Pipe jacking construction quality evaluation method and system based on machine learning

Through machine learning-based pipe hoist construction quality evaluation methods and systems, real-time monitoring and quality control problems in pipe hoist construction are solved, accurate evaluation and early warning of the construction process are achieved, and construction efficiency and quality are improved.

CN118966876APending Publication Date: 2024-11-15POWER CHINA KUNMING ENG CORP LTD
View PDF 0 Cites 7 Cited by

Patent Information

Application Number
CN202410998016.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-24
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

In the construction of sewage treatment pipelines, real-time monitoring and quality control of pipe top construction are difficult, and the existing technology cannot effectively solve the problems of timely warning and rectification of potential risks during the construction process.

Method used

Adopt the quality evaluation method and system of pipe top construction based on machine learning. By setting multiple working conditions categories and feedback time nodes, establishing quality evaluation models, monitoring construction parameters and environmental parameters in real time, adjusting evaluation indicators, timely warning and rectification, and optimizing Quality assessment model parameters to improve construction accuracy and efficiency.

Benefits of technology

Accurate quality evaluation and real-time monitoring of the pipe hoisting construction process, timely warning and rectification, improve construction efficiency and overall quality, and ensure the accuracy and effect of construction.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118966876A_ABST
    Figure CN118966876A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of pipe jacking construction quality evaluation, in particular to a pipe jacking construction quality evaluation method and system based on machine learning. Comprising the steps of generating multiple construction categories according to historical construction parameters, and establishing a quality evaluation model of each construction category; establishing a monitoring evaluation time axis according to the to-be-constructed plan of the pipe jacking, and generating a plurality of feedback time nodes according to the monitoring evaluation time axis; obtaining construction monitoring parameters and construction environment parameters according to a preset feedback time node, judging a construction category according to the construction environment parameters, and setting a corresponding quality evaluation model as a first-level quality evaluation model according to a judgment result; by setting a plurality of feedback time nodes of a plurality of working condition categories, adjusting corresponding evaluation indexes in time according to different working condition categories, performing periodic accurate evaluation and real-time monitoring on the pipe jacking construction quality, performing early warning and rectification on various potential construction risks in time, and ensuring the overall quality and the construction effect of pipe jacking construction.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of pipe jacking construction quality assessment, and in particular to a pipe jacking construction quality assessment method and system based on machine learning. Background Art

[0002] In the process of urban construction, sewage treatment pipelines are a very important infrastructure, and the quality of sewage treatment pipeline construction will have a huge impact on people's daily lives. In the construction of sewage treatment pipelines, pipe jacking construction is a very widely used construction technology.

[0003] As an advanced trenchless construction technology, pipe jacking is widely used in municipal pipelines, subways, public utilities, and hydropower applications due to its advantages of small construction footprint, environmental protection, and little disturbance to road traffic and buildings. During the pipe jacking construction process, how to conduct real-time monitoring and quality control of the pipe jacking construction has become a technical problem that needs to be solved urgently. Summary of the invention

[0004] The purpose of this application is: to solve the above-mentioned technical problems, this application provides a method and system for evaluating the quality of pipe jacking construction based on machine learning, aiming to improve the monitoring efficiency of the pipe jacking construction process, ensure the overall construction quality, and improve the construction effect.

[0005] In some embodiments of the present application, by setting multiple working condition categories and multiple feedback time nodes, the corresponding evaluation indicators are adjusted in time for different working condition categories, the quality of the pipe jacking construction is periodically and accurately evaluated and monitored in real time, and it is judged whether there are construction problems based on the real-time construction quality evaluation value, and various potential construction risks are warned and rectified in time to ensure the accuracy and construction efficiency of the pipe jacking construction process.

[0006] In some embodiments of the present application, based on the machine learning method, the working condition categories are continuously refined to improve the accurate evaluation of the construction quality at different stages of the pipe jacking construction process, and the monitoring parameters of each quality assessment model are optimized and screened, thereby improving the quality assessment model's analysis and evaluation accuracy for the pipe jacking construction quality, and timely warning of various potential construction risks. After the overall construction is completed, the overall construction quality evaluation value is generated to achieve an overall evaluation and analysis of the pipe jacking construction process, ensuring the overall quality and effect of the pipe jacking construction.

[0007] In some embodiments of the present application, a method for evaluating the quality of pipe jacking construction based on machine learning is provided, comprising:

[0008] Generate multiple construction categories based on historical construction parameters and establish quality assessment models for each construction category;

[0009] Establish a monitoring and evaluation timeline according to the pipe jacking construction plan, and generate multiple feedback time nodes based on the monitoring and evaluation timeline;

[0010] Obtain construction monitoring parameters and construction environment parameters according to preset feedback time nodes, determine the construction category according to the construction environment parameters, and set the corresponding quality assessment model as a first-level quality assessment model according to the determination result;

[0011] The construction quality evaluation value is generated based on the construction monitoring parameters and the first-level quality assessment model.

