A positioning screening method and system for construction engineering quality hidden dangers

Abstract

The application discloses a kind of construction engineering quality hidden danger positioning screening method and system, the method is by obtaining the structure, material, environment and so on parameter of construction engineering quality hidden danger and classification, data are fused using multidimensional heterogeneous building hidden danger data screening model;Based on the initial search population established by optimized bat positioning algorithm, set fitness function to evaluate individual, update population according to iteration rule until meet termination condition and output hidden danger positioning result.Introduce data fusion, fitness calculation and multiple models in the process, fully consider parameter weight, correlation and coupling relationship.System includes six units such as parameter acquisition, data fusion processing, and each unit cooperates with each other.The application effectively integrates multi-source data, accurately locates construction engineering quality hidden danger, overcomes the defects of insufficient data utilization and inaccurate positioning of traditional technology, realizes efficient and accurate screening of construction engineering quality hidden danger.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of building hidden danger positioning screening, and in particular to a positioning screening method and system for building engineering quality hidden dangers. BACKGROUND

[0002] With the rapid development of the construction industry, the scale and complexity of construction projects are constantly rising, and the importance of engineering quality hidden danger positioning screening is increasingly prominent. Traditional building engineering quality detection relies on manual inspection, which is not only inefficient and difficult to meet the detection needs of large-scale projects, but also greatly affected by human factors, making it difficult to guarantee accuracy and reliability.

[0003] In the prior art, although some automatic screening methods use data processing and algorithm technology, there are still many shortcomings. On the one hand, for multi-dimensional heterogeneous quality hidden danger data in building engineering, existing methods lack effective fusion means and cannot fully exploit the correlation between different types of data, resulting in the value of data not being fully utilized, which in turn affects the comprehensiveness and accuracy of quality hidden danger screening. On the other hand, in terms of positioning algorithms, traditional algorithms often cannot accurately adapt to the complex and variable environment and parameter characteristics of building engineering, easily falling into local optimal solutions, making it difficult to quickly and accurately locate the real position of quality hidden dangers, and reducing screening efficiency and effectiveness.

[0004] Therefore, there is an urgent need for a more efficient and accurate building engineering quality hidden danger positioning screening method and system. SUMMARY

[0005] In order to overcome the shortcomings and deficiencies of the prior art, the present application provides a positioning screening method and system for building engineering quality hidden dangers.

[0006] The technical solution adopted by the present application is a positioning screening method for building engineering quality hidden dangers, which comprises:

[0007] Step S1: Obtain various parameters of building engineering quality hidden dangers, divide the obtained various parameters into three categories of structure parameters, material parameters and environmental parameters according to data types, the structure parameters cover building component size deviation and structure connection node condition; the material parameters include material strength and durability index; the environmental parameters involve temperature and humidity and load condition;

[0008] Step S2: Use a multi-dimensional heterogeneous building hidden danger data screening model to perform data fusion processing on the divided structure parameters, material parameters and environmental parameters, and integrate different dimensions and different types of data into a data set with correlation;

[0009] Step S3: Based on the optimized bat positioning algorithm and the fused data set, an initial search population is established in the spatial dimension of the construction project, and the initial search position and search speed are determined;

[0010] Step S4: Setting a fitness function based on various parameters of the construction project quality hazards, and performing fitness evaluation on each individual in the initial search population to determine the degree of matching between the individual and the potential quality hazards;

[0011] Step S5: According to the iterative rules of the optimized bat positioning algorithm, the initial search population is iteratively updated, and the position and speed of the individual are continuously adjusted to gradually approach the true location of the quality hazards of the construction project;

[0012] Step S6: When the preset iteration termination condition is met, the position of the individual with the best fitness in the current search population is output, which is the positioning result of the quality hidden danger of the construction project.

[0013] Furthermore, in step S2, the data fusion process of the multi-dimensional heterogeneous building hidden danger data screening model adopts the following model formula:

[0014] ,

[0015] in, is the fused data set; Indicates the number of data types, in this scenario , corresponding to structural parameters, material parameters and environmental parameters; For the The weight coefficient of each type of data is set according to the importance of each type of data to the screening of quality hazards in construction projects; Indicates the Class data; is the adjustment coefficient, which is used to control the influence of the correlation between data on the fusion results; For the Class data and The covariance of class data reflects the degree of coordinated changes between data; and Respectively Class data and The variance of the class data measures the degree of dispersion of the data.

[0016] Furthermore, in step S4, the fitness function constructs the following model formula:

[0017] ,

[0018] in, For individuals The fitness value of the number of quality hidden trouble parameters of the construction project; the importance coefficient of the quality hidden trouble parameter, which is set according to the key degree of the parameter to the quality hidden trouble judgment; the individual corresponding to the quality hidden trouble parameter; the predicted value of the quality hidden trouble parameter.

[0019] Further, in the iterative updating process of the optimized bat positioning algorithm, the individual position updating formula is:

[0020] ,

[0021] wherein, is the position of the individual at the moment; is the position of the individual at the moment; is the speed of the individual at the moment; is the position updating step coefficient, which is used to control the amplitude of position updating; is a random number between 0 and 1, and the individual speed updating formula is:

[0022] ,

[0023] wherein, is the speed of the individual at the moment; is the speed inertia coefficient, which reflects the inheritance degree of the individual to the previous speed; is the speed adjustment coefficient, which controls the trend of the individual moving to the global optimal position .

