An engineering quality safety risk intelligent assessment method and system

Abstract

The application discloses an engineering quality safety risk intelligent evaluation method and system, relates to the technical field of engineering quality safety evaluation, adopts a plurality of sensors to collect data, breaks the limitation that only a single index is relied on to judge risks in traditional structure monitoring, realizes comprehensive perception of the structure state across parameters and components, solves the problem that structure risks are often evolved by multiple factors, and improves overall identification precision. A sliding window and a periodic variation coefficient calculation mechanism are introduced, so that the system can dynamically extract fluctuation characteristics of structure responses, and identify different evolution forms of short-time fluctuation, medium-term creep and long-term mutation. The system makes up for the defects of traditional methods, such as insufficient abnormal precursor identification ability and serious lag reaction. An abnormal mutation index AEI and an evolution acceleration factor EAF are constructed, so that the system no longer only takes whether a value at a moment exceeds a threshold as a basis, but identifies the formation process and development path of risks, and improves risk process perception ability.

Description

Technical Field

[0001] This invention relates to the field of engineering quality and safety assessment technology, specifically to an intelligent assessment method and system for engineering quality and safety risks. Background Technology

[0002] Engineering quality and safety assessment, as an important application area integrating civil engineering, structural engineering, and intelligent monitoring technology, has received increasing attention in recent years due to the growing complexity of large-scale infrastructure construction. Within this macro-level field, structural health monitoring technology has become a crucial means of ensuring the safety of projects throughout their entire lifecycle. Its core task is to dynamically monitor key structural components through multi-source sensors to promptly identify potential deterioration and abnormal evolution trends.

[0003] Currently, in structural health monitoring, the identification of abnormal risks mostly relies on threshold judgment based on single sensor data or isolated anomaly detection. Conventional methods ignore the coupling and temporal evolution characteristics between multiple parameters in structural behavior. This single-point monitoring mechanism has significant drawbacks: on the one hand, structural anomalies are usually gradual processes driven by multi-parameter coupling, and single-point anomalies cannot accurately reflect the overall trend; on the other hand, traditional static threshold methods cannot capture the evolution path of gradual and sudden risks, and are prone to "silent monitoring blind spots before abrupt changes."

[0004] The fundamental reason for these shortcomings lies in the fact that traditional monitoring methods focus more on identifying instantaneous outliers, while lacking a holistic understanding of the structural response time series in both spatial and temporal dimensions. In practical engineering applications, parameters such as uneven settlement of the foundation, tilting of the superstructure, and stress fluctuations are often not synchronously abruptly changed, but rather exhibit a latent evolutionary characteristic of "gradual change followed by abrupt change". Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides an intelligent assessment method and system for engineering quality and safety risks, solving the problems mentioned in the background section.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solution: an intelligent assessment system for engineering quality and safety risks, comprising a time-series data acquisition and preprocessing module, a data variation analysis module, an abnormal mutation index fusion module, a driving factor identification module, a dynamic determination module for early warning thresholds, and a trend evolution report generation module;

[0007] The time-series data acquisition and preprocessing module collects engineering quality data through sensors, fits it into the original dataset W, and performs preprocessing to obtain the quality dataset ZW.

[0008] The data variation analysis module calculates the degree of local variation of the data in the quality dataset ZW within a fixed period T and obtains the coefficient of variation Ψ.

[0009] The abnormal mutation index fusion module fuses the obtained coefficients of variation Ψ to construct the abnormal mutation index AEI;

[0010] The driving factor identification module identifies the acceleration of the abnormal mutation index AEI, determines whether there are mutation driving points in the time series of the abnormal mutation index AEI, obtains the evolution acceleration factor EAF, and identifies potential hidden dangers.

[0011] The early warning threshold dynamic judgment module jointly judges the abnormal mutation index AEI and the evolution acceleration factor EAF based on historical statistics and adaptive dynamic thresholds to obtain the risk level;

[0012] The trend evolution report generation module generates an execution strategy based on the acquired abnormal mutation index AEI, evolution acceleration factor EAF, and risk level.

[0013] Preferably, the time-series data acquisition and preprocessing module includes a parameter acquisition and fitting unit and a data preprocessing unit;

[0014] The parameter acquisition and fitting unit collects engineering quality data through sensors deployed at the location of the engineering structure, including the secondary settlement increment Adz of the foundation, the micro-drift value of crack width Acw, the stress rebound frequency deviation value Are, and the component tilt angle gradient Asp, and fits it into the original dataset W.

[0015] The secondary settlement increment Adz of the foundation was acquired using multi-point displacement gauges and fiber optic displacement sensors.

[0016] Specifically, it is obtained by the difference between the secondary settlement increment Adz of the foundation at time t and the secondary settlement increment Adz of the foundation at time t-1;

[0017] The micro-drift value Acw of the crack width was acquired using a crack gauge and a resistive crack width sensor.

[0018] The crack width micro-drift value Acw is obtained using the following formula:

[0019] ;

[0020] In the formula, Ct represents the real-time measured value of the crack width at time t, and Cref represents the reference width during the initial stabilization period of the crack;

[0021] The stress rebound frequency offset value Are was acquired using a stress wave frequency sensor and a high-frequency structural response sensor.

[0022] The stress rebound frequency offset value Are is obtained by the following formula:

[0023] ;

[0024] In the formula, fr(t) represents the instantaneous resonant frequency during the rebound phase after unloading at time t, and fs(t) represents the background frequency during the steady phase at time t.

[0025] The component tilt gradient rate Asp was acquired using a MEMS tilt sensor and a fiber optic inclinometer.

[0026] The gradient rate of component inclination Asp is obtained using the following formula:

[0027] ;

[0028] In the formula, JD(t) represents the tilt angle of the component at time t, JD(t-1) represents the tilt angle of the component at time t-1, and Δt represents the time interval;

[0029] The data preprocessing unit cleans and normalizes the original dataset W to obtain a high-quality dataset ZW;

[0030] Cleaning identifies outliers in the original dataset W by using adjacent transition judgment and local linear fitting residual thresholding.

[0031] For any data Wo in the original dataset W, construct a local linear fit and obtain the fitted value Wofit(t).

[0032] Where Wo represents the o-th data item in the original dataset W;

[0033] The fitted value Wofit(t) is obtained using the following formula:

[0034] Wofit(t) = at + b;

[0035] In the formula, Wofit(t) represents the fitted value, specifically the fitted value estimated by the linear fitting function for the o-th data in the original dataset W at time t, a represents the fitting slope term, and b represents the fitting intercept term.

[0036] Calculate the fitting residual E based on the fitted value Wofit(t) and determine whether it is abnormal;

[0037] The fitting residual E is obtained using the following formula:

[0038] E(t)=Wo(t)-Wofit(t);

[0039] In the formula, E(t) represents the fitting residual at time t, and Wo(t) represents the o-th data item in the original dataset W at time t;

[0040] The formula for judgment is as follows:

[0041] ;

[0042] In the formula, Ejump represents the jump tolerance factor used to adjust the judgment sensitivity, which is usually set between 1.5 and 2.0. The larger the value, the more conservative the judgment. Δ(t-1, t+1) represents the magnitude of the difference between adjacent points.

[0043] If the residual at the current point is larger than the average jump between adjacent points, it indicates that the point may be an "outlier" and needs to be removed.

[0044] Normalization is performed by processing the data in the original dataset W using the max-min normalization method to obtain the quality dataset ZW;

[0045] ;

[0046] In the formula, ZWo represents the o-th data in the quality dataset ZW, Wo represents the o-th data in the original dataset W, minWo represents the valley value of the o-th data in the original dataset W, and maxWo represents the peak value of the o-th data in the original dataset W.

