Big data-based construction engineering quality monitoring management method and system

Abstract

The application discloses a building engineering quality monitoring management method and system based on big data, and relates to the technical field of building engineering. The method carries out preprocessing operations including data time alignment, direction and speed data standardization, sudden disturbance data denoising correction, structure settlement data filtering and smoothing, sound spectrum data frequency domain filtering and data format unification on disturbance data in a big data monitoring management platform to obtain a standard disturbance data set, and further extracts a standard disturbance feature vector set. On this basis, a disturbance chain scoring model and a structure response scoring model are constructed, and a disturbance chain scoring result Rgchain and a structure response scoring result Rstruct are calculated. Through linkage modeling of disturbance behavior and structure response, quantitative analysis and trend judgment of disturbance risk are effectively realized, and potential flow damage channels or structure settlement abnormal trends that may be formed locally on the bottom plate can be identified in advance.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of building engineering, in particular to a building engineering quality monitoring management method and system based on big data. BACKGROUND

[0002] The field of building engineering, especially urban underground structure engineering, is an important infrastructure construction field under the rapid advancement of urbanization. In underground structure construction, deep foundation raft structure is an important component for supporting large volume load and isolating underground water disturbance, and its stability is directly related to the safe service life of the structure. Especially in complex construction scenes such as subway tunneling section, high-rise building basement and deep foundation pit, the influence of underground water dynamic conditions on the safety of raft structure is increasingly significant. Therefore, efficient monitoring and risk identification of underground water disturbance state and raft response during construction have become important research and practice problems for underground engineering quality control.

[0003] The existing underground structure quality monitoring method at the present stage usually focuses on the monitoring of underground water level, pore pressure and structure static settlement, and has not established a quantitative linkage model for dynamic factors such as "water speed disturbance-flow direction change-sudden behavior". Especially in the face of underground water mutation events such as instantaneous gushing speed surge, local flow direction deflection and frequent sudden disturbance, the traditional method can only find the result settlement, but cannot identify the potential risk before the event occurs. In addition, most of the monitoring data are processed in isolation, without the ability of joint modeling across parameters and dimensions, lacking of early warning and intervention executability, leading to lag of engineering response strategy and hidden dangers of structure damage.

[0004] With the underground structure construction entering deep complex hydrological environment, the original water stop and bearing system is facing stronger gushing disturbance challenges. The speed anomaly, flow instability or hydraulic gradient mutation of underground water will form the evolution process of bottom plate erosion and enclosure imbalance, leading to local instability, leakage and even structure perforation of raft structure. If these dynamic disturbance processes cannot be identified in time, not only the integrity of the foundation structure will be damaged, but also the chain risks such as settlement concentration and foundation bearing capacity attenuation will be triggered, thereby endangering the safe operation of the whole underground structure system. SUMMARY

[0005] In view of the deficiencies of the prior art, the present application provides a building engineering quality monitoring management method and system based on big data, which solves the problems mentioned in the background art.

[0006] To achieve the above purpose, the present application is realized by the following technical scheme:

[0007] S1, arranging collection points in the underground structure construction area and setting a sensor group to collect disturbance data of the underground structure construction area in real time, and then transmitting the disturbance data to a big data monitoring management platform;

[0008] S2. In the big data monitoring and management platform, the disturbance data is preprocessed to obtain a standard disturbance dataset, and features are extracted from the standard disturbance dataset to obtain a standard disturbance feature vector set.

[0009] S3. Construct a perturbation chain scoring model based on the standard perturbation feature vector set, calculate the perturbation chain scoring result Rgchain, and preset the perturbation threshold Rth to conduct a preliminary comparative evaluation of the perturbation chain scoring result Rgchain.

[0010] S4. Based on the preliminary comparative evaluation results, a repair strategy is triggered. The repair strategy is to extract the settlement vector set and calculate the structural response score Rstruct by combining it with the disturbance chain score result Rgchain.

[0011] S5. Substitute the disturbance chain score result Rgchain and the structural response score result Rstruct into the comprehensive score function model to calculate the comprehensive quality evolution score result Rsq. Perform a second comparison and evaluation between the comprehensive quality evolution score result Rsq and the preset response interval threshold, and execute the corresponding control strategy.

[0012] Preferably, S1 includes S11 and S12;

[0013] S11. The collection points include collection points P1, P2, P3, and P4. Corresponding sensors are configured within each collection point to collect disturbance data of the underground structure construction area in real time. This disturbance data includes instantaneous groundwater velocity, groundwater flow direction, hydraulic gradient jump frequency, abrupt changes in hydraulic head amplitude, raft settlement displacement, settlement rate, dominant frequency offset, and the number of abnormal frequency spectrum changes. Wherein:

[0014] The collection point P1 is located in the pore water distribution area at a depth of 20 meters below the raft structure. An electromagnetic groundwater velocity sensor and a flow direction determination module are installed at the collection point P1 to collect raw data on the instantaneous flow velocity and flow direction of the groundwater.

[0015] The sampling point P2 is located in the soil layer area at the junction of the raft edge and the water-stop curtain structure. A pore water pressure jump frequency detection sensor is installed at the sampling point P2 and embedded in the soil layer of the retaining edge in a radial array. It is used to collect the frequency of hydraulic gradient jump and the amplitude of water head change jump per unit time.

[0016] The acquisition point P3 is set at the central axis and symmetrical corner of the raft structure. A fiber optic grating settlement gauge is installed at acquisition point P3 to collect the raw data of the settlement displacement and settlement change rate of the raft structure.

[0017] The collection point P4 is connected to the interface area through a curtain structure arranged 3 meters below the raft, and a passive acoustic anomaly identification module with a sound wave reflection sensor and integrated FFT analysis function is installed at the collection point P4 to collect the main frequency offset value of the reflected sound wave energy spectrum of the reflected sound wave of the underground structure interface area and the number of spectrum anomaly mutations per unit time;

[0018] S12, through the data transmission module built-in all collection points, using Ethernet, the data transmission module is wirelessly connected with the big data monitoring management platform, and the disturbance data is uploaded to the big data monitoring management platform.

[0019] Preferably, the S2 comprises S21;

[0020] S21, pre-processing the disturbance data in the big data monitoring management platform to obtain a standard disturbance data set, the pre-processing including data time alignment processing, direction class and speed class data standardization, sudden class disturbance data denoising correction, structure settlement data filtering and smoothing, acoustic spectrum class data frequency domain filtering and data format unification;

[0021] The data time alignment processing aligns the disturbance data collection time uploaded by different collection points through a multi-source data unified timestamp mapping algorithm to construct a disturbance data set with consistent time sequence;

[0022] The direction class and speed class data standardization performs unit dimension normalization and coordinate system unification on the underground water instantaneous flow rate and underground water flow direction through a normalization function and direction angle conversion algorithm to eliminate the influence of flow field spatial direction heterogeneity;

[0023] The sudden class disturbance data denoising correction performs curve smoothing processing on the raft settlement displacement and settlement rate data through a cubic spline fitting and high-pass filtering method to suppress local jittering errors caused by construction disturbance and measurement point drift;

[0024] The acoustic spectrum class data frequency domain filtering filters out high-frequency interference signals in the reflected sound wave main frequency offset value and spectrum anomaly mutation number through a fast Fourier transform FFT and band-pass filtering technology, and retains the effective frequency band change trend;

[0025] The data format unification unifies the original values of the disturbance data with different dimensions, units and intervals to a standard interval with a mean value of 0 and a standard deviation of 1 through a Z-score standardization method using numerical range standardization, eliminates the dimension influence in all disturbance data, and then outputs as a standard disturbance data set using a structured data packaging rule.