[0024] Further, before the fusion processing of the data, based on each parameter of the construction engineering quality hidden trouble, the multi-dimensional heterogeneous building hidden trouble data screening model is used to detect the abnormal value of the data, and the abnormal value judgment formula is:

[0025] ,

[0026] wherein, is the abnormal value identification of the data, 1 indicating an abnormal value and 0 indicating a normal value; is the individual data; is the mean of the data set; is the standard deviation of the data set; is the abnormal value judgment threshold, which is set according to the characteristics of the construction project quality hidden danger parameters.

[0027] Furthermore, in step S4, when evaluating the fitness of an individual, various parameters of the quality hazards of the construction project are combined, the coupling relationship between the parameters is considered, and a coupling correction coefficient is introduced. The fitness correction formula is:

[0028] ,

[0029] in, For the corrected individual The fitness value of Individual before correction The fitness value of is the number of parameter pairs with coupling relationship; For the The correction weight coefficient of the coupling relationship; For the The coupling strength coefficient of the coupling relationship is determined by analyzing the correlation degree between the quality hidden danger parameters of the construction project.

[0030] Furthermore, the step S3 includes the following steps:

[0031] Step S3.1: Determine the size of the initial search population of the optimized bat positioning algorithm based on the spatial dimensions of the construction project and the distribution range of the quality hazard parameters, so that the initial search population can cover the spatial areas where quality hazards may exist in the construction project;

[0032] Step S3.2: In the three-dimensional spatial coordinate system of the construction project, assign initial position coordinates to each individual in the initial search population in a random distribution manner to ensure the randomness and diversity of the initial positions;

[0033] Step S3.3: Set an initial search speed for each individual in the initial search population. The magnitude and direction of the initial speed are set according to the changing trend and spatial characteristics of the construction quality hazard parameters to guide the individual to conduct an effective search in space.

[0034] Step S3.4: Initialize and verify the initial search population to check whether the initial position and speed of the individuals meet the actual space and parameter constraints of the construction project. If not, adjust them.

[0035] Furthermore, the step S4 includes the following steps:

[0036] Step S4.1: According to the parameters of the quality hidden danger of the construction project, the weight and standard threshold of each parameter in the fitness function are determined, and the weight and threshold are set based on the design specification and quality acceptance standard of the construction project;

[0037] Step S4.2: Map the individual position information in the initial search population to the construction project quality hidden danger parameters, and obtain the predicted value of each individual for each quality hidden danger parameter;

[0038] Step S4.3: According to the calculation formula of the fitness function, the predicted value of each individual is calculated with the standard threshold, and the preliminary fitness value of the individual is obtained;

[0039] Step S4.4: The preliminary fitness value is normalized to make the fitness value in a unified numerical interval, which is convenient for the fitness comparison and screening between individuals.

[0040] Further, the step S5 comprises the following steps:

[0041] Step S5.1: Calculate the distance between individuals in the current search population, and judge the similarity between individuals according to the individual distance, which provides the basis for subsequent information exchange and position update;

[0042] Step S5.2: According to the iteration rule of the optimized bat positioning algorithm, the speed of the individual with slow search speed and low fitness is enhanced, and the search efficiency and exploration ability are improved;

[0043] Step S5.3: The position of the individual close to the global optimal position is fine-tuned to avoid falling into local optimal solution, so that it can more accurately approach the real position of the construction project quality hidden danger;

[0044] Step S5.4: According to the fitness value and search history record of the individual, the related coefficients in the individual position and speed update formula are dynamically adjusted to optimize the search process.

[0045] A positioning and screening system for construction project quality hidden danger, comprising:

[0046] A parameter acquisition unit is used to acquire various parameters of the construction project quality hidden danger and classify them according to data types;

[0047] A data fusion processing unit connected with the parameter acquisition unit is used to use a multi-dimensional heterogeneous construction hidden danger data screening model to perform data fusion processing on the classified parameters;

[0048] A population initialization unit connected with the data fusion processing unit is used to establish an initial search population based on the fused data based on the optimized bat positioning algorithm;

[0049] A fitness evaluation unit, connected to the population initialization unit, is used to set a fitness function based on various parameters of the construction project quality hidden dangers and perform fitness evaluation on the initial search population;

[0050] an iterative updating unit, connected to the fitness evaluation unit, for iteratively updating the search population according to the iterative rule of the optimized bat positioning algorithm;

[0051] The result output unit is connected to the iterative update unit and is used to output the individual position with the best fitness, that is, the positioning result of the quality hidden danger of the construction project, when the preset iteration termination condition is met.

[0052] Beneficial effects: The present invention proposes a method and system for locating and screening quality hazards of construction projects. The method utilizes a multi-dimensional heterogeneous building hazard data screening model, and through a specific data fusion formula, fully considers factors such as weights and correlations between data, integrates different types of parameters such as structure, material, and environment into a data set with correlation, and performs outlier detection, deeply mines the value of data, and ensures the comprehensiveness and accuracy of data processing. In terms of positioning algorithms, based on the optimized bat positioning algorithm, a unique fitness function is constructed and modified in combination with the parameter coupling relationship. At the same time, innovative position and speed update formulas and refined iterative operations on the search population are used, such as dynamic adjustment coefficients, avoiding local optimal strategies, etc., to accurately adapt to the complex environment and parameter characteristics of construction projects, and quickly and accurately locate the true location of quality hazards. The present invention significantly improves the efficiency and accuracy of screening quality hazards of construction projects through a series of steps such as parameter classification, data fusion, population initialization, fitness evaluation, iterative update and result output, and cooperates with the various functional units of the system to operate in coordination, providing reliable technical support for the quality and safety management of construction projects. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 is a flow chart of the method steps of the present invention;

