and the construction maintenance data.

[0049] Preferably, step S5 comprises:

[0050] S51. In the early stage of house construction, construct an initial node set of the house based on the potential high-risk nodes, wherein the initial node set includes a component node set, a material node set, and a construction process node set;

[0051] S52: Processing the initial node set based on the initial game model to obtain the final state of each of the initial nodes, and obtaining the first potential high-risk node set in the initial construction stage in combination with the final state of the initial nodes;

[0052] S53: Processing the first potential high-risk node set based on the mid-term game model, identifying potential quality problems in the mid-term of housing construction, and determining the expansion trend of the quality problems, to obtain a second potential high-risk node set in the mid-term of construction;

[0053] S54: Processing the second potential high-risk node set based on the late game model, combining the status and historical data of each second potential high-risk node, to obtain the high-risk node set.

[0054] Preferably, step S52 includes:

[0055] S521: In the initial stage of construction, construct an initial game model based on the initial node set;

[0056] S522: adjusting the risk state of each of the initial nodes based on the initial game model, and obtaining an initial risk state value of each of the initial nodes;

[0057] S523: Determine whether the initial risk state value of the initial node is greater than an initial risk threshold;

[0058] S524: In response to the initial risk state value of the initial node being greater than the initial risk threshold, the initial node is used as a first potential high-risk node, and all the first potential high-risk nodes constitute a first potential high-risk node set;

[0059] The initial risk state value of the initial node is obtained by the following formula:

[0060]

[0061] Among them, w ij is the initial node n i With the initial node n j The weight between ij is the initial node n i With the initial node n j The initial game payoff between ij is the initial node n i With the initial node n j The interaction cost between is the initial node n i The initial risk state value after the tth initial game, is the initial node n i The initial risk state value after the t+1th initial game, N i is an initial node set; the initial risk threshold is associated with the construction process data, the construction material data and the construction maintenance data at the initial stage of construction.

[0062] Wherein, step S53 comprises:

[0063] S531: In the middle of construction, constructing a mid-term game model based on the first potential high-risk node set;

[0064] S532: adjusting the risk status of each of the first potential high-risk nodes based on the mid-term game model, and obtaining a mid-term risk status value of each of the first potential high-risk nodes;

[0065] S533: Determine whether the mid-term risk state value of the first potential high-risk node is greater than a mid-term risk threshold;

[0066] S534: In response to the mid-term risk state value of the first potential high-risk node being greater than the mid-term risk threshold, the first potential high-risk node is used as a second potential high-risk node, and all the second potential high-risk nodes constitute a second potential high-risk node set;

[0067] The mid-term risk status value of the first potential high-risk node is obtained by the following formula:

[0068]

[0069] Among them, α is the construction pressure influence coefficient in the middle of construction, β is the construction adjustment coefficient in the middle of construction, P k is the external pressure coefficient during the middle construction period, S k is an adjustment coefficient in the middle stage of construction, and the middle stage risk threshold is associated with the construction process data, the construction material data and the construction maintenance data in the middle stage of construction.

[0070] Wherein, step S54 comprises:

[0071] S541: In the later stage of construction, constructing a later stage game model based on the second potential high-risk node set;

[0072] S542: adjusting the risk status of each of the second potential high-risk nodes based on the late-stage game model, and obtaining a late-stage risk status value of each of the second potential high-risk nodes;

[0073] S543: Determine whether the late-stage risk state value of the second potential high-risk node is greater than a late-stage risk threshold;

[0074] S544: In response to the later risk state value of the second potential high-risk node being greater than the later risk threshold, the second potential high-risk node is used as a final high-risk node, and all the final high-risk nodes constitute the final high-risk node set;

[0075] The later risk state value of the second potential high-risk node is obtained by the following formula:

[0076]

[0077] in, The second potential high-risk node n l The value of the late risk state after the t+1th late game, The second potential high-risk node n l The historical risk status value, w lp The second potential high-risk node n l and the second potential high-risk node n p The weight between l is the second potential high-risk node set, x p The second potential high-risk node n p The current late-stage risk state value is , γ is the construction adjustment coefficient in the late stage of construction; the late-stage risk threshold is associated with the construction process data, the construction material data and the construction maintenance data in the late stage of construction.

[0078] Preferably, step S6 comprises:

[0079] S61: Based on potential high-risk nodes R i and high-risk node set N r ={n r1 ,n r2 ,…,n rk}, construct a subgraph G containing high-risk nodes r (R i ,N r ,E r ,W r );

[0080] S62: In subgraph G r (R i ,N r ,E r ,W r ) uses the second shortest path algorithm to identify the influence chain set P between each high-risk node = {p1, p2, ..., p m};

[0081] S63: Calculate each influence chain p in the influence chain set P according to formula (10) i The comprehensive risk value R pi ;

[0082] Among them, the comprehensive risk value R pi It is expressed by the following formula:

[0083]

[0084] Among them, E ris the edge set between high-risk nodes, W r is the corresponding weight matrix; the influence chain p i represents the second shortest path from one high-risk node to another high-risk node; R j is a high-risk node n j Comprehensive risk assessment value of L ij is a high-risk node n i and high-risk node n j The path length between them, β1 is the adjustment parameter of the influencing factor; A j is a high-risk node n j The set of adjacent nodes, R max is the maximum comprehensive risk assessment value among all high-risk nodes, R k For path p i Medium and high risk nodes k The comprehensive risk assessment value.