[0012] In some embodiments of the present application, when establishing a quality assessment model for each construction category, it includes

[0013] Traverse historical construction parameters and generate multiple construction categories;

[0014] Establish a construction category sequence A, A = (a1, a2...an), where ai is the i-th construction category; n is the number of construction categories;

[0015] Select the target construction category in sequence according to the construction category sequence A;

[0016] Set multiple evaluation indicators for target construction categories;

[0017] Generate a training set data package based on historical construction parameters and all evaluation indicators of the target construction category;

[0018] Establish a quality assessment model for the target construction category based on the training set data package;

[0019] Establish a quality assessment model sequence B, B = (b1, b2...bn), where bi is the quality assessment model corresponding to the i-th construction category;

[0020] Set the update parameters of the quality assessment model series.

[0021] In some embodiments of the present application, when determining the construction category according to the construction environment parameters, it includes:

[0022] Obtain the construction environment parameters of the current feedback time node;

[0023] Generate similarity evaluation values ​​between construction environment parameters and each construction category according to a preset similarity evaluation model;

[0024] Establish a similarity evaluation value sequence F of the current feedback time node, F = (f1, f2...fn), where fn is the similarity evaluation value of the construction environment parameter at the current feedback time node and the i-th construction category;

[0025] The quality assessment model of the construction category corresponding to the maximum value fmax in the similarity evaluation value sequence F is selected as the first-level quality assessment model.

[0026] In some embodiments of the present application, the generating of the construction quality evaluation value includes:

[0027] Obtain the first-level quality assessment model of the current feedback time node;

[0028] Generate an evaluation index sequence C according to the first-level quality assessment model, C = (c1, c2...cm), where ci is the i-th evaluation index and m is the number of evaluation indicators required by the first-level quality assessment model;

[0029] Generate reference values ​​for various evaluation indicators based on the construction monitoring parameters at the current feedback time node;

[0030] Establish a reference value sequence D of the current feedback time node, D = (d1, d2 ... dm), where di is the reference value of the i-th evaluation indicator at the current feedback time node;

[0031] Generate the construction quality evaluation value g at the current feedback time node;

[0032]

[0033] Among them, e1 is the preset first weight coefficient, e2 is the preset second weight coefficient, Q1 is the preset first fixed coefficient, Q2 is the preset second fixed coefficient, αi is the influencing factor of the i-th evaluation indicator; d'i is the standard reference value of the i-th evaluation indicator.

[0034] In some embodiments of the present application, further comprising:

[0035] Determine whether to generate an early warning instruction based on the construction quality evaluation value g at the current feedback time node;

[0036] A first construction quality evaluation value threshold G1 and a second construction quality evaluation value threshold G2 are preset, and G1<G2;

[0037] If g<G1, a first-level warning instruction is generated;

[0038] If G1≤g<G2, a second-level warning instruction is generated;

[0039] If g≥G2, no warning instruction is generated.

[0040] In some embodiments of the present application, further comprising:

[0041] When the overall construction is completed;

[0042] Obtain the construction quality evaluation value of each feedback time node according to the monitoring and evaluation timeline;

[0043] Establish a construction quality evaluation value sequence G, G = (g1, g2...gr), where gi is the construction quality evaluation value of the i-th feedback time node, and r is the number of feedback time nodes in the monitoring and evaluation time axis;

[0044] Generate an overall construction evaluation value h according to the construction instruction evaluation value sequence G;

[0045]

[0046] Wherein, e3 is the preset third weight coefficient, e4 is the preset fourth weight coefficient; Q3 is the preset third fixed coefficient, Q4 is the preset fourth fixed coefficient; △ g is the average value of all data in the construction quality evaluation value series G.

[0047] In some embodiments of the present application, when setting the update parameters of the quality assessment model series, it includes:

[0048] Generate a working condition category update parameter according to a preset update time node, and generate a first-level update instruction of a quality assessment model series according to the working condition category update parameter;

[0049] Obtain the historical call parameters of each quality assessment model according to the preset update time node, and generate secondary update instructions;

[0050] The secondary update instruction includes:

[0051] Select the target quality assessment model in turn;

[0052] Obtain historical call parameters of the target quality assessment model;

[0053] Generate monitoring indicator screening parameters of the target quality assessment model based on historical call parameters;

[0054] According to the screening results, the updated parameters of the target quality assessment model are generated;

[0055] Generate update parameters of each quality assessment model in the quality assessment model sequence B in sequence.

