A method, apparatus, device, and medium for safety rating of a concrete roof deck

Abstract

The present application relates to a kind of concrete roof panel safety rating method, device, equipment and medium, belong to industrial building structure diagnosis treatment technical field, its method includes: obtaining at least one influence factor of any component;At least one influence factor of component is preprocessed, and the decision tree algorithm in the rating model trained is handled to the influence factor after pre-processing, obtains at least one rating decision result of component;At least one rating decision result is handled by voting by the random forest algorithm in the rating model trained, and the safety rating result of component is obtained according to the voting result;Based on the safety rating result of multiple components of concrete roof panel, the safety rating result of concrete roof panel is obtained.The reliability of concrete roof panel is evaluated by the present application, and the structural defects or deficiencies of each component of concrete roof panel are found in time, which helps to take targeted maintenance measures in time, thereby prolonging the overall service life of building.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of industrial building structure diagnosis and treatment, and in particular to a concrete roof panel safety rating method, device, equipment and medium. BACKGROUND

[0002] The traditional industrial building structure diagnosis and treatment method has a technical development bottleneck and needs to be reformed. The detection of the concrete roof panel is usually completed by detection experts through on-site investigation and measurement. The task is heavy, the detection means is diverse, the structure is complex, and the like. Long construction period, large personnel investment, strong experience dependence and the like are prone to difficulties. The experience dependence is high during data processing. There is no standard processing standard and process. The detection means is one-time only. Only the component condition at the time of detection can be reflected. The safety of the structure cannot be monitored in real time. Safety accidents cannot be prevented. SUMMARY

[0003] To solve the above problems, the present application provides a concrete roof panel safety rating method, device, equipment and medium.

[0004] The first aspect of the embodiment of the present application provides a concrete roof panel safety rating method. The concrete roof panel includes a plurality of components, which includes:

[0005] At least one influence factor of any component is obtained. The influence factor includes: reinforced concrete damage information, structure information of the concrete roof panel, size information of the concrete roof panel, material strength of the concrete roof panel, displacement deformation information of the concrete roof panel and carbonation depth information of the concrete roof panel;

[0006] The at least one influence factor of the component is preprocessed to obtain a preprocessed influence factor;

[0007] The preprocessed influence factor is processed through a decision tree algorithm in a trained rating model to obtain at least one rating decision result of the component;

[0008] At least one rating decision result is processed through a random forest algorithm in the trained rating model to obtain a voting result. The safety rating result of the component is obtained according to the voting result;

[0009] Based on the safety rating results of the plurality of components of the concrete roof panel, the safety rating result of the concrete roof panel is obtained.

[0010] Optionally, the at least one influence factor of the component is preprocessed to obtain a preprocessed influence factor. The preprocessed influence factor is processed through a decision tree algorithm in a trained rating model to obtain at least one rating decision result of the component, which specifically includes:

[0011] Encode the feature data of at least one influencing factor of the concrete roof panel to obtain encoded feature data, assign values to the encoded feature data and mark the rating labels to obtain the encoded feature data and the rating labels corresponding to the encoded feature data;

[0012] Determine the encoded feature data and the rating labels corresponding to the encoded feature data by the decision tree algorithm in the trained rating model to obtain a rating decision result.

[0013] Optionally, historical original data of the concrete roof panel is obtained; wherein the historical original data includes at least one influencing factor of each component of the concrete roof panel;

[0014] Data cleaning is performed on the historical original data to obtain target feature data;

[0015] A data set of the rating model is constructed according to the target feature data of each component;

[0016] The data set is divided into a training set and a test set;

[0017] The initial rating model is trained through the training set to obtain a trained rating model;

[0018] The trained rating model is verified through the test set to obtain a trained rating model.

[0019] Optionally, the data cleaning on the historical original data to obtain the target feature data specifically includes:

[0020] The feature data in the influencing factor of each component is classified to obtain classified feature data;

[0021] The classified feature data is encoded to obtain encoded feature data, and the encoded feature data is assigned values and marked rating labels to obtain the target feature data.

[0022] Optionally, it further includes:

[0023] The corresponding rating standard value is matched according to the numerical value of the feature data; wherein the rating standard value is determined according to the feature data of components of different safety levels in a preset identification table;

[0024] The safety level corresponding to the rating standard value is obtained;

[0025] The safety level is encoded as the rating label corresponding to the feature data.

[0026] Optionally, it further includes:

[0027] In the case where the influence factor is missing, the missing values of the historical original data are filled by a random forest-based data imputation algorithm to obtain optimized historical original data.

[0028] Optionally, the reinforced concrete damage information includes defect information or damage information of the reinforced concrete.

[0029] The structural information of the concrete roof panel includes load information of the concrete roof panel.

[0030] The size information of the concrete roof panel includes cross-sectional size information of a flange of the concrete roof panel and steel bar size information of the concrete roof panel.

[0031] The material strength of the concrete roof panel includes concrete compressive strength and concrete hardness.

[0032] The displacement deformation information of the concrete roof panel includes longitudinal deviation, lateral deviation, inclination value, measured height, inclination rate, span, relative support height difference in the middle of the span, deflection, deformation rate, and vertical deformation of the truss beam.