[0085] Preferably, step S7 includes:

[0086] S71: Based on the comprehensive risk value R pi For each impact chain p i Generate a housing quality prediction model, the output of which is the expansion probability of potential quality problems during the housing construction process;

[0087] S72: Real-time monitoring of house construction is performed based on the expansion probability of all influencing chains. If it is detected that the expansion probability meets the preset conditions, an early warning is automatically issued and a house quality problem report is generated;

[0088] The expansion probability is expressed by the following formula:

[0089]

[0090] The preset condition is expressed by the following formula:

[0091] P exp (p i )>T exp (12);

[0092] Among them, P exp (p i ) is the influence chain p i The expansion probability, T exp is the abnormal probability threshold, and m is the total number of impact chains in the impact chain set P.

[0093] In a second aspect, the embodiment of the present application provides a housing quality problem detection system based on graph theory association analysis, including:

[0094] The data acquisition module is used to acquire the construction process data, construction material data and construction maintenance data of the house construction, and build a basic data set of the house;

[0095] A feature extraction module is used to perform multi-dimensional feature extraction on the house basic data set to obtain a house key factor set consisting of multiple key factors related to the construction process, construction materials and construction maintenance;

[0096] An association graph construction module is used to perform graph theory modeling on the set of key housing factors to obtain an association graph model of housing quality influencing factors;

[0097] A graph theory analysis module, used to dynamically adjust the association graph model and perform graph theory analysis to obtain a set of potential high-risk nodes;

[0098] A dynamic game module, used for processing the potential high-risk node set by adopting a multi-stage evolutionary game model to obtain a high-risk node set;

[0099] An influence chain generation module is used to process the high-risk nodes of the high-risk node set and their influence chains using a second shortest path algorithm to obtain an influence chain set of housing quality problems and a comprehensive risk value of each influence chain;

[0100] The detection and early warning module is used to build a house quality prediction model based on the set of influence chains of house quality problems and the comprehensive risk value of each influence chain, and to monitor house construction in real time. If an abnormal state of a risk node is detected, an early warning will be automatically issued and a house quality problem report will be generated.

[0101] Compared with the prior art, the present invention has the following beneficial effects:

[0102] (1) A comprehensive association graph model was established, which dynamically identifies high-risk nodes by integrating construction process, material performance and maintenance data. It not only provides a graphical representation of the interrelationships between various factors, improving the ability to identify potential housing quality risks, but also can adjust node weights and edge association strengths in real time to ensure effective monitoring and management of housing quality risks at different stages of construction. This approach enables construction teams to respond to potential quality issues more quickly and accurately, thereby improving overall construction quality and safety;

[0103] (2) A multi-stage evolutionary game model was constructed, and the construction data analysis was performed in combination with the association graph model at the early, middle and late stages of construction. In the early stage of construction, the game between nodes will focus on the construction quality, material performance and standard adaptability of the construction process. In the middle stage of construction, the focus will be on how the nodes choose response strategies under changing external conditions to reduce the risk of quality problems expanding. In the later stage, the game will combine historical data and real-time monitoring results to further evaluate the construction quality of high-risk nodes and form a more accurate risk assessment model.

[0104] (3) The introduction of the second shortest path algorithm can effectively identify high-risk nodes and their impact chains, and accurately predict the expansion path of housing quality problems during the construction process. This enables the system to quickly locate potential problem sources in complex network structures, significantly improves the early warning efficiency, and provides more timely decision-making support for construction management.

[0105] In summary, the dynamic adjustment mechanism of the present invention allows the system to flexibly adjust the node weights and the correlation strength of the edges according to the real-time data changes during the construction process. Through game analysis based on construction pressure and external conditions, each node can automatically select the optimal response strategy, so that the system has a higher level of adaptability and intelligence when dealing with sudden quality problems, reducing the need for manual intervention and improving management efficiency. The present invention comprehensively utilizes historical data and game results to form a quality risk assessment model for the later stage of construction. Through in-depth analysis of high-risk nodes, it can identify the expansion trend of potential quality problems, so as to take preventive measures in advance when the construction is about to be delivered, which not only reduces the cost of later rework caused by quality problems, but also greatly improves the safety and reliability of house delivery. BRIEF DESCRIPTION OF THE DRAWINGS

[0106] By referring to the following drawings, the exemplary embodiments of the present invention can be more completely understood. The drawings are used to provide a further understanding of the embodiments of the present application, and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present invention and do not constitute a limitation of the present invention. In the drawings, the same reference numerals generally represent the same parts or steps.

[0107] Figure 1 A flowchart of a housing quality problem detection method based on graph theory association analysis provided according to an exemplary embodiment of the present application;

[0108] Figure 2 It is a structural schematic diagram of a housing quality problem detection system based on graph theory association analysis provided according to an exemplary embodiment of the present application. DETAILED DESCRIPTION

[0109] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0110] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.

[0111] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0112] In addition, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0113] An embodiment of the present application provides a method for determining a three-dimensional posture of a projection device, which is described below in conjunction with the accompanying drawings.

[0114] Reference Figure 1 , which shows a housing quality problem detection method based on graph theory association analysis provided by some embodiments of the present application, comprising the following steps:

[0115] S1: Obtain the construction process data, construction material data and construction maintenance data of the house construction to build the basic data set of the house;

[0116] In this embodiment, based on a high-rise residential construction project in a certain city in June 2023, the total construction area of ​​the project is 50,000 square meters, including multiple 20-story residential buildings, and the estimated construction period is 24 months. In the initial stage of the project, a large amount of basic housing data was collected through the information system, including key factors of construction material data (concrete strength, steel quality, construction process standards), construction process data (foundation, wall, roof construction steps) and construction maintenance data. After sorting and classifying, these data are converted into nodes to construct an association graph model for housing quality problems.