[0047] Preferably, the data variation analysis module extracts the local variation intensity for each parameter in the quality dataset ZW and obtains the variation coefficient Ψ. The steps for obtaining the coefficient Ψ are as follows:

[0048] S1. Set a fixed period T and construct a multi-scale nested window: T1=T, T2=2T, T3=0.5T;

[0049] Setting up multi-scale nested windows (T1, T2, T3) is not arbitrary, but rather to extract variation features from time series data from multiple perspectives, thereby improving the sensitivity and adaptability to different structural anomaly evolution patterns.

[0050] The multi-scale nested window (T1, T2, T3) design is an asymmetric dynamic ratio construction method used to integrate the differences between long-term stable trends and short-term violent disturbances, thereby achieving a comprehensive perception of mutation precursors, anomaly intensity, and evolutionary trends through time-series statistics without relying on additional hardware.

[0051] T1 is used as the principal period to calculate the coefficient of variation Ψ on the basic time scale. o (T1) is used to capture local risk change characteristics; the fixed period T here is usually determined based on structural characteristics or monitoring frequency, such as 1 day, 1 week, and 1 month.

[0052] T2=2T is used to construct a stable trend benchmark for the mutation enhancement factor VAF; doubling the period can smooth out mutations and capture trend stability; when constructing the mutation enhancement factor VAF, T2 provides a relatively smooth low-frequency trend;

[0053] This helps to identify the magnitude of the difference between mutations and normal trends, and improves the ability to detect potential slow-mutation risks.

[0054] T3=0.5T is used to enhance sensitivity to abnormal spike fluctuations; it enhances the response to sudden small-scale anomalies by shortening the period; compared with T2, it can form a contrast gradient of mutation rate, constituting the molecule of the mutation enhancement factor;

[0055] It is suitable for identifying "early signs of change" and "local disturbances".

[0056] S2. For the o-th data ZWo in the quality dataset ZW, calculate and obtain the sliding coefficient of variation Ψo within a fixed period T1.

[0057] The sliding coefficient of variation Ψo is obtained by the following formula:

[0058] ;

[0059] In the formula, Ψo(T1) represents the sliding coefficient of variation of the o-th data in the quality dataset ZW within a fixed period T1, σZWo(T1) represents the standard deviation of the o-th data in the quality dataset ZW within a fixed period T1, μZWo(T1) represents the mean of the o-th data in the quality dataset ZW within a fixed period T1, and eo represents a non-zero constant.

[0060] S3. By introducing the variation enhancement factor (VAF), the sensitivity to different trend states is improved;

[0061] The mutation enhancement factor (VAF) is obtained using the following formula:

[0062] ;

[0063] In the formula, ln represents the logarithmic function, Ψo(T2) represents the sliding coefficient of variation of the o-th data in the quality dataset ZW within a fixed period T2, and Ψo(T3) represents the sliding coefficient of variation of the o-th data in the quality dataset ZW within a fixed period T3.

[0064] S4. Obtain the coefficient of variation Ψ;

[0065] Ψ = Ψo(T1) * VAF;

[0066] In the formula, Ψ represents the coefficient of variation, including the coefficient of variation of secondary settlement increment of foundation ΨAdz, the coefficient of variation of crack width micro-drift value ΨAcw, the coefficient of variation of stress rebound frequency deviation value ΨAre, and the coefficient of variation of component tilt angle gradient rate ΨAsp.

[0067] Preferably, the abnormal mutation index fusion module performs nonlinear structural aggregation on the obtained secondary settlement increment variation coefficient ΨAdz, crack width micro-drift value variation coefficient ΨAcw, stress rebound frequency deviation value variation coefficient ΨAre, and component tilt angle gradient variation coefficient ΨAsp to obtain the abnormal mutation index AEI;

[0068] The Abnormal Mutation Index (AEI) is obtained using the following formula:

[0069] ;

[0070] In the formula, ln represents the exponential function, A represents the nonlinear enhancement factor (where the value of the nonlinear enhancement factor A is 1.3), and arctan represents the arctangent function;

[0071] The formula for obtaining the Abnormal Mutation Index (AEI) achieves the mapping of multiple types of abnormal risks into a single, measurable mutation index (AEI) by performing nonlinear transformation and unified aggregation on four structural variation coefficients with different physical meanings.

[0072] Various parameters are processed by nonlinear functions to unify the risk response scale; each function selects different transformation logic and risk adjustment curves based on parameter characteristics.

[0073] The abnormal mutation index (AEI) is no longer obtained by conventional linear superposition. Instead, different mapping functions are designed for each parameter based on the engineering physical behavior mechanism. Each transformation takes into account both the goals of "early identification" and "late mutation regulation".

[0074] ΨAdz 2 This indicates that the foundation settlement fluctuates and tends to accumulate slowly; as the value increases, the risk multiplies.

[0075] This indicates that the opening and closing of cracks is unstable. Low values ​​serve as an early warning sign, while high values ​​are suppressed, highlighting early fluctuations.

[0076] Ln(1+ΨAre) represents the stress rebound frequency deviation; high-frequency disturbances are often accompanied by local microcracks.

[0077] arctanΨAsp represents a gradual change in tilt angle, controlling the saturation response at high tilt angles;

[0078] The obtained abnormal mutation index AEI is compared with the preset abnormal mutation threshold Tyc to obtain the mutation status;

[0079] Mutation states are obtained by matching in the following way:

[0080] When the abnormal mutation index AEI is less than the abnormal mutation threshold Tyc*0.5, it indicates that the mutation state is stable and normal monitoring is required.

[0081] When the abnormal mutation threshold Tyc*0.5 ≤ abnormal mutation index AEI < abnormal mutation threshold Tyc, it indicates a reversible abnormality, localized intensified monitoring, and trend analysis is initiated.

[0082] When the abnormal mutation threshold Tyc ≤ abnormal mutation index AEI < abnormal mutation threshold Tyc*1.5, it indicates a precursor to a coordinated mutation and an early warning is issued.

[0083] When the abnormal mutation threshold Tyc*1.5 ≤ the abnormal mutation index AEI, it indicates a high-risk state of multi-source structural mutation, and the clinical detection mechanism is activated.

[0084] Preferably, the driving factor identification module includes an acceleration identification unit and a mutation driving point determination unit;

[0085] The acceleration identification unit constructs evolutionary acceleration a by extracting the second-order change features of the anomalous mutation index AEI;

[0086] The evolutionary acceleration 'a' is obtained using the following formula:

[0087] ;

[0088] In the formula, AEI(t) represents the anomalous mutation index at time t, AEI(t+1) represents the anomalous mutation index at time t+1, AEI(t-1) represents the anomalous mutation index at time t-1, and Δt represents the time interval.

[0089] By introducing the risk growth inertia coefficient GX, the obtained evolution acceleration a is corrected to obtain the evolution acceleration factor EAF;

[0090] The risk growth inertia coefficient GX is obtained using the following formula:

[0091] ;

[0092] In the formula, AEI(t-3) represents the anomalous mutation index at time t-3, ae represents a positive number, and V(t) represents the first derivative of the anomalous mutation index AEI; it is obtained by the ratio of the difference between the anomalous mutation index AEI(t) at time t and the anomalous mutation index AEI(t-1) at time t-1 to the time interval Δt.

[0093] The evolution acceleration factor EAF is obtained by multiplying the risk growth inertia coefficient GX by the evolution acceleration a.

[0094] Preferably, the mutation driving point determination unit calculates and obtains the evolution threshold Teaf based on the evolutionary fluctuation of the evolution acceleration factor EAF;

[0095] The evolution threshold Teaf is obtained using the following formula:

[0096] ;

[0097] In the formula, μEAF represents the mean of the evolution acceleration factor, σEAF represents the standard deviation of the evolution acceleration factor, and af represents the sensitivity coefficient;

[0098] The evolution acceleration factor (EAF) is compared with the evolution threshold (Teaf) to determine the risk status.

