Construction method and system of civil air defense project detection information management model and readable storage medium

Abstract

The invention provides a construction method and system of a civil air defense project detection information management model and a readable storage medium, the construction method of the civil air defense project detection information management model comprises the following steps: S1, obtaining detection data of a civil air defense project, the detection data at least comprising structure parameters and environmental parameters; s2, generating a detection information model according to the detection data, wherein the detection information model is used for representing the state of the civil air defense project; and S3, processing the detection information model through a pre-established management platform to obtain a comprehensive management result which is used for representing the operation state of the civil air defense project. According to the invention, intelligent and dynamic management of civil air defense project detection information can be realized, and the management efficiency and the decision level are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of civil air defense information processing, and in particular to a civil air defense engineering detection information management model construction method and system and a readable storage medium. BACKGROUND

[0002] Civil air defense engineering detection information management faces complex challenges such as multi-dimensional data processing, dynamic environment adaptation, and decision optimization. Traditional static management mode is difficult to effectively cope with the diversity and variability of detection data, resulting in low information utilization efficiency and inability to timely identify key risks. How to quickly extract valuable feature information from massive detection data and dynamically adjust management strategies according to actual conditions has become a core problem that needs to be solved. Especially in the case of detection data interweaving in multiple dimensions such as structural state, environmental parameters, and equipment operation, how to accurately assess the importance of each indicator and make reasonable resource allocation decisions is crucial to improving the safety and reliability of civil air defense engineering. At the same time, the complexity and uncertainty of the detection environment also pose challenges to the adaptability of the management model. How to build an intelligent management system that can self-learn and optimize to adapt to changing detection scenarios and newly emerging risk factors is the key to achieving efficient management of civil air defense engineering detection information. The solution to this series of problems not only involves data processing and model construction, but also needs to consider various constraints in actual engineering applications. How to ensure the reliability and explainability of the decision while ensuring management efficiency is a major technical problem faced by civil air defense engineering detection information management. SUMMARY

[0003] Therefore, the present application provides a civil air defense engineering detection information management model construction method, system and readable storage medium, which can realize intelligent and dynamic management of civil air defense engineering detection information and improve management efficiency and decision-making level.

[0004] In one aspect, the present application provides a civil air defense engineering detection information management model construction method, which comprises the following steps: S1: obtaining detection data of civil air defense engineering, wherein the detection data at least includes structural parameters and environmental parameters; S2: generating a detection information model according to the detection data, wherein the detection information model is used to represent the state of civil air defense engineering; S3: processing the detection information model through a pre-established management platform to obtain a comprehensive management result, wherein the comprehensive management result is used to represent the running state of civil air defense engineering.

[0005] As described above, the aspect and any possible implementation manner further provides an implementation manner, wherein the detection data of civil air defense engineering obtained in S1 specifically comprises: S11: Collecting original data of the civil air defense engineering by a sensor device, the original data including structural stress values, displacement values and environmental temperature and humidity values; S12: Preprocessing the original data to obtain standardized detection data, the preprocessing including data cleaning and format conversion; S13: Determining the structural parameters and the environmental parameters according to the standardized detection data.

[0006] According to any possible implementation of the aspect as described above, an implementation is further provided, and the S2 of generating a detection information model according to the detection data specifically includes: S21: Determining mechanical characteristics of the civil air defense engineering according to the structural parameters, the mechanical characteristics including stress distribution and deformation trend; S22: Determining environmental adaptability of the civil air defense engineering according to the environmental parameters, the environmental adaptability including durability and stability; S23: Fusing the mechanical characteristics and the environmental adaptability by a preset modeling algorithm to generate the detection information model, the modeling algorithm including finite element analysis and data fitting.

[0007] According to any possible implementation of the aspect as described above, an implementation is further provided, and the S3 of processing the detection information model by a pre-established management platform specifically includes: S31: Classifying the detection information model by the management platform to obtain multiple state categories, the state categories including normal state and abnormal state; S32: Extracting key features in the detection information model for each state category, the key features including stress abnormal points and environmental over-limit points; S33: Generating the comprehensive management result according to the key features, the comprehensive management result including state evaluation result and early warning information.

[0008] According to any possible implementation of the aspect as described above, an implementation is further provided, and the S12 of preprocessing the original data specifically includes: S121: Screening the original data by a preset threshold to obtain effective data, the threshold including stress range and temperature and humidity range; S122: Performing denoising processing on the effective data to obtain smooth data, the denoising processing including signal processing based on wavelet transform; S123: Performing format conversion on the smooth data by data normalization to generate the standardized detection data, the data normalization including linear mapping and standardization processing.

[0009] According to the aspect and any possible implementation manner as described above, further provided is an implementation manner, wherein the step S23 of fusing the mechanical characteristic and the environmental adaptability by using a preset modeling algorithm specifically comprises: S231: performing meshing on the mechanical characteristic by using a finite element analysis algorithm to obtain a mechanical characteristic mesh; S232: performing curve fitting on the environmental adaptability by using a data fitting algorithm to obtain an environmental adaptability curve; S233: mapping the mechanical characteristic mesh and the environmental adaptability curve according to a preset fusion rule to generate a multi-dimensional detection information model, wherein the fusion rule comprises weight distribution and feature superposition.