[0117] Specifically, step S1 includes:

[0118] S11. Obtain construction process data, construction material data and construction maintenance data of the house construction; construct a construction process data set D s , Construction material dataset D m and construction maintenance dataset D n ;

[0119] S12: Construction process dataset D s , Construction material dataset D m and construction maintenance dataset D n Perform standardization to construct the basic housing data set D base ;

[0120] Among them, D s ={d s1 ,d s2 ,…,d sp}, d si is the construction time, material usage and construction process parameters of the i-th stage of the construction process data, 1≤i≤p;

[0121] Among them, D m ={d m1 ,d m2 ,…,d mk}, d mj Material type, durability and compressive strength of construction material data, 1≤j≤k;

[0122] Among them, D n ={d n1 ,d n2 ,…,d nl}, d nq The maintenance time, maintenance content, maintenance personnel and use of repair materials and equipment involved in the construction maintenance data, 1≤q≤l;

[0123] Among them, D base ={D s ,D m ,Dn}.

[0124] S2: Extract multi-dimensional features from the basic housing data set to obtain a set of key housing factors consisting of multiple key factors related to construction process, construction materials, and construction maintenance;

[0125] In this embodiment, based on the professional knowledge of experts in the field of house construction, feature extraction is performed on the house basic data set based on the three dimensions of construction process, construction materials and construction maintenance to obtain multiple key factors related to the three dimensions of the construction process, the construction materials and the construction maintenance. The multiple key factors constitute the house key factor set.

[0126] S3: Graph modeling is performed on the set of key factors of a house to obtain a correlation graph model of factors affecting house quality;

[0127] Specifically, the step includes:

[0128] S31: Model each key factor in the house key factor set as a graph theory node;

[0129] S32: Based on the interaction relationship between the key factors in the construction process, construction materials and construction maintenance, the association relationship between the graph nodes is obtained, and the association relationship between the graph nodes includes the edge set E of the association relationship between the graph nodes in the construction process s , the edge set E of the association relationship between the graph theory nodes of construction materials m and construct and maintain the edge set E of the association relationship between the graph nodes n ;

[0130] S33: Based on the association relationship between graph theory nodes, a correlation graph model G(N, E, W) of factors affecting housing quality is constructed;

[0131] Where N = N s ∪N m ∪N n is the graph node set of all key factors, N s is the graph node set of key factors in the construction process, N m is the graph node set of key factors of construction materials, N s is the graph node set of key factors of construction maintenance; E = E s ∪E m ∪E n is the edge set of the association relationship between all graph theory nodes, W is the weight matrix of the association strength between key factors, and the weight matrix W is expressed by the following formula:

[0132] W=(w si,mj w mj,nq wsi,nq ) (1);

[0133] Among them, w si,mj represents the association weight between the construction process graph theory node and the construction material graph theory node, w mj,nl represents the weight between the construction material graph theory node and the construction maintenance graph theory node, w si,nl Represents the weight between the construction process graph theory nodes and the construction maintenance graph theory nodes.

[0134] S4: Dynamically adjust the association graph model and perform graph theory analysis to obtain a set of potential high-risk nodes;

[0135] Specifically, step S4 includes:

[0136] S41: According to formula (2), for each graph node n in the association graph model G(N,E,W) i The weight w i Make dynamic adjustments;

[0137] S42: dynamically adjust the weight of each edge in the association graph model G(N, E, W) according to formula (3);

[0138] S43: Calculate the centrality C of each graph node according to formula (4) i ;

[0139] S44: Obtain the clustering coefficient of each graph node according to formula (5);

[0140] S45: Analyze the centrality C of the graph nodes according to formula (6) i and clustering coefficient γ i , obtain the comprehensive risk assessment value of each graph node;

[0141] S46: Screening graph nodes whose comprehensive risk assessment values ​​are greater than a preset risk threshold to obtain a set of potential high-risk nodes;

[0142] Wherein, formula (2) is expressed as follows:

[0143]

[0144] Wherein, formula (3) is expressed as follows:

[0145]

[0146] Wherein, formula (4) is expressed as follows:

[0147]

[0148] Wherein, formula (5) is expressed as follows:

[0149]

[0150] Wherein, formula (6) is expressed as follows:

[0151]

[0152] in, is the graph node n i The weight after the t+1th iteration, w ij is the graph node n i With graph theory node n j The weight between ij is the graph node n i With graph theory node n j The impact factor between i is the graph node n i The set of directly connected neighbor graph nodes; For edge e ij The weight after the t+1th iteration, α1 is the adjustment coefficient, Δ ij Represents edge e ij The corresponding change value of the correlation factor; a is the graph node n i is the number of actual connection edges between the nodes in the neighbor graph theory, b is the number of possible connection edges between all neighbor graph theory nodes; λ1, λ2, and λ3 are risk weight coefficients.

[0153] Specifically, the preset risk threshold is associated with the construction process data, the construction material data and the construction maintenance data. It is used as a screening criterion to effectively identify potential high-risk nodes in the model analysis for further processing. In this embodiment, the setting of the preset risk threshold is based on the following principles: (1) Data-driven: Combined with the actual distribution of risk assessment values ​​during the construction process, a reasonable range is set through statistical analysis (such as mean, standard deviation or quantile). For example, in the standardized range of risk assessment values ​​of [0,1], a high quantile (such as the 90% quantile) is usually selected as the threshold to ensure that the screened nodes are concentrated in the part with higher risk values; (2) Scenario adaptability: The specific value of the threshold will be adjusted according to the scale and risk tolerance of the construction project. For example, in large-scale projects with a large number of nodes, the threshold can be appropriately set low (such as 0.7-0.8) to capture more potential risks; in scenarios with low tolerance for false alarms, the threshold can be increased (such as 0.8-0.9); (3) Dynamic adjustment: This application supports dynamic updating of thresholds based on real-time data to respond to changes in risk distribution during construction and ensure that screening criteria are flexibly adjusted according to actual conditions. The preset risk threshold of this application is combined with the characteristics of construction data and the requirements of construction scenarios, and usually ranges from 0.6 to 0.9. The specific value needs to be adaptively changed according to the actual application scenario.