[0099] Risk status is obtained through the following methods:

[0100] When the evolution acceleration factor EAF ≤ the evolution threshold Teaf, it indicates a normal steady state;

[0101] When the evolution acceleration factor EAF is greater than the evolution threshold Teaf, it indicates a risky state, a potential hazard, and an early warning is issued.

[0102] Preferably, the early warning threshold dynamic determination module includes a joint index construction unit and a risk level classification unit;

[0103] The joint index construction unit jointly constructs the obtained evolution acceleration factor EAF and abnormal mutation index AEI to obtain the joint index ARIA;

[0104] The combined indicator ARIA is obtained using the following formula:

[0105] ;

[0106] In the formula, r1 represents the risk intensity adjustment factor, and r2 represents the risk acceleration adjustment factor.

[0107] Preferably, the risk level classification unit extracts all values ​​of the joint indicator ARIA from historical periods to form a set Har;

[0108] Threshold extraction was performed on the set Har using the mean method and the standard deviation method to obtain the index threshold THR;

[0109] The threshold THR is obtained by summing the mean of the joint indicator ARIA with 0.5 times the standard deviation of the joint indicator ARIA.

[0110] The risk level is obtained by comparing the combined indicator ARIA with the indicator threshold THR.

[0111] Risk levels are obtained through matching in the following ways:

[0112] When the combined indicator ARIA < 0.5 * indicator threshold THR, it indicates the first risk level, a safe state.

[0113] When 0.5 * indicator threshold THr ≤ joint indicator ARIA < indicator threshold THr, it indicates the second risk level, and a controllable anomaly has occurred.

[0114] When the indicator threshold THr ≤ the joint indicator ARIA < 1.5 * indicator threshold THr, it indicates the third risk level, and the precursor to a sudden change has appeared.

[0115] When 1.5 * the indicator threshold THr ≤ the joint indicator ARIA, it indicates the fourth risk level, an abnormally high risk state.

[0116] Preferably, the trend evolution report generation module performs trend identification on the time series composed of the Abnormal Mutation Index (AEI) and the Evolutionary Acceleration Factor (EAF), and constructs a time series curve of the Abnormal Mutation Index (AEI) and an acceleration curve of the Evolutionary Acceleration Factor (EAF).

[0117] Time series curve of Abnormal Mutation Index (AEI): The AEI curve is plotted according to the timestamp order, and the peak point, inflection point and risk level segment color band are marked on the graph.

[0118] The Evolutionary Accelerator Factor (EAF) acceleration curve includes the length of the continuously rising segment of EAF and the time interval from the peak value of EAF to the mutation point, reflecting the response hysteresis.

[0119] The length of the continuously rising EAF (Evolution Acceleration Factor) indicates the duration of accelerated evolution, expressed in hours, and is used to assess the accelerated preparation period before the outbreak of structural risks.

[0120] The time interval from the peak of the Evolution Acceleration Factor (EAF) to the mutation point represents the time difference from the maximum acceleration to the extreme point of the AEI, reflecting the response hysteresis.

[0121] Based on the time series curve of the Abnormal Mutation Index (AEI) and the acceleration curve of the Evolutionary Acceleration Factor (EAF), and combined with the risk level, an execution strategy is generated.

[0122] The execution strategy is obtained by matching in the following ways:

[0123] When the risk level is at its highest, the trend of the Abnormal Mutation Index (AEI) and the Evolution Acceleration Factor (EAF) are stable, indicating a stable state. The appropriate strategy is to perform routine inspections and periodic sampling.

[0124] When the risk level is at level 2, the abnormal mutation index AEI shows a slight fluctuation, and the evolution acceleration factor EAF shows an occasional slight increase, indicating that there is a disturbance and no continuous evolution. The strategy to be implemented is to increase the monitoring frequency and start the local trend tracking algorithm.

[0125] When the risk level is at level 3, the abnormal mutation index (AEI) shows a significant upward trend and the evolution acceleration factor (EAF) shows a continuous upward trend, indicating that cumulative instability is forming. The implementation strategy is: prepare local maintenance plans, re-inspect the structure, and assess whether it will affect the function.

[0126] When the risk level is at level four, the abnormal mutation index AEI approaches its extreme value, and the evolution acceleration factor EAF has reached its peak, indicating critical evolution and impending mutation. The appropriate strategies are: structural unloading, personnel evacuation, and work stoppage.

[0127] A method for intelligent assessment of engineering quality and safety risks includes the following steps:

[0128] Step 1: The time series data acquisition and preprocessing module collects engineering quality data through sensors, fits it into the original dataset W, and performs preprocessing to obtain the quality dataset ZW.

[0129] Step 2: The data variation analysis module calculates the degree of local variation of the data in the quality dataset ZW within a fixed period T, and obtains the coefficient of variation Ψ.

[0130] Step 3: The abnormal mutation index fusion module fuses the obtained coefficients of variation Ψ to construct the abnormal mutation index AEI;

[0131] Step 4: The driving factor identification module identifies the acceleration of the abnormal mutation index AEI, determines whether there are mutation driving points in the time series of the abnormal mutation index AEI, obtains the evolution acceleration factor EAF, and identifies potential hidden dangers.

[0132] Step 5: The early warning threshold dynamic judgment module jointly judges the abnormal mutation index AEI and the evolution acceleration factor EAF based on historical statistics to obtain the risk level.

[0133] Step Six: The Trend Evolution Report Generation Module generates an execution strategy based on the acquired Abnormal Mutation Index (AEI), Evolution Acceleration Factor (EAF), and Risk Level.

[0134] This invention provides an intelligent assessment method and system for engineering quality and safety risks, which has the following beneficial effects:

[0135] (1) During system operation, multiple types of sensors are used to collect data, breaking the limitation of relying on a single indicator to judge risk in traditional structural monitoring, and realizing comprehensive perception of structural status across parameters and components. This solves the problem that structural risks often evolve from multiple factors, improving the overall identification accuracy. By introducing a sliding window and periodic variation coefficient calculation mechanism, the system can dynamically extract the fluctuation characteristics of structural response and identify different evolutionary forms such as "short-term fluctuations," "medium-term creep," and "long-term mutations." This makes up for the shortcomings of traditional methods, such as insufficient ability to identify abnormal precursors and severe delayed response.

[0136] An Abnormal Mutation Index (AEI) and an Evolutionary Acceleration Factor (EAF) were constructed, moving beyond simply relying on whether a value exceeds a threshold at a given moment. Instead, these indicators identify the formation process and development path of risks, enhancing the system's ability to "perceive" risks. Dynamic early warning thresholds are constructed using historical stable period data, allowing the system to adapt personalized judgment criteria to different structures and environments, avoiding false alarms or missed alarms based on fixed thresholds.

[0137] (2) Data preprocessing not only performs routine missing / noise cleaning, but also introduces the "adjacent jump judgment" and "local linear fitting residual threshold" methods to identify trend constraints for outliers. Compared with the static upper and lower limit elimination method, this approach is more suitable for capturing the starting point of abrupt changes in time-series anomalies, improving the system's resistance to interference data and its ability to retain potential risk signals. By using the maximum-minimum normalization method to stretch and compress various engineering parameters within a range, data from different physical dimensions can be processed uniformly, providing consistent support for the basic data structure for subsequent variation analysis and trend fusion, and solving the problem of difficult cross-dimensional data fusion.

[0138] By collecting structural micro-variation parameters instead of static indicators, signals of structural performance degradation can be detected earlier, allowing for timely intervention and mitigation of potential hazards, effectively shortening the risk identification lag period. By using a cleaning algorithm to remove sudden data jumps in the structural response, data stability and the reliability of analysis results are effectively improved, preventing false alarms or missed alarms.

