Engineering supervision quality evaluation method and system based on big data

Abstract

The invention relates to the technical field of project quality monitoring, and particularly discloses a project supervision quality evaluation method and system based on big data, and the method comprises the steps: collecting construction paying-off point data and excavation depth data in real time, respectively calculating a paying-off deviation abnormal coefficient and an excavation depth deviation accumulated value, and combining an overall related influence coefficient to evaluate the project supervision quality. Constructing a dynamic monitoring model, comprehensively evaluating the actual influence of the paying-off precision on the excavation depth, identifying abnormal points in the construction process in real time through comprehensive and time sequence analysis, dynamically monitoring the deviation trend, and triggering an early warning mechanism when a comprehensive abnormal coefficient exceeds a threshold value. And marking an abnormal area and guiding a construction party to adjust a setting-out coordinate and an excavation depth so as to reduce the influence of accumulated deviation.

Description

Technical Field

[0001] The present invention relates to the technical field of engineering quality monitoring, and particularly to an engineering supervision quality assessment method and system based on big data. Background Art

[0002] During the construction process of engineering supervision, setting out and excavation are key links to ensure construction quality. The accuracy of setting out directly affects the accuracy of subsequent construction operations, and the deviation of excavation depth may lead to problems with the safety and stability of the construction structure. Therefore, real-time monitoring and deviation control of construction setting out points and excavation depth are the core means to ensure project quality. With the wide application of big data technology and intelligent monitoring equipment, achieving precise control of the construction process through efficient data collection and analysis technology has become the main development direction of the industry. However, how to effectively identify abnormal points in construction using multi-source data and dynamically adjust construction operations to improve the quality control level remains an urgent problem to be solved.

[0003] The existing technology has the following deficiencies:

[0004] Current construction monitoring methods usually rely on manual inspection and single-device data collection, suffering from problems such as low monitoring accuracy, slow response speed, and insufficient comprehensive data analysis ability. There is a lack of a quantitative evaluation method for the correlation between setting out accuracy and excavation depth, and it is impossible to accurately identify the actual impact of setting out errors on construction quality, resulting in a lack of clear basis for construction adjustment; secondly, there is a lack of a real-time dynamic monitoring mechanism, making it difficult to timely detect abnormal points during the construction process and trigger effective early warnings. It often relies on manual judgment or post-event adjustment, which not only delays the optimization opportunity but also reduces the efficiency of construction quality management. Therefore, the existing technology still has deficiencies in the comprehensiveness, dynamics, and guidance of data analysis. Summary of the Invention

[0005] The purpose of the present invention is to provide an engineering supervision quality assessment method and system based on big data to solve the problems in the above background.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] An engineering supervision quality assessment method based on big data includes:

[0008] S1: During the construction process, real-time collect construction setting out point data and excavation depth data;

[0009] S2: Analyze the construction setting out point data, judge the error between the setting out point and the design value through deviation calculation, and identify abnormal points of setting out accuracy in combination with the characteristics of the construction area;

[0010] S3: Analyze the excavation depth data, calculate the deviation of the excavation depth data, calculate the cumulative deviation value based on the numerical difference between the excavation depth and the designed depth, comprehensively evaluate the accuracy of the construction layout data and the deviation data of the excavation depth, and analyze the influence of the layout accuracy on the excavation depth;

[0011] S4: According to the comprehensive influence analysis results, identify the abnormal points during the construction process, make construction adjustments to the deviation influence during the construction process, dynamically monitor the deviation trend during the construction process based on time series analysis, trigger the early warning mechanism through the comprehensive quality evaluation coefficient, and optimize the construction quality.

[0012] As a further solution of the present invention: Analyze the construction layout point data, judge the error between the layout point and the designed value through deviation calculation, and identify the abnormal points of the layout accuracy in combination with the characteristics of the construction area, specifically including:

[0013] Collect the construction layout point data in real time to obtain the actual coordinate values of each layout point;

[0014] Calculate the deviation value of each layout point, compare it with the designed value to obtain the deviation error;

[0015] Calculate the environmental influence coefficient to correct the deviation calculation, adjust the accuracy of the layout data, and analyze the influence of the correction on the deviation;

[0016] According to the analysis results, calculate the layout deviation abnormal coefficient, judge whether the layout deviation abnormal coefficient is greater than or equal to the preset threshold. If so, record it as an abnormal point of the layout accuracy. If not, record it as a non-abnormal point of the layout accuracy.

[0017] As a further solution of the present invention: The process of obtaining the environmental influence coefficient is as follows:

[0018] Obtain real-time environmental data, including: environmental humidity data H, air pressure data P, land type S, and environmental temperature data T;

[0019] Establish a Bayesian network model, where the environmental humidity data H, air pressure data P, land type S, and environmental temperature data T are used as input nodes, and the construction deviation D is used as the output node. The environmental influence coefficient K env is used as the target variable;

[0020] Define the conditional probability distribution between each node according to the environmental data, specifically including:

[0021] The conditional probability P(H|T) of environmental humidity at a given temperature, the conditional probability P(P|T,H) of air pressure at a given temperature and humidity, the conditional probability P(S|H,P,T) of soil type at a given humidity, air pressure and temperature, and the conditional probability P(D|H,P,T,S) of construction deviation at a given humidity, air pressure, temperature and soil type;

[0022] Calculate the joint probability of all environmental factors and construction deviations. The calculation expression is:

[0023] P(H,P,T,S,D) = P(H|T)·P(P|T,H)·P(S|H,P,T)·P(D|H,P,T,S);

[0024] In the formula, P(H,P,T,S,D) represents the joint probability;

[0025] Calculate the environmental impact coefficient through the joint probability. The calculation expression is:

[0026] K env = ∑ H,P,T,S P(H,P,T,S,D)·f(H,P,T,S,D);

[0027] Among them, K env represents the environmental impact coefficient, and f(H,P,T,S,D) represents the weighting function.

