High-rise building construction shutdown risk assessment system based on hybrid intelligent algorithm

Abstract

The invention provides a high-rise building construction shutdown risk assessment system based on a hybrid intelligent algorithm, relates to the technical field of engineering, and aims to improve the accuracy and efficiency of shutdown risk identification. The method comprises the following steps: constructing a risk assessment index system covering factors such as environment, personnel, materials, management and the like; a G1 method and a CRITIC method are adopted to determine a weight, and a sparrow search algorithm is combined to realize adaptive combination weighting; performing grade evaluation on the construction shutdown risk by using the improved cloud model; a mathematical model is established for verification through the Tene chaotic mapping and golden sine function improved SABO algorithm in combination with an actual building engineering project. Experimental results show that the system has high practicability and precision in the aspects of multi-dimensional risk factor identification and intelligent analysis, and is suitable for high-rise building construction risk dynamic assessment and strategy formulation.

Description

Technical field:

[0001] The present invention relates to the field of engineering technology, and more particularly to a high-rise building construction shutdown risk assessment system based on a hybrid intelligent algorithm. Background technology:

[0002] With the acceleration of urbanization and the increasing scarcity of land resources, high-rise buildings have become an essential component of urban development. They not only conserve land resources but also improve the efficiency of urban space utilization. According to data released by the Council on Tall Buildings and Urban Habitat (CTBUH) on March 7, 2024, 1,000 buildings 200 meters and above were completed worldwide in 2015. However, since 2023, 2,269 buildings 200 meters and above and 232 supertall buildings (300 meters and above) have been completed worldwide. Clearly, this number has more than doubled in just eight years.

[0003] Given the long construction periods, frequent overlapping operations, complex construction processes, and numerous participating organizations of high-rise buildings, the construction process presents significant safety hazards. Therefore, to reduce construction risks and improve efficiency, it is necessary to conduct safety risk assessments and research on preventive measures. According to the latest statistics from the Ministry of Housing and Urban-Rural Development, the number of construction accidents and casualties has remained high in recent years, reaching a peak in 2019. Although there was a slight decline in 2020, the overall level remains significant. Statistics also show that the average number of deaths per accident remained relatively stable between 2016 and 2020, indicating a relatively constant severity of accidents. However, the overall number of accidents fluctuated significantly, posing a continuous threat to construction efficiency and personnel safety. As one of the country's pillar industries, the construction industry consistently has the highest accident rate of all sectors, not only impacting construction progress but also imposing significant social and economic costs.

[0004] Compared to typical multi-story residential buildings, high-rise buildings better meet the housing needs of urban residents. However, their increasing structural complexity places higher demands on foundational construction, leading to increased risk of fatalities and significant losses. Consequently, building safety management is becoming increasingly difficult. Furthermore, high-rise buildings often require longer construction periods, require more sophisticated technical requirements, involve more complex processes, and are more difficult to control construction resources. Consequently, they are more susceptible to accidents, face a wider range of risks, and their relationships are more complex. Therefore, risk assessment systems are essential considerations for high-rise buildings.

[0005] Therefore, in order to overcome the above problems, a high-rise building construction shutdown risk assessment system based on a hybrid intelligent algorithm is studied, aiming to improve the identification of risks in the construction process, and to take timely precautions against them, reduce accidents in the project, and improve the level and efficiency of risk management, which has important practical significance and application value. Summary of the invention:

[0006] The present invention aims to provide a high-rise building construction shutdown risk assessment system based on a hybrid intelligent algorithm. This system aims to address existing assessment methods' issues, including incomplete risk identification, subjective weight determination, low assessment accuracy, and delayed policy feedback. By constructing an assessment framework that integrates subjective and objective data analysis, intelligent weight optimization, and risk grading, the system comprehensively identifies key risk factors that may lead to construction shutdowns in high-rise buildings and scientifically categorizes risk levels. Furthermore, the present invention introduces an improved swarm intelligence optimization algorithm, enhancing the accuracy of risk weight calculation and the global convergence of the assessment model, effectively overcoming the vulnerability of traditional algorithms to local optima. Furthermore, the system integrates a risk prevention strategy generation module that automatically outputs corresponding countermeasures based on assessment results, creating a closed-loop system from risk identification and assessment to policy formulation. Deployment of this system can significantly improve risk prevention and control capabilities and construction management efficiency in high-rise building construction projects, reducing the economic losses and safety hazards associated with sudden construction shutdowns.