[0139] (3) The construction of the abnormal mutation index AEI is no longer a conventional weighted synthesis or simple averaging, but a deep coupling model of the coefficient of variation by combining nonlinear functions such as exponential function and arctangent function, which strengthens the response degree when the abnormal co-evolution of each dimension is strengthened, solves the traditional problem of "signal weakening after multi-source index fusion", and significantly improves the system's analytical power for the abnormal evolution of high-order structures.

[0140] By comparing the Abnormal Mutation Index (AEI) with a preset abnormal mutation threshold and classifying it into four mutation states, continuous closed-loop control from "monitoring" to "response" is achieved. This mechanism can not only identify the abnormality itself, but also classify and judge its development direction and controllability, thereby enhancing the system's dynamic adaptability and response accuracy.

[0141] (4) By calculating the current threshold based on the dynamic distribution characteristics of the evolution acceleration factor (EAF), a risk state discrimination mechanism that is adaptive to the structure is realized. The system can dynamically adjust its sensitivity according to different structures and different monitoring cycles to adapt to the risk judgment needs of different engineering scenarios. The system is no longer limited to judging whether it is abnormal, but further locates the starting point of abnormal growth, that is, the "starting node of risk evolution". This capability enables the system to identify the "driving point" before it reaches a state of full mutation, realize accurate early warning and early intervention, and significantly improve the depth of the system in identifying the precursors of mutation. Attached Figure Description

[0142] Figure 1 This is a flowchart illustrating the intelligent assessment system for engineering quality and safety risks according to the present invention.

[0143] Figure 2 This is a schematic diagram illustrating the steps of an intelligent assessment method for engineering quality and safety risks according to the present invention.

[0144] Figure 3 This is a schematic diagram of the process for obtaining the joint indicators of the present invention;

[0145] Figure 4 This is a histogram of the abnormal mutation index of the present invention;

[0146] Figure 5 This is a time-series trend chart of the abnormal mutation index of the present invention;

[0147] Figure 6 This is a time-series curve of the evolution acceleration factor of the present invention. Detailed Implementation

[0148] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0149] Example 1: This invention provides an intelligent assessment system for engineering quality and safety risks. Please refer to [link / reference]. Figures 1 to 6It includes a time-series data acquisition and preprocessing module, a data variation analysis module, an abnormal mutation index fusion module, a driving factor identification module, a dynamic judgment module for early warning thresholds, and a trend evolution report generation module;

[0150] The time-series data acquisition and preprocessing module collects engineering quality data through sensors, fits it into the original dataset W, and performs preprocessing to obtain the quality dataset ZW.

[0151] The data variation analysis module calculates the degree of local variation of the data in the quality dataset ZW within a fixed period T and obtains the coefficient of variation Ψ.

[0152] The abnormal mutation index fusion module fuses the obtained coefficients of variation Ψ to construct the abnormal mutation index AEI;

[0153] The driving factor identification module identifies the acceleration of the abnormal mutation index AEI, determines whether there are mutation driving points in the time series of the abnormal mutation index AEI, obtains the evolution acceleration factor EAF, and identifies potential hidden dangers.

[0154] The early warning threshold dynamic judgment module jointly judges the abnormal mutation index AEI and the evolution acceleration factor EAF based on historical statistics and adaptive dynamic thresholds to obtain the risk level;

[0155] The trend evolution report generation module generates an execution strategy based on the acquired abnormal mutation index AEI, evolution acceleration factor EAF, and risk level.

[0156] In this embodiment, multiple types of sensors are used to collect data, breaking the limitation of relying on a single indicator to judge risk in traditional structural monitoring, and realizing comprehensive perception of structural status across parameters and components. This solves the problem that structural risks often evolve due to multiple factors, improving overall identification accuracy. By introducing a sliding window and periodic variation coefficient calculation mechanism, the system can dynamically extract the fluctuation characteristics of structural response and identify different evolutionary forms such as "short-term fluctuations," "medium-term creep," and "long-term abrupt changes." This overcomes the shortcomings of traditional methods, such as insufficient ability to identify abnormal precursors and severe lag in response.

[0157] An Abnormal Mutation Index (AEI) and an Evolutionary Acceleration Factor (EAF) were constructed, moving beyond simply relying on whether a value exceeds a threshold at a given moment. Instead, these indicators identify the formation process and development path of risks, enhancing the system's ability to "perceive" risks. Dynamic early warning thresholds are constructed using historical stable period data, allowing the system to adapt personalized judgment criteria to different structures and environments, avoiding false alarms or missed alarms based on fixed thresholds.

[0158] By utilizing multi-dimensional parameter fusion and trend evolution mechanisms, it is possible to identify "gradual hidden dangers" or "precursors of mutation" that are difficult for traditional systems to detect in advance. The dual-indicator mechanism of the Abnormal Mutation Index (AEI) and the Evolution Acceleration Factor (EAF) takes into account both risk intensity and development speed, giving the system a natural ability to suppress false alarms, minor alarms, and delayed responses. The dynamic threshold algorithm is constructed based on the actual project, does not rely on manual settings, and is applicable to various structural types such as bridges, tunnels, and high-rise buildings.

[0159] Example 2 is an explanation of Example 1. Please refer to the example provided. Figure 1 and Figure 3 Specifically: the time series data acquisition and preprocessing module includes a parameter acquisition and fitting unit and a data preprocessing unit;

[0160] The parameter acquisition and fitting unit collects engineering quality data through sensors deployed at the location of the engineering structure, including the secondary settlement increment Adz of the foundation, the micro-drift value of crack width Acw, the stress rebound frequency deviation value Are, and the component tilt angle gradient Asp, and fits it into the original dataset W.

[0161] The secondary settlement increment Adz of the foundation was acquired by multi-point displacement gauges and fiber optic displacement sensors.

[0162] The micro-drift value Acw of the crack width was acquired by a crack gauge and a resistive crack width sensor;

[0163] The stress rebound frequency offset value Are was acquired by a stress wave frequency sensor and a high-frequency structural response sensor.

[0164] The component tilt gradient rate Asp was acquired using a MEMS tilt sensor and a fiber optic inclinometer.

[0165] The data preprocessing unit cleans and normalizes the original dataset W to obtain a high-quality dataset ZW;

[0166] Cleaning identifies outliers in the original dataset W by using adjacent transition judgment and local linear fitting residual thresholding.

[0167] Normalization is performed by processing the data in the original dataset W using the max-min normalization method to obtain the quality dataset ZW;

[0168] ;

[0169] In the formula, ZWo represents the o-th data in the quality dataset ZW, Wo represents the o-th data in the original dataset W, minWo represents the valley value of the o-th data in the original dataset W, and maxWo represents the peak value of the o-th data in the original dataset W.

[0170] This embodiment introduces four specific structural behavior quantification parameters: secondary settlement increment Adz, crack width micro-drift value Acw, stress rebound frequency deviation value Are, and component tilt angle gradient rate Asp. Compared with traditional structural monitoring methods that focus on single values, these parameters have stronger dynamic response characteristics and gradient identification capabilities, enabling earlier detection of latent evolutionary signs of "transition from stability". The parameters are collected using sensor types commonly used in current engineering monitoring, such as fiber optic sensors, multi-point displacement gauges, and MEMS tilt sensors. This allows for in-situ upgrades without the need for additional hardware, ensuring the system's deployability and cost-effectiveness in engineering projects, and meeting the core innovation requirement of "no need for additional sensors".