[0056] In some embodiments of the present application, a pipe jacking construction quality assessment system based on machine learning is provided, comprising:

[0057] The central control unit is used to generate multiple construction categories based on historical construction parameters and establish a quality assessment model for each construction category;

[0058] A monitoring unit is used to establish a monitoring and evaluation timeline according to the pipe jacking construction plan, and generate multiple feedback time nodes according to the monitoring and evaluation timeline;

[0059] The monitoring unit includes a plurality of monitoring submodules, and the monitoring submodules are used to collect construction monitoring parameters and construction environment parameters;

[0060] An evaluation unit, used to obtain construction monitoring parameters and construction environment parameters according to preset feedback time nodes;

[0061] The evaluation unit is also used to determine the construction category according to the construction environment parameters, set the corresponding quality evaluation model as a first-level quality evaluation model according to the determination result, and generate a construction quality evaluation value according to the construction monitoring parameters and the first-level quality evaluation model;

[0062] The central control unit comprises:

[0063] The first processing module is used to traverse the historical construction parameters and generate multiple construction categories;

[0064] Establish a construction category sequence A, A = (a1, a2...an), where ai is the i-th construction category; n is the number of construction categories;

[0065] Select the target construction category in sequence according to the construction category sequence A;

[0066] Set multiple evaluation indicators for target construction categories;

[0067] Generate a training set data package based on historical construction parameters and all evaluation indicators of the target construction category;

[0068] Establish a quality assessment model for the target construction category based on the training set data package;

[0069] Establish a quality assessment model sequence B, B = (b1, b2...bn), where bi is the quality assessment model corresponding to the i-th construction category;

[0070] A first updating module, used to generate a working condition category update parameter according to a preset update time node, and generate a first-level update instruction of a quality assessment model series according to the working condition category update parameter;

[0071] The first updating module is also used to obtain the historical calling parameters of each quality assessment model according to the preset update time node, and generate a secondary update instruction;

[0072] The secondary update instruction includes:

[0073] Select the target quality assessment model in turn;

[0074] Obtain historical call parameters of the target quality assessment model;

[0075] Generate monitoring indicator screening parameters of the target quality assessment model based on historical call parameters;

[0076] According to the screening results, the updated parameters of the target quality assessment model are generated;

[0077] Generate update parameters of each quality assessment model in the quality assessment model sequence B in sequence.

[0078] In some embodiments of the present application, the evaluation unit includes:

[0079] The first judgment module is used to obtain the construction environment parameters of the current feedback time node;

[0080] Generate similarity evaluation values ​​between construction environment parameters and each construction category according to a preset similarity evaluation model;

[0081] Establish a similarity evaluation value sequence F of the current feedback time node, F = (f1, f2...fn), where fn is the similarity evaluation value of the construction environment parameter at the current feedback time node and the i-th construction category;

[0082] The quality assessment model of the construction category corresponding to the maximum value fmax in the similarity evaluation value series F is selected as the first-level quality assessment model;

[0083] The first evaluation module is used to obtain a first-level quality evaluation model of the current feedback time node;

[0084] Generate an evaluation index sequence C according to the first-level quality assessment model, C = (c1, c2...cm), where ci is the i-th evaluation index and m is the number of evaluation indicators required by the first-level quality assessment model;

[0085] Generate reference values ​​for various evaluation indicators based on the construction monitoring parameters at the current feedback time node;

[0086] Establish a reference value sequence D of the current feedback time node, D = (d1, d2 ... dm), where di is the reference value of the i-th evaluation indicator at the current feedback time node;

[0087] Generate the construction quality evaluation value g at the current feedback time node;

[0088]

[0089] Wherein, e1 is the preset first weight coefficient, e2 is the preset second weight coefficient, Q1 is the preset first fixed coefficient, Q2 is the preset second fixed coefficient, αi is the influencing factor of the i-th evaluation index; d'i is the standard reference value of the i-th evaluation index;

[0090] The early warning module is used to determine whether to generate an early warning instruction according to the construction quality evaluation value g at the current feedback time node;

[0091] A first construction quality evaluation value threshold G1 and a second construction quality evaluation value threshold G2 are preset, and G1<G2;

[0092] If g<G1, a first-level warning instruction is generated;

[0093] If G1≤g<G2, a second-level warning instruction is generated;

[0094] If g≥G2, no warning instruction is generated.

[0095] In some embodiments of the present application, the evaluation unit further includes:

[0096] A second evaluation module, when the overall construction is completed, the second evaluation module obtains the construction quality evaluation value of each feedback time node according to the monitoring and evaluation time axis;

[0097] Establish a construction quality evaluation value sequence G, G = (g1, g2...gr), where gi is the construction quality evaluation value of the i-th feedback time node, and r is the number of feedback time nodes in the monitoring and evaluation time axis;

[0098] Generate an overall construction evaluation value h according to the construction instruction evaluation value sequence G;

[0099]

[0100] Wherein, e3 is the preset third weight coefficient, e4 is the preset fourth weight coefficient; Q3 is the preset third fixed coefficient, Q4 is the preset fourth fixed coefficient; △ g is the average value of all data in the construction quality evaluation value series G.

[0101] Compared with the prior art, the pipe jacking construction quality assessment method and system based on machine learning in the embodiment of the present application has the following beneficial effects:

[0102] By setting multiple working condition categories and multiple feedback time nodes, the corresponding evaluation indicators are adjusted in time for different working condition categories, and the quality of pipe jacking construction is periodically and accurately evaluated and monitored in real time. It is also possible to determine whether there are construction problems based on the real-time construction quality evaluation values, and to promptly issue early warnings and rectify various potential construction risks, thereby ensuring the accuracy and efficiency of the pipe jacking construction process.