[0028] As a further solution of the present invention: The process of obtaining the abnormal coefficient of wire laying deviation is as follows:

[0029] During the construction process, the actual coordinate values of each wire laying point are collected in real time:

[0030] X i =(x i1 ,x i2 ,...,x in );

[0031] Among them, i is the wire laying point number, i is a positive integer greater than 0, n is the dimension of the coordinates, n is a positive integer greater than 0, and X i represents the actual coordinate value of the i-th wire laying point;

[0032] Obtain the designed coordinate values of each wire laying point:

[0033] X design,i =(x design,i1 ,x design,i2 ,...,x design,in );

[0034] Among them, X design,i represents the designed coordinate value of the i-th wire laying point;

[0035] Calculate the deviation value between each setting-out point and the corresponding design coordinate. The calculation formula is as follows:

[0036] δ i =||X i -X design,i ||;

[0037] Among them, δ i represents the deviation value of the i-th setting-out point;

[0038] Construct an isolation forest model. Through the construction and random division of multiple trees, calculate the isolation depth h of each setting-out point i , and the isolation depth h of each tree i,j is the isolation path length of the setting-out point on the corresponding tree. Calculate the average isolation depth through multiple trees. The calculation formula is as follows:

[0039]

[0040] In the formula, t represents the number of trees in the isolation forest, j represents the trees in the isolation forest, and h i,j represents the isolation depth of the i-th point on the j-th tree, represents the average isolation depth;

[0041] According to the average isolation depth, calculate the anomaly score of each setting-out point. The calculation formula is as follows:

[0042]

[0043] Among them, is a constant, and score(X i ) represents the anomaly score of the i-th setting-out point;

[0044]

[0045] In the formula, α i represents the setting-out deviation anomaly coefficient of the i-th setting-out point, and max(δ) represents the maximum value of the deviation values of all setting-out points.

[0046] As a further solution of the present invention: Analyze the excavation depth data, calculate the deviation of the excavation depth data, and calculate the cumulative deviation value according to the numerical difference between the excavation depth and the design depth. Specifically, it includes:

[0047] During the construction process, collect the excavation depth data and the design depth data of each construction point in real time;

[0048] For each construction point, calculate the deviation value between the actual excavation depth and the design depth. The calculation formula is as follows:

[0049] s a =Hactual,a -H design,a ;

[0050] In the formula, s a represents the deviation value of the a-th point, and H actual,a represents the actual excavation depth of the a-th point, and H design,a represents the designed excavation depth of the a-th point, and a represents the excavation point;

[0051] Construct a training data set, and use the designed depth data of each construction point as the input feature Z a , and the deviation value s a as the target variable to form a training data set;

[0052] Use the random forest regression algorithm to train the model. Based on the designed depth data features, predict the deviation value through multiple decision trees. The output deviation prediction value of each tree is:

[0053] In the formula, represents the deviation prediction of the b-th tree for the a-th point, and Z a represents the input feature of the a-th point, and E b represents the b-th decision tree, and b represents the number of decision trees;

[0054] Obtain the final deviation prediction value by averaging the prediction results of all decision trees. The calculation expression is:

[0055]

[0056] In the formula, represents the final deviation prediction value of the a-th point, and B represents the total number of decision trees;

[0057] Calculate the sum of the deviation prediction values of all construction points to obtain the cumulative deviation value.

[0058] As a further solution of the present invention: comprehensively evaluate the accuracy of the construction layout data and the deviation data of the excavation depth, and analyze the influence of the layout accuracy on the excavation depth, specifically including:

[0059] Obtain the layout deviation abnormal coefficient of each layout point and the corresponding cumulative deviation value of each excavation point;

[0060] Construct an input data set and a target output data set

[0061] where s g represents the cumulative deviation value of the excavation depth, g represents the total number of acquisition points, h cg represents the layout deviation abnormal coefficient, and rg Denote the actual preset influence coefficient of the setting-out accuracy on the excavation depth;

[0062] The said acquisition points denote the setting-out acquisition and corresponding excavation acquisition points;

[0063] Construct a support vector machine regression model, and the calculation expression is:

[0064]

[0065] In the formula, ρ o Denote the support vector coefficient, K(Z o , Z) denote the kernel function, d denote the bias term, o denote the acquisition point, where the calculation expression of the kernel function is:

[0066] K(Z o , Z) = exp(-γ||Z o -Z|| 2 );

[0067] In the formula, γ denote the adjustment parameter of the kernel function;

[0068] Through the trained support vector machine regression model, predict the influence value of the setting-out accuracy on the excavation depth, and the calculation expression is:

[0069]

[0070] In the formula, Denote the predicted influence value of the p-th acquisition point;

[0071] According to the predicted influence value, calculate the overall correlation influence coefficient, and the calculation expression is:

[0072]

[0073] In the formula, R impact Denote the overall correlation influence coefficient;

[0074] The said overall correlation influence coefficient is used to evaluate the influence of the setting-out accuracy on the excavation depth.