[0171] Data preprocessing not only performs routine missing / noise cleaning but also introduces "adjacent jump determination" and "local linear fitting residual thresholding" methods to identify outliers with trend constraints. Compared to static upper and lower limit removal methods, this approach is more suitable for capturing the onset points of abrupt changes in time-series anomalies, improving the system's robustness to interference data and its ability to retain potential risk signals. By using the max-min normalization method to perform interval stretching and compression on various engineering parameters, data from different physical dimensions can be processed uniformly, providing consistent data structure support for subsequent variation analysis and trend fusion, and solving the problem of difficult cross-dimensional data fusion.

[0172] By collecting structural micro-variation parameters instead of static indicators, signals of structural performance degradation can be detected earlier, allowing for timely intervention and mitigation of potential hazards, effectively shortening the risk identification lag period. By using a cleaning algorithm to remove sudden data jumps in the structural response, data stability and the reliability of analysis results are effectively improved, preventing false alarms or missed alarms.

[0173] Example 3 is an explanation of Example 2. Please refer to the example provided. Figure 1 and Figure 4 Specifically: The data variation analysis module extracts the local variation intensity for each parameter in the quality dataset ZW and obtains the variation coefficient Ψ. The steps for obtaining the coefficient Ψ are as follows;

[0174] S1. Set a fixed period T and construct a multi-scale nested window: T1=T, T2=2T, T3=0.5T;

[0175] S2. For the o-th data ZWo in the quality dataset ZW, calculate and obtain the sliding coefficient of variation Ψo within a fixed period T1.

[0176] The sliding coefficient of variation Ψo is obtained by the following formula:

[0177] ;

[0178] In the formula, Ψo(T1) represents the sliding coefficient of variation of the o-th data in the quality dataset ZW within a fixed period T1, σZWo(T1) represents the standard deviation of the o-th data in the quality dataset ZW within a fixed period T1, μZWo(T1) represents the mean of the o-th data in the quality dataset ZW within a fixed period T1, and eo represents a non-zero constant.

[0179] S3. By introducing the variation enhancement factor (VAF), the sensitivity to different trend states is improved;

[0180] The mutation enhancement factor (VAF) is obtained using the following formula:

[0181] ;

[0182] In the formula, ln represents the logarithmic function, Ψo(T2) represents the sliding coefficient of variation of the o-th data in the quality dataset ZW within a fixed period T2, and Ψo(T3) represents the sliding coefficient of variation of the o-th data in the quality dataset ZW within a fixed period T3.

[0183] S4. Obtain the coefficient of variation Ψ;

[0184] Ψ = Ψo(T1) * VAF;

[0185] In the formula, Ψ represents the coefficient of variation, including the coefficient of variation of secondary settlement increment of foundation ΨAdz, the coefficient of variation of crack width micro-drift value ΨAcw, the coefficient of variation of stress rebound frequency deviation value ΨAre, and the coefficient of variation of component tilt angle gradient rate ΨAsp.

[0186] The abnormal mutation index fusion module performs nonlinear structural aggregation on the obtained secondary settlement increment variation coefficient ΨAdz, crack width micro-drift value variation coefficient ΨAcw, stress rebound frequency deviation value variation coefficient ΨAre, and component tilt angle gradient variation coefficient ΨAsp to obtain the abnormal mutation index AEI, as shown in Table 1.

[0187] The Abnormal Mutation Index (AEI) is obtained using the following formula:

[0188] ;

[0189] In the formula, ln represents the exponential function, A represents the nonlinear enhancement factor, and arctan represents the arctangent function;

[0190] Specific examples:

[0191] Table 1: Calculation table of Abnormal Mutation Index (AEI);

[0192]

[0193] Group 1, Group 2, Group 3, Group 4, and Group 5 represent the five sets of data collected.

[0194] The obtained abnormal mutation index AEI is compared with the preset abnormal mutation threshold Tyc to obtain the mutation status;

[0195] Mutation states are obtained by matching in the following way:

[0196] When the abnormal mutation index AEI is less than the abnormal mutation threshold Tyc*0.5, it indicates that the mutation state is stable and normal monitoring is required.

[0197] When the abnormal mutation threshold Tyc*0.5 ≤ abnormal mutation index AEI < abnormal mutation threshold Tyc, it indicates a reversible abnormality, localized intensified monitoring, and trend analysis is initiated.

[0198] When the abnormal mutation threshold Tyc ≤ abnormal mutation index AEI < abnormal mutation threshold Tyc*1.5, it indicates a precursor to a coordinated mutation and an early warning is issued.

[0199] When the abnormal mutation threshold Tyc*1.5 ≤ the abnormal mutation index AEI, it indicates a high-risk state of multi-source structural mutation, and the clinical detection mechanism is activated.

[0200] In this embodiment, by setting a fixed period T and constructing a multi-scale nested window, local variation analysis is performed, significantly overcoming the problem of "loss of trend details due to fixed period / single-scale judgment" in traditional methods. This method enables the system to identify long-period fluctuation trends at the macro scale and capture local disturbances at the micro scale, achieving multi-dimensional dynamic analysis of anomalous evolution. Traditional structural monitoring often assesses fluctuation amplitude based on the standard deviation under a fixed threshold, easily overlooking "gradual abrupt changes" signals. This system, by enhancing the local sliding variability, introduces a "variation enhancement factor (VAF)" to dynamically amplify the trend differences between parameters, allowing anomalous states to be captured and amplified early even without drastic numerical jumps.

[0201] The construction of the Abnormal Mutation Index (AEI) is no longer a conventional weighted synthesis or simple averaging. Instead, it combines nonlinear functions such as exponential functions and arctangent functions to deeply couple and model the coefficient of variation, thereby enhancing the response to the co-evolution of anomalies in various dimensions. This solves the traditional problem of "signal weakening after the fusion of multiple source indices" and significantly improves the system's analytical power for the evolution of anomalies in higher-order structures.

[0202] By comparing the Abnormal Mutation Index (AEI) with a preset abnormal mutation threshold and classifying it into four mutation states, continuous closed-loop control from "monitoring" to "response" is achieved. This mechanism can not only identify the abnormality itself, but also classify and judge its development direction and controllability, thereby enhancing the system's dynamic adaptability and response accuracy.

[0203] Multi-scale variation analysis and index fusion mechanisms enable the system to provide early warnings by detecting subtle deviations in trend fluctuations before risks become apparent, effectively reducing the probability of sudden accidents and enhancing the proactive safety management capabilities of structural operations. The system does not rely on fixed thresholds or single time windows; instead, it constructs a dynamic threshold and state determination framework based on the multi-dimensional characteristics of time series data. Therefore, it can flexibly adapt to different operating conditions, structural types, and varying environmental conditions, enhancing the system's versatility and scalability.

[0204] The four-level mutation status allows managers to initiate strategies such as enhanced monitoring, trend tracking, early repair, or emergency reinforcement as needed, enabling precise resource allocation and optimized management costs.

[0205] Example 4 is an explanation of Example 3. Please refer to the example provided. Figure 1 and Figure 6 Specifically: the driving factor identification module includes an acceleration identification unit and a mutation driving point determination unit;

[0206] The acceleration identification unit constructs evolutionary acceleration a by extracting the second-order change features of the anomalous mutation index AEI;

[0207] The evolutionary acceleration 'a' is obtained using the following formula:

[0208] ;

[0209] In the formula, AEI(t) represents the anomalous mutation index at time t, AEI(t+1) represents the anomalous mutation index at time t+1, AEI(t-1) represents the anomalous mutation index at time t-1, and Δt represents the time interval.

[0210] By introducing the risk growth inertia coefficient GX, the obtained evolution acceleration a is corrected to obtain the evolution acceleration factor EAF;

[0211] The risk growth inertia coefficient GX is obtained using the following formula:

[0212] ;

[0213] In the formula, AEI(t-3) represents the anomalous mutation index at time t-3, ae represents a positive number, and V(t) represents the first derivative of the anomalous mutation index AEI;

[0214] The evolution acceleration factor EAF is obtained by multiplying the risk growth inertia coefficient GX by the evolution acceleration a.