[0033] The carbonation depth information of the concrete roof panel includes a measured value of the carbonation depth of the concrete and an average value of the carbonation depth of the concrete.

[0034] The second aspect of the embodiments of the present application provides a concrete roof panel safety rating device, which comprises:

[0035] An initial data acquisition module is configured to acquire at least one influence factor of at least one component. The influence factor includes reinforced concrete damage information, structural information of a concrete roof panel, size information of the concrete roof panel, material strength of the concrete roof panel, displacement deformation information of the concrete roof panel, and carbonation depth information of the concrete roof panel.

[0036] A preprocessing module is configured to preprocess the at least one influence factor of the component to obtain preprocessed influence factors.

[0037] A decision module is configured to process the preprocessed influence factors by a decision tree algorithm in a trained rating model to obtain rating decision results.

[0038] A rating module is configured to vote at least one rating decision result by a random forest algorithm in the trained rating model to obtain a voting result, and obtain a safety rating result of the component according to the voting result.

[0039] The third aspect of the embodiments of the present application provides an electronic device, which comprises a memory and a processor, wherein

[0040] The memory is configured to store a program.

[0041] The processor is coupled to the memory and is configured to execute the program stored in the memory to implement the steps of the concrete roof slab safety rating method according to any of the foregoing schemes.

[0042] The computer readable storage medium according to the fourth aspect of the embodiments of the present application is configured to store computer readable programs or instructions, and the programs or instructions are configured to implement the steps of the concrete roof slab safety rating method according to any of the foregoing schemes when executed by a processor.

[0043] According to the technical scheme provided in the embodiments of the present application, the influence factors of multiple components of a concrete roof slab are comprehensively considered, and a decision tree algorithm and a random forest algorithm are combined based on the influence factors of the components to perform safety assessment on the reliability of the concrete roof slab, so that structural defects or deficiencies of each component of the concrete roof slab can be found in time, accidents such as collapse caused by structural hazards can be avoided, targeted maintenance measures can be taken in time, and the overall service life of the building can be prolonged. Meanwhile, the data processing flow of the structural diagnosis and treatment of the concrete roof slab is more efficient, and the dependence on personnel experience is reduced. BRIEF DESCRIPTION OF DRAWINGS

[0044] In order to more clearly illustrate the technical scheme of the embodiments of the present application, the drawings needed in the description of the embodiments of the present application will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0045] Figure 1 is a step flowchart of a concrete roof slab safety rating method provided in the embodiments of the present application;

[0046] Figure 2 is a step flowchart of preprocessing of component information and rating decision processing provided in the embodiments of the present application;

[0047] Figure 3 is a step flowchart of a rating model training process provided in the embodiments of the present application;

[0048] Figure 4 is a decision tree structure schematic diagram provided in the embodiments of the present application;

[0049] Figure 5 is a step flowchart of data cleaning provided in the embodiments of the present application;

[0050] Figure 6 is a step flowchart of determining a rating label provided in the embodiments of the present application;

[0051] Figure 7 is a structural block diagram of a concrete roof panel safety rating device provided in an embodiment of the present application;

[0052] Figure 8 is a hardware structural block diagram of an electronic device provided in each embodiment of the present application. DETAILED DESCRIPTION

[0053] In order to make the above-mentioned objectives, characteristics and advantages of the present application more apparent, clear and easy to understand, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0054] Referring to Figure 1 , a step flowchart of a concrete roof panel safety rating method is shown, which can be applied to industrial building structure diagnosis and treatment. The concrete roof panel includes a plurality of components, such as Figure 1 , which can specifically include the following steps:

[0055] Step S101, acquiring at least one influence factor of any component; the influence factor includes: reinforced concrete damage information, structure information of the concrete roof panel, size information of the concrete roof panel, material strength of the concrete roof panel, displacement deformation information of the concrete roof panel, and carbonation depth information of the concrete roof panel;

[0056] In this embodiment, the reinforced concrete damage information, the structure information of the concrete roof panel, the size information of the concrete roof panel, the material strength of the concrete roof panel, the displacement deformation information of the concrete roof panel, and the carbonation depth information of the concrete roof panel are acquired through field investigation and / or detection of the concrete roof panel.

[0057] Step S102, preprocessing at least one influence factor of the component to obtain a preprocessed influence factor;

[0058] In this embodiment, preprocessing at least one influence factor of the component can include:

[0059] The feature data in the influence factor of the component is data cleaned, repeated records in the data set are checked and deleted, errors in the data such as input errors and measurement errors are identified and corrected to ensure the accuracy of the data; the format and dimension of the same type of feature data are unified.

[0060] The cleaned feature data is converted by data encoding for subsequent rating model processing.