[0007] In order to achieve the above object, the present invention adopts the following technical solutions:

[0008] An improved subtraction average optimization algorithm based on architectural engineering design problems includes the following specific steps:

[0009] Step 1: Identify risk factors and conduct data analysis on actual high-rise building construction projects, extract key indicators from four aspects: environment, personnel, materials, and management, and establish a systematic shutdown risk assessment indicator system and mathematical modeling foundation;

[0010] Step 2: Use the order relationship analysis method (G1) and the standard correlation method (CRITIC) to determine the subjective and objective weights. By introducing the Tene chaotic map in the improved sparrow search algorithm (SSA), the initial population and dynamic search path of the optimization algorithm are generated to achieve adaptive combination weight determination;

[0011] Step 3: Use the golden sine function mechanism to improve the algorithm's ability to escape local optimality and enhance the global search performance and convergence efficiency of the risk assessment model;

[0012] Step 4: Build a cloud-based risk assessment system, classify construction suspension risks into five levels, and conduct simulation tests using actual project data to verify the effectiveness of the assessment method.

[0013] Step 5: Generate a risk prevention strategy model based on the assessment results, use the improved Subtraction Average Optimization (SABO) swarm intelligence algorithm to formulate response measures, and verify the effectiveness of the strategy based on engineering cases to achieve an integrated closed-loop design from assessment to strategy output.

[0014] Preferably, in step 1, risk factor identification and data analysis are performed on actual cases of high-rise building construction projects, key indicators are extracted from four aspects, namely environment, personnel, materials, and management, and a systematic shutdown risk assessment indicator system and mathematical modeling basis are established.

[0015] Preferably, in step 2, the G1 method and the CRITIC method are used to determine the subjective and objective weights, and the Tene chaotic map in the improved sparrow search algorithm (SSA) is introduced to generate the initial population and dynamic search path of the optimization algorithm to achieve adaptive combination weight determination. The specific process is as follows:

[0016] Step 2.1: Determine subjective weights using the G1 method:

[0017] The G1 method, also known as the ordinal relationship analysis method, is a subjective weighting method modified from the analytic hierarchy process (AHP). The G1 method weights indicators based on the ordinal relationships established by decision makers between different indicators. While computationally simpler than the AHP method, its results are also highly subjective. The main steps are as follows:

[0018] (1) Determine the order relationship

[0019] First, experts should establish a high-rise building project shutdown risk assessment index system {A1, A2, ..., A n}Select a most critical indicator and record it as A i ; Secondly, continue to select those indicators that are considered the most important among the remaining indicators and mark them as A -1 , and repeat this process to get the importance ranking: A i >A i -1>…>A n-i , so an order relationship of importance can be determined.

[0020] (2) Determine the importance between adjacent indicators

[0021] Let the adjacent index A k-1 and A k The importance ratio between them is r k ,but

[0022]

[0023] Where A kis the weight of the kth indicator.

[0024] (3) Calculate the weight of each attribute

[0025] First, the corresponding weights of the four first-level indicators in the high-rise building construction shutdown risk system are calculated. Through scientific evaluation methods, the importance of each first-level indicator in the entire risk system is determined. Similarly, the same calculation method is used for each second-level indicator under each first-level indicator to determine their respective weights to further refine and quantify the risk assessment system. The specific calculation formula is as follows:

[0026]

[0027] A j+1 =r j *A j (j=1,2,…,n-1) (3)

[0028] Step 2.2: Determine objective weights using the CRITIC method:

[0029] The CRITIC method is an objective weighting method. Its basic idea is to determine the objective weights of evaluation indicators based on the ability to distinguish and the degree of conflict between evaluation indicators. Compared with the entropy weight method and the coefficient of variation method, this method more fully considers the objective characteristics of the data. It not only considers the impact of indicator differences on weights, but also considers the conflict between indicators. In the CRITIC method, volatility is usually expressed by standard deviation, while conflict is expressed by correlation coefficient. Although the CRITIC method can reflect the relationship between data, it does not fully reflect the differences between data. Therefore, it is necessary to introduce the coefficient of variation to improve and optimize the CRITIC method. The specific steps are as follows:

[0030] (1) Constructing the original judgment matrix

[0031] Assume that we have m evaluation samples and n evaluation indicators, and construct an m×n judgment matrix A, where a ij Indicates the importance of indicator i relative to indicator j.

[0032]

[0033] (2) Standardization Processing The judgment matrix A′ is standardized to obtain the standardized matrix A′:

[0034]

[0035] maxX ij minX ii Where i and i are the maximum and minimum values of the i-th row in the original matrix A, respectively.

[0036] (3) Calculate the standard deviation σ of each indicator i , correlation coefficient ρ ij According to the standardized matrix A′, calculate the standard deviation σ of each indicator i The correlation coefficient ρ between the ij .