[0075] As a further solution of the present invention: The specific steps of identifying abnormal points in the construction process according to the comprehensive influence analysis result and making construction adjustments to the deviation influence in the construction process include:

[0076] By real-time collecting the setting-out deviation abnormal coefficient, the overall correlation influence coefficient and the deviation cumulative value of each excavation point, establish a dynamic monitoring data matrix; According to the numerical fluctuation characteristics in the construction process, set the weights of the setting-out deviation abnormal coefficient and the deviation cumulative value as w 1 And w 2, combined with the overall correlation influence coefficient for comprehensive calculation to obtain the comprehensive anomaly coefficient, and the calculation expression is:

[0077]

[0078] In the formula, E abn represents the comprehensive anomaly coefficient, g represents the total number of collection points, o represents the collection point, and s o represents the cumulative deviation value of the o-th collection point, and α co represents the wire laying deviation anomaly coefficient of the o-th collection point, R impact represents the overall correlation influence coefficient, and w 1 and w 2 represent the preset weight factors;

[0079] Judge whether the comprehensive anomaly coefficient is greater than or equal to the preset threshold. If so, it is recorded as a construction anomaly point; if not, it is recorded as a construction normal point;

[0080] For the identified anomaly points, reduce the influence of the cumulative deviation by adjusting the wire laying coordinate values and re-conducting excavation corrections.

[0081] As a further solution of the present invention: Dynamically monitor the deviation trend during construction based on time series analysis, trigger an early warning mechanism through a comprehensive quality evaluation coefficient, and optimize the construction quality, specifically including:

[0082] During the construction process, real-time collect the wire laying deviation anomaly coefficient, the cumulative excavation depth deviation value, and the overall correlation influence coefficient during construction, and record the dynamic changes of the data in the form of time series;

[0083] Analyze the deviation trend during construction by continuously calculating the comprehensive anomaly coefficient of each construction point;

[0084] When the comprehensive anomaly coefficient exceeds the preset threshold, the system triggers an early warning mechanism, marks the abnormal area, and identifies potential quality problems during construction;

[0085] According to the results of the early warning mechanism, the construction party immediately takes optimization measures, including: calibrating the wire laying position of the construction and adjusting the excavation depth operation.

[0086] The engineering supervision quality evaluation system based on big data includes:

[0087] A data acquisition module, which is used to collect real-time data of the wire laying position and excavation depth during construction;

[0088] A wire laying anomaly identification module, which is used to analyze the data of the wire laying position during construction, judge the error between the wire laying position and the design value through deviation calculation, and identify the wire laying accuracy anomaly points in combination with the characteristics of the construction area;

[0089] Construction deviation and impact assessment module, which is used to analyze the excavation depth data, calculate the deviation of the excavation depth data, calculate the cumulative deviation value according to the numerical difference between the excavation depth and the design depth, comprehensively evaluate the accuracy of the construction layout data and the deviation data of the excavation depth, and analyze the impact of the layout accuracy on the excavation depth;

[0090] Comprehensive analysis and dynamic monitoring assessment module, which identifies abnormal points during the construction process according to the comprehensive impact analysis results, makes construction adjustments to the deviation impact during the construction process, dynamically monitors the deviation trend during the construction process based on time series analysis, and triggers an early warning mechanism through the comprehensive quality evaluation coefficient to optimize the construction quality.

[0091] Advantages of the present invention:

[0092] (1) By constructing an evaluation system with the layout deviation abnormal coefficient, the cumulative excavation depth deviation value, and the overall correlation influence coefficient as the core, the present invention realizes a comprehensive quantitative analysis of the mutual influence between the layout accuracy and the excavation depth during the construction process. Combining the real-time data collection on the construction site and applying precise calculation and comprehensive analysis techniques, the present invention can dynamically capture the layout deviation caused by environmental changes or operation errors during the construction process and its cumulative influence on the excavation depth. By establishing a multi-dimensional correlation model, it can not only intuitively evaluate the error range of the layout accuracy but also clarify its potential influence on subsequent construction processes, providing a scientific basis for construction adjustment. Especially in complex construction scenarios, this method significantly improves the controllability of construction quality through accurately positioning abnormal points and the trend early warning mechanism, ensuring the overall stability of the project and high-quality delivery;

[0093] (2) The present invention constructs a dynamic monitoring model of the comprehensive abnormal coefficient, fuses the layout deviation abnormal coefficient, the cumulative excavation depth deviation value, and the overall correlation influence coefficient according to the set weights, and comprehensively monitors the dynamic change trend of the deviation during the construction process in combination with time series analysis. Through multi-dimensional analysis of the real-time collected data, this model can accurately capture construction abnormal points, identify the deviation fluctuation law, and dynamically trigger an early warning mechanism based on the change of the comprehensive abnormal coefficient, providing an efficient abnormal detection and response solution for construction quality management. When the system detects that the deviation exceeds the threshold, it can promptly mark the abnormal area and provide optimization suggestions to the construction party, including correcting the layout point and adjusting the excavation operation, so as to quickly repair the problem area. Through real-time monitoring and precise deviation evaluation, the present invention significantly improves the automation and scientific nature of construction quality management, ensuring the high-quality execution of the construction process and the reliability of the final project result. Description of the Drawings

[0094] The present invention will be further described below in conjunction with the accompanying drawings.