[0061] Step S103, processing the pre-processed impact factors through the decision tree algorithm in the trained rating model to obtain at least one rating decision result of the component;

[0062] In the embodiment, through the decision tree algorithm in the trained rating model, the first-level branch node, the second-level branch node and the third-level branch node of the concrete roof panel component can be confirmed;

[0063] The first-level branch node is the impact factor;

[0064] The second-level branch node is the steel reinforced concrete damage information, the structure information of the concrete roof panel, the size information of the concrete roof panel, the material strength of the concrete roof panel, the displacement deformation information of the concrete roof panel and the carbonation depth information of the concrete roof panel;

[0065] The third-level branch node is the specific numerical value or description content of the steel reinforced concrete damage information, the structure information of the concrete roof panel, the size information of the concrete roof panel, the material strength of the concrete roof panel, the displacement deformation information of the concrete roof panel and the carbonation depth information of the concrete roof panel;

[0066] According to the first-level branch node, the second-level branch node and the third-level branch node of the decision tree of the concrete roof panel component and the decision tree algorithm, the input impact factor data is processed to obtain the rating decision result of at least one impact factor.

[0067] In an embodiment, the specific numerical value or description content of the steel reinforced concrete damage information, the structure information of the concrete roof panel, the size information of the concrete roof panel, the material strength of the concrete roof panel, the displacement deformation information of the concrete roof panel and the carbonation depth information of the concrete roof panel is not only one, and one rating decision result is generated according to different specific numerical value or description content, so there is one or more impact factors.

[0068] Step S104, voting processing at least one rating decision result through the random forest algorithm in the trained rating model to obtain a voting result, and obtaining a safety rating result of the component according to the voting result;

[0069] In the embodiment, at least one rating decision result is obtained based on the decision tree, and the random forest in the trained rating model is used to vote process a plurality of rating decision results to obtain a safety rating result of the component, and the overfitting defect of the decision tree can be solved by combining the random forest algorithm.

[0070] Step S105, obtaining the safety rating result of the concrete roof panel based on the safety rating results of the plurality of components of the concrete roof panel.

[0071] In this embodiment, a plurality of influence factors of any component of the concrete roof panel are obtained, the plurality of influence factors are preprocessed to facilitate subsequent rating model processing, the preprocessed influence factors are processed by the decision tree algorithm in the trained rating model, at least one rating decision result of the component is obtained, the at least one rating decision result is voted by the random forest algorithm in the trained rating model, a voting result is obtained, and the safety rating result of the component is obtained according to the voting result; and the safety rating result of the concrete roof panel is obtained in combination with the safety rating results of the plurality of components of the concrete roof panel.

[0072] By comprehensively considering the influence factors of the plurality of components of the concrete roof panel, the safety of the concrete roof panel is evaluated based on the influence factors of the components combined with the decision tree algorithm and the random forest algorithm, so that structural defects or deficiencies of each component of the concrete roof panel can be found in time, accidents such as collapse caused by structural hazards can be avoided, targeted maintenance measures can be taken in time, and the overall service life of the building can be prolonged. At the same time, the data processing flow of the structure diagnosis and treatment of the concrete roof panel is more efficient, and the dependence on personnel experience is reduced.

[0073] In one embodiment, referring to Figure 2 , a step flow chart of preprocessing and rating decision processing of component information is shown, as Figure 2 , which can specifically include the following steps:

[0074] Step S201, encoding feature data of at least one influence factor of the concrete roof panel to obtain encoded feature data, assigning and labeling rating labels to the encoded feature data to obtain the encoded feature data and the rating labels corresponding to the encoded feature data;

[0075] Step S202, determining the encoded feature data and the rating labels corresponding to the encoded feature data by the decision tree algorithm in the trained rating model to obtain a rating decision result.

[0076] The technical scheme of the embodiment of the present application obtains clear data structure and clear data relationship through preprocessing, and the decision tree and the random forest model can better learn the patterns and rules in the data when the data is clear, thereby improving the accuracy of prediction.

[0077] In one embodiment, referring to Figure 3 , a step flow chart of the rating model training process is shown, asFigure 3 As shown, the method can specifically include the following steps:

[0078] Step S301, obtaining historical original data of the concrete roof panel; wherein the historical original data includes at least one influence factor of each component of the concrete roof panel;

[0079] Step S302, performing data cleaning on the historical original data to obtain target feature data; performing data conversion and / or grouping of similar data on the cleaned feature data, defining corresponding codes for different data categories, defining units of the detection data, converting different data into a uniform unit, and arranging into a data table, as shown in Table 1.

[0080] Table 1

[0081]

[0082] Step S303, constructing a data set of the rating model according to the target feature data of each component;

[0083] Step S304, dividing the data set into a training set and a test set;

[0084] Step S305, training an initial rating model through the training set to obtain a trained rating model;

[0085] In the present embodiment, the information increment corresponding to the branch point after classification of the decision tree in the initial rating model is:

[0086] ,

[0087] wherein D is the current data set; a is the feature currently considered; and V is the number of different values of the feature a. is the sub-data set of the feature a with the value v.

[0088] Ent(D) is the information entropy of the data set D, and the calculation formula is:

[0089] ,

[0090] wherein, is the proportion of the kth class of samples in the data set D.

[0091] In an embodiment, the information entropy value of the root node is calculated according to the training data, for example: there are 8 A-class safety ratings and 9 B-class safety ratings. Then the information entropy of the root node is Ent(D) =-1*(8 / 17)*log(8 / 17)-(9 / 17)log(9 / 17)=0.998.

