Prefabricated building construction safety risk dynamic early warning system, early warning method and storage medium

Abstract

The application discloses a kind of assembled building construction safety risk dynamic early warning system, early warning method, storage medium;It belongs to the technical field of building construction safety analysis software, and its technical points are in that S100, the stable force, disturbance force, state variable aggregate observation value u obs‑1~ u obs‑N 、v obs‑1~ v obs‑N 、X obs‑1~ X obs‑N Of successive N preset periods are acquired;S200, mapping parameter a u 、b u 、a v 、b v 、a x 、b x Are calculated;S300, bifurcation set discriminant △ N And stable force value u N Of the current period to be evaluated are calculated, and the safety state of the current period to be evaluated is risk early warning according to △ N 、u N .Using the technical scheme of the present application provides a quantifiable analysis method for construction safety management.

Description

Technical Field

[0001] This invention relates to the field of construction safety analysis software technology, and more specifically, to a dynamic early warning system, early warning method, and storage medium for safety risks in prefabricated building construction. Background Technology

[0002] Prefabricated building construction exhibits significantly different procedural coupling and spatial dependence compared to traditional cast-in-place construction in its component transportation, hoisting, assembly, and joint connection processes. Under the interaction of multiple factors, safety risks can easily accumulate and ultimately lead to accidents. In complex conditions such as high-altitude hoisting and multi-trade cross-operations, the construction system displays typical characteristics of dynamic evolution and nonlinear mutation, thus making research on dynamic risk evolution analysis increasingly urgent.

[0003] In recent years, catastrophe theory, as an effective tool for studying how "continuous quantitative change leads to discontinuous qualitative change" in a system, has begun to find initial applications in the field of engineering safety. For example:

[0004] Reference 1: “Zhang Longshen, Chen Weike, Gao Shuang, Research on Construction Risk Assessment Based on Factor Analysis and Mutation Series Method [J]. Value Engineering. 2020, 39(05): 98-100” introduces the mutation series method into construction risk assessment.

[0005] Reference 2: “Li Jue, Han Mei. Cusp catastrophe model and its application in the mechanism of building collapse accidents [J]. Journal of Railway Science and Engineering, 2022, 19(09): 2766-2775” constructed a cusp catastrophe model for building collapse accidents.

[0006] Reference 3: "Chen Weike, Ding Yu, Wu Xiaoyan. Three-state evolution mechanism of building construction system under the perspective of system thinking and catastrophe [J]. Journal of Systems Science, 2020, 28(02): 49-53" discusses the three-state evolution mechanism of building construction system.

[0007] However, existing construction safety risk assessment methods are mostly static evaluations or geared towards single accident types, making it difficult to capture the sudden qualitative changes in a system under continuous quantitative changes. Although catastrophe theory has been initially applied to the field of construction safety, a quantitative analysis method capable of systematically describing the evolution of construction safety risks from quantitative to qualitative changes has not yet been developed, and an operational, graded early warning method is also lacking. Summary of the Invention

[0008] The purpose of this invention is to address the shortcomings of the prior art by providing a dynamic early warning system for safety risks in prefabricated building construction.

[0009] Another objective of this invention is to provide a dynamic early warning method for safety risks in prefabricated building construction.

[0010] Another object of the present invention is to provide a storage medium.

[0011] A method for analyzing the abrupt evolution of safety risks in prefabricated building construction includes the following steps:

[0012] S100, obtain aggregated observations u of the stabilizing force, disturbance force, and state variables for N consecutive preset periods. obs-1~ u obs-N v obs-1~ v obs-N X obs-1~ X obs-N The Nth preset period is the current evaluation period;

[0013] For the i-th preset period, the measured values ​​of each index in the stability force index set, disturbance force index set, and state variable index set are collected, standardized, and then weighted and summed to obtain u. obs-i v obs-i X obs-i ;

[0014] S200, Calculate mapping parameter a u b u a v b v a x b x ;

[0015] The Levenberg-Marquardt algorithm is used to iteratively solve for the mapping parameter a, with the objective of minimizing the sum of squared residuals. u b u a v b v a x b x : ;

[0016] S300, Calculate the bifurcation set discriminant Δ for the current evaluation period. N and the stabilizing force value u N According to △ N u N Provide risk warnings for the safety status of the current evaluation period;

[0017] △ N =4 (a u ·u obs-N +b u ) 3 +27 (a) v ·v obs-N +b v ) 2 ;

[0018] u N =a u ·u obs-N +b u ;

[0019] If △ N >0 and u N >0, the current evaluation period is in a safe state;

[0020] If △ N >0 and u N ≤0, the current evaluation period is in an unstable alert state;

[0021] If △ N ≤0, the current evaluation period is in an unstable state.