[0154] S5: using a multi-stage evolutionary game model to process the potential high-risk node set to obtain a high-risk node set;

[0155] Specifically, through the aforementioned graph theory analysis, it is found that the correlation strength between the concrete used in the wall construction process and the material properties is gradually decreasing. In order to deal with this problem, a multi-stage evolutionary game model is applied to dynamically adjust the construction-related data. The multi-stage evolutionary model of this embodiment includes three parts: an initial game model, a mid-term game model, and a late-term game model, which respectively conducts game analysis on the data of the three time periods of the initial construction period, the mid-term construction period, and the late-term construction period of the project.

[0156] Specifically, step S5 includes:

[0157] S51. In the early stage of house construction, construct an initial node set of the house based on potential high-risk nodes, wherein the initial node set includes a component node set, a material node set, and a construction process node set;

[0158] Specifically, the component node set, material node set and construction process node set are divided based on the potential high-risk nodes. The division of these node sets is mainly based on the following logic:

[0159] Component node set: Potential high-risk nodes related to key components of the house (such as walls, foundations, etc.).

[0160] Material node sets: Potentially high-risk nodes related to the properties of materials used in the construction process (e.g., compressive strength, durability, etc.).

[0161] Construction process node set: Potentially high-risk nodes related to construction methods and technologies (such as process standards, equipment usage, etc.).

[0162] The above node division method is to provide structured data input and lay the foundation for the subsequent analysis of the game model.

[0163] S52: Processing the initial node set based on the initial game model to obtain the final state of each initial node, and obtaining the first potential high-risk node set in the initial construction stage in combination with the final state of the initial node;

[0164] In this example, the state of a node refers to the current risk quantification value of the node, which is used to dynamically describe the degree of risk change of the node during the game process. This state is not a fixed weight, but a dynamic parameter that gradually changes according to the interaction between the node and other nodes and external conditions. Through game analysis, the state value of the node will be updated with iterations, thereby reflecting the possible risk accumulation and change trend of the node in the early stage of construction. Specifically, the state value comprehensively considers the following factors: the interaction relationship between the node and the adjacent nodes, which includes the closeness of the association between the nodes and the intensity of risk propagation. The benefits obtained by the node in the interaction and the cost to be paid. If a node has low benefits and high costs, its risk status will gradually increase. The average risk level of the nodes in the entire network is used to measure whether a node has a significantly higher risk status than other nodes. The dynamic adjustment process of the state is to gradually discover which nodes' risk characteristics will increase significantly over time through multiple calculations.

[0165] In this embodiment, at the initial stage of construction, the following processing steps are included:

[0166] S521: In the initial stage of construction, an initial game model is constructed based on the initial node set;

[0167] S522: adjusting the risk state of each of the initial nodes based on the initial game model, and obtaining an initial risk state value of each of the initial nodes;

[0168] S523: Determine whether the initial risk state value of the initial node is greater than the initial risk threshold;

[0169] S524: In response to the initial risk state value of the initial node being greater than the initial risk threshold, the initial node is used as a first potential high-risk node, and all first potential high-risk nodes constitute a first potential high-risk node set;

[0170] Among them, the initial risk state value of the initial node is obtained by the following formula:

[0171]

[0172] Among them, w ij is the initial node n i With the initial node n j The weight between ij is the initial node n i With the initial node n j The initial game payoff between ij is the initial node n i With the initial node n j The interaction cost between is the initial node n i The initial risk state value after the tth initial game, is the initial node n i The initial risk state value after the t+1th initial game, N i is an initial node set; the initial risk threshold is associated with the construction process data, the construction material data and the construction maintenance data at the initial stage of construction.

[0173] Specifically, after multiple analysis iterations, if the risk status of some nodes continues to increase and is higher than the average level of other nodes, these nodes are further screened as high-risk nodes in the early stage of construction.

[0174] In this embodiment, in the middle stage of construction, the following processing steps are included:

[0175] S531: In the middle of construction, a mid-term game model is constructed based on the first potential high-risk node set;

[0176] S532: adjusting the risk status of each first potential high-risk node based on the mid-term game model, and obtaining a mid-term risk status value of each first potential high-risk node;

[0177] S533: Determine whether the mid-term risk state value of the first potential high-risk node is greater than the mid-term risk threshold;

[0178] S534: In response to the mid-term risk state value of the first potential high-risk node being greater than the mid-term risk threshold, the first potential high-risk node is used as a second potential high-risk node, and all second potential high-risk nodes constitute a second potential high-risk node set;

[0179] The mid-term risk status value of the first potential high-risk node is obtained by the following formula:

[0180]

[0181] Among them, α is the construction pressure influence coefficient in the middle of construction, which defines the construction pressure borne by the node as an external disturbance term; β is the construction adjustment coefficient in the middle of construction, which is based on the optimal strategy selected by the node according to the principle of profit maximization, which the enterprise dares not to adopt. The strategy set includes "maintaining the status quo", "strengthening adaptation", "temporary avoidance", etc.

[0182] P k is the external pressure coefficient during the middle construction period, S k is an adjustment coefficient in the middle stage of construction, which is determined according to the node selection strategy. The middle stage risk threshold is associated with the construction process data, the construction material data and the construction maintenance data in the middle stage of construction.