[0092] Assume that the first level branch node is the size information of the concrete panel, the size information has three categories: a class, b class, and c class, a class has 6, b class has 5, and c class has 6,

[0093] Among them, three of the "a class" are A class safety ratings, and three are B class safety ratings. Then Ent(a class) = - (3 / 6log3 / 6+3 / 6log3 / 6) = 1;

[0094] Among the "b class", one is an A class safety rating, and four are B class safety ratings. Then Ent(b class) = - (1 / 5log1 / 5+4 / 5log4 / 5) = 0.722;

[0095] Among the "c class", four are A class safety ratings, and two are B class safety ratings. Then Ent(c class) = - (4 / 6log4 / 6+2 / 6log2 / 6) = 0.918;

[0096] Considering that the probabilities of the three size information are 6 / 17, 5 / 17, and 6 / 17, if the first level branch node selects "size information", the entropy value after the size information classification is: (6 / 17*1+5 / 17*0.722+6 / 17*0.918) = 0.889.

[0097] Then if the size information is selected as the first level branch, the corresponding information increment is Gain(size information): 0.998-0.889=0.109.

[0098] In the same way, the information increment of other influence factors can be calculated. The influence factor with the largest information increment is selected as the first level branch node, and the second level branch node and the third level branch node are determined in turn.

[0099] In the same way, the trained decision tree can be obtained by iterative training. The structure of the decision tree taking the size information as an example is shown in Figure 4 .

[0100] The information gain of different influence factors is calculated, the influence factor with the largest information gain is selected as the first level branch node, and the second level branch node and the third level branch node are determined in turn. After the branch nodes are determined, the trained decision tree is obtained by cyclic training of the training data.

[0101] The decision result output by the decision tree is input into the random forest model for cyclic training to obtain the trained random forest. The trained decision tree and the random forest are determined as the trained rating model.

[0102] Step S306, verify the model of the trained rating model by the test set, and obtain the trained rating model.

[0103] The technical scheme of the embodiment of the application improves the consistency and accuracy of data by preprocessing the original data through classification and coding. The preprocessing of cleaning and sorting data can make the structure more clear, facilitating understanding and interpretation. Through the processed data, the operation accuracy of the rating model can be improved.

[0104] The rating model is trained by the training set and the test set to determine the optimal algorithm and parameter combination, and finally determine the optimal model parameters of the rating model.

[0105] In one embodiment, referring to Figure 5 , a step flow chart of data cleaning is shown, as Figure 5 , which can specifically include the following steps:

[0106] Step S501, classify the feature data in the influence factor of each component to obtain classified feature data;

[0107] Step S502, encode the classified feature data to obtain encoded feature data, assign and mark the rating label to the encoded feature data, and obtain target feature data.

[0108] In this embodiment, the feature data in the influence factor of each component is classified, wherein the influence factor is classified in the data category in Table 1, the influence factor is first classified to obtain basic information, structural action, defect or damage, and size, etc., and then the influence factor in the data category, that is, the feature data, is classified into the corresponding data category. The classified feature data is encoded, and the feature data is standardized and coded. Different feature data is converted into a unified unit, which is convenient for subsequent rating model processing.

[0109] In one embodiment, referring to Figure 6 , a step flow chart of determining the rating label is shown, as Figure 6 , which can specifically include the following steps:

[0110] Step S601, according to the numerical value of the feature data, match the corresponding rating standard value; wherein the rating standard value is determined according to the feature data of the components of different safety levels in the preset identification table; in this embodiment, the preset identification table can be the national standard "Industrial Building Reliability Identification Standard".

[0111] Step S602, obtaining the safety level corresponding to the rating standard value;

[0112] Step S603, encoding the safety level as the rating label corresponding to the feature data.

[0113] Through the determination of the rating label, the subsequent rating model can accurately distinguish the security level to which the feature data belongs.

[0114] In an embodiment, a method for missing value filling is provided, which can specifically include the following steps:

[0115] In the case where the influence factor has missing values, the missing values of the historical original data are filled by a random forest-based data imputation algorithm, to obtain optimized historical original data. In an embodiment, the data imputation algorithm can be a MissForest (missing data filling method based on random forest) algorithm.

[0116] The missing value filling can improve the integrity of the data. A complete data set can avoid analysis bias caused by missing data, so that the rating model can learn and predict using more information.

[0117] In an embodiment, the steel reinforced concrete damage information includes defect information or damage information of the steel reinforced concrete.

[0118] The structural information of the concrete roof panel includes load information of the concrete roof panel.

[0119] The size information of the concrete roof panel includes cross-sectional size information of a flange of the concrete roof panel and steel size information of the concrete roof panel.

[0120] The material strength of the concrete roof panel includes concrete compressive strength and concrete hardness.

[0121] The displacement deformation information of the concrete roof panel includes longitudinal deviation, lateral deviation, inclination value, measured height, inclination rate, span, relative support height difference in the middle of the span, deflection, deformation rate, and vertical deformation of the truss beam.

[0122] The carbonation depth information of the concrete roof panel includes a measured value of the carbonation depth of the concrete and an average value of the carbonation depth of the concrete.

[0123] By integrating the feature data of the influence factors, a method for evaluating the reliability of the concrete roof panel based on the feature data of the components, combined with the decision tree algorithm and the random forest algorithm, is proposed. This method can timely detect structural defects or deficiencies of each component of the concrete roof panel, avoid accidents such as collapse caused by structural hazards, and help to take targeted maintenance measures in time, thereby prolonging the overall service life of the building. At the same time, the data processing flow of the structure diagnosis and treatment of the concrete roof panel is more efficient, and the dependence on personnel experience is reduced.