[0103] Based on the machine learning method, the working condition categories are continuously refined to improve the accurate evaluation of the construction quality at different stages of the pipe jacking construction process, and the monitoring parameters of each quality assessment model are optimized and screened, thereby improving the analysis and evaluation accuracy of the quality assessment model for the pipe jacking construction quality, and timely warning of various potential construction risks. After the overall construction is completed, the overall construction quality evaluation value is generated to achieve an overall evaluation and analysis of the pipe jacking construction process, ensuring the overall quality and effect of the pipe jacking construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0104] Figure 1 It is a flow chart of a method for evaluating the quality of pipe jacking construction based on machine learning in a preferred embodiment of the present application. DETAILED DESCRIPTION

[0105] The specific implementation methods of the present application are further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present application but are not intended to limit the scope of the present application.

[0106] In the description of the present application, it should be understood that the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0107] The terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0108] In the description of this application, 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 this application can be understood according to specific circumstances.

[0109] like Figure 1 As shown, a method for evaluating the quality of pipe jacking construction based on machine learning in a preferred embodiment of the present application includes:

[0110] S101: Generate multiple construction categories based on historical construction parameters and establish a quality assessment model for each construction category;

[0111] S102: Establish a monitoring and evaluation timeline according to the pipe jacking construction plan, and generate multiple feedback time nodes according to the monitoring and evaluation timeline;

[0112] S103: Acquire construction monitoring parameters and construction environment parameters according to a preset feedback time node, determine the construction category according to the construction environment parameters, and set the corresponding quality assessment model as a first-level quality assessment model according to the determination result;

[0113] S104: Generate a construction quality evaluation value according to the construction monitoring parameters and the first-level quality assessment model.

[0114] Specifically, when establishing the quality assessment model for each construction category, it includes

[0115] Traverse historical construction parameters and generate multiple construction categories;

[0116] Establish a construction category sequence A, A = (a1, a2...an), where ai is the i-th construction category; n is the number of construction categories;

[0117] Select the target construction category in sequence according to the construction category sequence A;

[0118] Set multiple evaluation indicators for target construction categories;

[0119] Generate a training set data package based on historical construction parameters and all evaluation indicators of the target construction category;

[0120] Establish a quality assessment model for the target construction category based on the training set data package;

[0121] Establish a quality assessment model sequence B, B = (b1, b2...bn), where bi is the quality assessment model corresponding to the i-th construction category;

[0122] Set the update parameters of the quality assessment model series.

[0123] Specifically, multiple working condition evaluation indicators are extracted based on historical parameters, and multiple value intervals for a single working condition evaluation indicator are established in turn. Based on random combinations, multiple working condition categories are generated, and corresponding quality assessment models are set for different working condition categories, thereby achieving accurate evaluation of the quality of pipe jacking construction.

[0124] Specifically, the working condition evaluation indicators include but are not limited to parameters such as construction environment, construction volume, and construction category.

[0125] It can be understood that in the above embodiment, by setting multiple working condition categories and multiple feedback time nodes, the corresponding evaluation indicators are adjusted in time for different working condition categories, the quality of pipe jacking construction is periodically and accurately evaluated and monitored in real time, and it is judged whether there are construction problems based on the real-time construction quality evaluation value, and various potential construction risks are warned and rectified in time to ensure the accuracy and construction efficiency of the pipe jacking construction process.

[0126] In a preferred embodiment of the present application, when determining the construction category according to the construction environment parameters, it includes:

[0127] Obtain the construction environment parameters of the current feedback time node;

[0128] Generate similarity evaluation values ​​between construction environment parameters and each construction category according to a preset similarity evaluation model;

[0129] Establish a similarity evaluation value sequence F of the current feedback time node, F = (f1, f2...fn), where fn is the similarity evaluation value of the construction environment parameter at the current feedback time node and the i-th construction category;

[0130] The quality assessment model of the construction category corresponding to the maximum value fmax in the similarity evaluation value sequence F is selected as the first-level quality assessment model.

[0131] Specifically, when generating the construction quality evaluation value, it includes:

[0132] Obtain the first-level quality assessment model of the current feedback time node;

[0133] Generate an evaluation index sequence C according to the first-level quality assessment model, C = (c1, c2...cm), where ci is the i-th evaluation index and m is the number of evaluation indicators required by the first-level quality assessment model;

[0134] Generate reference values ​​for various evaluation indicators based on the construction monitoring parameters at the current feedback time node;

[0135] Establish a reference value sequence D of the current feedback time node, D = (d1, d2 ... dm), where di is the reference value of the i-th evaluation indicator at the current feedback time node;

[0136] Generate the construction quality evaluation value g at the current feedback time node;

[0137]

[0138] Among them, e1 is the preset first weight coefficient, e2 is the preset second weight coefficient, Q1 is the preset first fixed coefficient, Q2 is the preset second fixed coefficient, αi is the influencing factor of the i-th evaluation indicator; d'i is the standard reference value of the i-th evaluation indicator.

[0139] Specifically, the evaluation indicators include but are not limited to construction progress, construction volume, construction deviation, potential risks, displacement of soil and underground pipelines, construction requirements and construction deviation and other parameters.