[0022] Furthermore, the S300 also includes:

[0023] If the current evaluation period is in a safe state, continue with standard construction measures;

[0024] If the current evaluation period is in an unstable alert state, organize special risk identification and assessment, reduce the intensity of high-altitude operations, mark, isolate or replace unqualified components, and reinforce deformed or loose supports;

[0025] If the current evaluation period is in an unstable state, immediately activate the emergency plan, suspend work, or partially suspend work.

[0026] Furthermore, S100 includes the following sub-steps:

[0027] S101, collects the measured values ​​of each index in the set of stability force index, set of disturbance force index, and set of state variable index according to a preset cycle;

[0028] The stability index set includes: personnel risk identification capability, emergency plan management capability, component quality, and temporary support strength;

[0029] The set of disturbance indicators includes: human-machine operation at height, equipment failure, process complexity, and personnel fatigue;

[0030] The set of state variable indicators includes: weather and climate, civilized construction, and site layout;

[0031] S102, calculate the standardized values ​​of each index in the stability force index set, disturbance force index set, and state variable index set;

[0032] The standardized value of the indicator = (the measured value of the indicator - the benchmark value of the indicator) / the characteristic scale coefficient of the indicator;

[0033] S103, use DEMATEL to determine the weights of each index: the sum of the weights of each index in the stability index set is 1, the sum of the weights of each index in the disturbance index set is 1, and the sum of the weights of each index in the state variable index set is 1.

[0034] S104, weighted summation yields the aggregated observation value u of the ith preset period's stability force, disturbance force, and state variables. obs-i v obs-i X obs-i .

[0035] Furthermore, N≥20.

[0036] A dynamic early warning system for safety risks in prefabricated building construction, comprising:

[0037] Data acquisition module: used to acquire measured values ​​of the stability force index set, disturbance force index set, and state variable index set within N consecutive preset periods;

[0038] Standardization calculation module: used to calculate the standardized values ​​of each indicator;

[0039] Aggregate calculation module: used to obtain aggregated observations u of stabilizing force, disturbance force, and state variables for N consecutive preset periods by weighted summation based on DEMATEL weights. obs-1~ u obs-N v obs-1~ v obs-N X obs-1~ X obs-N ;

[0040] Parameter calibration module: used to solve for the mapping parameter b using the Levenberg-Marquardt algorithm. u a v b v a x b x ;

[0041] Discriminant Calculation Module: Used to calculate the bifurcation set discriminant Δ for the current evaluation period. N and stabilizing force value u N ;

[0042] Early warning output module: used to output based on Δ N and u N Output the warning level.

[0043] Furthermore, the standardized value of each indicator = (the measured value of the indicator - the benchmark value of the indicator) / the characteristic scale coefficient of the indicator.

[0044] Furthermore, the parameter calibration module operates as follows:

[0045] The Levenberg-Marquardt algorithm is used to iteratively solve for the mapping parameter a, with the objective of minimizing the sum of squared residuals. u b u a v b v a x b x : .

[0046] Furthermore, the working method of the discriminative calculation module is as follows:

[0047] △ N =4 (a u ·u obs-N +b u ) 3 +27 (a) v ·v obs-N +b v )2;

[0048] u N =a u ·u obs-N +b u .

[0049] Furthermore, the working method of the early warning output module is as follows:

[0050] If △ N >0 and u N >0, the current evaluation period is in a safe state;

[0051] If △ N >0 and u N ≤0, the current evaluation period is in an unstable alert state;

[0052] If △ N ≤0, the current evaluation period is in an unstable state.