[0037]

[0038] Where A′ is the mean of the elements in row i; Cov(A′,A′ j ) is the covariance between the i-th row and the j-th row in the standard matrix.

[0039] (4) Calculate the coefficient of variation of each indicator using the standard deviation and mean to calculate the CV of each indicator j :

[0040]

[0041] In the formula is the mean of the jth index in the standardized matrix.

[0042] (5) Calculating the independence coefficient is usually used to measure the degree of influence of each indicator on other indicators, which can be expressed by the correlation coefficient.

[0043]

[0044] (6) Calculation of comprehensive coefficient

[0045]

[0046] (7) Calculate the index weight ω j

[0047]

[0048] Step 2.3: Adaptive combination weighting:

[0049] The weight vector obtained by G1 method is G=(g1,g2,…,gn) T , the weight vector obtained by the improved CRITIC method is C=(c1,c2,…,c n ) T , the combined weight is Z=(z1,z2,…,z n ) T In order to ensure the scientificity and adaptability of the weights, this paper applies the above-mentioned SSA algorithm for adaptive combination. The function model of this paper is:

[0050]

[0051] Where αi is the adaptive weight of the i-th indicator; m is the total number of weighting methods; k is the number of indicators; K ij is the rate of change between the weights of the i+1th and ith indicators of the jth weighting method; i is the indicator i = 1, 2,…, k.

[0052] Preferably, in step 3, the golden sine function mechanism is used to improve the algorithm's ability to escape from local optimality and enhance the global search performance and convergence efficiency of the risk assessment model. The specific process is as follows:

[0053] The golden sine function is derived from the mathematical properties of the golden section and its expression is:

[0054]

[0055] To enhance the algorithm's global search capabilities, this paper introduces the golden sine function. By combining the golden ratio and the periodicity of the sine function, the golden sine function provides an effective search method, preventing the algorithm from falling into local optimality. Its formula is as follows:

[0056]

[0057] Among them, r1 and r2 are the golden ratios; i is the current number of iterations.

[0058] Preferably, in step 4, a risk assessment system based on a cloud model is constructed, the construction suspension risk is divided into five levels, and simulation tests are performed through actual engineering data to verify the effectiveness of the assessment method.

[0059] Preferably, in step 5, a risk prevention strategy model is generated based on the assessment results, countermeasures are formulated using the improved SABO group intelligent algorithm, and the effectiveness of the strategy is verified in combination with engineering cases, thus achieving an integrated closed-loop design from assessment to strategy output. The specific process is as follows:

[0060] The Subtraction Average Optimization (SABO) algorithm is a novel swarm-based intelligent optimization algorithm. It guides the search process by gradually reducing the search space and calculating the difference between the current optimal individual and the swarm average, thereby improving optimization efficiency and accuracy. The core idea of SABO is to leverage the relative positional relationships between individuals to gradually narrow the search range. By calculating the difference between the current optimal individual and the swarm average, the algorithm adjusts the search direction and range, gradually approaching the global optimal solution.

[0061] (1) Algorithm initialization

[0062] The solution space of engineering optimization problems is also called the search space. The search particles in the search space are randomly initialized. The specific formula is:

[0063]

[0064] x i,d =lb d +r i,d ·(ub d -lb d ), (i=1,2,…,n,d=1,2,…,m) (16)

[0065] Where X is the SABO overall matrix; X i is the i-th individual; d is the d-th dimension in the search space; n is the number of individuals; m is the number of decision variables; r i、d is a random number in the interval [0,1]; ub d lb d are the upper and lower bounds of the d-th dimension decision variable, respectively. Each search particle corresponds to a solution to the optimization problem, and their fitness function value set is represented as a vector F. The specific formula is as follows:

[0066]

[0067] Where F is the fitness function; F i is the fitness value corresponding to the i-th search particle.

[0068] (2) "-v" method

[0069] The "-v" method refers to the process of updating the search particles by subtracting the mean difference

[50] . The specific definition is as follows:

[0070] A-vB=sgn(F(A)-F(B))(Av*B) (18)

[0071] Where v is a vector of dimension m, which is a random number generated from [1,2]; F(A) and F(B) are the values of the target functions A and B respectively; sgn is the signum function

[0072] (3) Update particle position

[0073] In the SABO algorithm, any particle X i The displacement in the search space is calculated by j The position update formula is as follows:

[0074]

[0075] In the formula For the i-th search agent X i The new position of n is the total number of search agents; r iis a vector of dimension m.

[0076] If the updated position is better, replace the original position with formula (2-19); otherwise, keep it as it is.

