Underground space disaster perception and risk intelligent decision analysis method and system

Abstract

The invention relates to the technical field of risk perception, in particular to an underground space disaster perception and risk intelligent decision analysis method and system. The method comprises the following steps: performing data acquisition on an underground space by using a sensor to generate underground space data; preprocessing the underground space data to obtain standard underground space data; performing multi-parameter feature extraction on the standard underground space data to obtain underground space multi-parameter features; performing underground space key disaster information identification on the underground space multi-parameter features to obtain key disaster information; therefore, through systematic data acquisition and analysis, the information processing efficiency and evaluation precision in underground space management are remarkably improved, and the problems of information delay, environmental restriction and insufficient evaluation in a traditional method are effectively solved. The innovation improves the disaster sensing ability, optimizes the timeliness and scientificity of risk decision, and provides more accurate decision support for managers.

Description

Technical Field

[0001] The present invention relates to the field of risk perception technology, and in particular to an underground space disaster perception and risk intelligent decision-making analysis method and system. Background Art

[0002] In the field of underground space disaster management, existing technologies are mainly focused on traditional monitoring and analysis methods, which often rely on the collection of static data and the evaluation of single factors. The monitoring equipment and sensors used in many existing systems cannot achieve real-time data updates, resulting in delayed judgment of disaster status and failure to timely reflect the dynamic changes of the underground space environment. Traditional data analysis methods are limited to basic statistics and historical data reviews, and it is difficult to effectively integrate multi-dimensional real-time data, thus ignoring the complexity and variability of disaster risks. Existing technologies have limitations in feature extraction and disaster status perception. Many systems lack effective multi-parameter feature extraction technology and fail to fully identify key factors closely related to underground space disasters, such as underground space humidity, groundwater level changes, and rock and soil mechanical properties. This leads to the fact that the assessment of disaster risk levels is often one-sided and cannot provide comprehensive risk warnings. Current disaster decision-making models are mostly based on linear assumptions and are difficult to adapt to complex underground space environments. When processing multiple disaster information, there is a lack of flexible decision-making mechanisms, which makes it impossible to respond to sudden disasters quickly and accurately. The existing technology also has significant deficiencies in data sharing and system integration. CN106246228B discloses a method and system for dynamic assessment of urban underground space waterlogging disasters. The method includes the following steps: collecting distribution information of urban underground space, and constructing a distribution model of urban underground space based on the distribution information; conducting historical waterlogging disaster assessment based on the distribution model to obtain historical waterlogging disaster assessment results; formulating waterlogging disaster prevention measures based on the historical waterlogging disaster assessment results, and constructing a preventive distribution model based on the distribution model. CN109815623A discloses a method for decision-making on urban underground space disasters based on fuzzy pattern recognition. The method includes the following steps: investigation and analysis of the current status of urban underground space disasters; construction of urban underground space evaluation indicators; determination of different levels of danger; and establishment of a disaster decision-making model. The urban underground space disaster decision-making method based on fuzzy pattern recognition proposed in CN109815623A has certain value in disaster risk assessment, but its limitation is that the method is relatively static, relying only on investigation and analysis and fuzzy logic to divide the danger level, and lacks the ability to analyze real-time data and predict risks for the dynamic environment of underground space. In addition, its disaster decision-making model lacks depth in information feedback and risk warning, and fails to fully utilize multi-source sensor data and deep learning models to accurately perceive disaster status and dynamically predict risks. Summary of the invention

[0003] Based on this, it is necessary to provide a method and system for underground space disaster perception and risk intelligent decision-making analysis to solve at least one of the above technical problems.

[0004] To achieve the above purpose, a method for underground space disaster perception and risk intelligent decision-making analysis is provided, the method comprising the following steps:

[0005] Step S1: using sensors to collect data from underground space to generate underground space data; preprocessing the underground space data to obtain standard underground space data;

[0006] Step S2: extracting multi-parameter features from standard underground space data to obtain multi-parameter features of underground space; identifying key disaster information of underground space from the multi-parameter features of underground space to obtain key disaster information;

[0007] Step S3: Perform disaster status perception analysis on key disaster information to obtain disaster status perception data; perform disaster risk level threshold assessment on the disaster status perception data to generate a disaster risk level threshold; perform risk degree calculation on the disaster status perception data according to the disaster risk level data to obtain disaster risk degree data;

[0008] Step S4: Conduct real-time disaster risk measurement on the multi-parameter characteristics of underground space through disaster risk degree data to obtain real-time data on underground space disaster risk; construct a disaster risk intelligent decision-making model based on the real-time data on underground space disaster risk to generate a disaster risk intelligent decision-making model; conduct intelligent disaster risk decision-making on key disaster information based on the disaster risk intelligent decision-making model to obtain a disaster risk decision-making report, thereby completing the underground space disaster perception and risk intelligent decision-making tasks.

[0009] The beneficial effect of the present invention is that the underground space data generated by using sensor technology for data collection provides basic information for subsequent analysis. This process ensures the comprehensiveness and real-time nature of the data and lays a solid foundation for the feature extraction of underground space. In the feature extraction stage, through multi-parameter analysis of standardized data, key data reflecting important characteristics of the underground environment can be extracted, including underground space humidity, groundwater level and geotechnical properties, which provide a basis for the identification of disaster information. It enables researchers to conduct in-depth analysis of key disaster information and generate accurate risk assessment standards through risk level threshold assessment. This process not only improves the sensitivity to disaster status, but also provides data support for the calculation of subsequent risk levels. By comprehensively analyzing the disaster status perception data, the system can monitor the risk changes of underground space in real time, generate disaster risk level data, and make the decision-making process more scientific and reliable. Through the construction of real-time disaster risk determination and risk intelligent decision-making model, dynamic management of disaster risk is realized. Based on real-time data input, the model can flexibly respond to various potential risks, provide accurate basis for decision-making of key disaster information, and generate corresponding disaster risk decision reports. This series of steps not only improves the accuracy and efficiency of underground space disaster monitoring, but also provides scientific support for the safety management of underground space. Therefore, the present invention significantly improves the information processing efficiency and assessment accuracy in underground space management through systematic data collection and analysis, and effectively solves the problems of information delay and insufficient assessment due to environmental constraints in traditional methods. This innovation improves disaster perception capabilities, optimizes the timeliness and scientificity of risk decision-making, and provides managers with more accurate decision-making support.