[0053] A storage medium having a computer program stored thereon, which, when executed by a processor, implements the aforementioned method.

[0054] The advantages of this application are:

[0055] First, the technical problem this application aims to solve is the inability to dynamically analyze and provide early warnings of safety risks during prefabricated building construction. To address this problem, this application employs the following method:

[0056] (1) Construct a three-category index system of stability force, disturbance force and state variable, and establish a quantitative mapping with the cusp catastrophe model by combining DEMATEL centrality weighting;

[0057] (2) Using field measurement data for N consecutive preset periods, with the goal of minimizing the sum of squared residuals of the equilibrium equation, the Levenberg-Marquardt algorithm is used to calibrate the mapping parameter a. u b u a v b v a x b x ;

[0058] (3) Using u obs-N v obs-N a u b u a v b v a x b x To solve △ N and u N Based on △ N and u N This is used to divide the safe and controllable zone (safe state), the instability warning zone (instability warning state), and the bistable abrupt change zone (instability state).

[0059] Second, based on 24 weeks of data verification from actual prefabricated projects, the early warning results obtained using the early warning method proposed in this application are 87.5% consistent with the on-site safety records. This provides quantifiable theoretical tools and engineering methods for the safety management of prefabricated building construction, which helps to identify early signs of sudden changes and reduce the accident rate. Attached Figure Description

[0060] The present invention will be further described in detail below with reference to the embodiments shown in the accompanying drawings, but this does not constitute any limitation on the present invention.

[0061] Figure 1 This is the architecture design diagram of the tip model of this application.

[0062] Figure 2 This is the quantitative standard rule for the standardization of indicators in this application.

[0063] Figure 3 It's 24 weeks. obs ,v obs ,X obs Result image.

[0064] Figure 4 This is a diagram showing the evolution of safety risks in prefabricated building construction over 24 weeks.

[0065] Figure 5 This is a dynamic evolution trajectory diagram of the risks in this application.

[0066] Figure 6It is a three-dimensional evolution and mutation trajectory diagram of the safe state during the unstable transition period W1-W4.

[0067] Figure 7 It is a three-dimensional evolution and mutation trajectory diagram of the safe state during the stable fluctuation period W5-W13.

[0068] Figure 8 It is a three-dimensional evolution and mutation trajectory diagram of the safety status during the high-risk period W14-W16.

[0069] Figure 9 It is a three-dimensional evolution and mutation trajectory diagram of the safe state during the recovery and adjustment period from W17 to W24.

[0070] Figure 10 This is a diagram showing the response measures and targeted intervention results for different security zones.

[0071] Figure 11 This is a flowchart of the dynamic early warning method for construction safety risks of prefabricated buildings proposed in this application. Detailed Implementation

[0072] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments.

[0073] The physical meanings of the symbols used below are explained as follows:

[0074] u obs : Aggregate observations of stabilizing forces.

[0075] v obs : Aggregated observations of disturbance forces.

[0076] X obs : Aggregate observations of state variables.

[0077] X1: Personnel risk identification capability.

[0078] X7: Emergency response plan management capability.

[0079] X9: Component quality.

[0080] X11: Temporary support strength.

[0081] X4: Human-machine operation at height.

[0082] X10: Equipment malfunction.

[0083] X15: Process complexity.

[0084] X3: Staff fatigue.

[0085] X17: Weather and climate.

[0086] X18: Civilized construction.

[0087] X19: Site layout.

[0088] V i，actual : The actual measured value of the i-th indicator in the current period.

[0089] V i，bench : The critical safety threshold benchmark value of the i-th indicator.

[0090] R i : The characteristic scale coefficient of the i-th index.

[0091] a u b u a v b v a x b x : Mapping parameters.

[0092] u obs-i v obs-i X obs-i The aggregated observations of the stabilizing force, the aggregated observations of the disturbance force, and the aggregated observations of the state variables are calculated in the i-th period.

[0093] N: The total number of cycles obtained, generally greater than 20.

[0094] u: Stabilizing force.

[0095] v: Disturbance force.

[0096] △: Bifurcation set discriminant.

[0097] △ N : The discriminant of the bifurcation set for the current evaluation period.