[0183] Specifically, in the mid-term game model, the rules for dynamically adjusting the node status are used by external construction pressure. The node status is affected by both the node income and the construction pressure. Its update logic is to add the adjustment factor of the construction pressure to the node status value based on the game income, and quantify the impact of the external environment on the node. Construction pressure includes environmental factors (such as weather), resource constraints (such as material delays), and schedule requirements (such as tight construction schedules).

[0184] In the later game model, it is necessary to introduce the historical correction rules of the node status and combine the historical data with the current status to evaluate the long-term risk of the node. The final state of the node can be calculated by weighting the historical state and the current risk state to form the final high-risk node set, which is used to generate the construction risk assessment report and provide targeted maintenance suggestions.

[0185] In this embodiment, in the later stage of construction, the following processing steps are included:

[0186] S541: In the later stage of construction, constructing a later stage game model based on the second potential high-risk node set;

[0187] S542: adjusting the risk status of each second potential high-risk node based on the late-stage game model, and obtaining a late-stage risk status value of each second potential high-risk node;

[0188] S543: Determine whether the late-stage risk state value of the second potential high-risk node is greater than the late-stage risk threshold;

[0189] S544: In response to the later risk state value of the second potential high-risk node being greater than the later risk threshold, the second potential high-risk node is treated as a high-risk node, and all high-risk nodes constitute a high-risk node set;

[0190] The later risk state value of the second potential high-risk node is obtained by the following formula:

[0191]

[0192] in, The second potential high-risk node n l The value of the late risk state after the t+1th late game, The second potential high-risk node n l The historical risk status value, w lp The second potential high-risk node n l and the second potential high-risk node n p The weight between l is the second potential high-risk node set, x p The second potential high-risk node n pThe current late-stage risk state value, γ is the construction adjustment coefficient in the late stage of construction, which is specifically related to the historical data of the node and the current data; the late-stage risk threshold is associated with the construction process data, the construction material data and the construction maintenance data in the late stage of construction.

[0193] S6: Use the second shortest path algorithm to process the high-risk nodes of the high-risk node set and their impact chains to obtain the impact chain set of the housing quality problem and the comprehensive risk value of each impact chain;

[0194] Specifically, the step includes:

[0195] S61: Based on potential high-risk nodes R i and high-risk node set N r ={n r1 ,n r2 ,…,n rk}, construct a subgraph G containing high-risk nodes r (R i ,N r ,E r ,W r );

[0196] S62: In subgraph G r (R i ,N r ,E r ,W r ) uses the second shortest path algorithm to identify the influence chain set P between each high-risk node = {p1, p2, ..., p m};

[0197] S63: Calculate each influence chain p in the influence chain set P according to formula (10) i The comprehensive risk value R pi ;

[0198] Among them, the comprehensive risk value R pi It is expressed by the following formula:

[0199]

[0200] Among them, E r is the edge set between high-risk nodes, W r is the corresponding weight matrix; the influence chain p i represents the second shortest path from one high-risk node to another high-risk node; R j is a high-risk node n j Comprehensive risk assessment value of L ij is a high-risk node n i and high-risk node n jThe path length between them, β1 is the adjustment parameter of the influencing factor; A j is a high-risk node n j The set of adjacent nodes, R max is the maximum comprehensive risk assessment value among all high-risk nodes, R k For path p i Medium and high risk nodes k The comprehensive risk assessment value.

[0201] S7: Based on the impact chain set of house quality problems and the comprehensive risk value of each impact chain, a house quality prediction model is constructed and house construction is monitored in real time. If an abnormal state of a risk node is detected, an early warning is automatically issued and a house quality problem report is generated.

[0202] Specifically, the step includes:

[0203] S71: Based on the comprehensive risk value R pi For each impact chain p i Generate a housing quality prediction model, the output of which is the expansion probability of potential quality problems during the housing construction process;

[0204] S72: Real-time monitoring of house construction is performed based on the expansion probability of all influencing chains. If it is detected that the expansion probability meets the preset conditions, an early warning is automatically issued and a house quality problem report is generated;

[0205] The expansion probability is expressed by the following formula:

[0206]

[0207] The preset condition is expressed by the following formula:

[0208] P exp (p i )>T exp (12);

[0209] Among them, P exp (p i ) is the influence chain p i The expansion probability, T exp is the abnormal probability threshold, and m is the total number of impact chains in the impact chain set P.

[0210] In this embodiment, the focus is on monitoring P exp (p i), especially the chain with a rapidly increasing probability of expansion, which indicates that the impact of the relevant high-risk nodes is significantly expanding. The abnormal probability threshold is usually determined based on historical project data or risk tolerance, indicating that the probability of chain expansion exceeds the acceptable range. The common value range is 0.3 to 0.5, depending on the risk level and construction complexity.

[0211] In a preferred embodiment, the following settings can be made, even if the current P exp (p i ) does not exceed the interval value, but if the expansion probability rises rapidly for many consecutive times, it can also be regarded as an abnormal state. exp (p i ) If the growth rate exceeds 20% in two consecutive iterations, it is considered abnormal. At the same time, if the number of nodes associated with a high expansion probability chain or the number of affected chains increases significantly, it can also be considered abnormal. The system can monitor the number of covered nodes and weight changes of the chain to determine whether the expansion impact is worsening.

[0212] Throughout the construction process, construction quality problem reports were generated in real time, detailing node status changes, relevant material data, and potential quality risks. For example, on September 30, 2023, the team generated a quality report on wall construction, pointing out the quality fluctuations of nodes due to humidity changes in the past two weeks, and providing timely rectification suggestions.