[0077]

[0078] Where F i 、 are X i 、 The objective function value of .

[0079] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention discloses a high-rise building construction shutdown risk assessment system based on a hybrid intelligent algorithm, which takes into account the global exploration in the early stage of the search and the local development in the later stage, dynamically adjusts the balance between exploration and development, and avoids the problems of "insufficient exploration" or "over-development" in traditional optimization methods. Description of the drawings:

[0080] Figure 1 General construction plan

[0081] Figure 2 This is a diagram of the risk assessment indicator system for high-rise building construction stoppage;

[0082] Figure 3 This is the overall risk cloud model diagram of the project;

[0083] Figure 4 The optimal solution for the three-dimensional space of changes in total project duration, total project cost, and total resource allocation; Specific implementation method:

[0084] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0085] The present invention is further described in detail below with reference to the accompanying drawings. The present invention proposes a high-rise building construction shutdown risk assessment system based on a hybrid intelligent algorithm, comprising the following steps: 1. Identifying and analyzing risk factors in actual high-rise building construction projects, extracting key indicators from four aspects, namely, environment, personnel, materials, and management, and establishing a systematic shutdown risk assessment indicator system and a mathematical modeling foundation; 2. Determining subjective and objective weights using the G1 method and the CRITIC method, and generating the initial population and dynamic search path of the optimization algorithm by introducing the Tene chaotic map in the improved sparrow search algorithm (SSA) to achieve adaptive combination weight determination; 3. Using the golden sine function mechanism to improve the algorithm's ability to escape local optima, thereby enhancing the global search performance and convergence efficiency of the risk assessment model; 4. Constructing a cloud-based risk assessment system, dividing construction shutdown risks into five levels, and conducting simulation tests using actual engineering data to verify the effectiveness of the assessment method; 5. Generating a risk prevention strategy model based on the assessment results, formulating countermeasures using the improved SABO group intelligent algorithm, and verifying the effectiveness of the strategy based on engineering cases, thereby achieving an integrated closed-loop design from assessment to strategy output.

[0086] Step 1: Identify risk factors and conduct data analysis on actual high-rise building construction projects, extract key indicators from four aspects: environment, personnel, materials, and management, and establish a systematic shutdown risk assessment indicator system and mathematical modeling foundation;

[0087] Step 2: Use the G1 method and CRITIC method to determine the subjective and objective weights. By introducing the Tene chaotic map in the improved sparrow search algorithm (SSA), the initial population and dynamic search path of the optimization algorithm are generated to achieve adaptive combination weight determination.

[0088] Step 3: Use the golden sine function mechanism to improve the algorithm's ability to escape local optimality and enhance the global search performance and convergence efficiency of the risk assessment model;

[0089] Step 4: Build a cloud-based risk assessment system, classify construction suspension risks into five levels, and conduct simulation tests using actual project data to verify the effectiveness of the assessment method.

[0090] Step 5: Generate a risk prevention strategy model based on the assessment results, use the improved SABO group intelligent algorithm to formulate response measures, and verify the effectiveness of the strategy based on engineering cases to achieve an integrated closed-loop design from assessment to strategy output.

[0091] In step 1, risk factor identification and data analysis are conducted on actual cases of high-rise building construction projects, key indicators are extracted from four aspects, namely environment, personnel, materials, and management, and a systematic shutdown risk assessment indicator system and mathematical modeling basis are established.

[0092] Based on the actual case, we selected Project S as the research object. The project type is a residential high-rise building with a contract period of 197 days and a contract amount of RMB 15.4554 million. The project covers an area of 7576.39 m2, a building height of 53.5 m, a total of 14 floors, a fire protection level of Level 2, a raft pile foundation, a shear wall structure, and a seismic fortification intensity of Level 7. The general construction plan of this example is as follows: Figure 1 shown.

[0093] The safety risk management of high-rise building projects is studied from the perspective of the entire life cycle, that is, a comprehensive analysis of the influencing factors is carried out in each stage of design, decision-making, construction and completion. According to the characteristics of each stage, the risk factors can be summarized into environmental, personnel, material and management factors. Therefore, the risk of high-rise building construction shutdown is taken as the target layer of evaluation, and environmental factors, personnel factors, material factors and management factors are set as primary indicators. In the selection of secondary indicators, 18 major risk factors were finally determined after comprehensive screening and analysis. These factors cover all aspects, can effectively reflect the potential risks of high-rise building construction shutdown, and provide a systematic basis for subsequent risk assessment. Based on this, the present invention constructs an evaluation index system for the risk of high-rise building construction shutdown, and the specific structure is as follows: Figure 2 shown.