[0054] Figure 2 It is a diagram of the system unit composition of the present invention. DETAILED DESCRIPTION

[0055] It should be noted that, unless there is a conflict, the embodiments in this application and the features in the embodiments can be combined with each other. The application is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0056] like Figure 1 As shown, a method for locating and screening hidden dangers in construction project quality is provided, the method comprising:

[0057] Step S1: Obtain various parameters of construction quality hidden dangers, and divide the obtained various parameters into three categories of structure parameters, material parameters and environmental parameters according to data types, wherein the structure parameters cover building component size deviation and structure connection node condition; the material parameters include material strength and durability index; and the environmental parameters relate to temperature and humidity and load condition;

[0058] Specifically, in step S1, various parameters of construction quality hidden dangers are obtained by professional measuring instruments, sensors and engineering data collection, etc. These parameters cover multiple aspects and are then strictly divided into three categories of structure parameters, material parameters and environmental parameters. Among them, the structure parameters focus on the construction characteristics, such as the building component size deviation directly reflecting the difference between the actual size of the component and the designed size, and the size deviation too large may affect the stability of the overall structure of the building; the structure connection node condition relates to the firmness of the node, and the node connection not tight or with defects will weaken the bearing capacity of the structure. The material parameters emphasize the performance of the material, and the material strength is a key indicator to measure the ability of the material to withstand external force, and the substandard strength will reduce the safety of the building; the durability index determines the ability of the material to resist environmental erosion in the long-term use, and the poor durability will shorten the service life of the building. The environmental parameters mainly consider the influence of external environment on the building, and the change of temperature and humidity may cause the change of physical and chemical properties of the material, such as the deformation of wood due to moisture; the load condition includes various weights and external forces borne by the building, and the load exceeding the design range will bring great pressure to the building structure.

[0059] This step classifies the obtained complex parameters systematically, so that the originally scattered data becomes orderly, providing a clear framework for subsequent data processing and analysis. Different types of parameters have different ways and degrees of influence on construction quality hidden dangers, and after classification, in-depth research and processing can be carried out according to the characteristics of each type of parameter. For example, in the subsequent data fusion process, different weights can be given according to the characteristics of various parameters, so that the fused data can more accurately reflect the actual quality condition of the construction project, thereby improving the accuracy and reliability of the quality hidden danger screening.

[0060] Step S2: Use a multi-dimensional heterogeneous building hidden danger data screening model to perform data fusion processing on the divided structure parameters, material parameters and environmental parameters, and integrate data of different dimensions and different types into a correlated data set;

[0061] Specifically, step S2 applies a multi-dimensional heterogeneous building hidden danger data screening model to implement data fusion processing on the structure parameters, material parameters and environmental parameters divided in step S1. In the implementation process, the model analyzes the internal relationship and interaction between different types of data according to specific rules and algorithms. Since these parameters have different dimensions and types, such as structure parameters are mostly geometric and mechanical data, material parameters focus on physical and chemical performance data, and environmental parameters involve meteorological and load data, the model needs to integrate them together using appropriate methods. Specifically, the model will consider the weight of the data, that is, according to the importance of each type of data to building engineering quality hidden danger screening, give each type of data a corresponding weight coefficient. The data with high importance occupies a larger proportion in the fusion process to highlight its key role in quality hidden danger judgment. At the same time, the model will also consider the correlation between data, by analyzing the covariance and variance of data and other indicators, understand the degree of cooperative change and dispersion between data, so as to more reasonably fuse the data and form a data set with correlation.

[0062] The generation of building engineering quality hidden danger is often the result of the joint action of multiple factors, and single type of data cannot fully reflect the quality status. This step combines different dimensions and types of data through data fusion, which can dig out the information and rules hidden behind the data and make full use of the value of the data. The data set after fusion contains more rich information, which can describe the quality status of building engineering from multiple angles, provide more comprehensive and accurate data support for the subsequent quality hidden danger positioning based on optimized bat positioning algorithm, and avoid the missed detection or misjudgment of quality hidden danger caused by one-sided data.

[0063] Step S3: Based on the optimized bat positioning algorithm, the data set after fusion is used as the basis to establish an initial search population in the spatial dimension of building engineering, determine the initial search position and search speed;

[0064] Specifically, step S3 is based on the optimized bat positioning algorithm, and the data set fused in step S2 is used as the basis to establish an initial search population in the spatial dimension of the construction project and determine the initial search position and speed. In an embodiment, first, according to the spatial size of the construction project, such as the length, width, height of the building and the layout of each floor and other information, and combining the distribution range of the quality hidden danger parameters, the size of the initial search population is comprehensively determined. Ensure that the population size can cover the spatial area where the quality hidden danger may exist in the construction project, and will not increase the calculation complexity and time cost due to the size being too large. Then, in the three-dimensional coordinate system of the construction project, each individual in the initial search population is given an initial position coordinate according to the principle of random distribution. This random distribution method ensures the diversity of the initial position, avoids the concentration of individuals in a certain area, and thus improves the possibility of searching for quality hidden dangers by the algorithm. Finally, according to the change trend and spatial characteristics of the quality hidden danger parameters of the construction project, the initial search speed of each individual is set. For example, if it is known that the environmental parameters in a certain area change dramatically and there may be quality hidden dangers, individuals near this area will be given a relatively large initial speed in order to search this area faster.