[0026] Preferably, the S2 further comprises S22;

[0027] S22, based on the pre-processed standard perturbation data set, extracting for feature extraction, obtaining a standard perturbation feature vector set;

[0028] The standard perturbation feature vector set includes a perturbation vector set and a subsidence vector set;

[0029] The perturbation vector set includes groundwater velocity change rate ΔVgw, flow direction spin angle Odiv, abnormal jump frequency Rhop and surge break jump peak value;

[0030] The subsidence vector set includes raft instantaneous subsidence displacement ΔZsup and reflection perturbation energy spectrum index Necho;

[0031] The abnormal jump frequency Rhop is extracted from the hydraulic gradient jump frequency and the water head change mutation amplitude in unit time by the statistical window function and the jump threshold analysis algorithm, the abnormal jump frequency Rhop represents the number of mutation events in unit time, and the surge break jump peak value represents the maximum water head disturbance amplitude;

[0032] The groundwater velocity change rate ΔVgw and the flow direction spin angle Odiv are obtained by applying the first-order difference of adjacent time to the sequence of groundwater instantaneous flow velocity to obtain the groundwater velocity change rate ΔVgw, and using the three-dimensional vector angle calculation formula to obtain the current flow direction spin angle Odiv based on the designed direction as the reference direction.

[0033] The reflection perturbation energy spectrum index Necho is extracted from the frequency offset value of the reflected sound wave and the number of abnormal spectrum mutations in unit time by the frequency offset sliding window analysis method and the power spectrum mutation detection, which is used to describe the comprehensive intensity of the sound spectrum disturbance;

[0034] The raft instantaneous subsidence displacement ΔZsup is extracted from the raft subsidence displacement and the subsidence rate by the subsidence displacement derivative calculation function.

[0035] Preferably, the S3 includes S31;

[0036] S31, based on the non-linear combination function, constructing a perturbation chain scoring model, extracting the perturbation vector set in the standard perturbation feature vector set as an input item input into the perturbation chain scoring model, and calculating the perturbation chain scoring result Rgchain;

[0037] The perturbation chain scoring result Rgchain is calculated by the following perturbation chain scoring model;

[0038] ;

[0039] In the formula, log represents the logarithmic function, and tan represents the tangent function.

[0040] Preferably, the S3 further comprises S32;

[0041] S32, based on the historical sample disturbance chain score result Rgchain, extract the disturbance chain score result Rgchain sample set of the critical structure abnormal response, perform mean processing to obtain the disturbance threshold Rth, and then preliminarily compare and evaluate the real-time obtained disturbance chain score result Rgchain with the disturbance threshold Rth to judge the current groundwater disturbance behavior; the specific evaluation content is as follows:

[0042] When the disturbance chain score result Rgchain is greater than the disturbance threshold Rth, it indicates that the groundwater disturbance chain is abnormal, and at this time the repair strategy is triggered;

[0043] When the disturbance chain score result Rgchain is less than or equal to the disturbance threshold Rth, it indicates that the groundwater disturbance behavior is normal, and at this time no intervention is needed to continue monitoring.

[0044] Preferably, the S4 further comprises S41;

[0045] S41, after triggering the repair strategy in the preliminary comparison and evaluation, the settlement vector set is combined with the disturbance chain score result Rgchain for combined calculation to output the structure response score Rstruct;

[0046] The structure response score Rstruct is calculated and output by the following algorithm formula:

[0047] ;

[0048] In the formula, represents the partial derivative, represents the partial derivative with respect to the time variable t, and exp represents the exponential function.

[0049] Preferably, the S5 comprises S51;

[0050] S51, by constructing a comprehensive score function model, the real-time obtained disturbance chain score result Rgchain and the structure response score result Rstruct are extracted, and the reflected disturbance energy spectrum index Necho is combined to input into the comprehensive score function model for calculation to output the comprehensive quality evolution score result Rsq, and the linkage influence strength of the underground disturbance behavior on the stability of the underground structure construction area is quantitatively analyzed;

[0051] The comprehensive quality evolution score result Rsq is calculated and output by the following constructed comprehensive score function model:

[0052] .

[0053] Preferably, the S5 further comprises S52;

[0054] S52, based on the historical underground engineering comprehensive quality evolution score result Rsq, corresponding normal working condition, local repair event and abnormal event, and based on the upper limit value of the normal working condition and the upper limit value of the local repair event, the response interval threshold is set; the response interval threshold includes a first response threshold F1 and a second response threshold F2;

[0055] The response interval threshold is compared with the comprehensive quality evolution score result Rsq for secondary comparison and evaluation, the stability of the underground construction area structure under the disturbance of underground water is judged, and corresponding control strategy is executed based on the secondary comparison and evaluation result; the specific evaluation content is as follows;

[0056] When the comprehensive quality evolution score result Rsq is less than or equal to the first response threshold F1, it means that the disturbance of the underground construction area is normal and does not affect the structure, and no intervention is needed to continue monitoring;

[0057] When the first response threshold F1 is less than the comprehensive quality evolution score result Rsq and the comprehensive quality evolution score result Rsq is less than or equal to the second response threshold F2, it means that the disturbance of the underground construction area is abnormal, and at this time the regional repair strategy is executed;

[0058] When the comprehensive quality evolution score result Rsq is greater than the second response threshold F2, it means that the disturbance of the underground construction area has a risk of damage, and at this time the overall intervention is executed;

[0059] The regional repair strategy optimizes the mechanical construction load by prompting grouting reinforcement, negative pressure drainage control and postponing the current section construction in the underground construction section area;

[0060] The overall intervention stops personnel and machinery from entering the current underground construction section area by issuing a closure instruction, and prompts structure emergency reinforcement, replacement of curtain water stop layer and trial vertical barrier.

[0061] The building engineering quality monitoring management system based on big data includes a disturbance data acquisition module, a disturbance data processing module, a disturbance analysis module, a structure response analysis module and a joint analysis module;

[0062] The disturbance data acquisition module acquires disturbance data of the underground structure construction area by arranging acquisition points and setting sensor groups in the underground structure construction area, and then transmits the disturbance data to the big data monitoring management platform;

[0063] The disturbance data processing module pre-processes the disturbance data in the big data monitoring management platform, obtains a standard disturbance data set, and extracts features from the standard disturbance data set to obtain a standard disturbance feature vector set;

[0064] The disturbance analysis module obtains a disturbance chain score result Rgchain by constructing a disturbance chain score model based on a standard disturbance feature vector set, and pre-sets a disturbance threshold Rth to preliminarily compare and evaluate the disturbance chain score result Rgchain.