[0094] In step 2, the G1 method and the CRITIC method are used to determine the subjective and objective weights. By introducing the Tene chaotic map in the improved sparrow search algorithm (SSA), the initial population and dynamic search path of the optimization algorithm are generated to achieve adaptive combination weight determination. The specific process is as follows:

[0095] S21: G1 method to determine subjective weights:

[0096] The G1 method, also known as the ordinal relationship analysis method, is a subjective weighting method modified from the analytic hierarchy process (AHP). The G1 method weights indicators based on the ordinal relationships established by decision makers between different indicators. While computationally simpler than the AHP method, its results are also highly subjective. The main steps are as follows:

[0097] (1) Determine the order relationship

[0098] First, experts should establish a high-rise building project shutdown risk assessment index system {A1, A2, ..., A n}Select a most critical indicator and record it as A i ; Secondly, continue to select those indicators that are considered the most important among the remaining indicators and mark them as A -1 , and repeat this process to get the importance ranking: A i >A i-1 >…>A n-i, so an order relationship of importance can be determined.

[0099] (2) Determine the importance between adjacent indicators

[0100] Let the adjacent index A k-1 and A k The importance ratio between them is r k ,but

[0101]

[0102] Where A k is the weight of the kth indicator.

[0103] (3) Calculate the weight of each attribute

[0104] First, the corresponding weights of the four first-level indicators in the high-rise building construction shutdown risk system are calculated. Through scientific evaluation methods, the importance of each first-level indicator in the entire risk system is determined. Similarly, the same calculation method is used for each second-level indicator under each first-level indicator to determine their respective weights to further refine and quantify the risk assessment system. The specific calculation formula is as follows:

[0105]

[0106] A j+1 =r j *A j (j=1,2,…,n-1) (3)

[0107] S22: CRITIC method to determine objective weights:

[0108] The CRITIC method is an objective weighting method. Its basic idea is to determine the objective weights of evaluation indicators based on the ability to distinguish and the degree of conflict between evaluation indicators. Compared with the entropy weight method and the coefficient of variation method, this method more fully considers the objective characteristics of the data. It not only considers the impact of indicator differences on weights, but also considers the conflict between indicators. In the CRITIC method, volatility is usually expressed by standard deviation, while conflict is expressed by correlation coefficient. Although the CRITIC method can reflect the relationship between data, it does not fully reflect the differences between data. Therefore, it is necessary to introduce the coefficient of variation to improve and optimize the CRITIC method. The specific steps are as follows:

[0109] (1) Constructing the original judgment matrix

[0110] Assume that we have m evaluation samples and n evaluation indicators, and construct an m×n judgment matrix A, where a ij Indicates the importance of indicator i relative to indicator j.

[0111]

[0112] (2) Standardization Processing The judgment matrix A′ is standardized to obtain the standardized matrix A′:

[0113]

[0114] Where maxX ij minX ii Where i and i are the maximum and minimum values of the i-th row in the original matrix A, respectively.

[0115] (3) Calculate the standard deviation σ of each indicator i , correlation coefficient ρ ij According to the standardized matrix A′, calculate the standard deviation σ of each indicator i The correlation coefficient ρ between the ij .

[0116]

[0117] Where A′ is the mean of the elements in row i; Cov(A′,A′ j ) is the covariance between the i-th row and the j-th row in the standard matrix.

[0118] (4) Calculate the coefficient of variation of each indicator using the standard deviation and mean to calculate the CV of each indicator j :

[0119]

[0120] In the formula is the mean of the jth index in the standardized matrix.

[0121] (5) Calculating the independence coefficient is usually used to measure the degree of influence of each indicator on other indicators, which can be expressed by the correlation coefficient.

[0122]

[0123] (6) Calculation of comprehensive coefficient

[0124]

[0125] (7) Calculate the index weight ω j

[0126]

[0127] S23: Adaptive combination weighting:

[0128] The weight vector obtained by G1 method is G=(g1,g2,…,gn)T , the weight vector obtained by the improved CRITIC method is C=(c1,c2,…,cn) T , the combined weight is Z=(z1,z2,…,zn) T In order to ensure the scientificity and adaptability of the weights, this paper applies the above-mentioned SSA algorithm for adaptive combination. The function model of this paper is:

[0129]

[0130] Where α i is the adaptive weight of the i-th indicator; m is the total number of weighting methods; k is the number of indicators; K ij is the rate of change between the weights of the i+1th and ith indicators of the jth weighting method; i is the indicator i = 1, 2,…, k.