[0095] Figure 1 It is a specific step flow block diagram of the engineering supervision quality assessment method based on big data of the present invention;

[0096] Figure 2 It is a flow block diagram of the engineering supervision quality assessment system based on big data in the present invention. Specific embodiments

[0097] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0098] Please refer to Figure 1 As shown, the present invention is an engineering supervision quality assessment method based on big data, including the following steps:

[0099] S1: During the construction process, the construction layout point data and the excavation depth data are collected in real time;

[0100] S2: Analyze the construction layout point data, judge the error between the layout point and the design value through deviation calculation, and identify the abnormal points of layout accuracy in combination with the characteristics of the construction area;

[0101] S3: Analyze the excavation depth data, calculate the deviation of the excavation depth data, calculate the cumulative deviation value according to the numerical difference between the excavation depth and the design depth, comprehensively evaluate the accuracy of the construction layout data and the deviation data of the excavation depth, and analyze the influence of the layout accuracy on the excavation depth;

[0102] S4: According to the comprehensive influence analysis result, identify the abnormal points during the construction process, make construction adjustments to the deviation influence during the construction process, dynamically monitor the deviation trend during the construction process based on time series analysis, and trigger the early warning mechanism through the comprehensive quality evaluation coefficient to optimize the construction quality.

[0103] In S1, during the construction process, the construction layout point data and the excavation depth data are collected in real time, specifically including:

[0104] During the construction process, the acquisition of construction layout point data is completed through high-precision surveying equipment such as total stations, GNSS (Global Navigation Satellite System) receivers, and electronic distance meters. These devices can obtain the actual coordinate values of the layout points in real time and transmit the collected data to the construction monitoring system through wireless data transmission. Total stations can provide high-precision coordinate measurements in complex construction environments, while GNSS receivers are suitable for large-scale construction sites and provide support for real-time positioning and navigation. Combined with electronic distance meters, the length of construction layout is accurately measured to ensure the accuracy and reliability of the layout data.

[0105] The acquisition of excavation depth data is carried out through real-time measurement by a laser distance meter. Laser distance meters are suitable for open-air construction sites. They measure the excavation depth through beam reflection and provide high-precision depth data. The data collected by all devices can be synchronized to the central monitoring system in real time through wireless transmission or wired network for subsequent analysis and quality assessment.

[0106] In S2, the construction layout point data is analyzed. The error between the layout point and the design value is judged through deviation calculation, and the abnormal points of layout accuracy are identified in combination with the characteristics of the construction area, specifically including:

[0107] The analysis of the construction layout point data, judging the error between the layout point and the design value through deviation calculation, and identifying the abnormal points of layout accuracy in combination with the characteristics of the construction area, specifically including:

[0108] Collect the construction layout point data in real time to obtain the actual coordinate values of each layout point;

[0109] Calculate the deviation value of each layout point, compare it with the design value to obtain the deviation error;

[0110] Calculate the environmental impact coefficient to correct the deviation calculation, adjust the accuracy of the layout data, and analyze the impact of the correction on the deviation;

[0111] According to the analysis results, calculate the abnormal coefficient of layout deviation, and judge whether the abnormal coefficient of layout deviation is greater than or equal to the preset threshold. If so, it is recorded as an abnormal point of layout accuracy; if not, it is recorded as non-abnormal layout accuracy.

[0112] The process of obtaining the environmental impact coefficient is as follows:

[0113] Obtain real-time environmental data, including: environmental humidity data H, air pressure data P, land type S, and environmental temperature data T;

[0114] Establish a Bayesian network model, where the environmental humidity data H, air pressure data P, land type S, and environmental temperature data T are used as input nodes, and the construction deviation D is used as the output node, and the environmental impact coefficient K envAs the target variable;

[0115] Define the conditional probability distribution between each node according to the environmental data, specifically including:

[0116] The conditional probability of environmental humidity at a given temperature P(H|T), the conditional probability of air pressure at a given temperature and humidity P(P|T,H), the conditional probability of soil type at a given humidity, air pressure and temperature P(S|H,P,T), the conditional probability of construction deviation at a given humidity, air pressure, temperature and soil type P(D|H,P,T,S);

[0117] Calculate the joint probability of all environmental factors and construction deviations, and the calculation expression is:

[0118] P(H,P,T,S,D) = P(H|T)·P(P|T,H)·P(S|H,P,T)·P(D|H,P,T,S);

[0119] In the formula, P(H,P,T,S,D) represents the joint probability;

[0120] Calculate the environmental impact coefficient through the joint probability, and the calculation expression is:

[0121] K env = Σ H,P,T,S P(H,P,T,S,D)·f(H,P,T,S,D);

[0122] Among them, K env represents the environmental impact coefficient, and f(H,P,T,S,D) represents the weighting function;

[0123] It should be noted that: the environmental impact coefficient is used to correct the accuracy of which data when calculating the setting-out deviation.

[0124] The process of obtaining the setting-out deviation abnormal coefficient is as follows:

[0125] During the construction process, the actual coordinate values of each setting-out point are collected in real time:

[0126] X i =(x i1 ,x i2 ,...,x in );

[0127] Among them, i is the setting-out point number, i is a positive integer greater than 0, n is the dimension of the coordinates, n is a positive integer greater than 0, and X i represents the actual coordinate value of the i-th setting-out point;

[0128] Obtain the designed coordinate values of each setting-out point:

[0129] Xdesign,i =(x design,i1 , x design,i2 ,..., x design,in );

[0130] Among them, X design,i represents the design coordinate value of the i-th setting-out point;

[0131] Calculate the deviation value between each setting-out point and the corresponding design coordinate. The calculation expression is:

[0132] δ i = ||X i - X design,i ||;