[0124] In an embodiment, the data preprocessing process includes feature design, data arrangement, and missing value filling.

[0125] Characteristic design process, if a certain influencing factor needs to be monitored by multiple detections on different parts of the same component to obtain corresponding situation analysis, in order to reflect all data characteristics of multiple detections, this influencing factor will take the mean and variance of its multiple physical quantities as characteristic values to participate in model construction, such as the tensile strength of the component; if a certain influencing factor needs to be monitored by detecting two parts of the same component to form a comparison and then obtain corresponding situation analysis, regardless of the detection sequence, in order to reflect the comparison and actual measurement value, this influencing factor will take the minimum value and difference value of its physical quantity as characteristic values for modeling, such as the measured value of the length of the angle steel; if a certain influencing factor belongs to classification, and needs to be combined with all categories for joint analysis to obtain the corresponding influencing factor situation, in order to reflect the joint decision of multiple types, this influencing factor should be re-set as a new category for multiple type characteristics combination, and single category and multiple type combination category as characteristics for modeling, such as steel reinforced concrete defects and damage.

[0126] Data arrangement process, the data is standardized and coded, the specific processing method is to specify the corresponding code for different data categories, specify the unit of detection data, and convert different data into a unified unit. Then the different original detection reports are arranged into data tables.

[0127] Missing value filling process, some data is often missing in the original data, which is processed in two cases. According to the above influencing factor table, 1) if the data of “3 defects and damage” (as shown in Table 1) is not recorded, according to experience, it represents that the structure does not exist this kind of defects and damage. 2) The other 8 categories of influencing factors use the MissForest algorithm to fill in the missing values. This algorithm can automatically detect missing data and fill it in. The MissForest algorithm here is a packaged algorithm model, without self-improvement.

[0128] In an embodiment, the data is divided into training set and test set by support vector machine algorithm, and the decision tree algorithm and random forest algorithm are used to train the test set data to determine the optimal algorithm and parameter combination, and finally the random forest algorithm is determined to be optimal. The training set accounts for 70% of the total data, which is mainly used for machine learning model establishment, parameter selection and cross-validation; the test set accounts for 30% of the total data, which is mainly used for verifying the classification accuracy of the machine learning model. 10-fold cross-validation divides the training set into 10 parts, selects 9 parts for training, and the remaining 1 part as cross-validation test model accuracy. According to the performance of the test set prediction results, we choose the random forest algorithm.

[0129] In some embodiments, the pre-processed data is model trained using a random forest algorithm to ultimately determine the safety rating of the concrete roof panel.

[0130] In the process of using the random forest algorithm, automatic parameter tuning is performed during the training process. According to the rating requirements of the concrete roof panel, the important parameters in the random forest algorithm are given a parameter range. Then, within each parameter range, values are taken at certain intervals, and different parameter combinations are formed. Then, in the test set, the optimal parameter combination is selected according to the result accuracy through cross-validation. This approach allows the optimal parameter combination of the algorithm to be not fixed. With subsequent new data entering, the model can continuously optimize its parameter combination to make the current prediction result optimal.

[0131] In processing data, we use three different algorithms: decision tree method, random forest and MissForest to ensure the accuracy and accuracy of data processing. These algorithms have different focuses in data preprocessing, model selection, model training and model application.

[0132] First, in the data preprocessing stage, we take the physical quantities of the factors affecting the concrete roof panel, such as reinforced concrete defects and damage, displacement, material strength and size measurement, as the basis, and combine with actual business experience, and comprehensively analyze the safety and usability of the concrete roof panel by rating the parameters such as fatigue cracks, track eccentricity, appearance quality and strain to construct the safety and reliability features of the component. The more important factor in the influencing factors of the correct rating of the concrete roof panel safety rating algorithm is the size of the concrete roof panel. We organize and summarize the data according to the feature design scheme to form the input data table. For multiple measurement data, we perform mean, variance, minimum value, difference value preprocessing, and classify the category data. To ensure the smooth progress of model selection and training, we ensure the integrity of the data and perform data cleaning, unify the format and dimension of similar feature data, and avoid the case of fuzzy categories or data ranges as feature data.

[0133] In processing the missing values of reinforced concrete defects and damage in the data, we default to no defects and damage, and fill in the corresponding missing values by setting a new category. For other feature missing conditions, we use the MissForest algorithm because it can better fit the actual situation and help model analysis, so we choose the MissForest algorithm to fill in the missing values of other features.

[0134] In model selection, we use the pre-processed dataset for model training. We use random sampling and 10-fold cross-validation to divide the dataset into training and test sets, where the training set accounts for 70% of the total data, mainly for machine learning model establishment, parameter selection and cross-validation; the test set accounts for 30% of the total data, mainly for verifying the classification accuracy of the machine learning model. 10-fold cross-validation divides the training set into 10 parts, selects 9 parts for training, and the remaining 1 part as cross-validation test model accuracy. By establishing models for various parameter combinations and verifying them, we select the best parameter combination.