[0140] Specifically, by setting the first fixed coefficient and the second fixed coefficient, each evaluation index in the model is normalized. The higher the construction instruction evaluation value is, the higher the quality of the current pipe jacking construction is, and the lower the possibility of construction risks is.

[0141] Specifically, it also includes:

[0142] Determine whether to generate an early warning instruction based on the construction quality evaluation value g at the current feedback time node;

[0143] A first construction quality evaluation value threshold G1 and a second construction quality evaluation value threshold G2 are preset, and G1<G2;

[0144] If g<G1, a first-level warning instruction is generated;

[0145] If G1≤g<G2, a second-level warning instruction is generated;

[0146] If g≥G2, no warning instruction is generated.

[0147] Specifically, the first-level warning instruction indicates that the current pipe jacking construction quality is unqualified and there are major safety hazards, and re-construction is required to avoid construction safety problems. The second-level warning instruction indicates that there is a certain amount of deviation from the expected requirements during the current pipe jacking construction process, and timely adjustments are required, and the subsequent construction parameters should be corrected in a timely manner to improve construction quality and reduce safety risks.

[0148] In a preferred embodiment of the present application, it also includes:

[0149] When the overall construction is completed;

[0150] Obtain the construction quality evaluation value of each feedback time node according to the monitoring and evaluation timeline;

[0151] Establish a construction quality evaluation value sequence G, G = (g1, g2...gr), where gi is the construction quality evaluation value of the i-th feedback time node, and r is the number of feedback time nodes in the monitoring and evaluation time axis;

[0152] Generate an overall construction evaluation value h according to the construction instruction evaluation value sequence G;

[0153]

[0154] Wherein, e3 is the preset third weight coefficient, e4 is the preset fourth weight coefficient; Q3 is the preset third fixed coefficient, Q4 is the preset fourth fixed coefficient; △ g is the average value of all data in the construction quality evaluation value series G.

[0155] Specifically, by presetting the third fixed coefficient and the fourth fixed coefficient, each parameter in the overall construction evaluation model is normalized so that they are in the same value range.

[0156] Specifically, after the overall construction is completed, an overall construction quality evaluation value is generated to achieve an overall evaluation and analysis of the pipe jacking construction process, thereby ensuring the overall quality and effect of the pipe jacking construction.

[0157] In a preferred embodiment of the present application, when setting the update parameters of the quality assessment model series, it includes:

[0158] Generate a working condition category update parameter according to a preset update time node, and generate a first-level update instruction of a quality assessment model series according to the working condition category update parameter;

[0159] Obtain the historical call parameters of each quality assessment model according to the preset update time node, and generate secondary update instructions;

[0160] Secondary update instructions include:

[0161] Select the target quality assessment model in turn;

[0162] Obtain historical call parameters of the target quality assessment model;

[0163] Generate monitoring indicator screening parameters of the target quality assessment model based on historical call parameters;

[0164] According to the screening results, the updated parameters of the target quality assessment model are generated;

[0165] Generate update parameters of each quality assessment model in the quality assessment model sequence B in sequence.

[0166] It is understandable that in the above embodiments, based on the machine learning method, the working condition categories are continuously refined, thereby improving the accurate evaluation of the construction quality at different stages of the pipe jacking construction process, and optimizing and screening the monitoring parameters of each quality assessment model, thereby improving the quality assessment model's analysis and evaluation accuracy for the pipe jacking construction quality, and timely warning of various potential construction risks. After the overall construction is completed, the overall construction quality evaluation value is generated to achieve an overall evaluation and analysis of the pipe jacking construction process, ensuring the overall quality and effect of the pipe jacking construction.

[0167] According to another preferred embodiment of a method for evaluating the quality of pipe jacking construction based on machine learning in any of the above preferred embodiments, a method for evaluating the quality of pipe jacking construction based on machine learning is provided in this preferred embodiment, comprising:

[0168] The central control unit is used to generate multiple construction categories based on historical construction parameters and establish a quality assessment model for each construction category;

[0169] A monitoring unit is used to establish a monitoring and evaluation timeline according to the pipe jacking construction plan, and generate multiple feedback time nodes according to the monitoring and evaluation timeline;

[0170] The monitoring unit includes a plurality of monitoring submodules, and the monitoring submodules are used to collect construction monitoring parameters and construction environment parameters;

[0171] An evaluation unit, used to obtain construction monitoring parameters and construction environment parameters according to preset feedback time nodes;

[0172] The evaluation unit is also used to judge the construction category according to the construction environment parameters, set the corresponding quality evaluation model as the first-level quality evaluation model according to the judgment result, and generate the construction quality evaluation value according to the construction monitoring parameters and the first-level quality evaluation model;

[0173] The central control unit includes:

[0174] The first processing module is used to traverse the historical construction parameters and generate multiple construction categories;

[0175] Establish a construction category sequence A, A = (a1, a2...an), where ai is the i-th construction category; n is the number of construction categories;

[0176] Select the target construction category in sequence according to the construction category sequence A;

[0177] Set multiple evaluation indicators for target construction categories;