[0213] After full data tracking and analysis, the effects of the traditional method and the method of the present invention were compared. The data showed that during the construction process, the average time for the traditional method to discover problems was 48 hours, while the method of the present invention shortened it to 12 hours. In addition, in terms of the response time after the occurrence of quality problems, the average response time of the traditional method was 36 hours, while the method of the present invention shortened it to 8 hours, significantly improving the construction safety and efficiency.

[0214] Compared with the prior art, the present invention has the following beneficial effects:

[0215] (1) A comprehensive association graph model was established, which dynamically identifies high-risk nodes by integrating construction process, material performance and maintenance data. It not only provides a graphical representation of the interrelationships between various factors, improving the ability to identify potential quality risks, but also can adjust node weights and edge association strengths in real time to ensure effective monitoring and management of quality risks at different stages of construction. This approach enables construction teams to respond to potential quality issues more quickly and accurately, thereby improving overall construction quality and safety;

[0216] (2) A multi-stage evolutionary game model was constructed, and the construction data analysis was performed in combination with the association graph model at the early, middle and late stages of construction. In the early stage of construction, the game between nodes will focus on the construction quality, material performance and standard adaptability of the construction process. In the middle stage, the focus will be on how the nodes choose response strategies under changing external conditions to reduce the risk of quality problems expanding. In the later stage, the game will combine historical data and real-time monitoring results to further evaluate the construction quality of high-risk nodes and form a more accurate risk assessment model.

[0217] (3) The introduction of the second shortest path algorithm can effectively identify high-risk nodes and their impact chains, and accurately predict the expansion path of quality problems during the construction process, so that the system can quickly locate potential problem sources in complex network structures, significantly improve the early warning efficiency, and provide more timely decision-making support for construction management.

[0218] In summary, the dynamic adjustment mechanism of the present invention allows the system to flexibly adjust the node weights and the correlation strength of the edges according to the real-time data changes during the construction process. Through game analysis based on construction pressure and external conditions, each node can automatically select the optimal response strategy, so that the system has a higher level of adaptability and intelligence when dealing with sudden quality problems, reducing the need for manual intervention and improving management efficiency. The present invention comprehensively utilizes historical data and game results to form a quality risk assessment model for the later stage of construction. Through in-depth analysis of high-risk nodes, it can identify the expansion trend of potential quality problems, so as to take preventive measures in advance when the construction is about to be delivered, which not only reduces the cost of later rework caused by quality problems, but also greatly improves the safety and reliability of house delivery.

[0219] In the above embodiment, a method is provided, and correspondingly, the present application also provides a system. The system provided in the embodiment of the present application can implement the above method, and the system can be implemented by software, hardware, or a combination of software and hardware. For example, the system may include integrated or separate functional modules or units to perform the corresponding steps in the above methods.

[0220] In some implementations of the embodiments of the present application, the system 20 provided in the embodiments of the present application is based on the same inventive concept as the method provided in the aforementioned embodiments of the present application and has the same beneficial effects.

[0221] like Figure 2 As shown, the system 20 may include:

[0222] The data acquisition module 201 is used to acquire the construction process data, construction material data and construction maintenance data of the house construction, and construct a basic data set of the house;

[0223] The feature extraction module 202 is used to perform multi-dimensional feature extraction on the house basic data set to obtain a house key factor set consisting of multiple key factors related to the construction process, construction materials and construction maintenance;

[0224] The association graph construction module 203 is used to perform graph theory modeling on the set of key factors of a house to obtain an association graph model of factors affecting house quality;

[0225] A graph analysis module 204, used to dynamically adjust the association graph model and perform graph analysis to obtain a set of potential high-risk nodes;

[0226] The dynamic game module 205 is used to process the potential high-risk node set using a multi-stage evolutionary game model to obtain a high-risk node set;

[0227] The influence chain generation module 206 is used to process the high-risk nodes of the high-risk node set and their influence chains using a second shortest path algorithm to obtain an influence chain set of housing quality problems and a comprehensive risk value of each influence chain;

[0228] The detection and early warning module 207 is used to build a house quality prediction model based on the impact chain set of house quality problems and the comprehensive risk value of each impact chain and to monitor the house construction in real time. If an abnormal state of a risk node is detected, an early warning is automatically issued and a house quality problem report is generated.

[0229] The system provided in this embodiment is used to execute the method disclosed in the above-mentioned embodiment, and has the same technical concept as the above-mentioned embodiment, so this embodiment will not be described in detail here.

[0230] It should be noted that the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the systems, methods and computer program products according to multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of the boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0231] Technicians in the relevant field can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0232] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.

[0233] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0234] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0235] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0236] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, ordinary technicians in this field should understand that they can still modify the technical solutions recorded in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application, and they should all be included in the scope of the claims and specification of the present application.

Claims

1. A housing quality problem detection method based on graph theory association analysis, characterized in that: The steps include: S1: Obtain the construction process data, construction material data and construction maintenance data of the house construction to build the basic data set of the house; S2: extracting multi-dimensional features from the basic house data set to obtain a house key factor set consisting of multiple key factors related to construction process, construction materials and construction maintenance; S3: Performing graph theory modeling on the set of key factors of the house to obtain a correlation graph model of factors affecting house quality; S4: dynamically adjusting the association graph model and performing graph theory analysis to obtain a set of potential high-risk nodes; S5: using a multi-stage evolutionary game model to process the potential high-risk node set to obtain a final high-risk node set; S6: using a second shortest path algorithm to process the final high-risk node of the final high-risk node set and its influence chain, to obtain an influence chain set of housing quality problems and a comprehensive risk value of each influence chain; S7: Based on the set of impact chains of the housing quality problem and the comprehensive risk value of each impact chain, a housing quality prediction model is constructed and housing construction is monitored in real time. If an abnormal state of a risk node is detected, an early warning is automatically issued and a housing quality problem report is generated.