[0215] The mutation driving point determination unit calculates and obtains the evolution threshold Teaf based on the evolutionary fluctuation of the evolution acceleration factor EAF;

[0216] The evolution threshold Teaf is obtained using the following formula:

[0217] ;

[0218] In the formula, μEAF represents the mean of the evolution acceleration factor, σEAF represents the standard deviation of the evolution acceleration factor, and af represents the sensitivity coefficient;

[0219] The evolution acceleration factor (EAF) is compared with the evolution threshold (Teaf) to determine the risk status.

[0220] Risk status is obtained through the following methods:

[0221] When the evolution acceleration factor EAF ≤ the evolution threshold Teaf, it indicates a normal steady state;

[0222] When the evolution acceleration factor EAF is greater than the evolution threshold Teaf, it indicates a risky state, a potential hazard, and an early warning is issued.

[0223] In this embodiment, a dynamic indicator called "evolutionary acceleration α" is introduced into structural health risk identification. This expands the traditional judgment method, which is based solely on the current risk status or historical averages, into a trend identification logic centered on "second-order change characteristics in time series." This mechanism enables the system to identify the direction and speed of rapid risk growth trends, effectively compensating for the slow response of static thresholds to sudden evolutionary trends.

[0224] This invention designs a "risk growth inertia coefficient" as a "weighted correction term" for the upward trend of structural risk. It fully considers the cumulative growth trend and development inertia of the abnormal mutation index, so that when the system determines the potential mutation driving point, it can not only capture the current rate of change, but also trace the past risk accumulation process, thus enhancing the historical depth and foresight of the judgment.

[0225] By calculating the current threshold based on the dynamic distribution characteristics of the Evolution Acceleration Factor (EAF), a risk state discrimination mechanism that adapts to structures is realized. The system can dynamically adjust its sensitivity according to different structures and monitoring cycles to adapt to the risk assessment needs of different engineering scenarios. The system is no longer limited to judging whether there is an anomaly, but further locates the starting point of abnormal growth, that is, the "starting node of risk evolution". This capability enables the system to identify the "driving point" before a full-scale mutation state is reached, achieving accurate early warning and proactive intervention, significantly improving the depth of the system in identifying precursors of mutation.

[0226] Based on the identification method of evolution acceleration, the system can also be keenly aware of risk states that are "rapid in speed but small in magnitude". Especially in micro-change behaviors such as early crack expansion, foundation deformation accumulation and stress micro-drift, it can accurately identify potential trends and realize the "one step forward" of risk identification.

[0227] The dynamic judgment mechanism integrates current changes, historical inertia, and fluctuation range, enabling the system not only to determine "whether there is an anomaly," but also "whether the anomaly will continue and intensify," effectively reducing false alarms and missed alarms and improving the reliability of the early warning mechanism. Utilizing an evolutionary threshold adaptive mechanism, the system can dynamically set the sensitivity level based on the risk evolution characteristics of the monitored object itself, making it applicable to various engineering types such as bridges, tunnels, and high-rise buildings, enhancing the portability and universality of the system's deployment.

[0228] The system not only outputs the current risk level, but also identifies the "risk evolution direction in the next few steps" in advance. It can provide quantitative support and scientific suggestions for management behaviors such as structural reinforcement, adjustment of inspection frequency, and optimization of construction plan, and build an intelligent decision-making closed loop based on data evolution.

[0229] Example 5 is an explanation of Example 4. Please refer to the example provided. Figure 5 and Figure 6 Specifically: the early warning threshold dynamic determination module includes a joint index construction unit and a risk level classification unit;

[0230] The joint index construction unit jointly constructs the obtained evolution acceleration factor EAF and abnormal mutation index AEI to obtain the joint index ARIA;

[0231] The combined indicator ARIA is obtained using the following formula:

[0232] ;

[0233] In the formula, r1 represents the risk intensity adjustment factor, and r2 represents the risk acceleration adjustment factor.

[0234] The risk level classification unit extracts all values ​​of the joint indicator ARIA from historical periods to form a set Har;

[0235] Threshold extraction was performed on the set Har using the mean method and the standard deviation method to obtain the index threshold THR;

[0236] The threshold THR is obtained by summing the mean of the joint indicator ARIA with 0.5 times the standard deviation of the joint indicator ARIA.

[0237] The risk level is obtained by comparing the combined indicator ARIA with the indicator threshold THR.

[0238] Risk levels are obtained through matching in the following ways:

[0239] When the combined indicator ARIA < 0.5 * indicator threshold THR, it indicates the first risk level, a safe state.

[0240] When 0.5 * indicator threshold THr ≤ joint indicator ARIA < indicator threshold THr, it indicates the second risk level, and a controllable anomaly has occurred.

[0241] When the indicator threshold THr ≤ the joint indicator ARIA < 1.5 * indicator threshold THr, it indicates the third risk level, and the precursor to a sudden change has appeared.

[0242] When 1.5 * the indicator threshold THr ≤ the joint indicator ARIA, it indicates the fourth risk level, an abnormally high risk state.

[0243] The trend evolution report generation module identifies the trend of the time series composed of the Abnormal Mutation Index (AEI) and the Evolution Acceleration Factor (EAF), and constructs the time series curve of the Abnormal Mutation Index (AEI) and the acceleration curve of the Evolution Acceleration Factor (EAF).

[0244] Time series curve of Abnormal Mutation Index (AEI): The AEI curve is plotted according to the timestamp order, and the peak point, inflection point and risk level segment color band are marked on the graph.

[0245] The Evolutionary Accelerator Factor (EAF) acceleration curve includes the length of the continuously rising EAF segment and the time interval from the EAF peak to the mutation point.

[0246] Based on the time series curve of the Abnormal Mutation Index (AEI) and the acceleration curve of the Evolutionary Acceleration Factor (EAF), and combined with the risk level, an execution strategy is generated.

[0247] The execution strategy is obtained by matching in the following ways:

[0248] When the risk level is at the first level, it indicates that the situation is stable, and the execution strategy is: normal inspection and periodic sampling.

[0249] When the risk level is at level two, it indicates that there is a disturbance but no continuous evolution. The strategy to be implemented is to increase the monitoring frequency and activate the local trend tracking algorithm.

[0250] When the risk level is at level three, it indicates that cumulative instability is forming. The strategy to be implemented is: prepare a local maintenance plan, re-inspect the structure, and assess whether it will affect the function.

[0251] When the risk level is at level four, it indicates a critical evolution and an impending mutation. The appropriate strategies are: structural unloading, personnel evacuation, and work stoppage.

[0252] This embodiment no longer treats the Abnormal Mutation Index (AEI) and Evolutionary Acceleration Factor (EAF) in isolation, but instead constructs a unified joint risk index (ARIA) to reflect the comprehensive performance of risk status in terms of both intensity and rate of change. This mechanism effectively overcomes the problems of narrow risk expression dimensions and fragmented information in traditional systems, making the system's judgment more holistic, trend-oriented, and developmental.

[0253] By statistically analyzing the ARIA (Advanced Risk Assessment) distribution over historical monitoring periods, the system automatically extracts assessment thresholds and classifies risks into four levels. This achieves dynamic, self-calibrated, and scenario-adaptive risk level classification, overcoming the problems of "failure in different structural scenarios" or "excessive warnings / delayed alarms" under fixed threshold modes. For the first time, the system establishes a direct mapping logic between risk identification, level judgment, and engineering response operations, clearly defining the corresponding handling schemes for each risk level (such as initiating trend algorithms, structural unloading, and personnel evacuation). This fills the long-standing gap in structural monitoring where "alarms are only triggered but no response is provided," forming an intelligent, interconnected closed loop of "perception—identification—response."