[0133] Among them, δ i represents the deviation value of the i-th setting-out point;

[0134] Construct an isolation forest model. Through the construction and random division of multiple trees, calculate the isolation depth h i of each setting-out point. The isolation depth h i,j of each tree is the isolation path length of the setting-out point on the corresponding tree. Calculate the average isolation depth through multiple trees. The calculation expression is:

[0135]

[0136] In the formula, t represents the number of trees in the isolation forest, j represents the trees in the isolation forest, h i,j represents the isolation depth of the i-th point on the j-th tree, represents the average isolation depth;

[0137] According to the average isolation depth, calculate the anomaly score of each setting-out point. The calculation expression is:

[0138]

[0139] Among them, is a constant, score(X i ) represents the anomaly score of the i-th setting-out point;

[0140]

[0141] In the formula, α i represents the setting-out deviation anomaly coefficient of the i-th setting-out point, and max(δ) represents the maximum value of the deviation values of all setting-out points.

[0142] It should be noted that: The setting-out deviation anomaly coefficient reflects the degree of setting-out deviation during the construction process, and the higher the value of the setting-out deviation anomaly coefficient, the higher the degree of setting-out deviation during the corresponding construction process.

[0143] In S3, the excavation depth data is analyzed, deviation calculation is performed on the excavation depth data, the cumulative deviation value is calculated based on the numerical difference between the excavation depth and the designed depth, and the accuracy of the construction layout data and the deviation data of the excavation depth are comprehensively evaluated to analyze the influence of the layout accuracy on the excavation depth. Specifically, it includes:

[0144] The analysis of the excavation depth data, the deviation calculation of the excavation depth data, and the calculation of the cumulative deviation value based on the numerical difference between the excavation depth and the designed depth specifically include:

[0145] During the construction process, the excavation depth data and the designed depth data of each construction point are collected in real time;

[0146] For each construction point, calculate the deviation value between the actual excavation depth and the designed depth. The calculation expression is:

[0147] s a =H actual,a -H design,a ;

[0148] In the formula, s a represents the deviation value of the a-th point, H actual,a represents the actual excavation depth of the a-th point, H design,a represents the designed excavation depth of the a-th point, and a represents the excavation point;

[0149] Construct a training data set, use the designed depth data of each construction point as the input feature Z a , and the deviation value s a as the target variable to form a training data set;

[0150] Use the random forest regression algorithm to train the model. Based on the designed depth data features, the deviation value is predicted through multiple decision trees. The output deviation prediction value of each tree is:

[0151] In the formula, represents the deviation prediction of the b-th tree for the a-th point, Z a represents the input feature of the a-th point, E b represents the b-th decision tree, and b represents the number of decision trees;

[0152] The final deviation prediction value is obtained by averaging the prediction results of all decision trees. The calculation expression is:

[0153]

[0154] In the formula, represents the final deviation prediction value of the a-th point, and B represents the total number of decision trees;

[0155] Calculate the sum of the deviation prediction values of all construction points to obtain the cumulative deviation value.

[0156] Comprehensively evaluate the accuracy of the construction layout data and the deviation data of the excavation depth, and analyze the influence of the layout accuracy on the excavation depth, specifically including:

[0157] Obtain the layout deviation abnormal coefficient of each layout point and the corresponding cumulative deviation value of each excavation point;

[0158] Construct an input data set and a target output data set

[0159] where s g represents the cumulative deviation value of the excavation depth, g represents the total number of acquisition points, h cg represents the layout deviation abnormal coefficient, and r g represents the actual preset influence coefficient of the layout accuracy on the excavation depth;

[0160] The acquisition points represent the layout acquisition and the corresponding excavation acquisition points;

[0161] Construct a support vector machine regression model, and the calculation expression is:

[0162]

[0163] In the formula, ρ o represents the support vector coefficient, K(Z o , Z) represents the kernel function, d represents the bias term, and o represents the acquisition points. Among them, the calculation expression of the kernel function is:

[0164] K(Z o , Z) = exp(-γ||Z o - Z|| 2 );

[0165] In the formula, γ represents the adjustment parameter of the kernel function;

[0166] Through the trained support vector machine regression model, predict the influence value of the layout accuracy on the excavation depth, and the calculation expression is:

[0167]

[0168] In the formula, represents the predicted influence value of the pth acquisition point;

[0169] According to the predicted influence value, calculate the overall correlation influence coefficient, and the calculation expression is:

[0170]

[0171] In the formula, R impact represents the overall correlation influence coefficient;

[0172] The overall correlation influence coefficient is used to evaluate the influence of the wire laying accuracy on the excavation depth.

[0173] It should be noted that: the cumulative deviation value reflects the deviation of the excavation depth, and the larger the cumulative deviation value, the higher the abnormal degree of the corresponding excavation depth. Moreover, when the value of the overall correlation influence coefficient is larger, it indicates that the influence degree of the inaccurate wire laying on the excavation depth is higher.