[0135] And the feature importance application can identify the contribution degree (importance) of the existing data features to the correct classification of the model during the training of the data classification prediction model, and a continuous variable with a threshold range of [0, 1] is used to quantify the feature importance. The closer to 1, the higher the importance of the feature to the correct classification, and vice versa. The closer to 0, the lower the importance of the feature to the correct classification, realizing data-oriented sorting of the correlation of the rating features of the three typical components.

[0136] In summary, the three algorithms of decision tree method, random forest and MissForest have their own focuses in data processing and model training. The decision tree method plays a role in feature selection and model interpretation due to its intuitiveness and interpretability; the random forest performs outstandingly in model training and prediction due to its strong classification ability and anti-overfitting ability; and MissForest provides an effective solution in the data preprocessing stage due to its advantage in handling missing values. By comprehensively using the three algorithms, we ensure the accuracy and accuracy of data processing and improve the prediction performance of the model.

[0137] In other embodiments, statistical parameters of load effect and resistance are determined by statistical analysis of measured data and previous fatigue test data of the concrete roof panel. These parameters include mean value, standard deviation, etc. of load effect and resistance, which are the basis for evaluating fatigue reliability. According to the calculation results, the recommended values of fatigue target reliability indicators and design expressions of partial coefficients are given. These indicators and coefficients are key parameters for designing and evaluating the fatigue reliability of concrete roof panels. The safety and usability obtained from fatigue and deformation are used to assess the reliability.

[0138] Referring to Figure 7 As shown in FIG. 7, a structural block diagram of a concrete roof panel safety rating device 700 according to an embodiment of the present application is shown, and as shown in Figure 7 As shown in FIG. 7, the device can specifically include the following modules:

[0139] The initial data acquisition module 701 is configured to acquire at least one influence factor of at least one component; the influence factor includes: reinforced concrete damage information, structural information of a concrete roof panel, size information of the concrete roof panel, material strength of the concrete roof panel, displacement deformation information of the concrete roof panel, and carbonation depth information of the concrete roof panel;

[0140] The preprocessing module 702 is configured to preprocess the at least one influence factor of the component to obtain a preprocessed influence factor;

[0141] The decision module 703 is configured to process the preprocessed influence factor by using a decision tree algorithm in the trained rating model to obtain a rating decision result;

[0142] The rating module 704 is configured to perform voting processing on at least one rating decision result by using a random forest algorithm in the trained rating model to obtain a voting result, and obtain a safety rating result of the component according to the voting result.

[0143] In an embodiment, the preprocessing module 702 is specifically configured to encode feature data of the at least one influence factor of the concrete roof panel to obtain encoded feature data, and assign values to the encoded feature data and mark a rating label to obtain the encoded feature data and a rating label corresponding to the encoded feature data;

[0144] The decision module 703 is specifically configured to perform decision processing on the encoded feature data and the rating label corresponding to the encoded feature data by using a decision tree algorithm in the trained rating model to obtain a rating decision result.

[0145] In an embodiment, the training module is further configured to acquire historical original data of the concrete roof panel; wherein the historical original data includes at least one influence factor of each component of the concrete roof panel;

[0146] The historical original data is subjected to data cleaning to obtain target feature data;

[0147] A data set of the rating model is constructed according to the target feature data of each component;

[0148] The data set is divided into a training set and a test set;

[0149] The initial rating model is trained by using the training set to obtain a trained rating model;

[0150] The trained rating model is verified by using the test set to obtain a trained rating model.

[0151] In an embodiment, the training module is specifically configured to classify the feature data in the influence factor of each component, and obtain classified feature data;

[0152] The classified feature data is encoded to obtain encoded feature data, and the encoded feature data is assigned and labeled with a rating label to obtain target feature data.

[0153] In an embodiment, it further includes a rating label acquisition module configured to match a corresponding rating standard value according to the numerical value of the feature data; wherein the rating standard value is determined according to the feature data of components of different safety levels in a preset identification table;

[0154] The safety level corresponding to the rating standard value is obtained;

[0155] The safety level is encoded as the rating label corresponding to the feature data.

[0156] In an embodiment, it further includes a data optimization module configured to fill the missing values of the historical original data by using a random forest-based data interpolation algorithm when the influence factor has missing values, and obtain optimized historical original data.

[0157] In an embodiment, the reinforced concrete damage information includes defect information or damage information of the reinforced concrete;

[0158] The structural information of the concrete roof panel includes load information of the concrete roof panel;

[0159] The size information of the concrete roof panel includes cross-sectional size information of a flange of the concrete roof panel and steel bar size information of the concrete roof panel;

[0160] The material strength of the concrete roof panel includes concrete compressive strength and concrete hardness;

[0161] The displacement deformation information of the concrete roof panel includes longitudinal deviation, lateral deviation, inclination value, measured height, inclination rate, span, relative support height difference in the middle of the span, deflection, deformation rate, and vertical deformation of the truss beam;

[0162] The carbonation depth information of the concrete roof panel includes a measured value of the carbonation depth of the concrete and an average value of the carbonation depth of the concrete.