[0010] Preferably, step S1 comprises the following steps:

[0011] Step S11: using sensors to collect data from underground space and generate underground space data;

[0012] Step S12: performing data cleaning on the underground space data to generate underground space cleaning data;

[0013] Step S13: performing data standardization processing on the underground space cleaning data to generate standard underground space data.

[0014] The present invention uses sensors to collect data from underground spaces, and the generated underground space data not only includes key indicators such as underground space humidity, groundwater level and geotechnical properties, but also covers the environmental characteristics of the space. This process ensures the comprehensiveness and accuracy of the information and provides necessary data support for subsequent analysis. Secondly, a data cleaning step is performed for the collected underground space data. This step significantly improves the data quality by removing noise, filling missing values ​​and correcting erroneous data. The cleaned data can reflect the real underground environmental conditions and provide a reliable basis for subsequent data analysis and decision-making. In the data standardization processing stage, consistency conversion is performed for data from different sources and formats to generate standard underground space data. The standardization process makes various types of data comparable, eliminates the deviation caused by heterogeneous data sources, and provides a unified benchmark for subsequent analysis. Standard underground space data not only enhances the availability and applicability of data, but also lays the foundation for multi-dimensional feature extraction and deep analysis, making subsequent disaster identification, risk assessment and other steps more scientific and accurate. The present invention improves the quality and consistency of underground space data, thereby enhancing the effectiveness of subsequent analysis and the scientific nature of decision-making. An effective data processing process makes real-time monitoring and risk management of underground space an implementable step, thereby better responding to potential environmental challenges and disaster risks, and ultimately achieving intelligent management of underground space.

[0015] Preferably, step S2 comprises the following steps:

[0016] Step S21: extracting multi-parameter features from standard underground space data to obtain multi-parameter features of the underground space;

[0017] Step S22: performing multi-feature combination processing on the underground space multi-parameter features to generate underground space multi-parameter combination data;

[0018] Step S23: using the underground space multi-parameter combination data to identify underground space key disaster information based on the underground space multi-parameter characteristics, and obtaining key disaster information, wherein the key disaster information includes underground space humidity data, groundwater level data and rock and soil mechanics property data.

[0019] The present invention can effectively capture and quantify the key environmental characteristics of underground space by extracting multi-parameter features from standard underground space data. These characteristics not only include physical indicators such as underground space humidity, groundwater level and geotechnical properties, but also cover potential environmental changes and disaster risks, thereby providing a comprehensive data basis for subsequent analysis. Multi-feature combination processing is performed to integrate different types of multi-parameter features into multi-parameter combination data of underground space. This process helps to comprehensively evaluate underground space from multiple dimensions, enhances the complementarity and correlation of data, and enables the analysis model to more comprehensively reflect the complexity of the underground environment. In addition, through combined processing, the potential interactions between the parameters can be identified, thereby providing more abundant input information for the construction of multivariate disaster models. The multi-parameter combination data of underground space is used to identify key disaster information. This process converts multi-parameter features into useful disaster warning information through intelligent data analysis and processing methods, making the identification of key disaster information more accurate and efficient. This not only provides a reliable reference basis for underground space management, but also can timely respond to changes in the underground environment, identify potential disaster risks in advance, and promote rapid response and effective intervention. In summary, the implementation of this series of steps has significantly improved the utilization efficiency of underground space data, thus playing an important role in disaster monitoring and management. Through systematic feature extraction and combination, the understanding and response capabilities of complex underground environments have been enhanced, providing more comprehensive support for scientific decision-making.

[0020] Preferably, step S3 comprises the following steps:

[0021] Step S31: performing disaster status perception analysis on key disaster information to obtain disaster status perception data;

[0022] Step S32: performing time series analysis on the disaster status perception data to generate a disaster status time series analysis result; performing disaster risk level threshold evaluation on the disaster status time series analysis result to generate a disaster risk level threshold;

[0023] Step S33: Calculate the risk level of the disaster status perception data according to the disaster risk level data to obtain disaster risk level data.

[0024] The present invention can effectively integrate multi-dimensional environmental data by performing state perception analysis on key disaster information, thereby generating disaster state perception data. This process not only enhances the real-time monitoring capability of underground environmental changes, but also lays a solid foundation for subsequent data analysis, ensuring the accuracy and reliability of the analysis results. The disaster state perception data is processed by time series analysis to generate disaster state time series analysis results. This analysis can not only reveal the dynamic characteristics of disaster state changes over time, but also identify potential cyclical trends and abnormal fluctuations. This analysis method based on the time dimension provides an important basis for understanding the timeliness of disaster occurrence and its association with external factors. At the same time, the disaster risk level threshold evaluation of the time series analysis results can clarify the risk interval under different disaster states, provide a quantitative reference basis for subsequent decision-making, and make the risk level evaluation more scientific and systematic. The risk level of the disaster state perception data is calculated according to the disaster risk level data, thereby obtaining the disaster risk level data. This process uses scientific calculation methods to convert complex disaster information into easy-to-understand risk level indicators, which is convenient for quickly evaluating the severity of disaster impacts. This quantified risk level data not only provides timely warning information to decision makers, but also provides important support for resource allocation, emergency response, etc., ensuring effective intervention and management when disasters occur.

[0025] Preferably, step S31 includes the following steps:

[0026] Step S311: Acquire historical underground space humidity data; compare the moisture retention capacity of the historical underground space humidity data with the underground space humidity data, and perform underground area spatial analysis in combination with the geographic information system to generate underground space moisture comparison position data; perform underground space disaster perception on the underground space moisture comparison position data to generate underground area underground space moisture perception data;

[0027] Step S312: Analyze the water level fluctuation amplitude of the groundwater level data to generate groundwater level fluctuation amplitude data; analyze the underground space particle compaction degree of the groundwater level fluctuation amplitude data to generate groundwater level sedimentation particle analysis data; use the underground space moisture comparison position data to perceive the water level disaster state of the groundwater level sedimentation particle analysis data to generate groundwater level disaster perception data;

[0028] Step S313: Perform triaxial shear test on the geotechnical mechanics property data to generate underground space shear strength data; perform consolidation test processing on the geotechnical mechanics property data to generate underground space porosity data; perform geotechnical consolidation load disaster perception on the underground space shear strength data and underground space porosity data to generate underground space geotechnical perception data;

[0029] Step S314: Perform multi-level analysis on the underground space moisture perception data, groundwater level disaster perception data and underground space rock and soil perception data in the underground area to generate disaster status perception data.