[0174] In S4, according to the comprehensive influence analysis result, identify the abnormal points in the construction process, make construction adjustments to the deviation influence in the construction process, and dynamically monitor the deviation trend in the construction process based on time series analysis. Trigger the early warning mechanism through the comprehensive quality evaluation coefficient to optimize the construction quality, specifically including:

[0175] The specific steps of identifying the abnormal points in the construction process and making construction adjustments to the deviation influence according to the comprehensive influence analysis result include:

[0176] By collecting the wire laying deviation abnormal coefficient, the overall correlation influence coefficient and the cumulative deviation value of each excavation point in real time, establish a dynamic monitoring data matrix; according to the numerical fluctuation characteristics in the construction process, set the weights of the wire laying deviation abnormal coefficient and the cumulative deviation value as w 1 and w 2 respectively, and perform comprehensive calculation in combination with the overall correlation influence coefficient to obtain the comprehensive abnormal coefficient. The calculation expression is:

[0177]

[0178] In the formula, E abn represents the comprehensive abnormal coefficient, g represents the total number of collection points, o represents the collection point, s o represents the cumulative deviation value of the o-th collection point, α co represents the wire laying deviation abnormal coefficient of the o-th collection point, R impact represents the overall correlation influence coefficient, w 1 and w 2 represent the preset weight factors;

[0179] Judge whether the comprehensive abnormal coefficient is greater than or equal to the preset threshold. If so, record it as an abnormal construction point; if not, record it as a normal construction point;

[0180] For the identified abnormal points, reduce the influence of the cumulative deviation by adjusting the wire laying coordinate value and re - performing excavation correction;

[0181] Dynamically monitoring the deviation trend during construction based on time series analysis, triggering an early warning mechanism through a comprehensive quality evaluation coefficient, and optimizing construction quality, specifically including:

[0182] During the construction process, the abnormal coefficient of construction setting-out deviation, the cumulative value of excavation depth deviation, and the overall relevant influence coefficient are collected in real time, and the dynamic changes of the data are recorded in the form of time series.

[0183] By continuously calculating the comprehensive abnormal coefficient of each construction point, the deviation trend during the construction process is analyzed.

[0184] When the comprehensive abnormal coefficient exceeds the preset threshold, the system triggers an early warning mechanism, marks the abnormal area, and identifies potential quality problems during the construction process.

[0185] According to the results of the early warning mechanism, the construction party immediately takes optimization measures, including: calibrating the construction setting-out points and adjusting the excavation depth operation.

[0186] It should be noted that: through the calculated comprehensive abnormal coefficient, it is judged whether each construction point in the construction process is abnormal, and dynamic monitoring and early warning of the engineering construction monitoring can be carried out.

[0187] Please refer to Figure 2 As shown, the engineering supervision quality evaluation system based on big data includes:

[0188] A data acquisition module, which is used to collect the construction setting-out point data and excavation depth data in real time during the construction process.

[0189] A setting-out anomaly identification module, which is used to analyze the construction setting-out point data, judge the error between the setting-out point and the design value through deviation calculation, and identify the setting-out accuracy anomaly points in combination with the characteristics of the construction area.

[0190] A construction deviation and influence evaluation module, which is used to analyze the excavation depth data, calculate the deviation of the excavation depth data, calculate the cumulative deviation value according to the numerical difference between the excavation depth and the design depth, comprehensively evaluate the accuracy of the construction setting-out data and the deviation data of the excavation depth, and analyze the influence of the setting-out accuracy on the excavation depth.

[0191] A comprehensive analysis and dynamic monitoring evaluation module, which identifies the abnormal points during the construction process according to the comprehensive influence analysis results, makes construction adjustments to the deviation influence during the construction process, dynamically monitors the deviation trend during the construction process based on time series analysis, and triggers an early warning mechanism through a comprehensive quality evaluation coefficient to optimize the construction quality.

[0192] Working principle of the present invention: By collecting construction layout point data and excavation depth data in real time, and combining total station, GNSS receiver, and laser rangefinder to obtain actual coordinate values and depth information, the accuracy of data collection is ensured. In the data analysis stage, first, the error between the layout point and the design value is obtained through deviation calculation. Considering influencing factors such as environmental humidity, air pressure, and soil type, the layout data is corrected by calculating the environmental impact coefficient, and further the layout deviation anomaly coefficient is calculated to identify abnormal points. At the same time, the cumulative value analysis of the excavation depth deviation is carried out, and the support vector machine regression model is used to predict the overall correlation between the layout accuracy and the excavation depth. By dynamically monitoring the change of the comprehensive anomaly coefficient, the deviation trend is analyzed based on time series, and the early warning mechanism is triggered in combination with the comprehensive quality evaluation coefficient to guide the adjustment and optimization of the construction process. Through the innovative calculation method of the comprehensive anomaly coefficient, the present invention establishes an accurate abnormal point identification and dynamic construction quality monitoring mechanism, improving the intelligent level and quality control ability of project supervision.

[0193] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to get a formula closest to the actual situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0194] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center in a wired or wireless (such as infrared, wireless, microwave, etc.) manner. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center containing one or more collections of available media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium (such as a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0195] It should be understood that the term "and / or" in this text is merely a description of the association relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone. Here, A and B can be singular or plural. Additionally, the character " / " in this text generally represents an "or" relationship between the associated objects before and after, but it may also represent an "and / or" relationship, and the specific meaning can be understood by referring to the context before and after.

[0196] It should be understood that in various embodiments of this application, the magnitude of the sequence numbers of the above processes does not imply the order of execution. The order of execution of each process should be determined by its function and internal logic, and should not impose any limitation on the implementation process of the embodiments of this application.

[0197] The above has described in detail one embodiment of the present invention, but the content described is only a preferred embodiment of the present invention and cannot be considered as limiting the scope of implementation of the present invention. Any equivalent changes and improvements made in accordance with the scope of the application of the present invention should still fall within the scope covered by the patent of the present invention.