[0010] According to the aspect and any possible implementation manner as described above, further provided is an implementation manner, wherein the step S33 of generating the comprehensive management result according to the key feature specifically comprises: S331: scoring the key feature by using a preset evaluation model to obtain a state score, wherein the evaluation model comprises a weighted average model; S332: determining a priority of the state category according to the state score and a preset rating rule, wherein the priority represents an importance degree of the state category; S333: if the priority is higher than a preset priority threshold, generating the early warning information, wherein the early warning information comprises an abnormal position and a processing suggestion; S334: generating the state evaluation result according to the state score and the early warning information.

[0011] According to the aspect and any possible implementation manner as described above, further provided is an implementation manner, wherein the structure parameter in the step S13 comprises a stiffness coefficient and a bearing capacity, and the environmental parameter comprises temperature and humidity.

[0012] According to the aspect and any possible implementation manner as described above, further provided is a construction system of a civil air defense engineering detection information comprehensive management, which is used for a construction method of a civil air defense engineering detection information management model, and the construction system of the civil air defense engineering detection information comprehensive management comprises: a data acquisition module, configured to acquire detection data of a civil air defense engineering, wherein the detection data at least comprises structure parameters and environmental parameters; a model generation module, configured to generate a detection information model according to the detection data, wherein the detection information model is used to represent a state of the civil air defense engineering; a processing representation module, configured to process the detection information model by using a pre-established management platform to obtain a comprehensive management result, wherein the comprehensive management result is used to represent an operation state of the civil air defense engineering.

[0013] According to the aspect and any possible implementation manner described above, further provided is a readable storage medium, wherein a program is stored on the readable storage medium, and the program performs the construction method of the civil air defense engineering detection information management model to obtain detection data of a civil air defense engineering, the detection data at least including structure parameters and environment parameters; generates a detection information model according to the detection data, the detection information model representing a state of the civil air defense engineering; processes the detection information model through a pre-established management platform to obtain a comprehensive management result, and the comprehensive management result is used to represent an operation state of the civil air defense engineering.

[0014] The technical scheme provided by the embodiment of the application can have the following beneficial effects. The application discloses a civil air defense engineering detection information management method, which acquires multi-dimensional detection data, extracts key features and dynamically allocates weights, and constructs a self-adaptive comprehensive management model. BRIEF DESCRIPTION OF DRAWINGS

[0015] Figure 1 The application further discloses a civil air defense engineering detection information management model construction method and system and a readable storage medium. DETAILED DESCRIPTION

[0016] In order to further understand the content of the application, the application is described in detail in combination with the drawings and embodiments.

[0017] As shown in the drawings, the application discloses a civil air defense engineering detection information management model construction method, which can specifically include the following steps. Figure 1 S101, acquiring a civil air defense engineering detection information set, the set including a plurality of detection data, each detection data containing a time stamp and detection parameters.

[0018] ​The original data record is obtained from the detection equipment of the civil air defense project, and the record contains detection values and related parameters corresponding to multiple time points to form an initial data set. The initial data set is arranged in time sequence, each detection value is matched with the corresponding time stamp, and a structured data group containing time stamp and detection parameter is generated. The structured data group is checked for integrity, and if the time stamp is missing or the detection parameter is abnormal, supplementary data is obtained from the pre-established backup record to ensure the continuity and accuracy of the data set. The structured data group after verification is integrated into a civil air defense engineering detection information set, which contains multiple detection data, each detection data contains time stamp and detection parameter, which is used for subsequent comprehensive management of civil air defense engineering information.

[0019] For example, in the actual civil air defense engineering detection scene, obtaining the original data record from the detection equipment is a key link. Suppose in a certain underground protective space, multiple sensors for monitoring air quality and structural safety are installed, which will record data such as environmental humidity and wall stress at regular intervals. Each sensor will generate corresponding detection values and related parameters at different time points, such as humidity and stress values at a certain time point. These data are initially stored in a scattered form in the equipment. By summarizing these scattered data, an initial data set containing multiple time point records is formed, laying the foundation for subsequent processing. The advantage of this is that it can fully grasp the output of the detection equipment, ensuring the comprehensiveness and integrity of the data source.

[0020] Specifically, it is particularly important to arrange the initial data set in time sequence. Suppose the initial data set contains humidity values and stress values at multiple time points, but these data may be disordered, or even part of the data at some time points may be mixed with other parameters. By comparing each detection value with the corresponding time stamp one by one, it is arranged in time sequence and ensures that each value corresponds to a time stamp, and finally a structured data group is generated. The advantage of this arrangement is that it can clearly show the trend of data change over time, facilitating subsequent analysis and processing.

[0021] For example, when checking the integrity of the structured data group, it may be found that the humidity value at some time point is missing, or the stress value at a certain time period is abnormally high. In this case, supplementary data at the corresponding time point can be extracted from the pre-stored backup record, such as missing humidity value from another backup sensor, or average value of adjacent time points to fill in abnormal data. The effect of this is to ensure the continuity of the data set and avoid affecting the accuracy of the overall analysis due to missing or abnormal data.