[0254] The system constructs a quantitative assessment of the pace of anomalous accelerated evolution by analyzing the "length of the continuous rising segment" and the "time from the peak to the mutation point" in the Evolutionary Acceleration Factor (EAF) curve. This effectively captures non-drastic change patterns that are difficult to identify, such as "slow-speed acceleration abrupt changes" or "gradual cumulative anomalies," and significantly improves the system's adaptability to different types of anomalous evolution paths.

[0255] By establishing a composite judgment model using multi-dimensional features, the system can effectively identify boundary-type risk states that are high in intensity but slow in evolution, and low in intensity but with extremely high acceleration, avoiding the occurrence of risk "missed judgment" or "delayed judgment". Each level of risk corresponds one-to-one with the execution strategy, enabling managers to adopt differentiated resource allocation, monitoring frequency adjustment, or construction rhythm control according to different risk levels, achieving refined risk response management and optimal resource allocation.

[0256] Example 6: An intelligent assessment method for engineering quality and safety risks. Please refer to... Figure 2 Specifically, it includes the following steps:

[0257] Step 1: The time series data acquisition and preprocessing module collects engineering quality data through sensors, fits it into the original dataset W, and performs preprocessing to obtain the quality dataset ZW.

[0258] Step 2: The data variation analysis module calculates the degree of local variation of the data in the quality dataset ZW within a fixed period T, and obtains the coefficient of variation Ψ.

[0259] Step 3: The abnormal mutation index fusion module fuses the obtained coefficients of variation Ψ to construct the abnormal mutation index AEI;

[0260] Step 4: The driving factor identification module identifies the acceleration of the abnormal mutation index AEI, determines whether there are mutation driving points in the time series of the abnormal mutation index AEI, obtains the evolution acceleration factor EAF, and identifies potential hidden dangers.

[0261] Step 5: The early warning threshold dynamic judgment module jointly judges the abnormal mutation index AEI and the evolution acceleration factor EAF based on historical statistics to obtain the risk level.

[0262] Step Six: The Trend Evolution Report Generation Module generates an execution strategy based on the acquired Abnormal Mutation Index (AEI), Evolution Acceleration Factor (EAF), and Risk Level.

[0263] In this embodiment, the method utilizes six consecutive steps—from raw data acquisition and anomaly trend identification to risk level classification and execution strategy output—to establish a complete structural safety assessment process, from "bottom-level perception" to "top-level decision-making." Unlike traditional monitoring systems that "monitor without judging" or "judge without controlling," this solution constructs a closed-loop logic covering "perception—fusion—identification—early warning—response," achieving truly intelligent structural risk assessment.

[0264] By integrating the temporal evolution characteristics of multidimensional structural state parameters and constructing a trend-sensitive composite index through variation analysis and anomaly fusion, this approach solves the problems of "fragmented judgment of various anomaly features" and "limitations of static thresholds" in traditional methods, comprehensively improving the systematicness and sensitivity of risk identification. By identifying the acceleration of the anomalous mutation index and determining its underlying driving point, the system can proactively identify the source and timing of the accelerated accumulation of structural risks, effectively addressing the gap in identifying the "precursor stage of mutation" in existing methods and enabling proactive judgment of engineering hazards.

[0265] The joint indicator ARIA is constructed based on historical data, and risk level is stratified by combining dynamic statistical thresholds. This enables the assessment system to adapt to different structural environments and states, breaking the problems of "generalization failure" and "poor scenario adaptability" in the traditional fixed early warning threshold mode.

[0266] Through anomaly trend identification and driver point location mechanisms, the system can identify the evolutionary tendency of problems before they develop into overt disasters, enabling early detection and response to potential hazards and improving the initiative and stability of engineering management. The method can effectively identify abrupt, gradual, periodic, or cumulative risk evolution paths, significantly improving the system's ability to distinguish between various types of structural instability evolution modes and expanding its adaptability to complex structural systems. Different intervention strategies are tailored to different risk levels, enabling the monitoring system not only to determine "whether it is abnormal" but also "how to respond," providing quantitative basis for resource allocation, inspection schedule adjustments, and reinforcement plan implementation.

[0267] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An engineered quality safety risk intelligent assessment system, characterized in that: The method comprises a time series data collection and preprocessing module, a data variation analysis module, an abnormal mutation index fusion module, a driving factor identification module, a warning threshold dynamic determination module and a trend evolution report generation module. The time series data collection and preprocessing module collects engineering quality data through a sensor, fits the data into an original data set W, and pre-processes the data to obtain a quality data set ZW. The data variation analysis module calculates the local variation degree of the data in the quality data set ZW within a fixed period T to obtain a variation coefficient Ψ. The data variation analysis module extracts the local variation intensity of each parameter in the quality data set ZW to obtain a variation coefficient Ψ, including a foundation secondary settlement increment variation coefficient ΨAdz, a crack width micro-drift value variation coefficient ΨAcw, a stress rebound frequency deviation value variation coefficient ΨAre and a component inclination gradual rate variation coefficient ΨAsp, and the obtaining steps are as follows. S1, set a fixed period T, and construct a multi-scale nested window: T1=T, T2=2T, T3=0.5T. S2, for the oth data ZWo in the quality data set ZW, calculate and obtain a sliding variation coefficient Ψo within the fixed period T1. The sliding variation coefficient Ψo is obtained by the following formula: ； In the formula, Ψo(T1) represents the sliding variation coefficient of the oth data in the quality data set ZW within the fixed period T1, σZWo(T1) represents the standard deviation of the oth data in the quality data set ZW within the fixed period T1, μZWo(T1) represents the mean of the oth data in the quality data set ZW within the fixed period T1, and eo represents a non-zero constant. S3, the sensitivity of different trend states is improved by introducing a variation enhancement factor VAF. The variation enhancement factor VAF is obtained by the following formula: ； In the formula, ln represents a logarithmic function, Ψo(T2) represents the sliding variation coefficient of the oth data in the quality data set ZW within the fixed period T2, and Ψo(T3) represents the sliding variation coefficient of the oth data in the quality data set ZW within the fixed period T3. S4, obtain the variation coefficient Ψ. Ψ = Ψo(T1) VAF; In the formula, Ψ represents the variation coefficient. The abnormal mutation index fusion module fuses the obtained variation coefficient Ψ to construct an abnormal mutation index AEI. The abnormal mutation index AEI is obtained by the following formula: ； In the formula, ln represents an exponential function, A represents a non-linear enhancement factor, and arctan represents an inverse tangent function. The obtained abnormal mutation index AEI is compared with a preset abnormal mutation threshold Tyc to obtain a mutation state. The driving factor identification module identifies the acceleration of the abnormal mutation index AEI to determine whether there is a mutation driving point in the time series of the abnormal mutation index AEI, obtains an evolution acceleration factor EAF, and identifies potential hidden dangers. The driving factor identification module comprises an acceleration identification unit and a mutation driving point determination unit. The acceleration identification unit extracts the second-order change feature of the abnormal mutation index AEI to construct an evolution acceleration a. The evolution acceleration a is obtained by the following formula: ； In the formula, AEI(t) represents the abnormal mutation index at time t, AEI(t+1) represents the abnormal mutation index at time t+1, AEI(t-1) represents the abnormal mutation index at time t-1, and Δt represents the time interval; The evolution acceleration a obtained is modified by introducing a risk growth inertia coefficient GX to obtain an evolution acceleration factor EAF; The risk growth inertia coefficient GX is obtained by the following formula: ； In the formula, AEI(t-3) represents the abnormal mutation index at time t-3, ae represents a positive number, and V(t) represents the first derivative of the abnormal mutation index AEI; The evolution acceleration factor EAF is obtained by the product of the risk growth inertia coefficient GX and the evolution acceleration a; The early warning threshold dynamic judgment module jointly judges the abnormal mutation index AEI and the evolution acceleration factor EAF based on the adaptive dynamic threshold of historical statistics to obtain a risk level; The trend evolution report generation module generates an execution strategy according to the obtained abnormal mutation index AEI, evolution acceleration factor EAF and risk level.