[0065] The structure response analysis module triggers a repair strategy based on the preliminary comparison and evaluation result, the repair strategy is obtained by extracting a settlement vector set and combining the disturbance chain score result Rgchain for calculation to obtain a structure response score Rstruct.

[0066] The joint analysis module obtains a comprehensive quality evolution score result Rsq by substituting the disturbance chain score result Rgchain and the structure response score result Rstruct into a comprehensive score function model, and performs secondary comparison and evaluation on the comprehensive quality evolution score result Rsq and a preset response interval threshold, and executes a corresponding control strategy.

[0067] The application provides a building engineering quality monitoring management method and system based on big data.

[0068] (1) The method realizes real-time collection of disturbance data such as underground water instantaneous flow rate, underground water flow direction, hydraulic gradient jump frequency, water head change mutation amplitude, raft settlement displacement, settlement rate, main frequency offset value and frequency spectrum abnormal mutation times by using the collection points P1 to P4 arranged in the underground structure construction area, combining the sensor group composed of an electromagnetic underground water flow rate sensor, a flow direction determination module, a pore water pressure jump frequency detection sensor, a fiber bragg grating settlement meter and an integrated FFT analysis function passive acoustic anomaly recognition module.

[0069] (2) The method obtains a standard disturbance data set by performing preprocessing operations including data time alignment, direction class and speed class data standardization, sudden jump class disturbance data denoising correction, structure settlement data filtering and smoothing, sound spectrum class data frequency domain filtering and data format unification on the disturbance data in the big data monitoring management platform, and further extracts a standard disturbance feature vector set. On this basis, a disturbance chain scoring model and a structure response scoring model are constructed respectively, and the disturbance chain scoring result Rgchain and the structure response scoring result Rstruct are calculated. By linking the disturbance behavior and the structure reaction, the quantitative analysis and trend judgment of the disturbance risk are effectively realized, the potential flow damage channel or structure settlement abnormal trend that may be formed locally on the bottom plate can be identified in advance, and the problems of lagging disturbance identification and single reaction mechanism in the prior art are solved.

[0070] (3) The method inputs the disturbance chain scoring result Rgchain and the structure response scoring result Rstruct into the comprehensive scoring function model constructed, combines the reflected disturbance energy spectrum index Necho, outputs the comprehensive quality evolution scoring result Rsq, and sets the first response threshold F1 and the second response threshold F2 for secondary comparison and evaluation. The above method realizes the complete control closed loop from early warning, judgment to disposal by coupling the scoring of the disturbance chain and the structure response and the response strategy linkage triggering, which can greatly improve the response efficiency and emergency disposal capacity of the disturbance event in the underground structure construction, and ensures the construction safety controllable, the intervention operation clear and the treatment measures timely. BRIEF DESCRIPTION OF DRAWINGS

[0071] Figure 1 The figure is a step schematic diagram of the building engineering quality monitoring management method based on big data of the present application.

[0072] Figure 2 The figure is a structure schematic diagram of the building engineering quality monitoring management system based on big data of the present application.

[0073] Figure 3 The figure is a disturbance data acquisition point layout schematic diagram. DETAILED DESCRIPTION

[0074] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0075] Embodiment 1, please refer to Figure 1 and Figure 3The application provides a building engineering quality monitoring management method based on big data, and the above object is achieved by the following technical scheme: comprising the following steps:

[0076] S1, collecting points are arranged in the underground structure construction area, and a sensor group is arranged to collect disturbance data of the underground structure construction area in real time, and then the disturbance data is transmitted to a big data monitoring management platform;

[0077] S2, the disturbance data is preprocessed in the big data monitoring management platform, a standard disturbance data set is obtained, and a standard disturbance feature vector set is obtained by feature extraction on the standard disturbance data set;

[0078] S3, a disturbance chain scoring model is constructed based on the standard disturbance feature vector set, a disturbance chain scoring result Rgchain is obtained by calculation, and a disturbance threshold Rth is preset, and the disturbance chain scoring result Rgchain is preliminarily compared and evaluated;

[0079] S4, a repair strategy is triggered based on the preliminary comparison and evaluation result, the repair strategy is calculated by extracting a settlement vector set and combining the disturbance chain scoring result Rgchain, and a structure response score Rstruct is obtained;

[0080] S5, the disturbance chain scoring result Rgchain and the structure response score result Rstruct are substituted into a comprehensive score function model, a comprehensive quality evolution score result Rsq is obtained by calculation, a secondary comparison and evaluation is performed between the preset response interval threshold and the comprehensive quality evolution score result Rsq, and a corresponding control strategy is executed.

[0081] In this embodiment, the method is realized by laying collection points P1 to P4 in the underground structure construction area, embedding special equipment such as sensors in different structure parts, collecting disturbance data in real time, and transmitting the data to the big data monitoring management platform through Ethernet wireless transmission. The big data monitoring management platform constructs a standard disturbance data set through preprocessing technology, and further extracts a disturbance vector set and a settlement vector set as a standard disturbance feature vector set. Based on the standard disturbance feature vector set, a disturbance chain scoring model is constructed, and a disturbance chain scoring result Rgchain is output. At the same time, a disturbance threshold Rth is set by statistically analyzing the mean value of 500 historical abnormal samples, and the real-time disturbance level is preliminarily evaluated. When the disturbance chain scoring result Rgchain is higher than the disturbance threshold Rth, the repair strategy module is automatically triggered, the raft instantaneous settlement displacement ΔZsup in the settlement vector set and the reflection disturbance energy spectrum index Necho are extracted, and the structure response score Rstruct is calculated and output in combination with the disturbance scoring result Rgchain, reflecting the response strength of the current structure. Further, the disturbance chain scoring result Rgchain and the structure response scoring result Rstruct are input into the comprehensive scoring function model through a nonlinear function to obtain a comprehensive quality evolution scoring result Rsq. The system further sets a first response threshold F1 and a second response threshold F2 based on historical working condition experience for secondary comparative evaluation, so as to automatically judge whether the regional repair strategy needs to be executed or the overall intervention state needs to be entered. Through the implementation of the above method, the linkage modeling and risk quantification between the disturbance behavior of underground water and the response of underground structure are effectively realized. The potential risks can be identified through disturbance scoring in the early stage of construction, and the settlement evolution trend can be captured in real time in the structure response stage. The comprehensive scoring mechanism makes the intervention strategy have the technical advantages of quantifiability, traceability and gradability, significantly improves the real-time performance, accuracy and response efficiency of underground structure health monitoring, avoids the dependence of traditional monitoring methods on single water level change, and thus realizes the pre-intervention and dynamic regulation of problems such as erosion, gushing and structure cracking that may occur during raft construction, and enhances the safety protection capability and construction intelligent level of building engineering in complex geological environment.

[0082] Embodiment 2, please refer to Figure 1 and Figure 3 Specifically, S1 includes S11 and S12.

[0083] S11, the collection point includes collection point P1, collection point P2, collection point P3 and collection point P4, and corresponding sensors are configured in the collection point to collect disturbance data of the underground structure construction area in real time. The disturbance data includes underground water instantaneous flow rate, underground water flow direction, water head change mutation amplitude, raft settlement displacement, settlement rate, main frequency offset value and frequency spectrum abnormal mutation times.