[0030] The present invention obtains historical underground space humidity data and compares the moisture retention capacity with the current underground space humidity data. This process not only makes the change of underground space humidity have a clearer understanding, but also combines the geographic information system (GIS) technology to conduct underground regional spatial analysis. The generated underground space moisture comparison position data provides an important spatial reference for in-depth analysis of underground space moisture distribution and its changes. This data generation process effectively enhances the spatial visualization of underground moisture conditions, allowing potential underground space humidity-related disasters to be identified earlier. The process of land settlement particle analysis for groundwater level data not only provides a quantitative assessment of groundwater level changes, but also deeply explores the disaster status caused by water level changes through combination with moisture comparison position data. This analysis method ensures the accuracy and reliability of water level disaster perception data, and provides data support for water resource management and underground space utilization. By performing triaxial shear test and consolidation test processing on rock and soil mechanical properties data, the underground space shear strength and porosity data generated provide an experimental basis for the perception of rock and soil disasters in underground space. This experimental data not only reflects the physical characteristics of underground space, but also provides a scientific basis for understanding the stability and risk resistance of underground space structures, thereby improving the early warning capability of underground disasters as a whole. The generated underground space moisture perception data, groundwater level disaster perception data and rock and soil perception data are analyzed at multiple levels, and the disaster status perception data finally generated can effectively integrate various factors to form a comprehensive underground disaster status assessment system. This integrated analysis not only improves the efficiency of data utilization, but also provides a more reliable basis for decision makers to formulate corresponding disaster response strategies, ensuring that potential underground disaster risks can be responded to in a timely and accurate manner.

[0031] Preferably, step S33 includes the following steps:

[0032] Step S331: compare the historical underground space humidity data and the underground space moisture perception data of the underground area to generate the underground space moisture comparison data of the underground area; calculate the underground space humidity risk index for the underground space moisture comparison data of the underground area using the disaster risk level threshold to generate the underground space humidity risk index;

[0033] Step S332: monitor the water level fluctuation amplitude of the groundwater level disaster sensing data to generate underground space water level monitoring data; use the disaster risk level threshold to perform abnormal disaster fluctuation assessment calculation on the underground space water level monitoring data to generate an underground space water level risk index;

[0034] Step S333: using the disaster risk level threshold to evaluate and calculate the collapse risk index of the underground space geotechnical sensing data to generate the underground space geotechnical risk index;

[0035] Step S334: weight the risk index of underground space humidity, the risk index of underground space water level and the risk index of underground space geotechnical risk to generate disaster risk degree data.

[0036] The present invention compares the historical underground space humidity data with the underground space moisture perception data of the underground area, and the underground area underground space moisture comparison data generated provides a data basis for in-depth analysis of underground space moisture changes. This comparison not only reveals the temporal and spatial variation trend of underground space humidity, but also lays the foundation for further risk index calculation. The calculation of the underground space humidity risk index using the disaster risk level threshold not only enhances the early warning capability of potential humidity disasters, but also provides specific indicators for decision makers to guide corresponding response measures. The underground water level monitoring data generated by monitoring the water level fluctuation amplitude of the underground water level disaster perception data makes the dynamic change of the water level visualized, providing a real-time basis for the management and protection of groundwater resources. By applying the disaster risk level threshold to the water level monitoring data to evaluate and calculate the abnormal disaster fluctuation, the underground space water level risk index generated provides a scientific basis for identifying and responding to abnormal water level changes. The implementation of this process ensures the timely perception and processing of disasters related to water level fluctuations. By evaluating and calculating the collapse risk index of the underground space geotechnical perception data, the underground space geotechnical risk index generated further enriches the dimension of underground space disaster risk. This assessment method combines the characteristics of geotechnical mechanics and provides data support for understanding the stability of underground space and the degree to which it is affected by external factors, so that potential geotechnical disaster risks can be identified and assessed in a timely manner. The disaster risk level data generated by merging the weights of various risk indexes integrates the influence of underground space humidity, groundwater level and geotechnical properties, making the assessment of disaster risks more comprehensive and accurate. The setting of weights reflects the relative importance of each factor in the overall risk assessment, ensuring the scientificity and rationality of the risk assessment results. This multi-level and multi-dimensional data integration not only improves the overall monitoring level of underground space disasters, but also provides a reliable basis for decision makers to support their decision-making and emergency response in complex environments.

[0037] Preferably, step S4 comprises the following steps:

[0038] Step S41: Real-time disaster risk determination is performed on the multi-parameter characteristics of underground space through disaster risk degree data to obtain real-time data on underground space disaster risk;

[0039] Step S42: constructing a disaster risk intelligent decision-making model based on the real-time data of underground space disaster risk to generate a disaster risk intelligent decision-making model;

[0040] Step S43: using the underground space multi-parameter feature as input to the disaster risk intelligent decision-making model to perform underground disaster multi-parameter prediction, generate disaster intelligent risk parameter feature prediction data, and perform update timestamp processing to replace the underground space multi-parameter feature;

[0041] Step S44: Make intelligent disaster risk decisions on key disaster information based on the disaster risk intelligent decision-making model to obtain a disaster risk decision report.