2. The method according to claim 1, characterized in that Step S1 includes: S11. Obtain construction process data, construction material data and construction maintenance data of the house construction; construct a construction process data set D s , Construction material dataset D m and construction maintenance dataset D n ; S12: Construction process dataset D s , Construction material dataset D m and construction maintenance dataset D n Perform standardization to construct the basic housing data set D base ; Among them, D s ={d s1 ,d s2 ,…,d sp }, d si is the construction time, material usage and construction process parameters of the i-th stage of the construction process data, 1≤i≤p; Among them, D m ={d m1 ,d m2 ,…,d mk }, d mj Material type, durability and compressive strength of construction material data, 1≤j≤k; Among them, D n ={d n1 ,d n2 ,…,d nl }, d nq The maintenance time, maintenance content, maintenance personnel and use of repair materials and equipment involved in the construction maintenance data, 1≤q≤l; Among them, D base ={D s ,D m ,D n }.

3. The method according to claim 2, characterized in that Step S2 includes: Based on the professional knowledge of experts in the field of house construction, feature extraction is performed on the house basic data set based on the three dimensions of construction process, construction materials and construction maintenance, and multiple key factors related to the three dimensions of the construction process, the construction materials and the construction maintenance are obtained. The multiple key factors constitute the house key factor set.

4. The method according to claim 3, characterized in that Step S3 includes: S31: Modeling each key factor in the house key factor set as a graph theory node; S32: Based on the interaction relationship between the construction process, the construction materials and the key factors in the construction maintenance, obtain the association relationship between the graph nodes, the association relationship between the graph nodes includes the edge set E of the association relationship between the graph nodes in the construction process s , the edge set E of the association relationship between the graph nodes of the construction materials m and the edge set E of the association relationship between the construction and maintenance graph nodes n ; S33: constructing a correlation graph model G(N, E, W) of factors affecting housing quality based on the correlation relationship between the graph theory nodes; Where N = N s ∪N m ∪N n is the graph node set of all key factors, N s is the graph node set of key factors in the construction process, N m is the graph node set of key factors of construction materials, N s is the graph node set of key factors of construction maintenance; E = E s ∪E m ∪E n is the edge set of the association relationship between all graph theory nodes, W is the weight matrix of the association strength between key factors, and the weight matrix W is expressed by the following formula: In=(in s i ,mj In mj,nq In s i ,nq ) (1); Among them, w si,mj represents the association weight between the construction process graph theory node and the construction material graph theory node, w mj,nl represents the weight between the construction material graph theory node and the construction maintenance graph theory node, w si,nl Represents the weight between the construction process graph theory nodes and the construction maintenance graph theory nodes.

5. The method according to claim 4, characterized in that Step S4 includes: S41: According to formula (2), for each graph node n in the association graph model G(N, E, W) i The weight w i Make dynamic adjustments; S42: Dynamically adjust the weight of each edge in the association graph model G(N, E, W) according to formula (3); S43: Calculate the centrality C of each graph node according to formula (4) i ; S44: Obtain the clustering coefficient of each graph node according to formula (5); S45: Analyze the centrality C of the graph nodes according to formula (6) i and clustering coefficient γ i , obtain the comprehensive risk assessment value of each graph node; S46: Screening graph nodes whose comprehensive risk assessment values ​​are greater than a preset risk threshold to obtain a set of potential high-risk nodes; Wherein, formula (2) is expressed as follows: Wherein, formula (3) is expressed as follows: Wherein, formula (4) is expressed as follows: Wherein, formula (5) is expressed as follows: Wherein, formula (6) is expressed as follows: in, is the graph node n i The weight after the t+1th iteration, w ij is the graph node n i With graph theory node n j The weight between ij is the graph node n i With graph theory node n j The impact factor between i is the graph node n i The set of directly connected neighbor graph nodes; For edge e ij The weight after the t+1th iteration, α1 is the adjustment coefficient, Δ ij Represents edge e ij The corresponding change value of the correlation factor; a is the graph node n i is the number of actual connection edges between the neighbor graph theory nodes, b is the number of possible connection edges between all neighbor graph theory nodes; λ1, λ2, λ3 are risk weight coefficients; the preset risk threshold is associated with the construction process data, the construction material data and the construction maintenance data.

6. The method according to claim 5, characterized in that Step S5 includes: S51. In the early stage of house construction, construct an initial node set of the house based on the potential high-risk nodes, wherein the initial node set includes a component node set, a material node set, and a construction process node set; S52: Processing the initial node set based on the initial game model to obtain the final state of each of the initial nodes, and obtaining the first potential high-risk node set in the initial construction stage in combination with the final state of the initial nodes; S53: Processing the first potential high-risk node set based on the mid-term game model, identifying potential quality problems in the mid-term of housing construction, and determining the expansion trend of the quality problems, to obtain a second potential high-risk node set in the mid-term of construction; S54: Processing the second potential high-risk node set based on the late game model, combining the status and historical data of each second potential high-risk node, to obtain the high-risk node set.