[0084] The collection point P1 penetrates the bearing soil layer and reaches the stable groundwater mainstream layer through the pore water distribution area set at a depth of 20 meters below the raft structure, and an electromagnetic groundwater flow rate sensor and a flow direction determination module are installed at the collection point P1 and vertically layered in multiple sampling wells to collect the instantaneous flow rate and the original data of the groundwater flow direction;

[0085] The collection point P2 is set in the soil layer region at the junction of the raft edge and the waterproof curtain structure, and a pore water pressure jump frequency detection sensor is installed at the collection point P2 and embedded in the surrounding edge soil layer in a radial dot matrix manner to collect the hydraulic gradient mutation frequency and the water head change mutation amplitude per unit time;

[0086] The collection point P3 is set at the central axis position and the symmetric corner position of the raft structure, and a fiber Bragg grating settlement meter is installed at the collection point P3 to collect the original data of the raft structure settlement displacement and the settlement change rate;

[0087] The collection point P4 is set at the curtain structure connection junction area 3 meters below the raft, and a sound wave reflection sensor and a passive acoustic anomaly identification module integrated with FFT analysis function are installed at the collection point P4 to collect the reflection sound wave energy spectrum main frequency offset value and the frequency spectrum anomaly mutation times per unit time of the reflected sound wave in the underground structure interface area;

[0088] S12, through the data transmission module built-in all sensors in the collection point, using Ethernet to wirelessly connect the data transmission module with the big data monitoring management platform, and upload the disturbance data to the big data monitoring management platform.

[0089] In this embodiment, the method effectively collects key disturbance parameters such as groundwater flow rate, hydraulic jump, structure settlement and sound spectrum anomaly by arranging collection points P1 to P4 at different key positions of the underground structure and configuring corresponding types of high-performance sensor modules according to the respective arrangement positions. Among them, the collection point P1 realizes multi-layer sampling of the pore water flow rate and flow direction under the raft structure; the collection point P2 is arranged at the junction area of the waterproof curtain, and is used to capture high-frequency hydraulic gradient changes and water head fluctuations; the collection point P3 is arranged at the axisymmetric position of the raft structure, and is used to accurately quantify the settlement trend; the collection point P4 is positioned at the junction area of the curtain under the raft, and obtains sound spectrum energy change information through sound wave reflection to reveal the disturbance situation of the hidden interface of the underground structure. Through the Ethernet wireless communication mechanism described in S12, it is ensured that the sensors of all collection points can transmit disturbance data to the big data monitoring management platform in real time, and a stable and efficient underground structure disturbance monitoring network is constructed. The technical effect of this embodiment is not only to significantly improve the collection density and spatial distribution integrity of disturbance data in the underground construction area, but also to realize parallel perception of coupled disturbance information such as multiple physical quantities, flow rate, water pressure, settlement, sound wave, etc. At the same time, the collection points are arranged scientifically, and the collection dimensions are complementary, effectively avoiding the problem of missing key risk signs due to single monitoring factor.

[0090] Embodiment 3, please refer to Figure 1 , in particular: S2 comprises S21;

[0091] S21, pre-processes the disturbance data in the big data monitoring management platform to obtain a standard disturbance data set, the pre-processing includes data time alignment processing, direction class and speed class data standardization, jump class disturbance data denoising correction, structure settlement data filtering and smoothing, sound spectrum class data frequency domain filtering and data format unification;

[0092] The data time alignment processing aligns the disturbance data collection time uploaded by different collection points through a multi-source data unified timestamp mapping algorithm, and constructs a disturbance data set with consistent time sequence;

[0093] The direction class and speed class data standardization performs unit dimension normalization and coordinate system unification on the instantaneous flow rate of groundwater and the flow direction of groundwater through a normalization function and a direction angle conversion algorithm, and eliminates the influence of spatial direction heterogeneity of the flow field;

[0094] The jump class disturbance data denoising correction performs curve smoothing processing on the raft settlement displacement and settlement rate data through a cubic spline fitting and high-pass filtering method, and suppresses local jittering errors caused by construction disturbance and measurement point drift;

[0095] The frequency domain filtering of the acoustic spectrum data filters out high-frequency interference signals in the main frequency offset value and the number of abnormal spectrum mutations of the reflected sound wave by using the fast Fourier transform (FFT) and the band-pass filtering technology, and retains the variation trend of the effective frequency band.

[0096] The data format is unified by using the Z-score standardization method of numerical range standardization to unify the original values of different dimensions, units and intervals of the disturbance data to the standard interval with a mean of 0 and a standard deviation of 1, eliminating the dimensional influence of all disturbance data, and then using the structured data packaging rule to output as a standard disturbance data set.

[0097] S2 also includes S22;

[0098] S22, based on the pre-processed standard disturbance data set, extracts the feature extraction to obtain a standard disturbance feature vector set;

[0099] The standard disturbance feature vector set includes a disturbance vector set and a subsidence vector set;

[0100] The disturbance vector set includes a groundwater velocity change rate ΔVgw, a flow direction spin angle Odiv, an abnormal jump frequency Rhop and a surge break jump peak value;

[0101] The subsidence vector set includes a raft instantaneous subsidence displacement ΔZsup and a reflected disturbance energy spectrum index Necho;

[0102] The abnormal jump frequency Rhop is extracted from the hydraulic gradient jump frequency and the water head change mutation amplitude in unit time by using the statistical window function and the jump threshold analysis algorithm, and the abnormal jump frequency Rhop and the surge break jump peak value. The abnormal jump frequency Rhop represents the number of mutation events in unit time, and the surge break jump peak value represents the maximum water head disturbance amplitude;

[0103] The groundwater velocity change rate ΔVgw and the flow direction spin angle Odiv are obtained by applying the first-order difference of adjacent time to the sequence of the instantaneous flow rate of groundwater, and the flow direction spin angle Odiv is obtained by using the three-dimensional vector angle calculation formula based on the designed direction as the reference direction.

[0104] The reflected disturbance energy spectrum index Necho is extracted from the main frequency offset value and the number of abnormal spectrum mutations in unit time by using the frequency offset sliding window analysis method and the power spectrum mutation detection, which is used to describe the comprehensive intensity of the acoustic spectrum disturbance;

[0105] The raft instantaneous subsidence displacement ΔZsup is extracted from the raft subsidence displacement and the subsidence rate by using the subsidence displacement derivative calculation function.