[0098] u N The stability of the current evaluation period, u N =a u ·u obs-N +b u .

[0099] Example 1

[0100] A dynamic early warning system for safety risks in prefabricated building construction includes the following steps:

[0101] S100: Obtain the standardized values ​​of each indicator in the stability force indicator set, disturbance force indicator set, and state variable indicator set on a weekly (or daily) basis. Then, use the DEMATEL centrality weighting to perform a weighted summation to obtain the weekly aggregated stability force observation value u. obsAggregated observations of disturbance dynamics v obs Aggregated observations of state variables X obs .

[0102] Taking a prefabricated frame-shear wall structure project (prefabrication rate 50%) as an example.

[0103] (1) Determination of each indicator set

[0104] Figure 1 The model architecture diagram of this application is shown. The macroscopic situation of construction safety can be abstracted into a single state variable x, which comprehensively reflects the current state of the system—safe, alert, or unstable—and characterizes the evolution of construction safety risks. The set of state variable indicators includes: weather and climate X17, civilized construction X18, and site layout X19.

[0105] Stability u (intrinsic vulnerability control variable): reflects the inherent risk level of the construction system itself in terms of structure, management, and technology. The stability index set includes: personnel risk identification ability X1, emergency plan management ability X7, component quality X9, and temporary support strength X11, which together determine the inherent stability of the system.

[0106] Disturbance force v (external disturbance intensity control variable): reflects the degree of impact of external factors such as environment, personnel behavior, and equipment failure on the system. The disturbance force index set includes: human-machine operation at height x4, equipment failure x10, process complexity x15, and personnel fatigue x3. These factors can amplify system fluctuations and reduce safety margins.

[0107] (2) For any i-th indicator, its standardized value S i The method to obtain it is as follows:

[0108] S i =(V i，actual -V i，bench ) / R i ;

[0109] Among them, V i，actual V represents the actual measured value of the i-th indicator in the current period; i，bench R is the critical safety threshold benchmark value for the i-th indicator; i is the feature scaling coefficient of the i-th index, used to normalize the difference to the model fit interval.

[0110] Positive indicator: S i When the value is greater than 0, the actual value is higher than the benchmark, contributing positively to the stabilizing force u; S i When the value is less than 0, the actual value is lower than the benchmark and is in an unsafe state.

[0111] Negative indicator: S iWhen the value is greater than 0, the actual value exceeds the baseline upper limit, increasing the risk and making a positive contribution to the disturbance force v; S i When the risk is less than 0, the risk is within a controllable range.

[0112] (3) The final result was formed after discussion and confirmation by the research group through a questionnaire survey. Figure 2 The quantitative standard rules are shown below. Specific explanations are as follows:

[0113] (3.1) The set of stability indexes includes:

[0114] Personnel risk identification ability is assessed on a 100-point scale, with a baseline value V. bench =75 points, feature scale coefficient R=25, which is a positive indicator;

[0115] Emergency response plan management capability is assessed using a 100-point scoring system for emergency response drills, with a baseline value of V. bench =85 points, feature scale coefficient R=15, which is a positive indicator;

[0116] Component quality, measured by the incoming inspection pass rate, is based on the benchmark value V. bench =98%, characteristic scaling coefficient R=2%, which is a positive indicator;

[0117] The strength of temporary supports is defined as the minimum percentage of the measured design values ​​among all temporary supports, with a reference value V. bench =100%, characteristic scale coefficient R=5%, which is a positive indicator.

[0118] (3.2) The set of disturbance indices includes:

[0119] The percentage of high-risk work hours for human-machine high-altitude operations is calculated as a baseline value V. bench =30%, feature scale coefficient R=20%, which is a negative indicator;

[0120] Equipment failures are counted in weekly failures (times / week), with a baseline value V. bench =1, feature scale coefficient R=3, which is a negative indicator;

[0121] Process complexity is scored using a process rationality rating (0-10 points), with a baseline value V. bench =5, feature scale coefficient R=3, which is a negative indicator;

[0122] Personnel fatigue is measured by a subjective fatigue score (1-7 points), with a baseline value V. bench =4, feature scale coefficient R=3, which is a negative index.