[0131] Preferably, in step 3, the golden sine function mechanism is used to improve the algorithm's ability to escape from local optimality and enhance the global search performance and convergence efficiency of the risk assessment model. The specific process is as follows:

[0132] The golden sine function is derived from the mathematical properties of the golden section and its expression is:

[0133]

[0134] To enhance the algorithm's global search capabilities, this paper introduces the golden sine function. By combining the golden ratio and the periodicity of the sine function, the golden sine function provides an effective search method, preventing the algorithm from falling into local optimality. Its formula is as follows:

[0135]

[0136] Among them, r1 and r2 are the golden ratios; i is the current number of iterations.

[0137] In step 4, a risk assessment system based on a cloud model is constructed, the construction suspension risk is divided into five levels, and simulation tests are carried out using actual engineering data to verify the effectiveness of the assessment method.

[0138] Based on the above example, after consulting with multiple high-rise building professionals, we categorized the projects into five levels based on their specific circumstances: "High Risk," "Higher Risk," "Medium Risk," "Lower Risk," and "Low Risk." These risk level evaluations are based on the domain [0,5], with corresponding intervals of [0,1), [1,2), [2,3), [3,4), and [4,5], respectively, to ensure the rationality and accuracy of the risk grading. The specific grading results are shown in Table 1.

[0139] Table 1 Evaluation standard cloud level

[0140]

[0141] From this, we can conclude that the comprehensive cloud digital eigenvalue of the high-rise building construction shutdown risk assessment is (1.1222, 0.4012, 0.0732). Subsequently, the corresponding cloud map of the overall project risk was generated on the system, as shown in the following figure: Figure 3 shown.

[0142] According to various risk assessments for high-rise building construction, environmental and human factors are both considered low risk, with assessment cloud parameters of 1.9160 and 1.9670, respectively. The assessment cloud parameter for material factors is 2.5856, indicating a medium risk. The assessment cloud parameter for management factors is 3.4948, which is considered a high risk, indicating significant management risks and requiring special attention. The overall project assessment cloud parameter is 1.1222, indicating a low risk level. However, management risks still need to be controlled to ensure smooth construction. The specific cloud parameters and risk levels are shown in Table 2.

[0143] Table 2 Cloud parameters and risk levels for high-rise building construction suspension risk assessment

[0144]

[0145] In step 5, a risk prevention strategy model is generated based on the assessment results, countermeasures are formulated using the improved SABO swarm intelligence algorithm, and the effectiveness of the strategy is verified in conjunction with engineering cases, achieving an integrated closed-loop design from assessment to strategy output. The specific process is as follows:

[0146] The Subtraction Average Based Optimizer (SABO) is a novel swarm-based intelligent optimization algorithm. It guides the search process by gradually reducing the search space and calculating the difference between the current optimal individual and the swarm average, thereby improving optimization efficiency and accuracy. The core idea of SABO is to leverage the relative positional relationships between individuals to gradually narrow the search range. By calculating the difference between the current optimal individual and the swarm average, the algorithm adjusts the search direction and range, gradually approaching the global optimal solution.

[0147] (1) Algorithm initialization

[0148] The solution space of engineering optimization problems is also called the search space. The search particles in the search space are randomly initialized. The specific formula is:

[0149]

[0150] x i ,d=lb d+r i,d ·(ub d -lb d ), (i=1,2,…,n,d=1,2,…,m) (16)

[0151] Where X is the SABO overall matrix; X i is the i-th individual; d is the d-th dimension in the search space; n is the number of individuals; m is the number of decision variables; r i、d is a random number in the interval [0,1]; ub d lb d are the upper and lower bounds of the d-th dimension decision variable, respectively. Each search particle corresponds to a solution to the optimization problem, and their fitness function value set is represented as a vector F. The specific formula is as follows:

[0152]

[0153] Where F is the fitness function; F i is the fitness value corresponding to the i-th search particle.

[0154] (2) "-v" method

[0155] The "-v" method refers to the process of updating the search particles by subtracting the mean difference

[50] . The specific definition is as follows:

[0156] A-vB=sgn(F(A)-F(B))(Av*B) (18)

[0157] Where v is a vector of dimension m, which is a random number generated from [1,2]; F(A) and F(B) are the values of the target functions A and B respectively; sgn is the signum function

[0158] (3) Update particle position

[0159] In the SABO algorithm, any particle X i The displacement in the search space is calculated by j The position update formula is as follows:

[0160]

[0161] In the formula For the i-th search agent X i The new position of n is the total number of search agents; r i is a vector of dimension m.

[0162] If the updated position is better, replace the original position with formula (2-19); otherwise, keep it as it is.

[0163]

[0164] Where F i 、 are X i 、 The objective function value of .