[0022] Specifically, integrating the checked structured data set into the final civil air defense engineering detection information set is the final link of the whole process. Assuming that through the previous processing, a complete data set containing multiple time points, humidity values and stress values at each time point, and other parameters has been obtained, these data are integrated into a unified set. Such a set not only contains timestamps and detection parameters, but also provides reliable data support for subsequent comprehensive management of civil air defense engineering. The benefit is to provide a basis for real-time monitoring of engineering status and historical data tracing, ensuring that managers can make decisions based on complete data.

[0023] S102, extracting multi-dimensional features from the civil air defense engineering detection information set, the features including structure state, environmental parameters and equipment operation data, to obtain a feature data set.

[0024] From the civil air defense engineering detection information set, obtain the original data, which contains the monitoring values of the structure state, the recorded values of the environmental parameters, and the real-time data of the equipment operation. Through the pre-established classification rules, these data are preliminarily grouped according to different dimensions to obtain the structure state data set, the environmental parameter data set and the equipment operation data set. For the structure state data set, the environmental parameter data set and the equipment operation data set, the key feature values of the data in each group are extracted. Through quantitative processing of the change range of the monitoring values in the structure state data set, the structure state feature set is obtained. The fluctuation of the recorded values in the environmental parameter data set is normalized to obtain the environmental parameter feature set. The stability of the real-time data in the equipment operation data set is evaluated to obtain the equipment operation feature set. The structure state feature set, the environmental parameter feature set and the equipment operation feature set are integrated to construct a multi-dimensional feature data set. Through the pre-set threshold, the data in each feature set is screened to retain the data items that meet the conditions and eliminate outliers to obtain a preliminarily optimized feature data set. For the preliminarily optimized feature data set, data standardization processing is performed to ensure that the feature values of different dimensions are within the same dimension range to form the final feature data set for subsequent comprehensive management of civil air defense engineering detection information.

[0025] For example, in actual civil air defense engineering detection information processing, the step of obtaining original data from the overall data set is crucial. Assuming that in a certain underground protective facility, the monitoring equipment records the wall stress change as the monitoring value of the structure state, and at the same time, the air humidity is collected as the recorded value of the environmental parameter, and the running power of the ventilation equipment is collected as the real-time data. Through the pre-established classification rules, the wall stress change can be classified into the structure state data set, the air humidity can be classified into the environmental parameter data set, and the ventilation equipment power can be classified into the equipment operation data set. The benefit of this is that complex data is classified and categorized, which facilitates subsequent targeted processing and improves the efficiency of data processing.

[0026] For example, in the stage of extracting key feature values, for the wall stress changes in the structure state data set, abnormal stress points can be identified by quantifying the daily fluctuation range, forming a structure state feature set. Similarly, for the air humidity data in the environmental parameter data set, the humidity values of different time periods can be adjusted to a unified scale through normalization processing, facilitating comparative analysis, and obtaining an environmental parameter feature set. For the ventilation equipment power in the equipment operation data set, the power abnormal fluctuation time period can be screened out by evaluating the operation stability, forming an equipment operation feature set. The benefit of this process is to extract valuable information from the original data, laying the foundation for subsequent integration.

[0027] For example, in the integration of multiple feature sets, the structure state feature set, the environmental parameter feature set, and the equipment operation feature set are combined into a multi-dimensional feature data set. Assuming that some stress data points are found to be outside the reasonable range after merging, these abnormal values can be removed through preset threshold screening to obtain a preliminary optimized feature data set. The effect of this is to ensure the reliability of the data and avoid abnormal values interfering with the subsequent analysis results.

[0028] For example, in the final standardization processing stage, for the preliminary optimized feature data set, different dimensional feature values such as stress values and humidity values can be adjusted to the same dimension range to form the final feature data set. The benefit of this link is to eliminate dimensional differences, making different types of data comparable in subsequent comprehensive management, providing a solid data foundation for the overall construction of civil air defense engineering detection information. Through the close connection of the above links, the processing process from the original data to the final feature data set is completely realized, ensuring the data quality and application value.

[0029] S103, adopting a preset feature screening rule to screen the feature data set to obtain a key feature subset.

[0030] The original feature data set is obtained from the civil air defense engineering detection information, and each feature in the data set is preliminarily arranged and divided into multiple feature categories for subsequent processing. For the divided feature categories, a preset feature screening rule is used for class-by-class filtering to extract features that meet the rule to form a preliminary screening feature set. The preliminary screening feature set is further sorted by feature importance, and a preset threshold is used to judge the relevance of each feature to obtain a key feature subset. The key feature subset is applied to the comprehensive management of civil air defense engineering detection information to construct the classification and evaluation basis of detection information, ensuring the effective integration of information.

[0031] Specifically, the content is as follows: In one possible implementation, the process of obtaining the original feature dataset from the civil air defense engineering detection information can be understood as a comprehensive collection and preliminary arrangement of various detection data.

[0032] For example, in the detection of civil air defense engineering, information such as structural strength, ventilation conditions, and protective layer thickness may be involved. These information initially exists in different formats and sources and needs to be unified and arranged into a dataset containing multiple attributes. Such arrangement helps subsequent classification and screening of data, ensuring that each feature can be accurately identified and classified. In this way, a foundation is laid for subsequent processing, avoiding information omission due to scattered data.

[0033] Specifically, when filtering each category according to the preset filtering rules, specific conditions can be set to extract features that meet the requirements.