[0106] In this embodiment, the method is aimed at a large amount of disturbance original data obtained from the acquisition point, and in the big data monitoring management platform, the time stamp alignment, physical property normalization, disturbance type data denoising, settlement data smoothing, sound spectrum data filtering and format standardization and other preprocessing operations are sequentially completed, and finally a standard disturbance data set with consistent time scale, unified unit dimension and removed interference signal is formed. Then, in S22, the disturbance vector set and the settlement vector set are extracted based on the standard disturbance data set, including the underground water velocity change rate ΔVgw, the flow direction spin angle Odiv, the abnormal jump frequency Rhop, the surge break jump peak value, the raft instantaneous settlement displacement ΔZsup and the reflection disturbance energy spectrum index Necho, so as to construct a standard disturbance feature vector set for subsequent model calling. The core purpose of this embodiment is to ensure that the input data of the subsequent disturbance scoring model has high accuracy, high representativeness and material interpretability through fine data preprocessing and multi-dimensional disturbance feature extraction, and fundamentally improves the recognition ability of the system to the evolution mechanism of underground disturbance. The beneficial effects are: on the one hand, the time alignment and format unification eliminate the heterogeneity problem of multi-source data, providing a high consistency data basis for disturbance modeling; on the other hand, with the help of targeted feature extraction algorithm, the core change signals of underground disturbance in speed mutation, flow direction offset, water pressure jump, structure settlement and sound spectrum anomaly can be accurately captured, effectively enhancing the sensitivity and discriminability of subsequent scoring and risk identification. Therefore, this step provides high robustness and high credibility data support for building engineering quality monitoring, further improving the intelligent level and effective response ability of the monitoring system.

[0107] Embodiment 4, please refer to Figure 1 , in particular: S3 includes S31;

[0108] S31, based on a nonlinear combination function, a disturbance chain scoring model is constructed, the disturbance vector set in the standard disturbance feature vector set is input into the disturbance chain scoring model as an input item, and the disturbance chain scoring result Rgchain is calculated and output;

[0109] The disturbance chain scoring result Rgchain is calculated and output by the following disturbance chain scoring model;

[0110] ;

[0111] In the formula, log represents the logarithmic function, and tan represents the tangent function;

[0112] The formula derivation is based on the combination quantization method of the aggregation behavior of discrete events in statistical physics; is a conventional structure for logarithmic compression processing of order of magnitude aggregation in information theory, The disturbance intensity is the combined energy of the rate of change of velocity and the frequency of sudden changes; is a simplified model commonly used in geotechnical engineering to measure the risk factor of the slope of the angle offset;

[0113] The derivation process is as follows: the square of the rate of change of groundwater velocity ΔVgw represents the disturbance energy; the square of the abnormal sudden change frequency Rhop is used to quantify the number of water surges; the sum of the two is used to construct the disturbance intensity term; the logarithm is used to compress and avoid data amplification; the angle term is introduced as a flow direction offset risk enhancement factor; the final combined disturbance chain score result Rgchain is obtained, which is dimensionless;

[0114] Dimension consistency analysis: (ΔVgw) 2 : (m / s²)² = m² / s 4 ; Rhop 2 : (1 / h)² = 1 / h²; : dimensionless; : the logarithmic function acts on the dimensionless value; the overall result is compressed by the logarithm and adjusted by the dimensionless; the result is dimensionless and used for scoring;

[0115] The actual physical meaning of the formula: the disturbance chain score result Rgchain is a scoring factor used to quantify the comprehensive intensity of the disturbance of the underground flow field, which comprehensively reflects the three key factors of groundwater flow rate change, sudden event density and flow direction offset; the larger the value, the stronger the "disturbance coupling risk" of the current underground structure; this score can identify the foundation disturbance and support instability risk caused by raft bottom water gushing, water head mutation and abnormal deviation of water flow direction in advance; in the subway excavation section or deep foundation pit construction, if the disturbance chain score result Rgchain rapidly increases in a short time, it is likely to indicate that a "potential flow failure channel" has been formed in the bottom plate; therefore, the disturbance chain score result Rgchain is the first physical quantity used for disturbance warning in the entire system, which is used as the starting condition for structure evaluation and control strategy.

[0116] S3 also includes S32;

[0117] S32, based on the disturbance chain score result Rgchain of the historical sample, extracts the disturbance chain score result Rgchain sample set of the critical structure abnormal response, processes the mean value to obtain the disturbance threshold Rth, and then preliminarily compares and evaluates the disturbance chain score result Rgchain obtained in real time with the disturbance threshold Rth to judge the current groundwater disturbance behavior; the specific evaluation content is as follows:

[0118] When the disturbance chain score result Rgchain is greater than the disturbance threshold Rth, it indicates that the groundwater disturbance chain is abnormal, and the repair strategy is triggered at this time;

[0119] ​When the disturbance chain score result Rgchain is less than the disturbance threshold Rth, it indicates that the groundwater disturbance behavior is normal, and no intervention is needed to continue monitoring.

[0120] In this embodiment, the method is based on the disturbance vector set in the extracted standard disturbance feature vector set to construct a disturbance chain score model, takes the groundwater velocity change rate ΔVgw, the abnormal jump frequency Rhop and the flow direction spin angle Odiv as input variables, uses a nonlinear combination function to calculate, and outputs the disturbance chain score result Rgchain. The model realizes the scientific characterization of the comprehensive intensity of the disturbance by fusing the three key dimensions of disturbance intensity, event frequency and direction deviation, and has good dimensional consistency and discrimination sensitivity through the adjustment of the logarithmic function and the tangent function. In S32, in order to realize the dynamic identification and classification of the disturbance state, the typical abnormal response characteristic values are extracted based on the Rgchain data in the 500 groups of historical samples, and the disturbance threshold Rth is set through statistical mean calculation. By comparing the real-time obtained disturbance chain score result Rgchain with the disturbance threshold Rth, the preliminary judgment of the current groundwater disturbance behavior is realized: when Rgchain exceeds Rth, it is determined as an abnormal disturbance chain, and the subsequent structure response modeling and repair strategy is triggered immediately; if Rgchain is not higher than Rth, it is considered that the disturbance is in an acceptable range, and the system remains in the monitoring state without intervention. The purpose of this implementation is to build a set of quantifiable, assessable and triggerable underground disturbance judgment mechanism, which has a high intelligent dynamic early warning capability. Its beneficial effects include: first, the coupling strength of multi-source disturbance features is quantified through the scoring model, avoiding the subjective bias of human experience judgment; second, the threshold is extracted by using historical big data, which improves the objectivity and accuracy of disturbance identification; third, the trigger mechanism of disturbance and structure strategy linkage control is realized, which significantly enhances the response efficiency and safety protection ability of the underground engineering quality monitoring system.