Claims

1. The engineering supervision quality assessment method based on big data is characterized by: The following steps are involved: S1: During the construction process, real-time data on construction layout points and excavation depth are collected; S2: Analyze the construction layout point data, determine the error between the layout point and the design value through deviation calculation, and identify the abnormal points of layout accuracy based on the characteristics of the construction area; S3: Analyze the excavation depth data, calculate the deviation of the excavation depth data, calculate the cumulative deviation value according to the numerical difference between the excavation depth and the design depth, comprehensively evaluate the accuracy of the construction layout data and the deviation data of the excavation depth, and analyze the influence of the layout accuracy on the excavation depth; S4: Based on the results of comprehensive impact analysis, identify abnormal points in the construction process, make construction adjustments to the deviation impacts during the construction process, and dynamically monitor the deviation trend during the construction process based on time series analysis. Trigger the early warning mechanism through the comprehensive quality evaluation coefficient to optimize the construction quality.

2. The engineering supervision quality assessment method based on big data according to claim 1 is characterized in that: The analysis of the construction layout point data, the determination of the error between the layout point and the design value by deviation calculation, and the identification of the layout accuracy abnormal points in combination with the construction area characteristics specifically include: Collect construction layout point data in real time to obtain the actual coordinate value of each layout point; Calculate the deviation value of each layout point and compare it with the design value to obtain the deviation error; Calculate the environmental impact coefficient to correct the deviation calculation, adjust the accuracy of the line setting data, and analyze the impact of the correction on the deviation; According to the analysis results, the abnormal coefficient of the line-laying deviation is calculated to determine whether the abnormal coefficient is greater than or equal to the preset threshold. If so, it is recorded as a line-laying accuracy abnormal point; if not, it is recorded as a non-line-laying accuracy abnormal point.

3. The engineering supervision quality assessment method based on big data according to claim 2 is characterized in that: The process of obtaining the environmental impact coefficient is as follows: Obtain real-time environmental data, including: environmental humidity data H, air pressure data P, land type S and environmental temperature data T; A Bayesian network model is established, in which the environmental humidity data H, air pressure data P, land type S and environmental temperature data T are used as input nodes, the construction deviation D is used as the output node, and the environmental impact coefficient K is used as the output node. env as the target variable; Define the conditional probability distribution between nodes based on the environmental data, including: The conditional probability of ambient humidity at a given temperature is P(H|T), the conditional probability of air pressure at a given temperature and humidity is P(P|T,H), the conditional probability of soil type at a given humidity, air pressure and temperature is P(S|H,P,T), and the conditional probability of construction deviation at a given humidity, air pressure, temperature and soil type is P(D|H,P,T,S); Calculate the joint probability of all environmental factors and construction deviations. The calculation expression is: P(H,P,T,S,D)=P(H|T)·P(P|T,H)·P(S|H,P,T)·P(D|H,P,T,S); Where P(H,P,T,S,D) represents the joint probability; The environmental impact coefficient is calculated by joint probability, and the calculation expression is: K env =∑ H,P,T,S P(H,P,T,S,D)·f(H,P,T,S,D); Among them, K env represents the environmental impact coefficient, and f(H,P,T,S,D) represents the weighting function.

4. The engineering supervision quality assessment method based on big data according to claim 2 is characterized in that: The process of obtaining the abnormal coefficient of the line setting deviation is as follows: During the construction process, the actual coordinate values ​​of each layout point are collected in real time: X i =(x i1 ,x i2 ,...,x in ); Where i is the number of the point, i is a positive integer greater than 0, n is the dimension of the coordinate, n is a positive integer greater than 0, X i Represents the actual coordinate value of the i-th laying-out point; Get the design coordinate value of each layout point: X design,i =(x design,i1 ,x design,i2 ,...,x design,in ); Among them, X design,i Represents the design coordinate value of the i-th layout point; Calculate the deviation between each layout point and the corresponding design coordinates. The calculation expression is: δ i =||X i -X design,i ||; Among them, δ i Indicates the deviation value of the i-th laying-out point; Construct an isolation forest model, and calculate the isolation depth h of each release point by constructing and randomly dividing multiple trees. i , the isolation depth of each tree is h i,j The length of the isolated path of the line point on the corresponding tree is calculated by multiple trees to get the average isolated depth. The calculation expression is: In the formula, t represents the number of trees in the isolation forest, j represents the number of trees in the isolation forest, and h i,j represents the isolation depth of the i-th point on the j-th tree, represents the average isolation depth; According to the average isolation depth, the anomaly score of each release point is calculated, and the calculation expression is: in, is a constant, score(X i ) represents the abnormal score of the i-th release point; In the formula, α i It represents the abnormal coefficient of the setting-out deviation of the ith setting-out point, and max(δ) represents the maximum deviation value of all setting-out points.

5. The engineering supervision quality assessment method based on big data according to claim 1 is characterized in that: The analysis of the excavation depth data, the deviation calculation of the excavation depth data, and the calculation of the deviation cumulative value according to the numerical difference between the excavation depth and the design depth specifically include: During the construction process, the excavation depth data and design depth data of each construction point are collected in real time; For each construction point, the deviation between the actual excavation depth and the design depth is calculated. The calculation expression is: s a =H actual,a -H design,a ; In the formula, s a Indicates the deviation value of the a-th point, H actual,a represents the actual excavation depth of the ath point, H design,a represents the designed excavation depth of the ath point, where a represents the excavation point; Construct a training data set and use the design depth data of each construction point as the input feature Z a , deviation value s a As the target variable, a training data set is formed; The model is trained using the random forest regression algorithm. Based on the design of deep data features, the deviation value is predicted through multiple decision trees, where the output deviation prediction value of each tree is: In the formula, represents the deviation prediction of the bth tree for the ath point, Z a represents the input feature of the a-th point, E b represents the b-th decision tree, and b represents the number of decision trees; The final deviation prediction value is obtained by averaging the prediction results of all decision trees. The calculation expression is: In the formula, represents the final deviation prediction value of the a-th point, and B represents the total number of decision trees; Calculate the sum of the deviation prediction values ​​of all construction points to obtain the cumulative deviation value.