[0034] For example, for structural strength-related features, a minimum strength standard can be set, and only features that meet or exceed this standard will be included in the preliminary screening feature set. This category-by-category filtering approach effectively reduces irrelevant feature interference, allowing subsequent processing to focus more on content closely related to civil air defense engineering detection. Such a filtering process not only improves the relevance of data processing but also provides a more representative data basis for further analysis.

[0035] In one possible implementation, when performing feature importance sorting on the preliminary screening feature set and using a preset threshold to determine relevance, the correlation between each feature and the detection target can be compared.

[0036] For example, when evaluating protective layer thickness features, if it is found that its correlation with protective effect is higher than that of other features, its priority will be raised, and it will be included in the key feature subset in combination with the preset threshold. This sorting and judging process ensures that the final feature subset is more representative and practical, providing a reliable basis for subsequent comprehensive management.

[0037] Specifically, when applying the key feature subset to the comprehensive management of civil air defense engineering detection information, it can be used as the core basis for classification and evaluation.

[0038] For example, in actual detection, the protective conditions of different areas can be classified according to the key feature subset, and appropriate maintenance or improvement measures can be developed. This application method makes the integration of detection information more systematic and provides data support for decision-making. Through such processing, not only the efficiency of information management is improved, but also the actual needs of civil air defense engineering are better met, ensuring the accuracy and practicality of detection results.

[0039] S104, weight calculation is performed on the features in the key feature subset by using an adaptive weight allocation algorithm to obtain a weighted feature set.

[0040] The key feature subset is obtained from the civil air defense engineering detection data, each feature in the key feature subset is preliminarily screened to determine its relevance to the detection information comprehensive management target, and a preliminarily screened feature set is obtained. For each feature in the preliminarily screened feature set, weight calculation is performed using a pre-established adaptive weight allocation mechanism, the importance of the feature is dynamically adjusted, and a weighted feature set is generated. Feature values directly related to civil air defense engineering detection information management are extracted from the weighted feature set, the feature values are sorted in combination with a pre-set threshold, and a feature priority list for comprehensive management is constructed. For high-priority features in the feature priority list, multi-dimensional data in the civil air defense engineering detection information is integrated to form a feature mapping relationship for comprehensive management, to support the construction of civil air defense engineering detection information comprehensive management.

[0041] Specifically, the process of obtaining the key feature subset from the civil air defense engineering detection data can be understood as extracting features related to core indicators such as protection performance and structural safety from a large amount of detection data.

[0042] In one possible implementation, the detection data can include data of multiple dimensions such as ventilation efficiency of underground space, sealing performance of protective door, and compressive strength of wall, etc. Through preliminary screening, redundant information unrelated to the comprehensive management target is removed, such as some non-critical environmental noise data, and then a preliminarily screened feature set is formed. The purpose of this is to focus on core indicators and ensure that the data processed subsequently is highly relevant to the management target, thereby improving the pertinence of data processing.

[0043] Next, the process of weight calculation for the preliminarily screened feature set.

[0044] It should be noted that the pre-established adaptive weight allocation mechanism is a method of dynamically adjusting the importance of features.

[0045] For example, in actual detection, the sealing performance of the protective door may have a greater impact on the overall protection effect in some scenarios, and therefore it is dynamically assigned a higher weight, while the ventilation efficiency may have a smaller impact in a specific environment, and the weight is correspondingly reduced, and finally a weighted feature set is generated. The role of this process is to highlight the impact of key features on comprehensive management and ensure that subsequent analysis is more in line with actual needs.

[0046] Further, the step of extracting feature values from the weighted feature set and sorting them can be understood as giving priority to features with higher weights, sorting in combination with a pre-set threshold, and forming a feature priority list.

[0047] For example, assuming that the sealing of the protective door has the highest weight, the eigenvalue of the feature will be extracted first and arranged in the front of the list, so as to be paid attention to first in management. The advantage of this link is to provide clear priority guidance for subsequent resource allocation and decision-making.

[0048] Finally, the step of integrating multi-dimensional data for the high-priority features in the feature priority list.

[0049] Specifically, the structural data, environmental data, etc. in the detection information can be associated with the high-priority features to form a feature mapping relationship.

[0050] For example, the sealing of the protective door is combined with the corresponding environmental humidity and pressure data to form a comprehensive mapping relationship diagram for guiding the detection information management of the civil air defense project. The advantage of this approach is that through the integration of multi-dimensional data, the management scheme is more comprehensive, which can support the comprehensive construction of the civil air defense project detection information from multiple angles, and improve the scientificity and practicality of management.

[0051] S105, if the weight of any feature in the weighted feature set exceeds the preset threshold, the feature is marked as a priority processing object, and a priority feature list is generated.

[0052] Obtain the civil air defense project detection information data, and extract the weighted feature set. According to the extracted weighted feature set, the weight value of each feature is calculated. If the weight value of any feature exceeds the preset threshold, the feature is marked as a priority processing object. Through the marked priority processing object, a priority feature list is generated. The priority feature list is used to extract a corresponding detection information subset. According to the detection information subset, a feature association matrix is constructed. If there is a strong association relationship in the feature association matrix, the priority feature list is updated. Through the updated priority feature list, a comprehensive management information model is generated. According to the comprehensive management information model, a detection information management result is output.