[0121] Embodiment 5, please refer to Figure 1 , in particular: S4 further includes S41;

[0122] S41, after triggering the repair strategy through preliminary comparison and evaluation, the settlement vector set and the disturbance chain score result Rgchain are combined and calculated to output the structure response score Rstruct;

[0123] The structure response score Rstruct is calculated and output by the following algorithm formula;

[0124] ;

[0125] In the formula, represents the partial derivative, represents the partial derivative with respect to the time variable t, and exp represents the exponential function;

[0126] Formula derivation basis: This formula is derived from the structural response modeling formula in mechanics and the cumulative settlement estimation method in earthquake engineering, is the first-order linear term of the underground structure response, is commonly used for nonlinear excitation growth simulation, i.e., seismic intensity and fatigue damage, is the energy spectrum abnormal fluctuation compensation term in geoacoustic modeling;

[0127] Formula derivation process: The raft instantaneous settlement displacement ΔZsup represents the current settlement value; represents the settlement change rate; their sum represents the current settlement trend of the structure; squaring represents its energy effect on the result; combined with the disturbance score item (Rgchain) 0.5 , the response is stimulated by an exponential function; the reflection disturbance energy spectrum index Necho is added to prevent high-frequency misjudgment; finally, the structure response score Rstruct is obtained: a dimensionless score used to represent the structure response trend;

[0128] Dimension consistency analysis: After data format unification processing of the disturbance data in S2, the instantaneous settlement displacement ΔZsup and the reflection disturbance energy spectrum index Necho are extracted from the processed disturbance data; therefore, the instantaneous settlement displacement ΔZsup and the reflection disturbance energy spectrum index Necho are dimensionless parameters, and are also dimensionless; : the exponential is a dimensionless excitation coefficient; the overall result is a dimensionless score item used for subsequent risk level determination;

[0129] Physical meaning of the formula: The structure response score Rstruct is a quantitative index of the current settlement state and disturbed response of the raft structure; among them: the sum of the two and the square reflect the "structure deformation trend intensity", i.e., whether the structure has accumulated subsidence; the exponential term introduces the influence of disturbance chain, so that the amplification sensitivity to structural changes is enhanced when the disturbance level is high; the reflection disturbance energy spectrum index Necho is a compensation term for invisible interface changes, such as cavitation of the curtain-raft contact surface and structure cracks; the larger the structure response score Rstruct value, the more the structure is responding to abnormal disturbance and the worse the settlement trend, which is a prediction of the structure's health status; if the structure response score Rstruct continuously rises over a period of time, it means that the disturbance has begun to conduct to the structure level, and there is an evolution trend of collapse, erosion, and voiding.

[0130] In this embodiment, after the disturbance chain score result Rgchain exceeds the disturbance threshold Rth and triggers the repair strategy, the system further extracts the key parameters in the settlement vector set, i.e. the raft instantaneous settlement displacement ΔZsup and the reflection disturbance energy spectrum index Necho, combines the disturbance chain score result Rgchain, constructs a structural response score formula model, and calculates the structural response score result Rstruct. The model takes the settlement trend as the basic variable, wherein the sum of ΔZsup and its time derivative reflects the rate and trend of structural settlement, and is processed by squaring to enhance the sensitivity to structural cumulative deformation; further, an exponential incentive term of the disturbance chain score result Rgchain is introduced to describe the amplification effect of the disturbance degree on the structural response; and the reflection disturbance energy spectrum index Necho is further fused as a compensation term of spectral anomaly, which effectively improves the identification ability of the model to invisible interface damage.

[0131] The main purpose of this step is to establish a multi-dimensional coupling driven structural response evaluation mechanism to realize quantitative prediction of the evolution trend of disturbance behavior to structural abnormality. The beneficial effects are: on the one hand, by fusing the real-time deformation data of the structure and the disturbance intensity parameters, the dynamic perception ability of the structural health state is improved; on the other hand, by using the exponential function modeling method to enhance the response sensitivity of the model to high-level disturbance, the evolution risks such as erosion and collapse can be effectively identified in advance; at the same time, the introduction of the energy spectrum index makes up for the hidden risk points that cannot be detected by traditional settlement monitoring, and realizes the quantitative judgment of the deep risks of the contact interface of the raft and the curtain and other key components.

[0132] Embodiment 6, please refer to Figure 1 , specifically: S5 includes S51;

[0133] S51, by constructing a comprehensive score function model, extracting the disturbance chain score result Rgchain and the structural response score result Rstruct obtained in real time, and combining the reflection disturbance energy spectrum index Necho, inputting into the comprehensive score function model, calculating and outputting the comprehensive quality evolution score result Rsq, and quantitatively analyzing the linkage influence intensity of the underground disturbance behavior on the stability of the underground structure construction area;

[0134] The comprehensive quality evolution score result Rsq is calculated and output by the following comprehensive score function model;

[0135] ;

[0136] The formula derivation is based on the nonlinear signal fusion algorithm and the risk redundancy suppression function; the numerator is the main effect term: disturbance x response; and the denominator is the stability term: to prevent false judgment and false excitation risk of sound waves;

[0137] Formula derivation process: multiply the disturbance chain score result Rgchain and the structure response score result Rstruct to form the total action score; use Inhibit the misexcitation caused by spectrum anomaly; the result is the comprehensive quality evolution score result Rsq, which is used to judge whether to perform risk response operation;

[0138] Dimension consistency analysis: the disturbance chain score result Rgchain, the structure response score result Rstruct and the reflected disturbance energy spectrum index Necho are all dimensionless, so the output result of the comprehensive quality evolution score result Rsq is dimensionless.

[0139] S5 also includes S52;

[0140] S52, based on the comprehensive quality evolution score result Rsq of the historical underground engineering, corresponding normal working condition, local repair event and abnormal event, and based on the upper limit value of the normal working condition and the upper limit value of the local repair event, sets the response interval threshold; the response interval threshold includes the first response threshold F1 and the second response threshold F2;

[0141] Compare the response interval threshold with the comprehensive quality evolution score result Rsq twice to judge the stability of the underground construction area structure under the disturbance of underground water, and perform the corresponding control strategy based on the twice comparison evaluation result; the specific evaluation content is as follows;

[0142] When the comprehensive quality evolution score result Rsq is less than or equal to the first response threshold F1, it means that the disturbance of the underground construction area is normal and does not affect the structure, so there is no need to intervene and continue to monitor;

[0143] When the first response threshold F1 is less than the comprehensive quality evolution score result Rsq and the comprehensive quality evolution score result Rsq is less than or equal to the second response threshold F2, it means that the disturbance of the underground construction area is abnormal, so the regional repair strategy is executed at this time;

[0144] When the comprehensive quality evolution score result Rsq is greater than the second response threshold F2, it means that the disturbance of the underground construction area has a risk of damage, so the comprehensive intervention is executed at this time;

[0145] The regional repair strategy optimizes the mechanical construction load by prompting grouting reinforcement, negative pressure pumping control and delaying the current section construction in the underground construction section area;

[0146] The comprehensive intervention stops personnel and machinery from entering the current underground construction section area by issuing a closure instruction, and prompts structure emergency reinforcement, replacement of curtain water stop layer and trial vertical barrier.

[0147] In this embodiment, the method multiplies the disturbance chain score result Rgchain and the structure response score result Rstruct as main effect items by a comprehensive score function model, introduces the echo disturbance spectrum index Necho as a redundancy suppression factor, constructs a nonlinear scoring model, and effectively outputs a comprehensive quality evolution score result Rsq. The score result is used to evaluate the coupling strength between the disturbance and the structure response, has significant discrimination ability and actual physical meaning, and can accurately reflect the conduction strength and influence trend of the disturbance behavior at the structure level. On this basis, in S52, the upper limit values of Rsq corresponding to the normal state and the local repair event are used to scientifically set the first response threshold F1 and the second response threshold F2 by using the historical underground engineering working condition samples, and the real-time calculation of the comprehensive quality evolution score result Rsq is compared with the two thresholds for secondary comparison and evaluation. Through the implementation of the above comprehensive scoring evaluation system, a full-chain closed-loop control mechanism from disturbance detection, structure response analysis to risk level judgment is realized. The beneficial effects mainly include: 1. breaking through the response logical relationship between the disturbance and the structure, improving the accuracy and forward-looking of risk identification; 2. introducing the sound spectrum index to improve the discrimination ability of hidden damage and effectively suppress the false excitation risk; 3. different intervention strategies are matched with different disturbance levels, so that the control measures are more targeted and effective.