[0163] The above embodiment provides a concrete roof panel safety rating device, which can realize the technical solutions described in the above concrete roof panel safety rating method embodiment. The principles of the specific implementation of the above modules or units can be referred to the corresponding content in the above concrete roof panel safety rating method embodiment, which will not be described here.

[0164] AsFigure 8 The present application also provides an electronic device 800, as shown. The electronic device 800 comprises a processor 801, a memory 802 and a display 803. Figure 8 Only some components of the electronic device 800 are shown, but it should be understood that all the components shown are not required, and more or less components can be implemented instead.

[0165] The memory 802 can be an internal storage unit of the electronic device 800, such as a hard disk or a memory of the electronic device 800, in some embodiments. The memory 802 can also be an external storage device of the electronic device 800, such as a plug-in hard disk, a SmartMediaCard (SMC), a Secure Digital (SD) card, a FlashCard, etc. equipped on the electronic device 800, in other embodiments.

[0166] Further, the memory 802 can include both an internal storage unit and an external storage device of the electronic device 800. The memory 802 is used to store application software and various data installed on the electronic device 800.

[0167] The processor 801 can be a Central Processing Unit (CPU), a microprocessor or other data processing chip, used to run program codes or process data stored in the memory 802, such as a concrete roof panel safety rating method in the present application, in some embodiments.

[0168] The display 803 can be an LED display, a liquid crystal display, a touch liquid crystal display, an Organic Light-Emitting Diode (OLED) touch, etc. in some embodiments. The display 803 is used to display information of the electronic device 800 and to display visualized user interfaces. The components 801-803 of the electronic device 800 communicate with each other through a system bus.

[0169] In some embodiments of the present application, when the processor 801 executes the concrete roof panel safety rating program in the memory 802, the steps in any of the concrete roof panel safety rating method embodiments can be implemented:

[0170] It should be understood that, in addition to the above functions, the processor 801 can also implement other functions when executing the concrete roof panel safety rating program in the memory 802, which can be specifically understood by referring to the descriptions of the corresponding method embodiments above.

[0171] Further, the embodiments of the present application do not make specific limitation on the type of the electronic device 800 mentioned above, and the electronic device 800 can be a portable electronic device such as a mobile phone, a tablet computer, a personal digital assistant (PDA), a wearable device, a laptop computer, etc. Exemplary embodiments of the portable electronic device include, but are not limited to, a portable electronic device running an IOS, an android, a microsoft or other operating system. The portable electronic device mentioned above can also be other portable electronic devices such as a laptop computer having a touch-sensitive surface (e.g., a touch panel), etc. It should also be understood that in some other embodiments of the present application, the electronic device 800 can also not be a portable electronic device, but a desktop computer having a touch-sensitive surface (e.g., a touch panel).

[0172] In yet another aspect, the present application also provides a non-transitory computer readable storage medium having stored thereon a computer program, which, when executed by a processor, implements a method for safety rating of a concrete roof panel as described above, the method comprising:

[0173] obtaining at least one influence factor of any component; preprocessing the at least one influence factor of the component to obtain a preprocessed influence factor; processing the preprocessed influence factor through a decision tree algorithm in the trained rating model to obtain at least one rating decision result of the component; performing voting processing on the at least one rating decision result through a random forest algorithm in the trained rating model to obtain a voting result, and obtaining a safety rating result of the component according to the voting result; and obtaining a safety rating result of the concrete roof panel based on safety rating results of a plurality of components of the concrete roof panel.

[0174] It can be understood by those skilled in the art that all or part of the processes of the above-mentioned embodiments can be completed by a computer program instructing relevant hardware, and the program can be stored in a computer readable storage medium. The computer readable storage medium includes a magnetic disk, an optical disk, a read-only memory or a random access memory, etc.

[0175] The above describes in detail a method for safety rating of a concrete roof panel, a device, an apparatus and a medium provided by the present application. The principles and implementation manners of the present application are described by using specific examples. The above description of the embodiments is only used to help understand the method of the present application and its core idea. Meanwhile, for those skilled in the art, the specific implementation manners and application ranges can be changed according to the idea of the present application. In summary, the content of the present description should not be understood as a limitation of the present application.

Claims

1. A method for safety rating of concrete roof slabs, characterized in that, The concrete roof slab comprises several components, including: Obtain at least one influencing factor for any component; the influencing factor includes: reinforced concrete damage information, structural information of the concrete roof panel, dimensional information of the concrete roof panel, material strength of the concrete roof panel, displacement and deformation information of the concrete roof panel, and carbonation depth information of the concrete roof panel. The reinforced concrete damage information includes: defect information or damage information of the reinforced concrete. The structural information of the concrete roof panel includes: load information of the concrete roof panel. The dimensional information of the concrete roof panel includes: cross-sectional dimensions of the flange of the concrete roof panel and dimensions of the reinforcing steel bars in the concrete roof panel. The material strength of the concrete roof panel includes: concrete compressive strength and concrete hardness. The displacement and deformation information of the concrete roof panel includes: longitudinal deviation, lateral deviation, tilt value, measurement height, tilt rate, span, mid-span relative support height difference, deflection, deformation rate, and vertical deformation of the truss beam. The carbonation depth information of the concrete roof panel includes: measured value of concrete carbonation depth and average value of concrete carbonation depth. Preprocessing at least one influencing factor of the component to obtain the preprocessed influencing factor includes: encoding the feature data of at least one influencing factor of the concrete roof panel to obtain encoded feature data; assigning values ​​and marking rating labels to the encoded feature data to obtain the encoded feature data and the rating labels corresponding to the encoded feature data. The rating labels of the influencing factors are processed by the decision tree algorithm in the trained rating model to obtain at least one rating decision result for the component. The random forest algorithm in the trained rating model is used to vote on at least one rating decision result to obtain a voting result, and the security rating result of the component is obtained based on the voting result. Based on the safety rating results of multiple components of the concrete roof panel, the safety rating result of the concrete roof panel is obtained; The process of determining the rating labels includes: The corresponding rating standard value is matched according to the numerical value of the feature data; wherein, the rating standard value is determined according to the feature data of components of different safety levels in the preset appraisal table; Obtain the security level corresponding to the rating standard value; The security level is encoded as a rating label corresponding to the feature data.