[0178] Generate a training set data package based on historical construction parameters and all evaluation indicators of the target construction category;

[0179] Establish a quality assessment model for the target construction category based on the training set data package;

[0180] Establish a quality assessment model sequence B, B = (b1, b2...bn), where bi is the quality assessment model corresponding to the i-th construction category;

[0181] A first updating module, used to generate a working condition category update parameter according to a preset update time node, and generate a first-level update instruction of a quality assessment model series according to the working condition category update parameter;

[0182] The first update module is also used to obtain the historical call parameters of each quality assessment model according to the preset update time node, and generate a secondary update instruction;

[0183] Secondary update instructions include:

[0184] Select the target quality assessment model in turn;

[0185] Obtain historical call parameters of the target quality assessment model;

[0186] Generate monitoring indicator screening parameters of the target quality assessment model based on historical call parameters;

[0187] According to the screening results, the updated parameters of the target quality assessment model are generated;

[0188] Generate update parameters of each quality assessment model in the quality assessment model sequence B in sequence.

[0189] In a preferred embodiment of the present application, the evaluation unit includes:

[0190] The first judgment module is used to obtain the construction environment parameters of the current feedback time node;

[0191] Generate similarity evaluation values ​​between construction environment parameters and each construction category according to a preset similarity evaluation model;

[0192] Establish a similarity evaluation value sequence F of the current feedback time node, F = (f1, f2...fn), where fn is the similarity evaluation value of the construction environment parameter at the current feedback time node and the i-th construction category;

[0193] The quality assessment model of the construction category corresponding to the maximum value fmax in the similarity evaluation value series F is selected as the first-level quality assessment model;

[0194] The first evaluation module is used to obtain a first-level quality evaluation model of the current feedback time node;

[0195] Generate an evaluation index sequence C according to the first-level quality assessment model, C = (c1, c2...cm), where ci is the i-th evaluation index and m is the number of evaluation indicators required by the first-level quality assessment model;

[0196] Generate reference values ​​for various evaluation indicators based on the construction monitoring parameters at the current feedback time node;

[0197] Establish a reference value sequence D of the current feedback time node, D = (d1, d2 ... dm), where di is the reference value of the i-th evaluation indicator at the current feedback time node;

[0198] Generate the construction quality evaluation value g at the current feedback time node;

[0199]

[0200] Wherein, e1 is the preset first weight coefficient, e2 is the preset second weight coefficient, Q1 is the preset first fixed coefficient, Q2 is the preset second fixed coefficient, αi is the influencing factor of the i-th evaluation index; d'i is the standard reference value of the i-th evaluation index;

[0201] The early warning module is used to determine whether to generate an early warning instruction according to the construction quality evaluation value g at the current feedback time node;

[0202] A first construction quality evaluation value threshold G1 and a second construction quality evaluation value threshold G2 are preset, and G1<G2;

[0203] If g<G1, a first-level warning instruction is generated;

[0204] If G1≤g<G2, a second-level warning instruction is generated;

[0205] If g≥G2, no warning instruction is generated.

[0206] Specifically, the assessment units also include:

[0207] The second evaluation module, when the overall construction is completed, obtains the construction quality evaluation value of each feedback time node according to the monitoring and evaluation time axis;

[0208] Establish a construction quality evaluation value sequence G, G = (g1, g2...gr), where gi is the construction quality evaluation value of the i-th feedback time node, and r is the number of feedback time nodes in the monitoring and evaluation time axis;

[0209] Generate an overall construction evaluation value h according to the construction instruction evaluation value sequence G;

[0210]

[0211] Wherein, e3 is the preset third weight coefficient, e4 is the preset fourth weight coefficient; Q3 is the preset third fixed coefficient, Q4 is the preset fourth fixed coefficient; △ g is the average value of all data in the construction quality evaluation value series G.

[0212] According to the first concept of the present application, by setting multiple working condition categories and multiple feedback time nodes, the corresponding evaluation indicators are adjusted in time for different working condition categories, and the quality of pipe jacking construction is periodically and accurately evaluated and monitored in real time. In addition, it is determined whether there are construction problems based on the real-time construction quality evaluation values, and various potential construction risks are warned and rectified in time to ensure the accuracy and efficiency of the pipe jacking construction process.

[0213] According to the second concept of this application, based on the method of machine learning, the categories of working conditions are continuously refined to improve the accurate evaluation of the construction quality at different stages of the pipe jacking construction process, and the monitoring parameters of each quality assessment model are optimized and screened, thereby improving the analysis and evaluation accuracy of the quality assessment model for the pipe jacking construction quality, and timely warning of various potential construction risks. After the overall construction is completed, the overall construction quality evaluation value is generated to achieve an overall evaluation and analysis of the pipe jacking construction process, ensuring the overall quality and effect of the pipe jacking construction.

[0214] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and substitutions can be made without departing from the technical principles of the present application. These improvements and substitutions should also be regarded as the scope of protection of the present application.