[0065] This step is the initial stage of using the optimized bat positioning algorithm to locate the quality hidden danger. The initial search population is established and the initial position and speed are reasonably determined, which lays the foundation for the subsequent search process of the algorithm. Suitable initial conditions can guide the algorithm to search more effectively in the space of the construction project, avoid blind search, and improve search efficiency. The diversity of the initial search population and the reasonable initial speed setting enable the algorithm to explore potential quality hidden dangers in a wider spatial range, increase the probability of discovering quality hidden dangers, and provide a good start for the subsequent gradual approximation of the real position of the quality hidden danger through fitness evaluation and iterative updating.

[0066] Step S4: According to the parameters of the quality hidden danger of the construction project, set the fitness function, evaluate the fitness of each individual in the initial search population, and judge the matching degree with the potential quality hidden danger;

[0067] Specifically, step S4 sets a fitness function according to the parameters of the construction quality hidden danger, evaluates the fitness of each individual in the initial search population established in step S3, and judges the matching degree with the potential quality hidden danger. In implementation, first, according to the parameters of the construction quality hidden danger, combined with the design specification and quality acceptance standard of the construction project, the weight and standard threshold of each parameter in the fitness function are determined. The setting of these weights and thresholds is based on the in-depth understanding and analysis of the quality requirements of the construction project. Different parameters have different key degrees in judging the quality hidden danger, so different weights are given, and the standard threshold is used as the basis for measuring whether the parameters meet the quality requirements. Then, the individual position information in the initial search population is mapped with the construction quality hidden danger parameters, and the predicted value of each quality hidden danger parameter corresponding to each individual is obtained through a specific calculation method. Next, according to the calculation formula of the fitness function, the predicted value of the individual is compared and calculated with the standard threshold, and the preliminary fitness value of the individual is obtained. Finally, in order to facilitate the comparison and selection of the fitness of individuals, the preliminary fitness value is normalized to make the fitness value in a unified numerical interval.

[0068] This step provides a method for quantitatively evaluating the matching degree of individuals and quality hidden dangers for the optimized bat positioning algorithm. Through the calculation of the fitness function, individuals that are closer to the potential quality hidden danger can be selected from a large number of individuals, providing a clear direction for the iterative update of the algorithm. Only by accurately evaluating the fitness of individuals can the algorithm continuously adjust the position and speed of individuals in the subsequent iteration process, gradually approaching the true position of the quality hidden danger, thereby realizing the accurate positioning of the construction quality hidden danger and improving the accuracy and reliability of the positioning screening.

[0069] Step S5: According to the iteration rules of the optimized bat positioning algorithm, the initial search population is iteratively updated, the position and speed of the individual are continuously adjusted, and the true position of the construction quality hidden danger is gradually approached;

[0070] Specifically, step S5 iteratively updates the initial search population evaluated in step S4 according to the iteration rules of the optimized bat positioning algorithm, continuously adjusts the position and speed of the individual, and gradually approaches the real position of the construction quality hidden danger. In the implementation process, first calculate the distance between individuals in the current search population, judge the similarity between individuals by distance, and determine the information exchange mode and range between individuals based on this. For individuals with slow search speed and low fitness, perform speed enhancement operation according to the iteration rules of the algorithm, adjust the size and direction of the speed, improve the search efficiency and exploration ability of the individual, so that it can more effectively search for potential quality hidden dangers in space. For individuals close to the global optimal position, in order to avoid falling into local optimal solution, fine-tune the position to make it jump out of the local optimal area and continue to approach the real position of the quality hidden danger more accurately. In addition, according to the fitness value and search history record of the individual, dynamically adjust the related coefficients in the individual position and speed update formula, such as the speed inertia coefficient and the speed adjustment coefficient, optimize the whole search process, so that the algorithm can better adapt to the complex requirements of construction quality hidden danger positioning.

[0071] This step is the core link of the optimized bat positioning algorithm to realize accurate positioning. Through continuous iteration and update, the algorithm can dynamically adjust the position and speed of the individual in the construction space according to the matching degree of the individual and the quality hidden danger, gradually reduce the search range, and improve the positioning accuracy. Each iteration is an optimization of the search result, so that the algorithm can overcome various interference factors in the complex construction environment and finally accurately find the real position of the quality hidden danger, so as to realize effective positioning screening of the construction quality hidden danger and ensure the quality and safety of the construction project.

[0072] Step S6: When the preset iteration termination condition is met, output the position of the individual with the optimal fitness in the current search population, which is the positioning result of the construction quality hidden danger.

[0073] Specifically, step S6 sets a clear termination condition. When the preset iteration termination condition is met, the position of the individual with the optimal fitness in the current search population is output, which is the positioning result of the construction engineering quality hidden danger. In the implementation, the preset iteration termination condition usually includes multiple cases. One common termination condition is to set a maximum number of iterations. When the number of iterations of the optimized bat positioning algorithm reaches the pre-set maximum number, the iteration process is stopped regardless of whether the absolutely accurate quality hidden danger position is found. This is to avoid wasting a large amount of computing resources and time due to the algorithm being trapped in an infinite loop. Another termination condition is to determine according to the change of the fitness value. When the fitness value of the optimal individual in the search population changes by less than a certain pre-set threshold in consecutive multiple iterations, it indicates that the algorithm has approached convergence, i.e., it is close to finding the true position of the quality hidden danger, at which time the iteration can also be terminated. Once the termination condition is met, the system outputs the position of the individual with the optimal fitness in the current search population, which is the position of the construction engineering quality hidden danger determined by the algorithm after a series of searches and iterations.