7. The method according to claim 6, characterized in that Step S52 includes: S521: In the initial stage of construction, construct an initial game model based on the initial node set; S522: adjusting the risk state of each of the initial nodes based on the initial game model, and obtaining an initial risk state value of each of the initial nodes; S523: Determine whether the initial risk state value of the initial node is greater than an initial risk threshold; S524: In response to the initial risk state value of the initial node being greater than the initial risk threshold, the initial node is used as a first potential high-risk node, and all the first potential high-risk nodes constitute a first potential high-risk node set; The initial risk state value of the initial node is obtained by the following formula: Among them, w ij is the initial node n i With the initial node n j The weight between ij is the initial node n i With the initial node n j The initial game payoff between ij is the initial node n i With the initial node n j The interaction cost between is the initial node n i The initial risk state value after the tth initial game, is the initial node n i The initial risk state value after the t+1th initial game, N i is an initial node set; the initial risk threshold is associated with the construction process data, the construction material data and the construction maintenance data at the initial stage of construction. Wherein, step S53 comprises: S531: In the middle of construction, constructing a mid-term game model based on the first potential high-risk node set; S532: adjusting the risk status of each of the first potential high-risk nodes based on the mid-term game model, and obtaining a mid-term risk status value of each of the first potential high-risk nodes; S533: Determine whether the mid-term risk state value of the first potential high-risk node is greater than a mid-term risk threshold; S534: In response to the mid-term risk state value of the first potential high-risk node being greater than the mid-term risk threshold, the first potential high-risk node is used as a second potential high-risk node, and all the second potential high-risk nodes constitute a second potential high-risk node set; The mid-term risk status value of the first potential high-risk node is obtained by the following formula: Among them, α is the construction pressure influence coefficient in the middle of construction, β is the construction adjustment coefficient in the middle of construction, P k is the external pressure coefficient during the middle construction period, S k is an adjustment coefficient in the middle stage of construction, and the middle stage risk threshold is associated with the construction process data, the construction material data and the construction maintenance data in the middle stage of construction. Wherein, step S54 comprises: S541: In the later stage of construction, constructing a later stage game model based on the second potential high-risk node set; S542: adjusting the risk status of each of the second potential high-risk nodes based on the late-stage game model, and obtaining a late-stage risk status value of each of the second potential high-risk nodes; S543: Determine whether the late-stage risk state value of the second potential high-risk node is greater than a late-stage risk threshold; S544: In response to the later risk state value of the second potential high-risk node being greater than the later risk threshold, the second potential high-risk node is used as a final high-risk node, and all the final high-risk nodes constitute the final high-risk node set; The later risk state value of the second potential high-risk node is obtained by the following formula: in, The second potential high-risk node n l The value of the late risk state after the t+1th late game, The second potential high-risk node n l The historical risk status value, w lp The second potential high-risk node n l and the second potential high-risk node n p The weight between l is the second potential high-risk node set, x p The second potential high-risk node n p The current late-stage risk state value is , γ is the construction adjustment coefficient in the late stage of construction; the late-stage risk threshold is associated with the construction process data, the construction material data and the construction maintenance data in the late stage of construction.

8. The method according to claim 7, characterized in that Step S6 includes: S61: Based on potential high-risk nodes R i and high-risk node set N r ={n r1 ,n r2 ,…,n rk }, construct a subgraph G containing high-risk nodes r (R i ,N r ,E r ,W r ); S62: In subgraph G r (R i ,N r ,E r ,W r ) uses the second shortest path algorithm to identify the influence chain set P between each high-risk node = {p1, p2, ..., p m }; S63: Calculate each influence chain p in the influence chain set P according to formula (10) i The comprehensive risk value Among them, the comprehensive risk value It is expressed by the following formula: Among them, E r is the edge set between high-risk nodes, W r is the corresponding weight matrix; the influence chain p i represents the second shortest path from one high-risk node to another high-risk node; R j is a high-risk node n j Comprehensive risk assessment value of L ij is a high-risk node n i and high-risk node n j The path length between them, β1 is the adjustment parameter of the influencing factor; A j is a high-risk node n j The set of adjacent nodes, R max is the maximum comprehensive risk assessment value among all high-risk nodes, R k For path p i Medium and high risk nodes k The comprehensive risk assessment value.

9. The method according to claim 8, characterized in that Step S7 includes: S71: Based on comprehensive risk value For each impact chain p i Generate a housing quality prediction model, the output of which is the expansion probability of potential quality problems during the housing construction process; S72: Real-time monitoring of house construction is performed based on the expansion probability of all influencing chains. If it is detected that the expansion probability meets the preset conditions, an early warning is automatically issued and a house quality problem report is generated; The expansion probability is expressed by the following formula: The preset condition is expressed by the following formula: P exp (p i )>T exp (12); Among them, P exp (p i ) is the influence chain p i The expansion probability, T exp is the abnormal probability threshold, and m is the total number of impact chains in the impact chain set P.

10. A housing quality problem detection system based on graph theory association analysis, characterized in that: include: The data acquisition module is used to acquire the construction process data, construction material data and construction maintenance data of the house construction, and build a basic data set of the house; A feature extraction module is used to perform multi-dimensional feature extraction on the house basic data set to obtain a house key factor set consisting of multiple key factors related to the construction process, construction materials and construction maintenance; An association graph construction module is used to perform graph theory modeling on the set of key housing factors to obtain an association graph model of housing quality influencing factors; A graph theory analysis module, used to dynamically adjust the association graph model and perform graph theory analysis to obtain a set of potential high-risk nodes; A dynamic game module, used for processing the potential high-risk node set by adopting a multi-stage evolutionary game model to obtain a high-risk node set; An influence chain generation module is used to process the high-risk nodes of the high-risk node set and their influence chains using a second shortest path algorithm to obtain an influence chain set of housing quality problems and a comprehensive risk value of each influence chain; The detection and early warning module is used to build a house quality prediction model based on the set of influence chains of house quality problems and the comprehensive risk value of each influence chain, and to monitor house construction in real time. If an abnormal state of a risk node is detected, an early warning will be automatically issued and a house quality problem report will be generated.

Images