[0148] Embodiment 7, please refer to Figure 2 , the building engineering quality monitoring management system based on big data includes a disturbance data acquisition module, a disturbance data processing module, a disturbance analysis module, a structure response analysis module and a joint analysis module;

[0149] The disturbance data acquisition module acquires disturbance data of the underground structure construction area in real time by arranging acquisition points and setting sensor groups in the underground structure construction area, and then transmits the disturbance data to the big data monitoring management platform;

[0150] The disturbance data processing module pre-processes the disturbance data in the big data monitoring management platform, obtains a standard disturbance data set, and extracts features from the standard disturbance data set to obtain a standard disturbance feature vector set;

[0151] The disturbance analysis module constructs a disturbance chain scoring model based on the standard disturbance feature vector set, calculates the disturbance chain score result Rgchain, and pre-sets a disturbance threshold Rth to preliminarily compare and evaluate the disturbance chain score result Rgchain;

[0152] The structure response analysis module triggers a repair strategy based on the preliminary comparison and evaluation result, the repair strategy extracts a settlement vector set, and calculates the structure response score Rstruct in combination with the disturbance chain score result Rgchain;

[0153] The joint analysis module calculates the comprehensive quality evolution score result Rsq by substituting the disturbance chain score result Rgchain and the structure response score result Rstruct into a comprehensive score function model, compares the preset response interval threshold value with the comprehensive quality evolution score result Rsq again, and executes a corresponding control strategy.

[0154] Specific examples:

[0155] Assumption: groundwater velocity change rate ΔVgw=0.3; flow direction spin angle Odiv=0.222, abnormal jump frequency Rhop=0.333;

[0156] raft instantaneous settlement displacement ΔZsup=0.4; reflected disturbance energy spectrum index Necho=0.3;

[0157] ;

[0158] ;

[0159] ;

[0160] Assuming that the disturbance interval threshold values are F1=0.3 and F2=0.55, then Rsq=0.2016≤0.30, and monitoring is continued;

[0161] Although embodiments of the present application have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the present application.

Claims

1. A method for construction engineering quality monitoring management based on big data, characterized in that: The method comprises the following steps: S1, arranging collection points in the underground structure construction area and setting a sensor group to collect disturbance data of the underground structure construction area in real time, and then transmitting the disturbance data to a big data monitoring management platform; S2, preprocessing the disturbance data in the big data monitoring management platform, obtaining a standard disturbance data set, and extracting features of the standard disturbance data set to obtain a standard disturbance feature vector set; S3, constructing a disturbance chain scoring model based on the standard disturbance feature vector set, calculating a disturbance chain scoring result Rgchain, and presetting a disturbance threshold Rth to preliminarily compare and evaluate the disturbance chain scoring result Rgchain; The S3 further comprises S32; S32, based on the disturbance chain scoring result Rgchain of the historical sample, extracting a disturbance chain scoring result Rgchain sample set of a critical structure abnormal response, performing mean value processing to obtain a disturbance threshold Rth, and then preliminarily comparing and evaluating the real-time obtained disturbance chain scoring result Rgchain with the disturbance threshold Rth to judge the current groundwater disturbance behavior; the specific evaluation content is as follows: When the disturbance chain scoring result Rgchain is greater than the disturbance threshold Rth, it indicates that the groundwater disturbance chain is abnormal, and at this time, the repair strategy is triggered; When the disturbance chain scoring result Rgchain is less than or equal to the disturbance threshold Rth, it indicates that the groundwater disturbance behavior is normal, and at this time, no intervention is needed to continue monitoring; S4, triggering the repair strategy based on the preliminary comparison and evaluation result, wherein the repair strategy is calculated by extracting a settlement vector set and combining the disturbance chain scoring result Rgchain to obtain a structure response score Rstruct; The S4 further comprises S41; S41, after triggering the repair strategy in the preliminary comparison and evaluation, combining the settlement vector set and the disturbance chain scoring result Rgchain to calculate and output the structure response score Rstruct; The structure response score Rstruct is calculated and output by the following algorithm formula: ； wherein denotes the partial derivative, denotes the partial derivative with respect to the time variable t, exp denotes the exponential function, ΔZsup denotes the raft instantaneous settlement displacement, Necho denotes the reflected disturbance energy spectrum index; S5, substituting the disturbance chain scoring result Rgchain and the structure response score result Rstruct into a comprehensive scoring function model to calculate a comprehensive quality evolution score result Rsq, performing secondary comparison and evaluation between a preset response interval threshold and the comprehensive quality evolution score result Rsq, and executing a corresponding control strategy; The S5 comprises S51; S51, constructing a comprehensive scoring function model, extracting the real-time obtained disturbance chain scoring result Rgchain and the structure response score result Rstruct, combining a reflected disturbance energy spectrum index Necho, inputting into the comprehensive scoring function model, calculating and outputting the comprehensive quality evolution score result Rsq, and quantitatively analyzing the linkage influence strength of the underground disturbance behavior on the stability of the underground structure construction area; The comprehensive quality evolution score result Rsq is calculated and output by the following constructed comprehensive scoring function model: 。 2. The big data based construction engineering quality monitoring management method according to claim 1, characterized in that: The S1 comprises S11 and S12; S11, the collection point includes collection point P1, collection point P2, collection point P3 and collection point P4, and corresponding sensors are configured in the collection points to collect disturbance data of the underground structure construction area in real time, and the disturbance data includes instantaneous flow rate of underground water, flow direction of underground water, water head gradient jump frequency, water head change mutation amplitude, raft settlement displacement, settlement rate, main frequency offset value and frequency spectrum abnormal mutation times; Wherein: The collection point P1 is through the pore water distribution area arranged at a depth of 20 meters below the raft structure, and an electromagnetic underground water flow rate sensor and a flow direction measuring module are installed at the collection point P1 to collect the original data of the instantaneous flow rate of underground water and the flow direction of underground water; The collection point P2 is through the soil layer area arranged at the junction of the raft edge and the waterproof curtain structure, and a pore water pressure jump frequency detection sensor is installed at the collection point P2 and embedded in the retaining edge soil layer in a radial dot matrix manner to collect the water head gradient mutation frequency and the water head change mutation amplitude per unit time; The collection point P3 is arranged at the central axis position and the symmetrical corner position of the raft structure, and a fiber bragg grating settlement meter is installed at the collection point P3 to collect the original data of the raft structure settlement displacement and the settlement change rate; The collection point P4 is arranged at the curtain structure connection junction area at a position 3 meters below the raft, and a sound wave reflection sensor and a passive acoustic anomaly recognition module integrated with FFT analysis function are installed at the collection point P4 to collect the main frequency offset value of the reflected sound wave energy spectrum of the reflected sound wave of the underground structure interface area and the frequency spectrum abnormal mutation times per unit time; S12, through the data transmission module built in the sensor in all collection points, the data transmission module is wirelessly connected with the big data monitoring management platform using Ethernet, and the disturbance data is uploaded to the big data monitoring management platform.