[0042] The present invention uses the disaster risk degree data to measure the real-time disaster risk of the underground space multi-parameter characteristics, and the obtained real-time data of underground space disaster risk provides a solid foundation for subsequent risk assessment and decision-making. This process ensures the real-time and effectiveness of the data, so that decision makers can quickly identify potential disaster risks and take corresponding measures to intervene in time. This not only improves the response speed to sudden disasters, but also effectively enhances the grasp of long-term risk trends. The process of constructing a disaster risk intelligent decision-making model based on the real-time data of underground space disaster risk provides data-driven decision support for complex risk management. The model integrates multi-dimensional data features, so that flexible and accurate decisions can be made in the dynamic environment where disasters occur. This model can not only process historical data, but also update in real time to reflect the latest risk status, thereby ensuring the scientificity and adaptability of decision-making. The multi-parameter characteristics of the underground space are used as input, and the multi-parameter prediction of underground disasters is performed through the risk intelligent decision-making model. The generated disaster intelligent risk parameter feature prediction data provides a forward-looking assessment for various potential disasters. This prediction process ensures the timeliness of the data by updating the timestamp processing, so that the prediction results can reflect real-time changes, thereby providing decision makers with the latest risk assessment information and effectively guiding their resource allocation and emergency response. Based on the disaster risk intelligent decision-making model, intelligent disaster risk decision-making is carried out on key disaster information, and the disaster risk decision-making report obtained integrates all the results of previous analysis and evaluation to form a systematic report. This report not only covers various risk assessment results, but also provides corresponding response strategies and suggestions to help decision makers formulate more scientific and reasonable emergency plans and management measures.

[0043] Preferably, step S42 includes the following steps:

[0044] Step S421: using the underground space multi-parameter features as the data input layer of the model to synchronize the data timestamps and generate an underground space model input layer;

[0045] Step S422: constructing a model feature extraction layer for the real-time data of underground space disaster risk to generate an underground space model feature extraction layer;

[0046] Step S423: Use the disaster status perception data as training data to train the underground space model input layer and the underground space model feature extraction layer, thereby generating a disaster risk intelligent decision-making model.

[0047] The present invention ensures the timeliness and consistency of input data by using the multi-parameter features of underground space as the data input layer of the model to synchronize the data timestamp. This process not only helps to eliminate potential errors caused by differences in data acquisition time, but also enhances the sensitivity of the model to real-time changing environments, thereby providing accurate basic data for subsequent analysis. The model feature extraction layer is constructed for the real-time data of underground space disaster risk. Through this feature extraction process, the most representative features can be identified from complex input data. This feature extraction not only enables the model to focus on key influencing factors, but also reduces the interference of redundant information, thereby improving the prediction accuracy and efficiency of the model. Through refined data processing, this step provides deep insights for the decision-making model, helps to identify potential risks and formulate corresponding countermeasures. The operation of training the number of model layers for the underground space model input layer and the underground space model feature extraction layer using disaster state perception data greatly enhances the learning ability and adaptability of the model. Through repeated learning of these training data, the model can gradually optimize its parameters and improve its response ability to complex disaster situations. This process not only realizes the dynamic update of the model, but also ensures that the model can adapt to the changing underground environment and has a strong generalization ability.

[0048] Preferably, step S44 includes the following steps:

[0049] Step S441: predicting the real-time data of underground space disaster risk based on the intelligent decision-making model for disaster risk, and generating underground space disaster prediction results, wherein the underground space disaster prediction results include underground space low-risk disaster data, underground space medium-risk disaster data, and underground space high-risk disaster data;

[0050] Step S442: When the underground space disaster prediction result is underground space low risk disaster data, threshold classification analysis is performed on the disaster risk level threshold based on the underground space data, an underground space low risk threshold data analysis report is generated, and a green label is performed, thereby completing the underground space low risk disaster strategy;

[0051] Step S443: When the underground space disaster prediction result is underground space medium risk disaster data, the underground space humidity data is monitored to generate medium risk underground space humidity analysis data; the groundwater level data is marked with abnormal increase, and abnormal underground space water level data is found during the generation; the medium risk underground space humidity analysis data and the underground space water level abnormal data are predicted by exponential smoothing method to generate an underground space medium risk prediction report, thereby completing the underground space medium risk disaster strategy;

[0052] Step S444: When the underground space disaster prediction result is underground space high-risk disaster data, the underground space moisture comparison position data is marked as an emergency outlier to generate underground space high-risk underground space humidity outlier data; the underground space geotechnical perception data is identified as key points, and positioned in combination with GIS to generate underground space high-risk key point data; the underground space high-risk underground space humidity outlier data and underground space high-risk key point data are reported to generate an underground space high-risk disaster report, and the report is sent using the cloud platform, thereby formulating an underground space high-risk disaster strategy;

[0053] Step S445: Merge the underground space low risk threshold data analysis report, the underground space medium risk prediction report and the underground space high risk disaster report to generate a disaster risk decision report.

[0054] The present invention generates prediction results of different risk levels by predicting the disaster results of underground space disaster risk real-time data based on the disaster risk intelligent decision-making model. This process realizes the classification and identification of low-risk, medium-risk and high-risk disasters in underground space, laying a data foundation for the formulation of subsequent targeted strategies. It reflects the flexibility and adaptability of data-driven decision-making. In the case of low risk, a low-risk threshold data analysis report is generated by classifying and analyzing the threshold of disaster risk level. This report not only provides decision makers with a clear risk cognition, but also enhances the visual management of low-risk status through green labeling, which is conducive to the rational allocation and efficient utilization of resources. In the case of medium risk, medium-risk underground space humidity analysis data and underground space water level abnormal data are generated by monitoring and analyzing underground space humidity data and groundwater level data. This operation provides managers with dynamic monitoring and forward-looking evaluation of medium-risk situations through risk prediction by exponential smoothing method, ensuring timely response measures to reduce potential losses. For high-risk scenarios, step S444 generates high-risk underground space humidity abnormal value data and high-risk key point data by emergency marking and key point identification of moisture comparison position data and geotechnical perception data. This process not only reveals the specific location of high-risk areas, but also sends high-risk disaster reports in a timely manner through the cloud platform, providing instant information support for emergency response and enhancing the timeliness of response measures. The merging of reports at each risk level generates a comprehensive disaster risk decision report. This summary information not only integrates data analysis results at different levels, but also provides decision makers with a global perspective on risk assessment. This series of measures makes the entire disaster management process more systematic and data-based, improves the scientificity and rationality of decision-making, and ensures rapid response and efficient handling in emergencies.