[0053] Specifically, the original data of structure deformation, leakage rate, equipment operation state, etc. are obtained from the civil air defense engineering detection database. The principal component analysis method is used to extract the weighted feature set, including crack width (0-10 mm), settlement (0-50 mm), ventilation efficiency (60%-100%), etc. The entropy weight method is used to calculate the weight value of each feature, and the preset threshold is set to 0.15. If the crack width weight reaches 0.18, it is marked as a priority processing object. The marked crack width, settlement, etc. are stored in the priority feature list, and the corresponding detection information subset such as crack location, development rate, etc. is extracted from the database through SQL query. The Pearson correlation coefficient is used to construct the feature correlation matrix. If the correlation coefficient between crack width and settlement exceeds 0.7, it is determined to be strongly correlated, and the settlement is added to the priority feature list. The updated list is used to train the random forest model, input the crack history data to predict the future expansion trend, and output the management results including risk level and maintenance suggestions.

[0054] S106, constructing a dynamic information management model according to the priority feature list to obtain a civil air defense engineering comprehensive management scheme.

[0055] Collect civil air defense engineering detection data and construct an initial data set. According to the initial data set, extract the priority feature list. Through the priority feature list, generate the feature weight matrix. If the feature weight matrix meets the preset threshold, perform data standardization processing to obtain the standardized feature set. According to the standardized feature set, construct a dynamic information management model. Use the dynamic information management model to perform data hierarchical classification and determine the hierarchical management scheme. Through the hierarchical management scheme, integrate the civil air defense engineering detection information to generate a comprehensive management data structure. According to the comprehensive management data structure, optimize the information flow path to obtain the civil air defense engineering comprehensive management scheme. Through the comprehensive management scheme, update the dynamic information management model to obtain the optimized management model.

[0056] Specifically, when collecting civil air defense engineering detection data, sensor network is used to collect parameters such as structural deformation, temperature and humidity, and gas concentration, with a sampling frequency of 1 Hz, and the initial data set is stored in the distributed database HBase. According to the initial data set, the random forest algorithm is used to calculate the feature importance, and the features with importance score ≥0.8 (such as crack width, CO2 concentration) are selected to generate a priority feature list. Through the priority feature list, the entropy weight method is used to calculate the weight of each feature to form a 5x5 weight matrix, and the matrix element range is 0.1~0.9. If the average value of the main diagonal of the weight matrix is >0.6, the Z-score algorithm is called to normalize the feature value, so that the data distribution conforms to N(0,1), and the standardized feature set is output. Based on the standardized feature set, an LSTM neural network model is built, with the number of hidden layer nodes set to 128 and the Dropout rate set to 0.2, and the model is built by training for 500 iterations. The model is used to cluster the detection data, and the DBSCAN algorithm (eps=0.5, min_samples=3) is used to divide the protection levels A / B / C into three categories, and the hierarchical management scheme is output. When integrating the hierarchical data, the graph database Neo4j is established to associate the relationship, the node attributes include the detection timestamp, device ID, and abnormal code, the edge weight is calculated according to the Euclidean distance, and the comprehensive management data structure is generated. According to the structure, the Dijkstra algorithm is used to calculate the shortest information path, and the path weight = transmission delay x data priority coefficient. Finally, the comprehensive management scheme is output. The scheme is imported into the model fine-tuning module, the Adam optimizer (lr=0.001) is used to update the LSTM parameters, and the model iteration is completed.

[0057] S107, real-time updating the civil air defense engineering comprehensive management scheme, obtaining new detection data and integrating it into the weighted feature set, and generating an updated management scheme.

[0058] Obtain new detection data from the civil air defense engineering detection equipment, preprocess the data to generate a first data set. The preprocessing includes format standardization and missing value filling of the detection data to generate the first data set. Extract the weighted features from the first data set to construct a feature matrix. Calculate the weighted values of each feature in the first data set to generate a feature matrix containing weighted features. Real-time update the feature matrix by integrating new detection data to generate a second feature matrix. After obtaining the new detection data, fuse the feature values of the new detection data with the existing features in the feature matrix to generate a second feature matrix. Based on the second feature matrix, update the civil air defense engineering comprehensive management scheme. Use the feature values in the second feature matrix to adjust the parameters of the management scheme to generate an updated management scheme.

[0059] For example, in the actual civil air defense engineering detection scenario, obtaining new data from detection equipment is a key link. Assuming that in a certain underground protective space, sensors collect real-time air quality, structural stress, and other information. These data may come from different types of equipment and have different formats. By standardizing the format of these data, such as unifying the timestamps output by different devices into a format, and filling in missing values, such as using the average or median value method to fill in the partially missing stress data, a unified first data set can be generated. The benefit of this is to ensure that the data for subsequent processing is consistent, laying the foundation for further analysis. Specifically, the process of extracting weighted features from the first data set and constructing a feature matrix can be understood as deep processing of data. Assuming that the first data set contains two types of indicators, air humidity and structural stress, according to historical experience or business requirements, a lower weight value can be assigned to humidity, while a higher weight value can be assigned to stress, because the latter has a more direct impact on safety. After constructing the feature matrix, each indicator of each detection point is quantified and integrated into a matrix, which facilitates subsequent updating and adjustment. The effect of this approach is to highlight the impact of key indicators and improve the relevance of data processing.