[0074] This step provides a clear result for the positioning of construction engineering quality hidden dangers, enabling engineering personnel to accurately know the location of the quality hidden danger, so that targeted measures can be taken for repair and treatment. By setting reasonable termination conditions, the execution efficiency of the algorithm is improved while ensuring the accuracy of positioning, avoiding unnecessary waste of computing resources. The output quality hidden danger positioning result provides an important basis for construction engineering quality and safety management, which helps to eliminate quality hidden dangers in a timely manner and ensure the normal use of construction engineering and personnel safety.

[0075] Preferably, in the step S2, the data fusion process of the multi-dimensional heterogeneous building hidden danger data screening model adopts the following model formula:

[0076]

[0077] wherein, is the fused data set; represents the number of data types, in the present scenario corresponding to the structure parameters, material parameters and environmental parameters; is the weight coefficient of the th data, which is set according to the importance of each type of data for construction engineering quality hidden danger screening; represents the th data; is the adjustment coefficient, used to control the influence degree of the correlation between data on the fusion result; is the covariance of the th data and the th data, reflecting the degree of cooperative change between data; and​ the variance of the first class data and the variance of the first class data, which measure the dispersion degree of the data.

[0078] Specifically, in the data fusion process of step S2, the multi-dimensional heterogeneous building hidden danger data screening model is used to realize the deep integration of different types of parameters. The model sets weight coefficients for structure, material and environment parameters according to the importance of the data, and the size of the weight coefficient depends on the key degree of each type of data to quality hidden danger screening. At the same time, the model introduces an adjustment coefficient to control the influence of the correlation between data on the fusion result. Through the analysis of the covariance and variance of the data, the degree of collaborative change and dispersion between the data is accurately captured, and finally a data set with strong correlation is formed. This fusion method not only retains the unique characteristics of each type of data, but also excavates the potential relationship between the data, significantly improving the utilization value of the data and providing a more comprehensive and accurate data basis for subsequent hidden danger positioning.

[0079] Preferably, in step S4, the fitness function is constructed as follows:

[0080] ,

[0081] wherein, is the fitness value of the individual; is the number of building engineering quality hidden danger parameters; is the importance coefficient of the first quality hidden danger parameter, which is set according to the key degree of the parameter to the quality hidden danger judgment; is the predicted value of the individual corresponding to the first quality hidden danger parameter; is the standard threshold value of the first quality hidden danger parameter.

[0082] Specifically, in the fitness evaluation of step S4, a fitness function based on multiple parameters is constructed. The function assigns different importance coefficients according to the key degree of each quality hidden danger parameter to the judgment result, and the setting of the importance coefficient strictly follows the building engineering design specification and quality acceptance standard. By comparing the individual predicted value with the standard threshold value, the fitness value reflecting the matching degree of the individual and the potential quality hidden danger is obtained. The innovation of this function is to use the form of absolute value reciprocal, so that the individual whose parameter deviates from the standard threshold value more obtains a lower fitness value, thereby more clearly distinguishing the advantages and disadvantages of different individuals, and providing accurate evaluation basis for the iterative optimization of the algorithm.

[0083] Preferably, in step S5, in the iterative update process of the optimized bat positioning algorithm, the individual position update formula is:​

[0084] ,

[0085] in, For the Individuals in Position at the moment; For the Individuals in Position at the moment; For the Individuals in the speed of the moment; is the position update step coefficient, which is used to control the amplitude of position update; is a random number between 0 and 1, and the individual speed update formula is:

[0086] ,

[0087] in, For the Individuals in the speed of the moment; is the velocity inertia coefficient, which reflects the degree to which the individual inherits the previous velocity; is the speed adjustment coefficient, which controls the individual to the global optimal position The trend of movement.

[0088] Specifically, during the iterative update process of step S5, the position and velocity update mechanisms of the optimized bat positioning algorithm were improved. During position updates, a random perturbation factor was introduced to enable individuals to explore in random directions based on their original velocity, effectively expanding the search range and preventing the algorithm from falling into a local optimum. Velocity updates combine the individual's historical velocity and global optimal position information. The velocity inertia coefficient controls the degree to which the individual inherits the previous velocity, while the velocity adjustment coefficient guides the individual toward the global optimal position. This dual-coefficient adjustment mechanism enables the algorithm to maintain a certain degree of exploratory power during the search process while ensuring convergence to the optimal solution, significantly improving the accuracy and efficiency of positioning.

[0089] Preferably, in step S2, before the data is fused, based on various parameters of construction quality hidden dangers, a multi-dimensional heterogeneous building hidden danger data screening model is used to perform outlier detection on the data. The outlier judgment formula is:

[0090] ,

[0091] in, For the The outlier value identification of each data, 1 represents an outlier and 0 represents a normal value; For the individual data; is the mean of the data set; is the standard deviation of the data set; is the abnormal value judgment threshold, which is set according to the characteristics of the construction project quality hidden danger parameters.

[0092] Specifically, during the data preprocessing phase of step S2, effective outlier detection is achieved based on a multidimensional, heterogeneous building hazard data screening model. This detection mechanism, based on the data's mean and standard deviation, sets a specific threshold to identify data that deviates from the mean by more than the threshold as outliers. The threshold setting fully considers the characteristics of construction quality hazard parameters, with different threshold standards for different types of parameters. This outlier detection method can promptly identify errors or mutations in data collection, ensuring the reliability of subsequent data fusion and analysis and preventing abnormal data from interfering with hazard location results.