[0165] In order to verify the performance of the system, actual data was used for testing. The results are as follows: Figure 4 shown. Figure 4 The Pareto optimal frontier of project management solutions is presented, including the total change in total project cost, total project duration, and resource allocation. These Pareto optimal points reveal how resource allocation changes under varying conditions of total project cost and total duration. These diagrams clearly demonstrate how to strike a balance between total cost, total duration, and resource allocation within a three-dimensional objective space. The results presented in the figures demonstrate that the improved SABO algorithm can effectively find the optimal solution between these three objectives, thereby optimizing the overall project schedule solution and achieving an optimal balance of objectives within the project management process.

[0166] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Reference can be made to the common and similar parts between the various embodiments. For the devices disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple, and the relevant parts can be referred to the method description.

[0167] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A high-rise building construction shutdown risk assessment system based on a hybrid intelligent algorithm, characterized by: The method comprises the following steps: Step 1: Identify risk factors and conduct data analysis on actual high-rise building construction projects, extract key indicators from four aspects: environment, personnel, materials, and management, and establish a systematic shutdown risk assessment indicator system and mathematical modeling foundation; Step 2: Use the order relationship analysis method (G1) and the standard correlation method (CRITIC) to determine the subjective and objective weights. By introducing the Tene chaotic map in the improved sparrow search algorithm (SSA), the initial population and dynamic search path of the optimization algorithm are generated to achieve adaptive combination weight determination; Step 3: Use the golden sine function mechanism to improve the algorithm's ability to escape local optimality and enhance the global search performance and convergence efficiency of the risk assessment model; Step 4: Build a cloud-based risk assessment system, classify construction suspension risks into five levels, and conduct simulation tests using actual project data to verify the effectiveness of the assessment method. Step 5: Generate a risk prevention strategy model based on the assessment results, formulate countermeasures using the improved Subtractive Average Optimization (SABO) swarm intelligence algorithm, and verify the effectiveness of the strategy using engineering cases, achieving an integrated closed-loop design from assessment to strategy output. in: In step 1, risk factor identification and data analysis are performed on actual cases of high-rise building construction projects, key indicators are extracted from four aspects, namely environment, personnel, materials, and management, and a systematic shutdown risk assessment indicator system and mathematical modeling foundation are established; In step 2, the G1 method and the CRITIC method are used to determine the subjective and objective weights. By introducing the Tene chaotic map in the improved sparrow search algorithm (SSA), the initial population and dynamic search path of the optimization algorithm are generated to achieve adaptive combination weight determination. The specific process is as follows: Step 2.1: Determine subjective weights using the G1 method: The G1 method, also known as the ordinal relationship analysis method, is a subjective weighting method improved from the analytic hierarchy process (AHP). The G1 method weights indicators based on the ordinal relationship established by decision makers between different indicators. It is not only simpler than the AHP method in terms of calculation, but its results are also full of subjectivity. The main steps are as follows: (1) Determine the order relationship First, experts should establish a high-rise building project shutdown risk assessment index system {A1, A2, ..., A n }Select a most critical indicator and record it as A i ; Secondly, continue to select the most important indicators from the remaining indicators and mark them as A-1, and repeat this process to get the importance ranking: A i >A i-1 >…>A n-i , so an order relationship of importance can be determined; (2) Determine the importance between adjacent indicators Let the adjacent index A k-1 and A k The importance ratio between them is r k ,but Where A k is the weight of the kth indicator; (3) Calculate the weight of each attribute First, the corresponding weights of the four first-level indicators in the high-rise building construction shutdown risk system are calculated. The importance of each first-level indicator in the entire risk system is determined through a scientific evaluation method. Similarly, the same calculation method is used for each second-level indicator under each first-level indicator to determine their respective weights, so as to further refine and quantify the risk assessment system. The specific calculation formula is as follows: A j+1 =r j *A j (j=1,2,…,n-1) (3) Step 2.2: Determine objective weights using the CRITIC method: The CRITIC method is an objective weighting method. Its basic idea is to determine the objective weights of evaluation indicators based on the ability to distinguish and the degree of conflict between evaluation indicators. Compared with the entropy weight method and the coefficient of variation method, this method more fully considers the objective characteristics of the data. It not only considers the impact of indicator differences on weights, but also considers the conflict between indicators. In the CRITIC method, volatility is usually expressed by standard deviation, while conflict is expressed by correlation coefficient. Although the CRITIC