[0060] For example, in the real-time updating of the feature matrix, when new detection data enters, such as a sudden increase in stress value in a certain area, its feature value can be fused with the data in the existing matrix to generate a second feature matrix. This process ensures the timeliness of the data, so that the matrix always reflects the latest engineering state. Such an updating mechanism can keep the management scheme in line with the actual situation, avoiding decision-making deviations caused by data lag.

[0061] Specifically, updating the comprehensive management scheme based on the second feature matrix can be seen as a process of converting data into practical guidance. Assuming that the second feature matrix shows that the stress value in a certain area is consistently high, the parameters of the management scheme may be adjusted, such as increasing the frequency of inspections in that area or deploying more resources for maintenance. Such adjustments directly utilize the feature values in the matrix, ensuring that the management scheme is highly consistent with the current engineering state, thereby improving overall management efficiency and response speed.

[0062] S108, evaluating the updated management scheme through a preset verification rule to obtain a comprehensive management performance indicator.

[0063] The initial management scheme data is obtained from the civil air defense engineering detection information, the data is structured and arranged, a preliminary management scheme framework is formed, and the classification and priority division of the detection information are included. According to the preliminary management scheme framework, the data integrity and consistency are checked in combination with the preset verification rules, and the updated management scheme is generated to ensure that the classification and priority division meet the actual needs. According to the updated management scheme, multi-dimensional evaluation is carried out according to the preset verification rules, the comprehensive management performance index is calculated, and the matching degree and priority execution efficiency of each classified information in the evaluation process are recorded. According to the comprehensive management performance index, the key matching degree and execution efficiency data are extracted for subsequent adjustment of the updated management scheme to ensure that the comprehensive management of civil air defense engineering detection information meets the expected target.

[0064] For example, in processing civil air defense engineering detection information, first, initial management scheme data needs to be extracted from a large amount of raw data. These data may include the status of protective facilities in different areas, detection time, and abnormal records. Through structured arrangement, these information is classified according to facility type and emergency level to form a preliminary management scheme framework. Such classification and priority division helps to quickly identify which facilities need priority attention, thereby improving management efficiency.

[0065] Specifically, after forming the preliminary management scheme framework, the data integrity and consistency need to be checked in combination with the preset verification rules. For example, check if the detection record of a protective facility is missing, or if there are contradictions in data from different sources. Through this process, the updated management scheme is generated to ensure accurate classification and priority division. The role of this step is to avoid management errors caused by data errors and lay a reliable foundation for subsequent evaluation.

[0066] In one embodiment, for the updated management scheme, multi-dimensional evaluation can be carried out according to the preset verification rules, and the comprehensive management performance index is calculated. For example, the evaluation may include checking the matching degree of classified information, i.e. whether a facility is correctly classified into the corresponding type; at the same time, the priority execution efficiency is evaluated, such as whether the facilities with high emergency level can be processed in the shortest time. Recording these evaluation results helps to find potential deficiencies in the management scheme and provides a basis for further optimization.

[0067] For example, for the comprehensive management performance index, key data such as matching degree and execution efficiency can be extracted. These data are used to update the management scheme to ensure that each detection information is properly handled. For example, if the matching degree of a certain type of facility is low, the classification standard may need to be reexamined; and if the execution efficiency is low, the resource allocation method may need to be adjusted. Through such adjustment, the comprehensive management of civil air defense engineering detection information can better meet the actual needs and ensure the continuous improvement of management effect. Such a way not only improves the utilization rate of data, but also provides more accurate reference for management decision-making.

[0068] S109, if the comprehensive management performance index is lower than the preset standard, adjusting the parameters of the adaptive weight distribution algorithm to regenerate the weighted feature set.

[0069] Real-time data of each region is obtained from the civil air defense engineering detection information, and the data is preliminarily sorted to form an initial information set for subsequent feature weighting processing. For the initial information set, a pre-established weighting rule is used to assign weights to different dimensions of the region data to generate a first weighted feature set for further comprehensive management performance evaluation. On the basis of the first weighted feature set, it is judged whether its comprehensive management performance index reaches the preset threshold, if not, the parameters of the weight assignment rule are adjusted to regenerate the second weighted feature set. For the second weighted feature set, the calculation of the comprehensive management performance index is re-performed to ensure that it meets the needs of the comprehensive management of civil air defense engineering detection information, and the optimization construction of the feature set is completed.

[0070] Specifically, the generated content is as follows: In one possible implementation, the process of obtaining real-time data of each region from civil air defense engineering detection information and preliminarily sorting can be understood as unified formatting and classification processing of multi-source information.

[0071] For example, in multiple regions of civil air defense engineering, there may be different types of data such as ventilation conditions, structural stability, and personnel density. After these data are collected by sensors or manually recorded, they need to be converted into a unified digital format to form an initial information set. The purpose of this is to facilitate subsequent weighting processing, ensure data consistency and operability, and also improve data processing efficiency.

[0072] In one possible implementation, the initial information set is assigned weights using a pre-established weighting rule to generate a first weighted feature set, which is actually a distinction of the importance of the data.

[0073] For example, the ventilation condition data of some key areas in the civil air defense project may need to be given a higher weight because it is directly related to personnel safety; while for some secondary non-key data of areas, the weight can be appropriately reduced. The weighted feature set generated in this way can more accurately reflect the priority of each area in comprehensive management, providing a reliable basis for subsequent evaluation. The advantage of this approach is that it can highlight the management needs of key areas and avoid the blindness of resource allocation.