[0055] In this specification, a system for underground space disaster perception and risk intelligent decision-making analysis is provided, which is used to execute the above-mentioned underground space disaster perception and risk intelligent decision-making analysis method. The underground space disaster perception and risk intelligent decision-making analysis system:

[0056] Data acquisition module: Use sensors to collect data from underground space and generate underground space data; pre-process underground space data to obtain standard underground space data;

[0057] Feature extraction module: extract multi-parameter features from standard underground space data to obtain multi-parameter features of underground space; identify key disaster information of underground space based on multi-parameter features of underground space to obtain key disaster information;

[0058] Disaster perception analysis module: conduct disaster status perception analysis on key disaster information to obtain disaster status perception data; conduct disaster risk level threshold assessment on disaster status perception data to generate disaster risk level threshold; calculate the risk level of disaster status perception data based on disaster risk level data to obtain disaster risk level data;

[0059] Risk decision-making module: Real-time disaster risk measurement of multi-parameter characteristics of underground space is carried out through disaster risk degree data to obtain real-time data of underground space disaster risk; a disaster risk intelligent decision-making model is constructed based on the real-time data of underground space disaster risk to generate a disaster risk intelligent decision-making model; intelligent disaster risk decision-making is carried out on key disaster information based on the disaster risk intelligent decision-making model to obtain a disaster risk decision-making report, thereby completing the underground space disaster perception and risk intelligent decision-making tasks.

[0060] The beneficial effect of the present invention is that the underground space data generated by using sensor technology for data collection provides basic information for subsequent analysis. This process ensures the comprehensiveness and real-time nature of the data and lays a solid foundation for the feature extraction of underground space. In the feature extraction stage, through multi-parameter analysis of standardized data, key data reflecting important characteristics of the underground environment can be extracted, including underground space humidity, groundwater level and geotechnical properties, which provide a basis for the identification of disaster information. It enables researchers to conduct in-depth analysis of key disaster information and generate accurate risk assessment standards through risk level threshold assessment. This process not only improves the sensitivity to disaster status, but also provides data support for the calculation of subsequent risk levels. By comprehensively analyzing the disaster status perception data, the system can monitor the risk changes of underground space in real time, generate disaster risk level data, and make the decision-making process more scientific and reliable. Through the construction of real-time disaster risk determination and risk intelligent decision-making model, dynamic management of disaster risk is realized. Based on real-time data input, the model can flexibly respond to various potential risks, provide accurate basis for decision-making of key disaster information, and generate corresponding disaster risk decision reports. This series of steps not only improves the accuracy and efficiency of underground space disaster monitoring, but also provides scientific support for the safety management of underground space. Therefore, the present invention significantly improves the information processing efficiency and assessment accuracy in underground space management through systematic data collection and analysis, and effectively solves the problems of information delay and insufficient assessment due to environmental constraints in traditional methods. This innovation improves disaster perception capabilities, optimizes the timeliness and scientificity of risk decision-making, and provides managers with more accurate decision-making support. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 A schematic diagram of the steps of an underground space disaster perception and risk intelligent decision-making analysis method;

[0062] Figure 2 for Figure 1 Detailed implementation steps of step S2 in the flowchart;

[0063] Figure 3 for Figure 1 Detailed implementation steps of step S3 in FIG.

[0064] Figure 4 for Figure 1 Detailed implementation steps of step S4 in FIG.

[0065] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0066] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0067] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0068] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0069] To achieve this, please refer to Figures 1 to 4 , a method for underground space disaster perception and risk intelligent decision-making analysis, the method comprising the following steps:

[0070] Step S1: using sensors to collect data from underground space to generate underground space data; preprocessing the underground space data to obtain standard underground space data;

[0071] Step S2: extracting multi-parameter features from standard underground space data to obtain multi-parameter features of underground space; identifying key disaster information of underground space from the multi-parameter features of underground space to obtain key disaster information;

[0072] Step S3: Perform disaster status perception analysis on key disaster information to obtain disaster status perception data; perform disaster risk level threshold assessment on the disaster status perception data to generate a disaster risk level threshold; perform risk degree calculation on the disaster status perception data according to the disaster risk level data to obtain disaster risk degree data;

[0073] Step S4: Conduct real-time disaster risk measurement on the multi-parameter characteristics of underground space through disaster risk degree data to obtain real-time data on underground space disaster risk; construct a disaster risk intelligent decision-making model based on the real-time data on underground space disaster risk to generate a disaster risk intelligent decision-making model; conduct intelligent disaster risk decision-making on key disaster information based on the disaster risk intelligent decision-making model to obtain a disaster risk decision-making report, thereby completing the underground space disaster perception and risk intelligent decision-making tasks.

[0074] The beneficial effect of the present invention is that the underground space data generated by using sensor technology for data collection provides basic information for subsequent analysis. This process ensures the comprehensiveness and real-time nature of the data and lays a solid foundation for the feature extraction of underground space. In the feature extraction stage, through multi-parameter analysis of standardized data, key data reflecting important characteristics of the underground environment can be extracted, including underground space humidity, groundwater level and geotechnical properties, which provide a basis for the identification of disaster information. It enables researchers to conduct in-depth analysis of key disaster information and generate accurate risk assessment standards through risk level threshold assessment. This process not only improves the sensitivity to disaster status, but also provides data support for the calculation of subsequent risk levels. By comprehensively analyzing the disaster status perception data, the system can monitor the risk changes of underground space in real time, generate disaster risk level data, and make the decision-making process more scientific and reliable. Through the construction of real-time disaster risk determination and risk intelligent decision-making model, dynamic management of disaster risk is realized. Based on real-time data input, the model can flexibly respond to various potential risks, provide accurate basis for decision-making of key disaster information, and generate corresponding disaster risk decision reports. This series of steps not only improves the accuracy and efficiency of underground space disaster monitoring, but also provides scientific support for the safety management of underground space. Therefore, the present invention significantly improves the information processing efficiency and assessment accuracy in underground space management through systematic data collection and analysis, and effectively solves the problems of information delay and insufficient assessment due to environmental constraints in traditional methods. This innovation improves disaster perception capabilities, optimizes the timeliness and scientificity of risk decision-making, and provides managers with more accurate decision-making support.