[0123] (3.3) The set of state variable indices includes:

[0124] Weather and climate are measured by the percentage of unfavorable weather working hours (%), with a baseline value V. bench=5%, feature scale coefficient R=10%, which is a positive indicator;

[0125] Civilized construction is scored using a specific evaluation system, with a baseline value of V. bench =80, feature scale coefficient R=20, which is a positive indicator;

[0126] The site layout is based on the hazard source rectification closure rate (%), with a benchmark value V. bench =95%, characteristic scale coefficient R=5%, which is a positive indicator.

[0127] based on Figure 2 The quantitative standards shown have the following weights for each indicator:

[0128] Stabilizing force U: X1 (0.469), X7 (0.125), X9 (0.226), X11 (0.181);

[0129] Disturbance force V: X3 (0.215), X4 (0.482), X10 (0.205), X15 (0.098);

[0130] State variables X: X17 (0.470), X18 (0.259), X19 (0.271).

[0131] Figure 3 This indicates the 24th week of the U obs ,v obs ,X obs result.

[0132] S200, Calculate mapping parameter a u b u a v b v a x b x ;

[0133] based on Figure 1 A cusp catastrophe model for the safety risks of prefabricated building construction is established, with the potential function being: V(x) = 0.25x 4 +0.5ux 2 +vx.

[0134] The equilibrium curve obtained from the potential function is: dV / dx = x 3 +ux+v=0 (equilibrium equation).

[0135] Based on this, a linear mapping is established between abstract control variables, state variables, and aggregated observations:

[0136] u=a u u obs +b u v=a vv obs +b v x=a x X obs +b x ;

[0137] Equilibrium equation: x 3 +ux+v=0, meaning that theoretically, for u in 24 weeks... obs ,v obs ,X obs All should satisfy the following relationship:

[0138] (a) x X obs +b x ) 3 +(a u u obs +b u (a) x X obs +b x )+a v v obs +b v =0.

[0139] However, due to errors, the right side of the equation is not strictly zero. Therefore, using the aggregated observation sequence from weeks 1 to 20 as the training set, and aiming to minimize the sum of squared residuals of the equilibrium equation, the Levenberg-Marquardt algorithm is used to iteratively optimize the six mapping parameters.

[0140] That is, calculate the mapping parameter a u b u a v b v a x b x The following method shall be adopted:

[0141] Using a series of aggregated observations over N consecutive weeks {(u obs-1 v obs-1 X obs-1}、……{(u obs-N v obs-N X obs-N Using the observed values ​​as the training set, and with the objective of minimizing the sum of squared residuals of the equations expressed in the form of the observed values, the Levenberg-Marquardt algorithm is used to iteratively optimize the six mapping parameters to obtain the calculated a. u b u a v b v a x b x ;

[0142] The square of the residuals in the equation expressed in terms of observed values ​​is: [(a x X obs-1 +b x ) 3 +(a u u obs-1 +b u (a) x X obs-1 +b x )+a v v obs-1 +b v ] 2 +……[(a x X obs-N +b x ) 3 +(a u u obs-N +b u (a) x X obs-N +b x )+a v v obs-N +b v ] 2 .

[0143] Will Figure 3 By solving the data in the table, we can obtain the project's a. u =0.848、b u = 0.154, a v =0.033、b v = 0.032, a x =1、b x = 0.288.

[0144] That is, the calculated variable mapping relationship is as follows:

[0145] u=0.848u obs 0.154;

[0146] v=0.033v obs 0.032;

[0147] x=X obs 0.288.

[0148] S300, Calculate the bifurcation set discriminant Δ for the current evaluation period. N (The Nth preset period is the current period to be evaluated), according to △ N Provide risk warnings for the safety status of the current evaluation period;

[0149] Singularity condition: d 2 V / dx 2 =3x 2 +u=0, that is, u=-3x 2 .

[0150] Let u=-3x 2 Substituting into the equilibrium equation x 3 +ux+v=0, therefore: v=2x 3 .

[0151] From u=-3x 2 v=2x 3 Therefore, u = -3(v / 2) 2 / 3 Two cubes on both sides, we have: u 3 =-27v 2 / 4(4u) 3 +27v 2 =0).