[0074] In a possible implementation, the process of determining whether the comprehensive management performance index meets the preset threshold based on the first weighted feature set can be regarded as a test of the current data processing effect.

[0075] For example, if it is found that the weighted feature values of some areas are low, resulting in the overall performance index not meeting the standard, the parameters of the weight assignment rule need to be adjusted to generate a second weighted feature set. This adjustment may be further improving the weight of some key data or reclassifying and assigning values to part of the data. In this way, the feature set can be dynamically optimized to ensure that it better meets the actual management needs and thus improves the overall performance.

[0076] In a possible implementation, the comprehensive management performance index is recalculated for the second weighted feature set, which is essentially a re-verification of the effect of the adjusted data.

[0077] For example, after adjusting the weight, the feature values of some areas may be significantly improved, so that the overall index meets the expectation. The benefit of this approach is that through repeated verification and optimization, it ensures that the final generated feature set can meet the needs of civil air defense engineering detection information comprehensive management and provide solid data support for subsequent decision-making. This continuous optimization process can effectively cope with management challenges in complex environments.

[0078] S1010, generating a final civil air defense engineering detection information management result according to the comprehensive management performance index.

[0079] The original detection data is acquired from the civil air defense engineering detection equipment, and denoising processing is performed on the original detection data to generate a first detection data set. The first detection data set contains time-sequenced detection parameter values. For the time-sequenced detection parameter values in the first detection data set, the change trend of each parameter in a preset time window is calculated to generate a parameter change matrix. The parameter change matrix records the fluctuation characteristics of each detection parameter. The fluctuation characteristics of each detection parameter are extracted from the parameter change matrix, and a weighted fluctuation score is calculated by combining the pre-established comprehensive management performance index weight to generate a second detection data set. The second detection data set reflects the comprehensive management priority of each detection parameter. For the comprehensive management priority in the second detection data set, a support vector machine algorithm is used for classification to generate a final civil air defense engineering detection information management result. The final result contains the priority ranking and management category of each detection parameter.

[0080] Specifically, the generation step is as follows: The original detection data is acquired from the civil air defense engineering detection equipment, and denoising processing is performed on the original detection data to generate a first detection data set. The first detection data set contains time-sequenced detection parameter values.

[0081] For the time-sequenced detection parameter values in the first detection data set, the change trend of each parameter in a preset time window is calculated to generate a parameter change matrix. The parameter change matrix records the fluctuation characteristics of each detection parameter.

[0082] The fluctuation characteristics of each detection parameter are extracted from the parameter change matrix, and a weighted fluctuation score is calculated by combining the pre-established comprehensive management performance index weight to generate a second detection data set. The second detection data set reflects the comprehensive management priority of each detection parameter.

[0083] For the comprehensive management priority in the second detection data set, a support vector machine algorithm is used for classification to generate a final civil air defense engineering detection information management result. The final result contains the priority ranking and management category of each detection parameter.

[0084] For example, the original detection data acquired from the civil air defense engineering detection equipment may contain parameters such as air quality and structural stress, and these data often have noise due to environmental interference. Denoising processing removes abnormal fluctuations through smoothing filtering method to generate a first detection data set, ensuring that the data reflects the true state. This step helps to improve the accuracy of subsequent analysis.

[0085] In one possible implementation, for the time-sequenced detection parameter values of the first detection data set, the change trend of the parameters in a preset time window is calculated.

[0086] For example, the fluctuation rate of air quality parameters within a day can be calculated by difference to generate a parameter change matrix, recording the dynamic characteristics of each parameter. This matrix provides basic data for subsequent priority evaluation.

[0087] Specifically, the fluctuation characteristics in the parameter change matrix can be calculated by extracting statistical quantities such as standard deviation, combined with pre-established comprehensive management performance index weights, such as air quality weight higher than structural stress weight, to calculate a weighted fluctuation score, generating a second detection data set. This data set reflects the management priority of each parameter, facilitating focus on key issues.

[0088] For example, the comprehensive management priority in the second detection data set can be used as input for a support vector machine algorithm, which classifies parameters into high, medium and low priority through classification, generating final civil air defense engineering detection information management results. This result includes priority ranking and management categories, which helps to efficiently allocate detection resources.

[0089] It should be noted that the support vector machine algorithm separates different priority data by constructing a hyperplane to ensure the reliability of the classification result, thereby achieving precise management.

[0090] The present application also provides a construction system for comprehensive management of civil air defense engineering detection information, which is used for the construction method of the civil air defense engineering detection information management model. The construction system for comprehensive management of civil air defense engineering detection information comprises: A data acquisition module for acquiring detection data of civil air defense engineering, wherein the detection data at least includes structural parameters and environmental parameters; A model generation module for generating a detection information model according to the detection data, wherein the detection information model is used to represent the state of civil air defense engineering; A processing representation module for processing the detection information model through a pre-established management platform to obtain a comprehensive management result, wherein the comprehensive management result is used to represent the running state of civil air defense engineering.