[0075] In the embodiment of the present invention, reference Figure 1 The above is a schematic diagram of the steps of an underground space disaster perception and risk intelligent decision-making analysis method of the present invention. In this example, the underground space disaster perception and risk intelligent decision-making analysis method includes the following steps:

[0076] Step S1: using sensors to collect data from underground space to generate underground space data; preprocessing the underground space data to obtain standard underground space data;

[0077] In the embodiment of the present invention, the specific technical means of using sensors to collect data from underground space include the combined use of multiple types of sensors, such as underground space humidity sensors, groundwater level sensors, and geotechnical sensors. These sensors generate rich underground space data by real-time monitoring of the physical and chemical characteristics of the underground space. These data include underground space humidity, groundwater level, temperature, pressure, and other related environmental parameters, which together constitute a comprehensive understanding of the underground environment. After the data collection is completed, the process of preprocessing the original underground space data is crucial. The preprocessing technology includes steps such as data cleaning, outlier detection, and data standardization. Data cleaning aims to remove noise and inconsistent data points, and identify and eliminate obvious erroneous values ​​through algorithms to ensure the accuracy and reliability of the data. In this process, statistical methods, such as the calculation of mean and standard deviation, are used to identify outliers and perform corresponding processing. Data standardization is a key step to unify the values ​​obtained by different sensors. By normalizing or standardizing the data, different types of data can be compared and analyzed under the same dimension, which involves converting the data into standard scores with the same mean and standard deviation, or scaling it to a specific range.

[0078] Step S2: extracting multi-parameter features from standard underground space data to obtain multi-parameter features of underground space; identifying key disaster information of underground space from the multi-parameter features of underground space to obtain key disaster information;

[0079] In the embodiment of the present invention, the specific technical means for extracting multi-parameter features from standard underground space data mainly include methods such as feature selection and feature transformation. The process of feature selection aims to screen out the parameters that have the greatest influence on subsequent analysis and decision-making from standard underground space data. This involves the application of statistical analysis and machine learning algorithms, such as principal component analysis (PCA) and correlation analysis, through which features with high correlation or importance can be identified, which can effectively reflect the physical and chemical state of the underground space. Feature transformation technology can be used to further process the extracted multi-parameter features for subsequent analysis. This includes data normalization, standardization, and logarithmic transformation, so that different features are comparable under the same dimension. Through these transformations, the scale differences between features can be reduced, thereby enhancing the sensitivity and accuracy of the model to data changes. After multi-parameter feature extraction, the standard underground space data finally forms a multi-parameter feature set containing multiple dimensional information. In the process of identifying key disaster information, commonly used technical means include machine learning algorithms (such as support vector machines, decision trees, random forests, etc.) and deep learning methods (such as convolutional neural networks). These algorithms can effectively distinguish different types of disaster information by training and learning multi-parameter features. In this process, the training data consists of historical disaster data and corresponding environmental characteristics. The model is optimized iteratively to continuously improve the accuracy of disaster information recognition. Through the extraction of multi-parameter features and the identification of key disaster information, the system can generate key disaster information about underground space, including underground space humidity, groundwater level and rock and soil mechanical properties.

[0080] Step S3: Perform disaster status perception analysis on key disaster information to obtain disaster status perception data; perform disaster risk level threshold assessment on the disaster status perception data to generate a disaster risk level threshold; perform risk degree calculation on the disaster status perception data according to the disaster risk level data to obtain disaster risk degree data;

[0081] In an embodiment of the present invention, historical disaster data and current key disaster information are deeply analyzed to identify potential disaster states. Specifically, state perception analysis uses time series analysis, cluster analysis, and anomaly detection technology. These technologies can reveal potential trends, periodic changes, and abnormal patterns in the data, thereby providing an important basis for subsequent risk assessment. When evaluating the threshold value of disaster risk levels for disaster state perception data, commonly used technical means include statistical analysis and threshold model construction. Through statistical analysis of historical disaster events, the risk level thresholds of various types of disasters can be determined. These thresholds are based on the analysis of data distribution characteristics, such as using quantiles, standard deviations, etc. to set corresponding risk level limits, so as to classify different disaster states. The establishment of the threshold model can also consider the influence of multidimensional characteristics to ensure the comprehensiveness and accuracy of risk assessment. The risk degree of disaster state perception data is calculated based on the disaster risk level data, involving technical means such as weighted average method, fuzzy logic system, or multi-attribute decision analysis. In this process, researchers can assign weights to different risk factors according to their importance, and comprehensively calculate a comprehensive risk index. The index can effectively reflect the comprehensive risk level of underground space and provide decision makers with clear risk assessment results. Through the application of this series of technical means, the disaster risk level data generated can not only reflect the complexity of the current disaster status, but also provide a scientific basis for future disaster warning and response.

[0082] Step S4: Conduct real-time disaster risk measurement on the multi-parameter characteristics of underground space through disaster risk degree data to obtain real-time data on underground space disaster risk; construct a disaster risk intelligent decision-making model based on the real-time data on underground space disaster risk to generate a disaster risk intelligent decision-making model; conduct intelligent disaster risk decision-making on key disaster information based on the disaster risk intelligent decision-making model to obtain a disaster risk decision-making report, thereby completing the underground space disaster perception and risk intelligent decision-making tasks.

[0083] In the embodiment of the present invention, the specific technical means used for real-time determination of disaster risk degree data, construction of risk intelligent decision-making model and generation of risk decision-making report cover a variety of data analysis and modeling methods. This process first relies on real-time data processing technology, and realizes rapid analysis of multi-parameter characteristics of underground space through stream data processing platform. The key technologies at this stage include data fusion and real-time calculation, which integrates data from different sensors to generate a comprehensive disaster risk assessment view to ensure timely acquisition of real-time data of disaster risk in underground space. The construction of disaster risk intelligent decision-making model based on these real-time data adopts machine learning and artificial intelligence algorithms. These algorithms include support vector machines, random forests and deep learning models, which can effectively learn and capture the complex relationship between underground space disaster characteristics and risk factors. In this process, the training process of the model relies on the combination of historical data and real-time data to ensure that the model has high prediction accuracy and generalization ability when facing actual disasters. In the process of generating disaster risk intelligent decision-making report, the construction of decision support system is involved. These systems can provide decision makers with intelligent decision support for key disaster information by integrating multiple decision rules and optimization algorithms. Specifically, the technical means at this stage include rule-based reasoning system, fuzzy logic and multi-objective optimization algorithm. These methods enable the system to comprehensively consider different risk factors and provide corresponding decision-making recommendations based on the preset risk level threshold, thus forming a complete disaster risk decision-making report. Ultimately, the effective combination of this series of technical means not only improves the ability of intelligent perception of underground space disasters, but also greatly improves the efficiency and accuracy of disaster risk management, allowing decision makers to respond quickly in a dynamically changing environment.