[0152] That is, the bifurcation set equation is: 4u 3 +27v 2 =0.

[0153] When Δ=4u 3 +27v 2 When Δ < 0, the system is in the abrupt change region, and the equilibrium equation has three real roots (two of which are stable and one is unstable). Small perturbations may cause a sudden change in state. When Δ > 0, there is only one real root, and the system is stable.

[0154] Based on △, the system can be divided into three typical security states:

[0155] Region I (Safe and Controllable Region): Δ>0 and u>0. The system is located outside the bifurcation set, the equilibrium surface is a single leaf, and there exists a unique highly safe and stable equilibrium state. The state changes continuously, and it has strong resistance to disturbances.

[0156] Region III (Instability Warning Zone): Δ>0 and u≤0. Although the system is still monostable, the stabilizing force u is negative, and the only equilibrium point on the equilibrium surface is located in the lower leaf (low safety level). The system is sensitive to external disturbances and is prone to further deterioration.

[0157] Region II (Bistable Transition Region): Δ≤0. The system enters the bifurcation set, where high-safety and low-safety states coexist. Even a small perturbation can cause the state to jump from the upper leaf to the lower leaf, manifesting as a sudden occurrence of a safety incident.

[0158] Accordingly, the aforementioned project-specific mutation model:

[0159] (X) obs 0.288)3 + (0.848u) obs 0.154)(X obs 0.288) + 0.033v obs 0.032 = 0.

[0160] Critical discriminant for sudden changes in project risk:

[0161] △=4 (0.848u) obs 0.154) 3 +27 (0.033v) obs 0.032) 2 .

[0162] Based on this, the Δ for all 24 weeks can be calculated (the Δ for all 24 weeks is calculated in this application to determine the correctness of the model; in actual use, only the Δ for the current evaluation period needs to be calculated).

[0163] That is, for the current preset period to be evaluated, its u obs-N u obs-N Substitute into the following formula to calculate △ N :

[0164] △ N =4 (a u u obs-N +b u ) 3 +27 (a) v v obs-N +b v ) 2 .

[0165] Figure 4 The model illustrates the evolution of safety risks during prefabricated building construction over 24 weeks. Comparing the early warning results with on-site safety records, the warning levels in 21 out of the 24 weeks matched the actual levels, achieving a consistency rate of 87.5%. Furthermore, model calculations show that the sum of squared residuals in the training set (weeks 1-20) was 5.8842 × 10⁻³, with a root mean square error (RMSE) of 0.0172. The absolute value of the residuals in most weeks was ≤0.1, indicating excellent fitting accuracy. The RMSE in the validation set (weeks 21-24) was 0.0284, with prediction accuracy close to that of the calculation set, showing no overfitting. This demonstrates the model's strong generalization ability and reliable extrapolation.

[0166] To visualize the system state transition path, the (uobs,vobs) values ​​for 24 weeks are mapped to the control plane, and a dynamic risk evolution trajectory diagram is drawn. Figure 5(This illustrates the dynamic evolution trajectory of the risk).

[0167] Unstable transition period (W1) W4) Δ is negative but its absolute value gradually decreases, and the system approaches the critical boundary; during the stable fluctuation period (W5) W13) Δ turns positive but the value is small, the system is on the critical edge; high-risk period (W14) W16) Week 14 Δ=0, Week 15 Δ= 0.001, state variable x from 0.366 suddenly jumped to 0.731, fully reproducing the entire process of cusp mutation: "critical → entering the bifurcation set → state jump → exiting the bifurcation set"; recovery adjustment period (W17) When W24)Δ remains positive, the system exits the bifurcation set, but the state variables are significantly lower than before the mutation, reflecting the new steady-state characteristics after the resilience is damaged.

[0168] Figure 6 This illustrates the unstable transition period W1. A three-dimensional evolution and abrupt change trajectory diagram of the safety state in stage W4. The trajectory distribution in this stage is relatively flat.

[0169] Figure 7 This illustrates the stable fluctuation period W5. A three-dimensional evolution and mutation trajectory diagram of the safe state in stage W13. During this stage, sample points move closer to the boundary of the bifurcation set.