[0091] The present application also provides a readable storage medium, wherein the readable storage medium stores a program, and the program, when executed, performs the construction method of the civil air defense engineering detection information management model to acquire detection data of civil air defense engineering, wherein the detection data at least includes structural parameters and environmental parameters; generate a detection information model according to the detection data, wherein the detection information model represents the state of civil air defense engineering; and process the detection information model through a pre-established management platform to obtain a comprehensive management result, wherein the comprehensive management result is used to represent the running state of civil air defense engineering.

[0092] It is apparent that a person skilled in the art can make various modifications and variations to the embodiments of the application without departing from the spirit and scope of the application. Therefore, the application is intended to cover the modifications and variations of this application provided they come within the scope of the appended claims and their equivalents.

Claims

1. A method for constructing a civil defense engineering inspection information management model, characterized in that, The method for constructing the civil defense engineering inspection information management model includes the following steps: S1: Obtain the inspection data of the civil defense project, wherein the inspection data includes at least structural parameters and environmental parameters; S2: Generate a detection information model based on the detection data, the detection information model being used to characterize the status of civil defense projects; S3: The detection information model is processed through a pre-established management platform to obtain a comprehensive management result, which is used to characterize the operational status of the civil defense project.

2. The method for constructing the civil defense engineering inspection information management model according to claim 1, characterized in that, The acquisition of detection data for civil defense projects in S1 specifically includes: S11: Collect raw data of civil defense projects through sensor devices, including structural stress values, displacement values, and ambient temperature and humidity values; S12: Preprocess the raw data to obtain standardized detection data. The preprocessing includes data cleaning and format conversion. S13: Determine the structural parameters and the environmental parameters based on the standardized test data.

3. The method for constructing the civil defense engineering inspection information management model according to claim 1, characterized in that, The generation of the detection information model based on the detection data in step S2 specifically includes: S21: Determine the mechanical characteristics of the civil defense project based on the structural parameters, wherein the mechanical characteristics include stress distribution and deformation trend; S22: Determine the environmental adaptability of the civil defense project based on the environmental parameters, wherein the environmental adaptability includes durability and stability; S23: The mechanical characteristics and environmental adaptability are fused together using a preset modeling algorithm to generate the detection information model. The modeling algorithm includes finite element analysis and data fitting.

4. The method for constructing the civil defense engineering inspection information management model according to claim 1, characterized in that, The processing of the detection information model through a pre-established management platform in step S3 specifically includes: S31: The detection information model is classified through the management platform to obtain multiple status categories, including normal status and abnormal status; S32: For each state category, extract key features from the detection information model, including stress anomaly points and environmental exceedance points; S33: Generate the comprehensive management result based on the key features, the comprehensive management result including status assessment results and early warning information.

5. The method for constructing the civil defense engineering inspection information management model according to claim 2, characterized in that, The preprocessing of the raw data in S12 specifically includes: S121: The raw data is filtered by a preset threshold to obtain valid data, wherein the threshold includes stress range and temperature and humidity range; S122: Denoise the effective data to obtain smooth data. The denoising process includes signal processing based on wavelet transform. S123: The smoothed data is format-converted through data normalization to generate the standardized detection data. The data normalization includes linear mapping and standardization processing.

6. The method for constructing the civil defense engineering inspection information management model according to claim 3, characterized in that, The process of fusing the mechanical features and environmental adaptability using a preset modeling algorithm in step S23 specifically includes: S231: The mechanical feature is meshed using a finite element analysis algorithm to obtain the mechanical feature mesh; S232: The environmental adaptability is curve-fitted using a data fitting algorithm to obtain an environmental adaptability curve; S233: According to the preset fusion rules, the mechanical feature mesh is mapped to the environmental adaptability curve to generate a multi-dimensional detection information model. The fusion rules include weight allocation and feature superposition.

7. The method for constructing the civil defense engineering inspection information management model according to claim 4, characterized in that, The generation of the comprehensive management result based on the key features in S33 specifically includes: S331: The key features are scored using a preset evaluation model to obtain a status score, wherein the evaluation model includes a weighted average model; S332: Determine the priority of the state category based on the state score and the preset rating rules, wherein the priority represents the importance of the state category; S333: If the priority is higher than the preset priority threshold, then the warning information is generated, and the warning information includes the abnormal location and handling suggestions; S334: Generate the status assessment result based on the status score and the warning information.

8. The method for constructing the civil defense engineering inspection information management model according to claim 2, characterized in that, The structural parameters in S13 include stiffness coefficient and load-bearing capacity, and the environmental parameters include temperature and humidity.

9. A system for constructing a comprehensive management system for civil defense engineering inspection information, used in the method for constructing a civil defense engineering inspection information management model according to any one of claims 1 to 8, characterized in that, The system for the comprehensive management of civil defense engineering inspection information includes: The data acquisition module is used to acquire the inspection data of civil defense projects, and the inspection data includes at least structural parameters and environmental parameters; The model generation module is used to generate a detection information model based on the detection data, and the detection information model is used to characterize the status of the civil defense project. The processing and characterization module is used to process the detection information model through a pre-established management platform to obtain a comprehensive management result, which is used to characterize the operational status of the civil defense project.

10. A readable storage medium, characterized in that, The readable storage medium stores a program, and when the program is executed, it performs the method for constructing the civil defense engineering inspection information management model as described in any one of claims 1 to 8.

Images