[0084] Preferably, step S1 comprises the following steps:

[0085] Step S11: using sensors to collect data from underground space and generate underground space data;

[0086] Step S12: performing data cleaning on the underground space data to generate underground space cleaning data;

[0087] Step S13: performing data standardization processing on the underground space cleaning data to generate standard underground space data.

[0088] In the embodiment of the present invention, in the data collection stage, sensors are used to monitor the underground space in real time. This process relies on the Internet of Things technology. By deploying various types of sensors (such as underground space humidity sensors, groundwater level sensors and geotechnical mechanics sensors), the system can continuously collect underground space data under different environmental conditions. These data contain rich physical and chemical characteristics, which help to fully understand the state of the underground environment. In the data cleaning stage, the generated underground space data often contains problems such as noise, missing values ​​and outliers, so a series of data cleaning technologies need to be adopted. This process includes means such as data denoising, missing value filling and outlier detection. Specifically, commonly used methods include threshold detection based on statistical analysis, interpolation methods (such as linear interpolation or spline interpolation) and machine learning models (such as isolated forest or local outlier factor algorithms). Through these methods, the cleaned data can significantly improve the data quality and reduce the impact on subsequent analysis results. Entering the data standardization processing stage, the purpose is to convert the cleaned data into a format with a unified standard to facilitate subsequent data analysis and modeling. Data standardization involves normalizing or standardizing each feature, such as converting the data to a distribution with a mean of 0 and a standard deviation of 1, or scaling the data to the interval [0,1].

[0089] As an example of the present invention, refer to Figure 2 As shown, in this example, step S2 includes:

[0090] Step S21: extracting multi-parameter features from standard underground space data to obtain multi-parameter features of the underground space;

[0091] Step S22: performing multi-feature combination processing on the underground space multi-parameter features to generate underground space multi-parameter combination data;

[0092] Step S23: using the underground space multi-parameter combination data to identify underground space key disaster information based on the underground space multi-parameter characteristics, and obtaining key disaster information, wherein the key disaster information includes underground space humidity data, groundwater level data and rock and soil mechanics property data.

[0093] In the embodiment of the present invention, in the multi-parameter feature extraction stage, the standard underground space data is analyzed and processed to extract multi-parameter features with significant representativeness. This process applies statistical analysis, signal processing and machine learning methods, such as principal component analysis (PCA) and feature selection techniques (such as LASSO regression) to reduce the data dimension while maintaining important information. Through these technical means, the system can extract characteristic variables closely related to the underground environmental state from complex data sets, such as key parameters such as underground space humidity, groundwater level, and geotechnical properties, which lays the foundation for subsequent analysis. Entering the multi-feature combination processing stage, the extracted multi-parameter features need to be integrated and optimized to generate more comprehensive underground space multi-parameter combination data. This process uses feature engineering methods, such as feature interaction, feature mapping and cluster analysis, and uses aggregation functions (such as mean, variance, maximum value, etc.) to form new combination features. This processing method not only helps to capture the relationship between each feature, but also improves the sensitivity and accuracy of the model in disaster risk identification. In the key disaster information identification stage, combined with the generated underground space multi-parameter combination data, the data is further analyzed by applying supervised learning and unsupervised learning models (such as support vector machines, decision trees, and random forests) to identify potential key disaster information. This process involves training models to identify and classify data of different risk levels, especially focusing on key indicators such as underground space humidity, groundwater level, and geotechnical properties.

[0094] As an example of the present invention, refer to Figure 3 As shown, in this example, step S3 includes:

[0095] Step S31: performing disaster status perception analysis on key disaster information to obtain disaster status perception data;

[0096] Step S32: performing time series analysis on the disaster status perception data to generate a disaster status time series analysis result; performing disaster risk level threshold evaluation on the disaster status time series analysis result to generate a disaster risk level threshold;

[0097] Step S33: Calculate the risk level of the disaster status perception data according to the disaster risk level data to obtain disaster risk level data.

[0098] In an embodiment of the present invention, in the disaster state perception analysis stage, features related to the disaster state can be extracted through in-depth analysis of key disaster information. This process uses data mining and pattern recognition technology, such as a classification model based on support vector machine (SVM), decision tree or neural network, to classify and score disaster information, thereby generating disaster state perception data. This method can effectively learn data features under different states by establishing a training data set, thereby improving the perception ability of potential disasters. In the time series analysis stage, the disaster state perception data is input into a time series model, such as an autoregressive integrated moving average (ARIMA) model or a long short-term memory (LSTM) network, to capture the dynamic characteristics of data changing over time. By modeling and predicting historical data, time series analysis can reveal the evolution trend and periodic fluctuations of the disaster state, thereby generating a disaster state time series analysis result. This analysis not only provides a basis for understanding the time pattern of disaster occurrence, but also lays the foundation for subsequent risk assessment. After generating the disaster state time series analysis result, the next risk level threshold assessment is to assess the risk level of the disaster by setting a series of thresholds. This involves extreme value theory or empirical distribution function (EDF) in statistics to identify and define boundary conditions for different risk levels and generate disaster risk level thresholds. The key to this process is to conduct in-depth analysis of historical disaster events, extract key indicators of high risk and low risk, and use these indicators to build corresponding risk assessment models. Based on the generated disaster risk level data, the risk level of the disaster status perception data is calculated. This calculation process involves weighted scoring models, fuzzy logic or Bayesian inference, etc., which generate disaster risk level data by comprehensively evaluating different indicators.