[0170] Figure 8 This indicates the high-risk period W14. A three-dimensional evolution and abrupt change trajectory diagram of the safe state in stage W16. The trajectory in this stage crosses a critical region and abruptly jumps.

[0171] Figure 9 This indicates the recovery and adjustment period W17. A three-dimensional evolution and mutation trajectory diagram of the W24 safe state. The trajectory in this stage gradually moves away from the bifurcation set.

[0172] like Figure 10 As shown, the processing measures for Zones I, II, and III are given:

[0173] Zone I: Routine monitoring, weekly inspections, maintaining the existing management system, and overall monitoring of core indicator trends.

[0174] Zone III: Intensive monitoring (twice a week), specialized diagnosis, and preventative intervention. Intervention strategies include: targeting... Organize specialized risk identification and assessment, strengthen safety briefings, and establish a risk early warning ledger. (Targeting...) Adjust work plans, reduce the intensity of work at height, increase protective measures, and strengthen on-site supervision. (Regarding...) Strengthen the acceptance inspection of incoming components, identify, isolate, or replace unqualified components, and trace the supplier's quality. (This is in response to...) : Review the design bearing capacity of temporary supports, reinforce deformed or loose supports, and increase the monitoring frequency.

[0175] Zone II: Immediately activate the emergency plan, halt work or partially halt work, forcibly increase stability and reduce disturbance forces, monitor daily, conduct a comprehensive investigation of stability and disturbance force indicators, and forcibly intervene in personnel risk identification capabilities, personnel and machinery working at heights, component quality, and temporary support strength. Regular construction can only resume after the safety status changes to Zone I.

[0176] A dynamic early warning system for safety risks in prefabricated building construction, comprising:

[0177] Data acquisition module: used to acquire measured values ​​of the stability force index set, disturbance force index set, and state variable index set within N consecutive preset periods;

[0178] Standardization Calculation Module: Used to calculate the standardized values ​​of each index based on pre-stored benchmark values ​​and feature scale coefficients;

[0179] Aggregate calculation module: used to obtain aggregated observations u of stabilizing force, disturbance force, and state variables for N consecutive preset periods by weighted summation based on DEMATEL weights. obs-1~ u obs-N v obs-1~ v obs-N X obs-1~ X obs-N ;

[0180] Parameter calibration module: used to solve for the mapping parameter b using the Levenberg-Marquardt algorithm. u a v b v a x b x ;

[0181] Discriminant Calculation Module: Used to calculate the bifurcation set discriminant Δ for the current evaluation period. N and stabilizing force value u N ;

[0182] Early warning output module: used to output based on Δ N and u N Output the warning level.

[0183] It should be noted that the preset cycle can be daily or weekly.

[0184] The above-described embodiments are preferred embodiments of the present invention and are only used to facilitate the illustration of the present invention. They are not intended to limit the present invention in any way. Any person skilled in the art who makes local modifications or alterations to the technical content disclosed in the present invention without departing from the scope of the technical features of the present invention shall still fall within the scope of the technical features of the present invention.

Claims

1. A dynamic early warning system for safety risks in prefabricated building construction, characterized in that, include: Data acquisition module: used to acquire the measured values ​​of the stability force index set, disturbance force index set, and state variable index set within N consecutive preset periods; Standardization calculation module: used to calculate the standardized values ​​of each indicator; Aggregate calculation module: used to obtain aggregated observations u of stabilizing force, disturbance force, and state variables for N consecutive preset periods by weighted summation based on DEMATEL weights. obs-1~ u obs-N v obs-1~ v obs-N X obs-1~ X obs-N ; Parameter calibration module: used to solve for the mapping parameter b using the Levenberg-Marquardt algorithm. u a v b v a x b x ; Discriminant Calculation Module: Used to calculate the bifurcation set discriminant Δ for the current evaluation period. N and stabilizing force value u N ; Early warning output module: used to output based on Δ N and u N Output the warning level.

2. The prefabricated building construction safety risk dynamic early warning system according to claim 1, characterized in that, The standardized value of each indicator = (the measured value of each indicator - the benchmark value of each indicator) / the characteristic scale coefficient of each indicator.