Claims

1. A method for evaluating pipe jacking construction quality based on machine learning, characterized in that: include: Generate multiple construction categories based on historical construction parameters and establish quality assessment models for each construction category; Establish a monitoring and evaluation timeline according to the pipe jacking construction plan, and generate multiple feedback time nodes based on the monitoring and evaluation timeline; Obtain construction monitoring parameters and construction environment parameters according to preset feedback time nodes, determine the construction category according to the construction environment parameters, and set the corresponding quality assessment model as a first-level quality assessment model according to the determination result; The construction quality evaluation value is generated based on the construction monitoring parameters and the first-level quality assessment model.

2. The method for evaluating pipe jacking construction quality based on machine learning according to claim 1, characterized in that: When establishing quality assessment models for each construction category, include Traverse historical construction parameters and generate multiple construction categories; Establish a construction category sequence A, A = (a1, a2...an), where ai is the i-th construction category; n is the number of construction categories; Select the target construction category in sequence according to the construction category sequence A; Set multiple evaluation indicators for target construction categories; Generate a training set data package based on historical construction parameters and all evaluation indicators of the target construction category; Establish a quality assessment model for the target construction category based on the training set data package; Establish a quality assessment model sequence B, B = (b1, b2...bn), where bi is the quality assessment model corresponding to the i-th construction category; Set the update parameters of the quality assessment model series.

3. The method for evaluating pipe jacking construction quality based on machine learning as claimed in claim 2, characterized in that: When judging the construction category based on construction environment parameters, it includes: Obtain the construction environment parameters of the current feedback time node; Generate similarity evaluation values ​​between construction environment parameters and each construction category according to a preset similarity evaluation model; Establish a similarity evaluation value sequence F of the current feedback time node, F = (f1, f2...fn), where fn is the similarity evaluation value of the construction environment parameter at the current feedback time node and the i-th construction category; The quality assessment model of the construction category corresponding to the maximum value fmax in the similarity evaluation value sequence F is selected as the first-level quality assessment model.

4. The method for evaluating pipe jacking construction quality based on machine learning as claimed in claim 3, characterized in that: The generation of the construction quality evaluation value includes: Obtain the first-level quality assessment model of the current feedback time node; Generate an evaluation index sequence C according to the first-level quality assessment model, C = (c1, c2...cm), where ci is the i-th evaluation index and m is the number of evaluation indicators required by the first-level quality assessment model; Generate reference values ​​for various evaluation indicators based on the construction monitoring parameters at the current feedback time node; Establish a reference value sequence D of the current feedback time node, D = (d1, d2 ... dm), where di is the reference value of the i-th evaluation indicator at the current feedback time node; Generate the construction quality evaluation value g at the current feedback time node; Among them, e1 is the preset first weight coefficient, e2 is the preset second weight coefficient, Q1 is the preset first fixed coefficient, Q2 is the preset second fixed coefficient, αi is the influencing factor of the i-th evaluation indicator; d'i is the standard reference value of the i-th evaluation indicator.

5. The method for evaluating pipe jacking construction quality based on machine learning as claimed in claim 4, characterized in that: Also includes: Determine whether to generate an early warning instruction based on the construction quality evaluation value g at the current feedback time node; A first construction quality evaluation value threshold G1 and a second construction quality evaluation value threshold G2 are preset, and G1<G2; If g<G1, a first-level warning instruction is generated; If G1≤g<G2, a second-level warning instruction is generated; If g≥G2, no warning instruction is generated.

6. The method for evaluating pipe jacking construction quality based on machine learning as claimed in claim 5, characterized in that: Also includes; When the overall construction is completed; Obtain the construction quality evaluation value of each feedback time node according to the monitoring and evaluation timeline; Establish a construction quality evaluation value sequence G, G = (g1, g2...gr), where gi is the construction quality evaluation value of the i-th feedback time node, and r is the number of feedback time nodes in the monitoring and evaluation time axis; Generate an overall construction evaluation value h according to the construction instruction evaluation value sequence G; Wherein, e3 is the preset third weight coefficient, e4 is the preset fourth weight coefficient; Q3 is the preset third fixed coefficient, Q4 is the preset fourth fixed coefficient; △ g is the average value of all data in the construction quality evaluation value series G.

7. The method for evaluating pipe jacking construction quality based on machine learning according to claim 6, characterized in that: When setting the update parameters of the quality assessment model series, include: Generate a working condition category update parameter according to a preset update time node, and generate a first-level update instruction of a quality assessment model series according to the working condition category update parameter; Obtain the historical call parameters of each quality assessment model according to the preset update time node, and generate secondary update instructions; The secondary update instruction includes: Select the target quality assessment model in turn; Obtain historical call parameters of the target quality assessment model; Generate monitoring indicator screening parameters of the target quality assessment model based on historical call parameters; According to the screening results, the updated parameters of the target quality assessment model are generated; Generate update parameters of each quality assessment model in the quality assessment model sequence B in sequence.