[0093] Preferably, in step S4, when evaluating the fitness of an individual, various parameters of the quality hazards of the construction project are combined, the coupling relationship between the parameters is considered, and a coupling correction coefficient is introduced. The fitness correction formula is:

[0094] ,

[0095] in, For the corrected individual The fitness value of Individual before correction The fitness value of is the number of parameter pairs with coupling relationship; For the The correction weight coefficient of the coupling relationship; For the The coupling strength coefficient of the coupling relationship is determined by analyzing the correlation degree between the quality hidden danger parameters of the construction project.

[0096] Specifically, the fitness assessment in step S4 further considers the coupling relationship between the parameters of construction quality hazards. By introducing a coupling correction coefficient, the original fitness value is corrected. During the correction process, the number of parameter pairs with coupling relationships is first determined, and then a corresponding correction weight coefficient and coupling strength coefficient are set for each pair of coupled parameters. The coupling strength coefficient is determined by analyzing the degree of correlation between the parameters and reflects the degree of mutual influence between the parameters. This correction mechanism makes the fitness assessment more consistent with the actual situation of construction projects. In actual projects, multiple parameters often interact and jointly influence the occurrence of quality hazards, thereby improving the accuracy of the assessment results.

[0097] Preferably, the step S3 comprises the following steps:

[0098] Step S3.1: According to the spatial size of the construction project and the distribution range of the quality defect parameters, the size of the initial search population of the optimized bat positioning algorithm is determined, so that the initial search population can cover the spatial area where the quality defects may exist in the construction project;

[0099] Step S3.2: In the three-dimensional spatial coordinate system of the construction project, initial position coordinates are assigned to each individual in the initial search population in a random distribution manner, ensuring the randomness and diversity of the initial positions;

[0100] Step S3.3: Set the initial search speed for each individual in the initial search population, and the size and direction of the initial speed are set according to the variation trend and spatial characteristics of the construction quality defect parameters, to guide the individual to search effectively in space;

[0101] Step S3.4: Perform initialization verification on the initial search population, check whether the initial position and speed of the individual meet the actual spatial and parameter constraints of the construction project, and adjust if not.

[0102] Specifically, step S3 includes four sub-steps. First, according to the spatial size of the construction project and the distribution range of the quality defect parameters, the size of the initial search population is accurately calculated to ensure that the population can cover all areas where defects may exist. Second, the initial position of the individual is randomly assigned in the three-dimensional coordinate system, and this random distribution strategy ensures the comprehensiveness and diversity of the search. Then, according to the variation trend and spatial characteristics of the quality defect parameters, the initial speed of the individual is set, so that the individual can search effectively in the direction where the defects are most likely to exist. Finally, the initial population is verified to ensure that the position and speed of each individual meet the actual spatial and parameter constraints of the construction project, and the individual that does not meet the requirements will be adjusted, thereby ensuring the rationality of the initial conditions of the algorithm.

[0103] Preferably, the step S4 comprises the following steps:

[0104] Step S4.1: According to the parameters of the construction quality defects, determine the weights and standard thresholds of each parameter in the fitness function, and the weights and thresholds are set based on the design specifications and quality acceptance standards of the construction project;

[0105] Step S4.2: Map the individual position information in the initial search population to the construction quality defect parameters to obtain the predicted values of each quality defect parameter corresponding to each individual;

[0106] Step S4.3: Calculate the predicted values of each individual and the standard threshold according to the calculation formula of the fitness function to obtain the preliminary fitness value of the individual;

[0107] Step S4.4: Normalize the preliminary fitness values ​​so that they are within a uniform numerical range, facilitating fitness comparison and screening between individuals.

[0108] Specifically, step S4 includes four sub-steps. First, based on the various parameters of the quality hazards of the construction project, combined with the design specifications and quality acceptance standards, the weights and standard thresholds of the parameters in the fitness function are determined to ensure the scientificity and authority of the evaluation standards. Secondly, through a specific mapping method, the individual's position information is converted into the predicted value of the corresponding quality hazard parameters, and the association between the position and the parameter is established. Then, according to the calculation formula of the fitness function, the predicted value of each individual is quantitatively compared with the standard threshold to obtain a preliminary fitness value. Finally, the preliminary fitness value is normalized to eliminate the influence of different parameter dimensions, so that the fitness values ​​of all individuals are in a unified comparable range, which is convenient for subsequent screening and iterative optimization.

[0109] Preferably, the step S5 comprises the following steps:

[0110] Step S5.1: Calculate the distance between individuals in the current search population and determine the similarity between individuals based on the individual distance, providing a basis for subsequent information exchange and location update;

[0111] Step S5.2: Based on the iterative rules of the optimized bat positioning algorithm, speed enhancement operations are performed on individuals with slow search speed and low fitness to improve their search efficiency and exploration ability;

[0112] Step S5.3: Fine-tune the positions of individuals close to the global optimal position to prevent them from falling into the local optimal solution, so that they can more accurately approach the true location of the quality hazards of the construction project;

[0113] Step S5.4: According to the individual's fitness value and search history, dynamically adjust the correlation coefficient in the individual position and speed update formula to optimize the search process.

[0114] Specifically, step S5 includes four sub-steps. First, the distance between individuals in the current search population is calculated, and the degree of similarity of individuals is judged through distance analysis to provide a basis for subsequent information exchange and position update. For individuals with slow search speed and low fitness, speed enhancement operation is implemented to improve their search efficiency and exploration ability by adjusting their speed and direction. For individuals close to the global optimal position, position fine-tuning is performed to prevent them from falling into the local optimal solution, ensuring that the true location of quality hazards can be approached more accurately. Finally, according to the individual's fitness value and search history, the correlation coefficient in the position and speed update formula is dynamically adjusted, so that the algorithm can adaptively optimize the search strategy according to the search situation and improve the accuracy and efficiency of positioning.