6. The engineering supervision quality assessment method based on big data according to claim 1 is characterized in that: The accuracy of the construction layout data and the deviation data of the excavation depth are comprehensively evaluated to analyze the influence of the layout accuracy on the excavation depth, specifically including: Obtain the abnormal coefficient of the line setting deviation of each line setting point and the corresponding cumulative deviation value of each excavation point; Constructing the input dataset and the target output dataset Among them, s g represents the cumulative deviation of the excavation depth, g represents the total number of collection points, and h cg Indicates the abnormal coefficient of line setting deviation, r g Indicates the actual preset influence coefficient of the setting-out accuracy on the excavation depth; The collection points represent the line setting collection and the corresponding excavation collection points; Construct a support vector machine regression model, and the calculation expression is: In the formula, ρ o represents the support vector coefficient, K(Z o ,Z) represents the kernel function, d represents the bias term, and o represents the acquisition point. The calculation expression of the kernel function is: C(Z o ,Z)=exp(-γ||Z o -Z|| 2 ); In the formula, γ represents the adjustment parameter of the kernel function; The influence value of layout accuracy and excavation depth is predicted by the trained support vector machine regression model. The calculation expression is: In the formula, represents the predicted impact value of the pth collection point; According to the predicted impact value, the overall relevant impact coefficient is calculated, and the calculation expression is: In the formula, R impact represents the overall correlation influence coefficient; The overall correlation influence coefficient is used to evaluate the influence of the setting-out accuracy on the excavation depth.

7. The engineering supervision quality assessment method based on big data according to claim 1 is characterized in that: The above-mentioned steps of identifying abnormal points in the construction process based on the comprehensive impact analysis results and making construction adjustments to the deviation impacts in the construction process specifically include: A dynamic monitoring data matrix is ​​established by real-time collection of the abnormal coefficient of line setting deviation, the overall relevant influence coefficient and the cumulative deviation value of each excavation point; according to the numerical fluctuation characteristics during the construction process, the weights of the abnormal coefficient of line setting deviation and the cumulative deviation value are set as w1 and w2 respectively, and the overall relevant influence coefficient is combined for comprehensive calculation to obtain the comprehensive abnormal coefficient. The calculation expression is: In the formula, E abn represents the comprehensive anomaly coefficient, g represents the total number of collection points, o represents the collection point, s o represents the cumulative deviation of the oth acquisition point, α co Indicates the abnormal coefficient of the line setting deviation at the oth acquisition point, R impact represents the overall correlation influence coefficient, w1 and w2 represent the preset weight factors; Determine whether the comprehensive abnormality coefficient is greater than or equal to a preset threshold value. If so, it is recorded as a construction abnormal point; if not, it is recorded as a construction normal point; For the identified abnormal points, the influence of the cumulative deviation can be reduced by adjusting the coordinate values ​​of the layout and re-excavating and correcting them.

8. The engineering supervision quality assessment method based on big data according to claim 1 is characterized in that: The method of dynamically monitoring the deviation trend during the construction process based on time series analysis and triggering the early warning mechanism through the comprehensive quality evaluation coefficient to optimize the construction quality specifically includes: During the construction process, the abnormal coefficient of construction layout deviation, the cumulative value of excavation depth deviation and the overall related influence coefficient are collected in real time, and the dynamic changes of data are recorded in the form of time series; By continuously calculating the comprehensive abnormal coefficient of each construction point, the deviation trend during the construction process is analyzed; When the comprehensive abnormality coefficient exceeds the preset threshold, the system triggers the early warning mechanism, marks the abnormal area and identifies potential quality problems in the construction process; Based on the results of the early warning mechanism, the construction party immediately took optimization measures, including: calibrating construction layout points and adjusting excavation depth operations.

9. The engineering supervision quality assessment system based on big data is characterized by: The method for evaluating the quality of engineering supervision based on big data as claimed in any one of claims 1 to 8 comprises: A data acquisition module, which is used to collect construction layout point data and excavation depth data in real time during the construction process; A wiring anomaly identification module is used to analyze the wiring point data of the construction, determine the error between the wiring point and the design value through deviation calculation, and identify the wiring accuracy anomaly points in combination with the characteristics of the construction area; A construction deviation and impact assessment module, which is used to analyze the excavation depth data, calculate the deviation of the excavation depth data, calculate the cumulative deviation value according to the numerical difference between the excavation depth and the design depth, comprehensively evaluate the accuracy of the construction layout data and the deviation data of the excavation depth, and analyze the impact of the layout accuracy on the excavation depth; A comprehensive analysis and dynamic monitoring and evaluation module identifies abnormal points in the construction process based on the comprehensive impact analysis results, makes construction adjustments to the deviation impacts in the construction process, and dynamically monitors the deviation trend in the construction process based on time series analysis, triggers an early warning mechanism through a comprehensive quality evaluation coefficient, and optimizes construction quality.

Images