method can reflect the relationship between data, it does not fully reflect the differences between data. Therefore, it is necessary to introduce the coefficient of variation to improve and optimize the CRITIC method. The specific steps are as follows: (1) Constructing the original judgment matrix Assume that we have m evaluation samples and n evaluation indicators, and construct an m×n judgment matrix A, where a ij Indicates the importance of indicator i relative to indicator j, (2) Standardization Processing The judgment matrix A′ is standardized to obtain the standardized matrix A′: Where maxX ij minX ii Where i and i are the maximum and minimum values of the i-th row in the original matrix A, respectively; (3) Calculate the standard deviation σ of each indicator i , correlation coefficient ρ ij According to the standardized matrix A′, calculate the standard deviation σ of each indicator i The correlation coefficient ρ between the ij , Where A′ is the mean of the elements in row i; Cov(A′,A′ j ) is the covariance between the i-th row and the j-th row in the standard matrix; (4) Calculate the coefficient of variation of each indicator using the standard deviation and mean to calculate the CV of each indicator j : In the formula is the mean of the jth index in the standardized matrix; (5) Calculating the independence coefficient is usually used to measure the degree of influence of each indicator on other indicators, which can be expressed by the correlation coefficient. (6) Calculation of comprehensive coefficient (7) Calculate the index weight ω j Step 2.3: Adaptive combination weighting: The weight vector obtained by G1 method is G=(g1,g2,…,g n ) T , the weight vector obtained by the improved CRITIC method is C=(c1,c2,…,c n ) T , the combined weight is Z=(z1,z2,…,z n ) T In order to ensure the scientificity and adaptability of the weights, this paper applies the above-mentioned SSA algorithm for adaptive combination. The function model of this paper is: Where α i is the adaptive weight of the i-th indicator; m is the total number of weighting methods; k is the number of indicators; K ij is the rate of change between the weights of the i+1th and ith indicators of the jth weighting method; i is the indicator i = 1, 2, ..., k; In step 3, the golden sine function mechanism is used to improve the algorithm's ability to escape local optimality and enhance the global search performance and convergence efficiency of the risk assessment model. The specific process is as follows: The golden sine function is derived from the mathematical properties of the golden section and its expression is: To enhance the global search capability of the algorithm, the present invention introduces the golden sine function. By combining the golden ratio and the periodicity of the sine function, the golden sine function can provide an effective search method to prevent the algorithm from falling into a local optimum. Its formula is as follows: Among them, r1, r2 are the golden ratio; i is the current iteration number; In step 4, a cloud-based risk assessment system is constructed to classify the construction suspension risk into five levels, and simulation tests are conducted using actual project data to verify the effectiveness of the assessment method. In step 5, a risk prevention strategy model is generated based on the assessment results. Response measures are formulated using the improved SABO swarm intelligence algorithm. The effectiveness of the strategy is verified by combining it with engineering cases, thus achieving an integrated closed-loop design from assessment to strategy output. The specific process is as follows: The Subtraction Average Optimization (SABO) algorithm is a new intelligent optimization algorithm based on populations. It guides the search process by gradually reducing the search space and calculating the difference between the current optimal individual and the population average, thereby improving optimization efficiency and accuracy. The core idea of SABO is to use the relative position relationship between individuals to gradually narrow the search range. The algorithm adjusts the search direction and range by calculating the difference between the current optimal individual and the population average, gradually approaching the global optimal solution. (1) Algorithm initialization The solution space of the engineering optimization problem is also called the search space. The search particles in the search space are randomly initialized. The specific formula is: x i,d =lb d +r i,d ·(ub d -lb d ),(i=1,2,…,n,d=1,2,…,m) (16) Where X is the SABO overall matrix; X i is the i-th individual; d is the d-th dimension in the search space; n is the number of individuals; m is the number of decision variables; r i、d is a random number in the interval [0,1]; ub d lb d are the upper and lower bounds of the d-th dimension decision variable, respectively. Each search particle corresponds to a solution to the optimization problem. Their fitness function value set is represented as a vector F. The specific formula is as follows: Where F is the fitness function; F i is the fitness value corresponding to the i-th search particle; (2) "-v" method The "-v" method refers to the process of updating the search particles by subtracting the mean difference [50], which is defined as follows: A-vB=sgn(F(A)-F(B))(Av*B) (18) Where v is a vector of dimension m, which is a random number generated from [1,2]; F(A) and F(B) are the values of the target functions A and B respectively; sgn is the signum function; (3) Update particle position In the SABO algorithm, any particle X i The displacement in the search space is calculated by j The position update formula is as follows: In the formula For the i-th search agent X i The new position of n is the total number of search agents; r i is a vector of dimension m, If the updated position is better, use formula (2-19) to replace the original position; otherwise, keep the original position. Where F i 、 are X i 、 The objective function value of .

Images