[0115] As Figure 2 shown, a positioning screening system for construction engineering quality hidden dangers, comprising:

[0116] A parameter acquisition unit is configured to acquire various parameters of construction engineering quality hidden dangers and classify them according to data types.

[0117] A data fusion processing unit is connected to the parameter acquisition unit and configured to perform data fusion processing on the classified parameters by using a multi-dimensional heterogeneous building hidden danger data screening model.

[0118] A population initialization unit is connected to the data fusion processing unit and configured to establish an initial search population based on the fused data according to an optimized bat positioning algorithm.

[0119] A fitness evaluation unit is connected to the population initialization unit and configured to set a fitness function according to various parameters of construction engineering quality hidden dangers and evaluate the fitness of the initial search population.

[0120] An iteration updating unit is connected to the fitness evaluation unit and configured to perform iteration updating on the search population according to the iteration rules of the optimized bat positioning algorithm.

[0121] A result output unit is connected to the iteration updating unit and configured to output the position of the individual with the optimal fitness, i.e., the positioning result of the construction engineering quality hidden danger, when the preset iteration termination condition is met.

[0122] The present application innovates from two core dimensions of data processing and algorithm optimization. In the aspect of data processing, the prior art cannot effectively integrate multi-source heterogeneous data, resulting in information isolation and one-sided screening. The present application uses a multi-dimensional heterogeneous building hidden danger data screening model to deeply fuse parameters such as structure, material and environment according to weights and correlation, and uses an outlier judgment formula to eliminate interference data, thereby constructing a data set with strong correlation and providing a comprehensive and accurate data basis for subsequent analysis, and completely changing the fragmentation and inefficiency of traditional data processing.

[0123] In the aspect of positioning algorithm, traditional algorithms often fall into local optimum due to their inability to adapt to complex building environments, making it difficult to accurately locate hidden dangers. The present application uses an optimized bat positioning algorithm, constructs a dedicated fitness function and introduces a coupling correction coefficient, fully considers the key degree of quality hidden danger parameters and the coupling relationship between parameters, and accurately evaluates the matching degree of individuals and hidden dangers. In the iteration process, the innovative position and speed updating formula is combined with individual distance calculation, speed enhancement, position fine-tuning and dynamic coefficient adjustment to make the algorithm quickly and stably approach the real position of the hidden danger, significantly improve the positioning efficiency and accuracy, and effectively solve the technical bottlenecks of poor adaptability and large positioning deviation of traditional algorithms.

[0124] In addition, the present positioning screening system forms a complete and closed screening process through the close cooperation of six functional units of parameter acquisition, data fusion processing, population initialization, fitness evaluation, iterative updating and result output. Each unit has its own function and cooperates with each other, from data source control to final result output, realizing the full-process automation and intelligentization of building engineering quality hidden danger screening. Compared with the prior art, the screening efficiency is greatly improved, the labor cost is reduced, a reliable and efficient technical support is provided for building engineering quality and safety management, and significant technical advantages and application value are shown.

[0125] In the description of the present application, it should be noted that unless otherwise explicitly specified and limited, the terms "arrangement", "installation", "connection", "connection", "fixing" should be understood broadly, for example, can be fixedly connected, can also be detachably connected, or integrally connected, can be mechanically connected, can also be electrically connected, can be directly connected, can also be indirectly connected through an intermediate medium, and can be the communication inside two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0126] Although the embodiments of the present application have been shown and described, it can be understood by those skilled in the art that various equivalent changes, modifications, replacements and variations of the embodiments can be made without departing from the principles and spirits of the present application, and the scope of the present application is defined by the appended claims and their equivalent scope.

Claims

1. A method for locating and screening hidden dangers in construction engineering quality, characterized in that: The method includes: Step S1: Obtain various parameters of construction quality hazards, and classify the obtained parameters into three categories according to data type: structural parameters, material parameters, and environmental parameters. The structural parameters include building component size deviations and structural connection node conditions; material parameters include material strength and durability indicators; and environmental parameters include temperature, humidity, and load conditions. Step S2: Using the multi-dimensional heterogeneous building hidden danger data screening model, data fusion processing is performed on the divided structural parameters, material parameters and environmental parameters, and data of different dimensions and types are integrated into a related data set; Step S3: Based on the optimized bat positioning algorithm and the fused data set, an initial search population is established in the spatial dimension of the construction project, and the initial search position and search speed are determined; Step S4: Setting a fitness function based on various parameters of the construction project quality hazards, performing fitness evaluation on each individual in the initial search population, and determining the degree of matching with the potential quality hazards; Step S5: According to the iterative rules of the optimized bat positioning algorithm, the initial search population is iteratively updated, and the position and speed of the individual are continuously adjusted to gradually approach the true location of the quality hazards of the construction project; Step S6: When the preset iteration termination condition is met, the location of the individual with the best fitness in the current search population is output, which is the location result of the construction project quality hidden danger; In step S2, the data fusion process of the multi-dimensional heterogeneous building hidden danger data screening model adopts the following model formula: ； in, is the fused data set; Indicates the number of data types, in this scenario , corresponding to structural parameters, material parameters and environmental parameters; For the The weight coefficient of each type of data is set according to the importance of each type of data to the screening of quality hazards in construction projects; Indicates the Class data; is the adjustment coefficient, which is used to control the influence of the correlation between data on the fusion results; For the Class data and The covariance of class data reflects the degree of coordinated changes between data; and Respectively Class data and The variance of the class data measures the degree of dispersion of the data; In step S4, the fitness function constructs the following model formula: ； in, For individuals The fitness value of is the number of construction engineering quality hidden danger parameters; For the The importance coefficient of each quality hazard parameter is set according to the criticality of the parameter to the quality hazard judgment; For individuals Corresponding to The predicted value of each quality risk parameter; For the The standard threshold value of each quality risk parameter.