2. The safety rating method for concrete roof panels according to claim 1, characterized in that, The process of preprocessing at least one influencing factor of the component to obtain preprocessed influencing factors, and then processing the preprocessed influencing factors using a decision tree algorithm in a trained rating model to obtain at least one rating decision result for the component, specifically includes: The feature data of at least one influencing factor of the concrete roof panel is encoded to obtain encoded feature data. The encoded feature data is then assigned a value and labeled with a rating tag to obtain the encoded feature data and the rating tag corresponding to the encoded feature data. The decision tree algorithm in the trained rating model is used to determine the coded feature data and the rating labels corresponding to the coded feature data, thereby obtaining the rating decision result.

3. A method for safety rating of concrete roof panels according to claim 1 or 2, characterized in that, Also includes: Obtain historical raw data of the concrete roof slab; wherein, the historical raw data includes: at least one influencing factor for each component of the concrete roof slab; The historical raw data is cleaned to obtain target feature data; A dataset for constructing a rating model is built based on the target feature data of each component; The dataset is divided into a training set and a test set; The initial rating model is trained using the training set to obtain the trained rating model; The trained rating model is validated using a test set to obtain the trained rating model.

4. The safety rating method for concrete roof panels according to claim 3, characterized in that, The step of cleaning the historical raw data to obtain target feature data specifically includes: Classify the feature data in the influence factors of each component to obtain the classified feature data; The classified feature data is encoded to obtain encoded feature data. The encoded feature data is then assigned values ​​and labeled with rating tags to obtain target feature data.

5. The safety rating method for concrete roof panels according to claim 4, characterized in that, Also includes: In the case of missing values ​​in the influencing factors, the missing values ​​in the historical original data are filled by a data imputation algorithm based on random forest to obtain optimized historical original data.

6. A safety rating device for concrete roof panels, characterized in that, include: The initial data acquisition module is used to acquire at least one influencing factor for any component; The influencing factors include: reinforced concrete damage information, concrete roof slab structural information, concrete roof slab dimensional information, concrete roof slab material strength, concrete roof slab displacement and deformation information, and concrete roof slab carbonation depth information. The reinforced concrete damage information includes: defect or damage information of the reinforced concrete. The concrete roof slab structural information includes: load information of the concrete roof slab. The concrete roof slab dimensional information includes: cross-sectional dimensions of the concrete roof slab flanges and dimensions of the reinforcing steel bars in the concrete roof slab. The concrete roof slab material strength includes: concrete compressive strength and concrete hardness. The concrete roof slab displacement and deformation information includes: longitudinal deviation, lateral deviation, tilt value, measurement height, tilt rate, span, mid-span relative support height difference, deflection, deformation rate, and vertical deformation of the truss beam. The concrete roof slab carbonation depth information includes: measured values ​​of concrete carbonation depth and average values ​​of concrete carbonation depth. The preprocessing module is used to preprocess at least one influencing factor of the component to obtain the preprocessed influencing factor, including: encoding the feature data of at least one influencing factor of the concrete roof panel to obtain encoded feature data, assigning values ​​and marking rating labels on the encoded feature data to obtain the encoded feature data and the rating labels corresponding to the encoded feature data. The decision module is used to process the rating labels of the influence factors through the decision tree algorithm in the trained rating model to obtain at least one rating decision result of the component. The rating module is used to process at least one rating decision result through a voting process using the random forest algorithm in the trained rating model to obtain a voting result, and to obtain a safety rating result for the component based on the voting result; and to obtain a safety rating result for the concrete roof panel based on the safety rating results of multiple components of the concrete roof panel. The process of determining the rating labels includes: The corresponding rating standard value is matched according to the numerical value of the feature data; wherein, the rating standard value is determined according to the feature data of components of different safety levels in the preset appraisal table; Obtain the security level corresponding to the rating standard value; The security level is encoded as a rating label corresponding to the feature data.

7. An electronic device, characterized in that, Including memory and processor, among which, The memory is used to store programs; The processor, coupled to the memory, is used to execute the program stored in the memory to implement the steps in the concrete roof panel safety rating method according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, Used to store computer-readable programs or instructions, which, when executed by a processor, can perform the steps in the safety rating method for concrete roof panels as described in any one of claims 1 to 5.

Images