3. The prefabricated building construction safety risk dynamic early warning system according to claim 1, characterized in that, The parameter calibration module works as follows: The Levenberg-Marquardt algorithm is used to iteratively solve for the mapping parameter a, with the objective of minimizing the sum of squared residuals. u b u a v b v a x b x : .

4. The prefabricated building construction safety risk dynamic early warning system according to claim 3, characterized in that, The working method of the discriminant calculation module is as follows: △ N =4(a u ·u obs-N +b u ) 3 +27(a v ·v obs-N +b v ) 2 ; u N =a u ·u obs-N +b u 。 5. The prefabricated building construction safety risk dynamic early warning system according to claim 4, characterized in that, The working method of the early warning output module is as follows: If △ N >0 and u N >0, the current evaluation period is in a safe state; If △ N >0 and u N ≤0, the current evaluation period is in an unstable alert state; If △ N ≤0, the current evaluation period is in an unstable state.

6. A method for analyzing the abrupt evolution of safety risks in prefabricated building construction, characterized in that, Includes the following steps: S100, obtain aggregated observations u of the stabilizing force, disturbance force, and state variables for N consecutive preset periods. obs-1~ u obs-N v obs-1~ v obs-N X obs-1~ X obs-N The Nth preset period is the current evaluation period; For the i-th preset period, the measured values ​​of each index in the stability force index set, disturbance force index set, and state variable index set are collected, standardized, and then weighted and summed to obtain u. obs-i v obs-i X obs-i ; S200, Calculate mapping parameter a u b u a v b v a x b x ; The Levenberg-Marquardt algorithm is used to iteratively solve for the mapping parameter a, with the objective of minimizing the sum of squared residuals. u b u a v b v a x b x : S300, Calculate the bifurcation set discriminant Δ for the current evaluation period. N and the stabilizing force value u N According to △ N u N Provide risk warnings for the safety status of the current evaluation period; △ N =4(a u ·u obs-N +b u ) 3 +27(a v ·v obs-N +b v ) 2 ; u N =a u ·u obs-N +b u ; If △ N >0 and u N >0, the current evaluation period is in a safe state; If △ N >0 and u N ≤0, the current evaluation period is in an unstable alert state; If △ N ≤0, the current evaluation period is in an unstable state.

7. The method for analyzing the sudden evolution of safety risks in prefabricated building construction according to claim 6, characterized in that, The S300 also includes: If the current evaluation period is in a safe state, continue with standard construction measures; If the current evaluation period is in an unstable alert state, organize special risk identification and assessment, reduce the intensity of high-altitude operations, mark, isolate or replace unqualified components, and reinforce deformed or loose supports; If the current evaluation period is in an unstable state, immediately activate the emergency plan, suspend work, or partially suspend work.

8. The method for analyzing the sudden evolution of safety risks in prefabricated building construction according to claim 6, characterized in that, S100 includes the following sub-steps: S101, collects the measured values ​​of each index in the set of stability force index, set of disturbance force index, and set of state variable index according to a preset cycle; The stability index set includes: personnel risk identification capability, emergency plan management capability, component quality, and temporary support strength; The set of disturbance indicators includes: human-machine operation at height, equipment failure, process complexity, and personnel fatigue; The set of state variable indicators includes: weather and climate, civilized construction, and site layout; S102, calculate the standardized values ​​of each index in the stability force index set, disturbance force index set, and state variable index set; The standardized value of the indicator = (the measured value of the indicator - the benchmark value of the indicator) / the characteristic scale coefficient of the indicator; S103, use DEMATEL to determine the weights of each index: the sum of the weights of each index in the stability index set is 1, the sum of the weights of each index in the disturbance index set is 1, and the sum of the weights of each index in the state variable index set is 1. S104, weighted summation yields the aggregated observation value u of the ith preset period's stability force, disturbance force, and state variables. obs-i v obs-i X obs-i .

9. The method for analyzing the sudden evolution of safety risks in prefabricated building construction according to claim 6, characterized in that, N≥20。 10. A storage medium, characterized in that, It stores a computer program that, when executed by a processor, implements the method as described in any one of claims 6 to 9.

Images