8. A pipe jacking construction quality assessment system based on machine learning, applying the pipe jacking construction quality assessment method based on machine learning according to any one of claims 1 to 7, characterized in that: include; The central control unit is used to generate multiple construction categories based on historical construction parameters and establish a quality assessment model for each construction category; A monitoring unit is used to establish a monitoring and evaluation timeline according to the pipe jacking construction plan, and generate multiple feedback time nodes according to the monitoring and evaluation timeline; The monitoring unit includes a plurality of monitoring submodules, and the monitoring submodules are used to collect construction monitoring parameters and construction environment parameters; An evaluation unit, used to obtain construction monitoring parameters and construction environment parameters according to preset feedback time nodes; The evaluation unit is also used to determine the construction category according to the construction environment parameters, set the corresponding quality evaluation model as a first-level quality evaluation model according to the determination result, and generate a construction quality evaluation value according to the construction monitoring parameters and the first-level quality evaluation model; The central control unit comprises: The first processing module is used to traverse the historical construction parameters and generate multiple construction categories; Establish a construction category sequence A, A = (a1, a2...an), where ai is the i-th construction category; n is the number of construction categories; Select the target construction category in sequence according to the construction category sequence A; Set multiple evaluation indicators for target construction categories; Generate a training set data package based on historical construction parameters and all evaluation indicators of the target construction category; Establish a quality assessment model for the target construction category based on the training set data package; Establish a quality assessment model sequence B, B = (b1, b2...bn), where bi is the quality assessment model corresponding to the i-th construction category; A first updating module, used to generate a working condition category update parameter according to a preset update time node, and generate a first-level update instruction of a quality assessment model series according to the working condition category update parameter; The first updating module is also used to obtain the historical calling parameters of each quality assessment model according to the preset update time node, and generate a secondary update instruction; The secondary update instruction includes: Select the target quality assessment model in turn; Obtain historical call parameters of the target quality assessment model; Generate monitoring indicator screening parameters of the target quality assessment model based on historical call parameters; According to the screening results, the updated parameters of the target quality assessment model are generated; Generate update parameters of each quality assessment model in the quality assessment model sequence B in sequence.

9. The machine learning-based pipe jacking construction quality assessment system according to claim 8, characterized in that: The evaluation unit comprises: The first judgment module is used to obtain the construction environment parameters of the current feedback time node; Generate similarity evaluation values ​​between construction environment parameters and each construction category according to a preset similarity evaluation model; Establish a similarity evaluation value sequence F of the current feedback time node, F = (f1, f2...fn), where fn is the similarity evaluation value of the construction environment parameter at the current feedback time node and the i-th construction category; The quality assessment model of the construction category corresponding to the maximum value fmax in the similarity evaluation value series F is selected as the first-level quality assessment model; The first evaluation module is used to obtain a first-level quality evaluation model of the current feedback time node; Generate an evaluation index sequence C according to the first-level quality assessment model, C = (c1, c2...cm), where ci is the i-th evaluation index and m is the number of evaluation indicators required by the first-level quality assessment model; Generate reference values ​​for various evaluation indicators based on the construction monitoring parameters at the current feedback time node; Establish a reference value sequence D of the current feedback time node, D = (d1, d2 ... dm), where di is the reference value of the i-th evaluation indicator at the current feedback time node; Generate the construction quality evaluation value g at the current feedback time node; Wherein, e1 is the preset first weight coefficient, e2 is the preset second weight coefficient, Q1 is the preset first fixed coefficient, Q2 is the preset second fixed coefficient, αi is the influencing factor of the i-th evaluation index; d'i is the standard reference value of the i-th evaluation index; The early warning module is used to determine whether to generate an early warning instruction according to the construction quality evaluation value g at the current feedback time node; A first construction quality evaluation value threshold G1 and a second construction quality evaluation value threshold G2 are preset, and G1<G2; If g<G1, a first-level warning instruction is generated; If G1≤g<G2, a second-level warning instruction is generated; If g≥G2, no warning instruction is generated.

10. The pipe jacking construction quality assessment system based on machine learning according to claim 9, characterized in that: The evaluation unit further comprises: A second evaluation module, when the overall construction is completed, the second evaluation module obtains the construction quality evaluation value of each feedback time node according to the monitoring and evaluation time axis; Establish a construction quality evaluation value sequence G, G = (g1, g2...gr), where gi is the construction quality evaluation value of the i-th feedback time node, and r is the number of feedback time nodes in the monitoring and evaluation time axis; Generate an overall construction evaluation value h according to the construction instruction evaluation value sequence G; Wherein, e3 is the preset third weight coefficient, e4 is the preset fourth weight coefficient; Q3 is the preset third fixed coefficient, Q4 is the preset fourth fixed coefficient; △ g is the average value of all data in the construction quality evaluation value series G.

Citation Information

Cited By

  • Organic electroluminescence driving circuit fault diagnosis method based on deep learning

    CN119181318A

  • Vacuum paraffin sealing control method for high-temperature superconducting magnet

    CN119620617A

  • Welding quality intelligent monitoring method and system

    CN119747955A

  • Pipe jacking construction risk analysis method and system based on BIM technology

    CN120125042A

  • A Risk Analysis Method and System for Jacking Pipe Construction Based on BIM Technology

    CN120125042B