2. A method for locating and screening hidden dangers in construction engineering quality according to claim 1, characterized in that: In step S5, during the iterative update process of the optimized bat positioning algorithm, the individual position update formula is: ； in, For the Individuals in Position at the moment; For the Individuals in Position at the moment; For the Individuals in the speed of the moment; is the position update step coefficient, which is used to control the amplitude of position update; is a random number between 0 and 1, and the individual speed update formula is: ； in, For the Individuals in the speed of the moment; is the velocity inertia coefficient, which reflects the degree to which the individual inherits the previous velocity; is the speed adjustment coefficient, which controls the individual to the global optimal position The trend of movement.

3. A method for locating and screening hidden dangers in construction engineering quality according to claim 1, characterized in that: In step S2, before the data is fused, based on the various parameters of the construction quality hidden dangers, a multi-dimensional heterogeneous building hidden danger data screening model is used to detect outliers on the data. The outlier judgment formula is: ； in, For the The outlier value identification of each data, 1 represents an outlier and 0 represents a normal value; For the individual data; is the mean of the data set; is the standard deviation of the data set; is the abnormal value judgment threshold, which is set according to the characteristics of the construction project quality hidden danger parameters.

4. A method for locating and screening hidden dangers in construction engineering quality according to claim 1, characterized in that: In step S4, when evaluating the fitness of an individual, various parameters of the quality hazards of the construction project are combined, the coupling relationship between the parameters is considered, and a coupling correction coefficient is introduced. The fitness correction formula is: ； in, For the corrected individual The fitness value of Individual before correction The fitness value of is the number of parameter pairs with coupling relationship; For the The correction weight coefficient of the coupling relationship; For the The coupling strength coefficient of the coupling relationship is determined by analyzing the correlation degree between the quality hidden danger parameters of the construction project.

5. The method for locating and screening hidden dangers in construction engineering quality according to claim 1, characterized in that: The step S3 comprises the following steps: Step S3.1: Determine the size of the initial search population of the optimized bat positioning algorithm based on the spatial dimensions of the construction project and the distribution range of the quality hazard parameters, so that the initial search population can cover the spatial areas where the construction project may have quality hazards; Step S3.2: Assign initial position coordinates to each individual in the initial search population in a random distribution in the three-dimensional space coordinate system of the construction project; Step S3.3: Set an initial search speed for each individual in the initial search population. The magnitude and direction of the initial speed are set according to the changing trend of the construction quality hidden danger parameters and the spatial characteristics, guiding the individual to search in space. Step S3.4: Initialize and verify the initial search population to check whether the initial position and speed of the individuals meet the actual space and parameter constraints of the construction project. If not, adjust them.

6. A method for locating and screening hidden dangers in construction engineering quality according to claim 1, characterized in that: The step S4 comprises the following steps: Step S4.1: Determine the weights and standard thresholds of the parameters in the fitness function based on the various parameters of the construction project quality hazards. The weights and thresholds are set based on the design specifications and quality acceptance standards of the construction project. Step S4.2: Map the individual position information in the initial search population to the construction project quality hazard parameters to obtain the predicted value of each quality hazard parameter corresponding to each individual; Step S4.3: Calculate the predicted value of each individual and the standard threshold according to the calculation formula of the fitness function to obtain the preliminary fitness value of the individual; Step S4.4: Normalize the preliminary fitness values ​​so that they are within a uniform numerical range, and perform fitness comparison and screening between individuals.

7. A method for locating and screening hidden dangers in construction engineering quality according to claim 1, characterized in that: The step S5 comprises the following steps: Step S5.1: Calculate the distance between individuals in the current search population and determine the similarity between individuals based on the individual distance, providing a basis for subsequent information exchange and location update; Step S5.2: According to the iterative rule of the optimized bat positioning algorithm, a speed enhancement operation is performed on the individuals with slow search speed and low fitness; Step S5.3: Fine-tune the positions of individuals close to the global optimal position to continuously approach the true location of the quality hazards of the construction project; Step S5.4: According to the individual's fitness value and search history, dynamically adjust the coefficients in the individual's position and speed update formula to optimize the search process.

8. A positioning and screening system for hidden dangers in construction engineering quality, characterized in that: include: A parameter acquisition unit is used to obtain various parameters of construction project quality hazards and classify them according to data types; A data fusion processing unit, connected to the parameter acquisition unit, is used to perform data fusion processing on the classified parameters using a multi-dimensional heterogeneous building hidden danger data screening model; A population initialization unit, connected to the data fusion processing unit, for establishing an initial search population based on the fused data based on an optimized bat positioning algorithm; A fitness evaluation unit, connected to the population initialization unit, is used to set a fitness function based on various parameters of the construction project quality hidden dangers and perform fitness evaluation on the initial search population; an iterative updating unit, connected to the fitness evaluation unit, for iteratively updating the search population according to the iterative rule of the optimized bat positioning algorithm; The result output unit is connected to the iterative update unit and is used to output the individual position with the best fitness, that is, the positioning result of the quality hidden danger of the construction project, when the preset iteration termination condition is met.

Images