3. The big data based construction engineering quality monitoring management method according to claim 2, characterized in that: The S2 includes S21; S21, the disturbance data is preprocessed in the big data monitoring management platform to obtain a standard disturbance data set, and the preprocessing includes data time alignment processing, direction class and speed class data standardization, jump class disturbance data denoising correction, structure settlement data filtering and smoothing, acoustic spectrum class data frequency domain filtering and data format unification; The data time alignment processing aligns the disturbance data collection time uploaded by different collection points through a multi-source data unified timestamp mapping algorithm to construct a disturbance data set with consistent time sequence; The direction class and speed class data standardization are normalized by a normalization function and a direction angle conversion algorithm, and the instantaneous flow rate of underground water and the flow direction of underground water are normalized and coordinated in a unit dimension to eliminate the influence of flow field spatial direction heterogeneity; The jump class disturbance data denoising correction is a curve smoothing processing method for raft settlement displacement and settlement rate data by using cubic spline fitting and high pass filtering, which can suppress local jitter error caused by construction disturbance and measurement point drift; The acoustic spectrum class data frequency domain filtering is a filtering method for high frequency interference signals in the reflected sound wave main frequency offset value and the frequency spectrum abnormal mutation times by using the fast Fourier transform FFT and the band pass filtering technology, which can retain the effective frequency band change trend. The data format is unified by adopting the Z-score standardization method of numerical range standardization to unify the original values of the disturbance data of different dimensions, units and intervals to the standard interval with a mean of 0 and a standard deviation of 1, eliminate the dimension influence in all disturbance data, and then output as a standard disturbance data set by adopting a structured data packaging rule.

4. The big data based construction engineering quality monitoring management method according to claim 3, characterized in that: The S2 further comprises S22; S22, based on the pre-processed standard disturbance data set, extracting for feature extraction to obtain a standard disturbance feature vector set; The standard disturbance feature vector set comprises a disturbance vector set and a subsidence vector set; The disturbance vector set comprises a groundwater velocity change rate ΔVgw, a flow direction spin angle Odiv, an abnormal jump frequency Rhop and a surge break jump peak value; The subsidence vector set comprises a raft instantaneous subsidence displacement ΔZsup and a reflection disturbance energy spectrum index Necho; The abnormal jump frequency Rhop is extracted from the hydraulic gradient jump frequency and the water head change mutation amplitude in unit time by a statistical window function and a jump threshold analysis algorithm, the abnormal jump frequency Rhop represents the number of mutation events in unit time, and the surge break jump peak value represents the maximum water head disturbance amplitude; The groundwater velocity change rate ΔVgw and the flow direction spin angle Odiv are obtained by applying a first-order difference between adjacent time points to a sequence of groundwater instantaneous flow velocity to obtain the groundwater velocity change rate ΔVgw, and using a three-dimensional vector angle calculation formula to obtain the current flow direction spin angle Odiv based on the designed direction as the reference direction. The reflection disturbance energy spectrum index Necho is extracted from the main frequency offset value of the reflection sound wave and the number of abnormal spectrum mutations in unit time by a frequency offset sliding window analysis method and a power spectrum mutation detection, and is used to describe the comprehensive intensity of the sound spectrum disturbance; The raft instantaneous subsidence displacement ΔZsup is extracted from the raft subsidence displacement and the subsidence rate by a subsidence displacement derivative calculation function.

5. The big data based construction engineering quality monitoring management method according to claim 4, characterized in that: The S3 comprises S31; S31, constructing a disturbance chain scoring model based on a nonlinear combination function, extracting the disturbance vector set in the standard disturbance feature vector set as an input item input into the disturbance chain scoring model, and calculating and outputting a disturbance chain scoring result Rgchain; The disturbance chain scoring result Rgchain is calculated and outputted by the following disturbance chain scoring model: ； In the formula, log represents a logarithmic function, and tan represents a tangent function.

6. The big data based construction engineering quality monitoring management method according to claim 1, characterized in that: The S5 further comprises S52; S52, based on the comprehensive quality evolution scoring result Rsq of the corresponding normal working condition, local repair event and abnormal event of the historical underground engineering, setting the upper limit value of the normal working condition and the upper limit value of the local repair event as a response interval threshold value; the response interval threshold value comprises a first response threshold value F1 and a second response threshold value F2; The response interval threshold value is compared with the comprehensive quality evolution scoring result Rsq for secondary comparison and evaluation to judge the stability of the underground construction area structure under the groundwater disturbance, and a corresponding control strategy is executed based on the secondary comparison and evaluation result; the specific evaluation content is as follows: When the comprehensive quality evolution score result Rsq is less than or equal to a first response threshold F1, it indicates that the disturbance of the underground construction area is normal and does not affect the structure, and no intervention is needed to continue monitoring; When the first response threshold F1 is less than the comprehensive quality evolution score result Rsq and the comprehensive quality evolution score result Rsq is less than or equal to a second response threshold F2, it indicates that the disturbance of the underground construction area is abnormal, and a regional repair strategy is executed at this time; When the comprehensive quality evolution score result Rsq is greater than the second response threshold F2, it indicates that the disturbance of the underground construction area has a risk of damage, and a comprehensive intervention is executed at this time; The regional repair strategy optimizes the mechanical construction load by prompting grouting reinforcement, negative pressure pumping control and postponing the current section construction in the underground construction area; The comprehensive intervention stops personnel and machinery from entering the current underground construction area by issuing a closure instruction, and prompts structure emergency reinforcement, replacement of curtain water stop layer and trial vertical barrier.

7. The building engineering quality monitoring management system based on big data, applied to the building engineering quality monitoring management method based on big data according to any one of claims 1-6, characterized in that: The disturbance data acquisition module, the disturbance data processing module, the disturbance analysis module, the structure response analysis module and the joint analysis module are included. The disturbance data acquisition module acquires disturbance data of the underground structure construction area in real time by arranging acquisition points and setting sensor groups in the underground structure construction area, and transmits the disturbance data to a big data monitoring management platform. The disturbance data processing module pre-processes the disturbance data in the big data monitoring management platform, obtains a standard disturbance data set, and extracts features from the standard disturbance data set to obtain a standard disturbance feature vector set. The disturbance analysis module constructs a disturbance chain scoring model based on the standard disturbance feature vector set, calculates a disturbance chain score result Rgchain, and pre-sets a disturbance threshold Rth to preliminarily compare and evaluate the disturbance chain score result Rgchain. The structure response analysis module triggers a repair strategy based on the preliminary comparison and evaluation result, and the repair strategy extracts a settlement vector set and combines the disturbance chain score result Rgchain to calculate a structure response score Rstruct. The joint analysis module substitutes the disturbance chain score result Rgchain and the structure response score result Rstruct into a comprehensive score function model to calculate a comprehensive quality evolution score result Rsq, compares the comprehensive quality evolution score result Rsq with a pre-set response interval threshold for a second time, and executes a corresponding control strategy.

Images