2. The system of claim 1, wherein: The time series data acquisition and preprocessing module includes a parameter acquisition and fitting unit and a data preprocessing unit; The parameter acquisition and fitting unit collects engineering quality data through sensors deployed at the position of the engineering structure, including the secondary settlement increment Adz of the foundation, the crack width micro-drift value Acw, the stress rebound frequency deviation value Are and the member inclination gradual rate Asp, and fits them into an original data set W; The secondary settlement increment Adz of the foundation is collected and obtained by a multipoint displacement meter and a fiber optic displacement sensor: The crack width micro-drift value Acw is collected and obtained by a crack meter and a resistance crack width sensor: The stress rebound frequency deviation value Are is collected and obtained by a stress wave frequency sensor and a high-frequency structure response sensor: The member inclination gradual rate Asp is collected and obtained by a MEMS inclination sensor and a fiber optic inclinometer: The data preprocessing unit cleans and normalizes the original data set W to obtain a quality data set ZW; Cleaning identifies abnormal points in the original data set W by using adjacent jump judgment and local linear fitting residual threshold method; Normalization processes the data in the original data set W by using the maximum and minimum normalization method to obtain the quality data set ZW; ； In the formula, ZWo represents the oth data in the quality data set ZW, Wo represents the oth data in the original data set W, minWo represents the valley value of the oth data in the original data set W, and maxWo represents the peak value of the oth data in the original data set W.

3. The system of claim 2, wherein: The abnormal mutation index fusion module nonlinearly aggregates the secondary settlement increment variation coefficient ΨAdz of the foundation, the crack width micro-drift value variation coefficient ΨAcw, the stress rebound frequency deviation value variation coefficient ΨAre and the member inclination gradual rate variation coefficient ΨAsp to obtain the abnormal mutation index AEI; The mutation state is matched and obtained by the following way: When the abnormal mutation index AEI < abnormal mutation threshold Tyc 0.5, it indicates that the mutation state is stable, and normal monitoring is required. When the abnormal mutation threshold Tyc 0.5 ≤ Abnormal Mutation Index AEI < Abnormal Mutation Threshold Tyc indicates reversible abnormality, local encryption monitoring, and trend analysis is started. When the abnormal mutation threshold Tyc≤abnormal mutation index AEI<abnormal mutation threshold Tyc 1.5, indicating a synergistic mutation precursor, issuing an early warning; When the abnormal mutation threshold Tyc 1.5≤ abnormal mutation index AEI, indicating a high-risk state of polygenetic structure mutation, starting the prompt inspection mechanism.

4. The system of claim 3, wherein: The mutation driving point judgment unit calculates and obtains an evolution threshold Teaf according to the evolution fluctuation of the evolution acceleration factor EAF; The evolution threshold Teaf is obtained by the following formula: ； In the formula, μEAF represents the mean of the evolution acceleration factor, σEAF represents the standard deviation of the evolution acceleration factor, and af represents the sensitivity coefficient; The evolution acceleration factor EAF is compared with the evolution threshold Teaf to determine the risk state; The risk state is obtained in the following manner: When the evolution acceleration factor EAF is less than or equal to the evolution threshold Teaf, it indicates a normal stable state; When the evolution acceleration factor EAF is greater than the evolution threshold Teaf, it indicates a risk state, indicating that a hidden danger exists and a warning is needed.

5. The system of claim 4, wherein: The early warning threshold dynamic determination module includes a joint index construction unit and a risk level division unit; The joint index construction unit constructs the evolution acceleration factor EAF and the abnormal mutation index AEI obtained to obtain the joint index ARIA; The joint index ARIA is obtained through the following formula: ； In the formula, r1 represents the risk intensity adjustment factor, and r2 represents the risk acceleration adjustment factor.

6. The system of claim 5, wherein: The risk level division unit extracts all values of the joint index ARIA from historical periods to form a set Har; The threshold value of the index THr is obtained by using the mean value method and the standard deviation method to extract the threshold value of the set Har; The index threshold THr is obtained by summing the mean value of the joint index ARIA and 0.5 times the standard deviation of the joint index ARIA; The joint index ARIA is compared with the index threshold THr to obtain the risk level; The risk level is matched and obtained in the following manner: when the combined indicator ARIA < 0.5 when the indicator threshold THr, indicates a first risk level, a safe state; when 0.5 an indicator threshold THr, the second risk level is indicated, a controllable abnormality occurs; when the indicator threshold THr < the combined indicator ARIA < 1.5 when the indicator threshold THr, a third risk level is indicated, a precursor to a mutation appears; When 1.5 An index threshold THr < joint index ARIA indicates a fourth risk level, an abnormally high risk state.

7. The system of claim 6, wherein: The trend evolution report generation module performs trend identification on the time series of the abnormal mutation index AEI and the evolution acceleration factor EAF, and constructs an abnormal mutation index AEI time series curve and an evolution acceleration factor EAF acceleration curve. The abnormal mutation index AEI time series curve: the abnormal mutation index AEI curve is drawn according to the time stamp order, and the peak point, inflection point and risk level segmented color band are marked in the graph; The evolution acceleration factor EAF acceleration curve includes the length of the continuous rising section of the evolution acceleration factor EAF and the time interval from the peak value of the evolution acceleration factor EAF to the mutation point; According to the abnormal mutation index AEI time series curve and the evolution acceleration factor EAF acceleration curve, an execution strategy is generated in combination with the risk level; The execution strategy is matched and obtained in the following manner: When in the first risk level, it indicates a stable state, and the execution strategy is: normal inspection, periodic sampling; When in the second risk level, it indicates that there is disturbance and no continuous evolution, and the execution strategy is: increase the monitoring frequency and start the local trend tracking algorithm; When in the third risk level, it indicates that cumulative instability is being formed, and the execution strategy is: local maintenance plan preparation, structure re-inspection, and evaluation of whether it affects the function; When in the fourth risk level, it indicates critical evolution and impending mutation, and the execution strategy is: structure unloading, personnel evacuation, and shutdown treatment.

8. An engineering quality safety risk intelligent assessment method applied to the engineering quality safety risk intelligent assessment system of any one of claims 1-7, characterized in that: The method comprises the following steps: Step one, the time series data acquisition and preprocessing module collects engineering quality data through sensors, fits it into an original data set W, and performs preprocessing to obtain a quality data set ZW; Step two, the data variation analysis module calculates the local variation degree of the data in the quality data set ZW within a fixed period T to obtain the variation coefficient Ψ; Step three, the abnormal mutation index fusion module fuses the obtained coefficient of variation Ψ to construct an abnormal mutation index AEI; Step four, the driving factor identification module identifies the acceleration of the abnormal mutation index AEI, judges whether there is a mutation driving point in the time series of the abnormal mutation index AEI, obtains an evolution acceleration factor EAF, and identifies potential hidden dangers; Step five, the early warning threshold dynamic judgment module jointly judges the abnormal mutation index AEI and the evolution acceleration factor EAF based on the adaptive dynamic threshold of historical statistics to obtain a risk level; Step six, the trend evolution report generation module generates an execution strategy according to the obtained abnormal mutation index AEI, evolution acceleration factor EAF and risk level.

Images