[0099] Preferably, step S31 includes the following steps:

[0100] Step S311: Acquire historical underground space humidity data; compare the moisture retention capacity of the historical underground space humidity data with the underground space humidity data, and perform underground area spatial analysis in combination with the geographic information system to generate underground space moisture comparison position data; perform underground space disaster perception on the underground space moisture comparison position data to generate underground area underground space moisture perception data;

[0101] Step S312: Analyze the water level fluctuation amplitude of the groundwater level data to generate groundwater level fluctuation amplitude data; analyze the underground space particle compaction degree of the groundwater level fluctuation amplitude data to generate groundwater level sedimentation particle analysis data; use the underground space moisture comparison position data to perceive the water level disaster state of the groundwater level sedimentation particle analysis data to generate groundwater level disaster perception data;

[0102] Step S313: Perform triaxial shear test on the geotechnical mechanics property data to generate underground space shear strength data; perform consolidation test processing on the geotechnical mechanics property data to generate underground space porosity data; perform geotechnical consolidation load disaster perception on the underground space shear strength data and underground space porosity data to generate underground space geotechnical perception data;

[0103] Step S314: Perform multi-level analysis on the underground space moisture perception data, groundwater level disaster perception data and underground space rock and soil perception data in the underground area to generate disaster status perception data.

[0104] In an embodiment of the present invention, after obtaining the historical underground space humidity data, by comparing the historical data with the real-time underground space humidity data, a statistical analysis method (such as a t-test or variance analysis) can be used to evaluate the moisture retention capacity of the underground space. This comparison process can not only reveal the moisture change trend of the underground space, but also use the geographic information system (GIS) for spatial analysis, and generate underground space moisture comparison position data in combination with geographical factors such as topography and land use. The application of GIS technology makes the visualization and analysis of spatial data more intuitive and efficient, so that the underground area underground space moisture perception data generated can reflect the potential underground space disaster risk. Secondly, when performing land settlement particle analysis on groundwater level data, the particle size analysis method and settlement model are used to deeply understand the impact of groundwater level changes on foundation stability. By analyzing the characteristics of settlement particles, the generated groundwater level settlement particle analysis data can provide strong support for the perception of water level disaster status. In addition, by using the relationship between the moisture comparison position data, the impact of groundwater level changes on underground space humidity can be further evaluated, thereby forming groundwater level disaster perception data. The integration and analysis of such data helps to identify abnormal changes in water levels and their potential disasters. Through triaxial shear tests on geotechnical mechanical properties, the shear strength data of underground space can be obtained, which is crucial for judging the stability and disaster resistance of underground space. At the same time, the results of the consolidation test can generate underground space porosity data, and the combination of the two provides an important basis for underground space geotechnical disaster perception. By comparing and analyzing the relationship between underground space shear strength and porosity, the instability risk of rock and soil can be identified and underground space geotechnical perception data can be generated. The underground area underground space moisture perception data, groundwater level disaster perception data and underground space geotechnical perception data are analyzed at multiple levels, and a comprehensive evaluation method or fuzzy comprehensive evaluation method is used. This method constructs a comprehensive evaluation model and weights data from different sources to generate disaster status perception data.

[0105] Preferably, step S33 includes the following steps:

[0106] Step S331: compare the historical underground space humidity data and the underground space moisture perception data of the underground area to generate the underground space moisture comparison data of the underground area; calculate the underground space humidity risk index for the underground space moisture comparison data of the underground area using the disaster risk level threshold to generate the underground space humidity risk index;

[0107] Step S332: monitor the water level fluctuation amplitude of the groundwater level disaster sensing data to generate underground space water level monitoring data; use the disaster risk level threshold to perform abnormal disaster fluctuation assessment calculation on the underground space water level monitoring data to generate an underground space water level risk index;

[0108] Step S333: using the disaster risk level threshold to evaluate and calculate the collapse risk index of the underground space geotechnical sensing data to generate the underground space geotechnical risk index;

[0109] Step S334: weight the risk index of underground space humidity, the risk index of underground space water level and the risk index of underground space geotechnical risk to generate disaster risk degree data.

[0110] In an embodiment of the present invention, by comparing the historical underground space humidity data and the underground space moisture perception data of the underground area, using statistical methods such as mean comparison, variance analysis or correlation analysis, underground area underground space moisture comparison data can be generated. Then, the underground space humidity risk index is calculated based on the disaster risk level threshold for these comparison data, and different thresholds are combined with actual data using fuzzy logic or weighted average method to obtain the risk index of underground space humidity. This process makes the quantitative assessment of underground space humidity risk more accurate and can reflect the potential risk of moisture imbalance and underground space quality deterioration. The water level fluctuation amplitude of the groundwater level disaster perception data is monitored, and time series analysis methods such as moving average or exponential smoothing are used to capture the trend and fluctuation of water level changes. The generated underground space water level monitoring data provides a basis for subsequent abnormal disaster fluctuation assessment. Subsequently, by comparing the monitoring data with the disaster risk level threshold, the underground space water level monitoring data is analyzed using an outlier detection method (such as Z-score or IQR method) to calculate and generate the underground space water level risk index. The use of this method can not only quantify the risk of water level fluctuations, but also timely identify the abnormal water level caused by it, providing a basis for emergency management. The disaster risk level threshold is used to evaluate the collapse risk index of underground space geotechnical sensing data. Multivariate regression analysis or logistic regression model is used to combine the mechanical properties of geotechnical and environmental factors to calculate the risk index of rock and soil collapse. This process can form a comprehensive risk assessment system by integrating geotechnical properties, humidity and other environmental factors, providing a scientific basis for the stability of underground space. Different risk indices are weighted and merged, and the risk index of underground space humidity, underground space water level risk index and underground space geotechnical risk index are integrated according to preset weights (0.4, 0.4 and 0.2) by weighted average method to generate comprehensive disaster risk level data.

[0111] As an example of the present invention, refer to Figure 4 As shown, in this example, step S4 includes:

[0112] Step S41: Real-time disaster risk determination is performed on the multi-parameter characteristics of underground space through disaster risk degree data to obtain real-time data on underground space disaster risk;

[0113] Step S42: constructing a disaster risk intelligent decision-making model based on the real-time data of underground space disaster risk to generate a disaster risk intelligent decision-making model;

[0114] Step S43: using the underground space multi-parameter feature as input to the disaster risk intelligent decision-making model to perform underground disaster multi-parameter prediction, generate disaster intelligent risk parameter feature prediction data, and perform update timestamp processing to replace the underground space multi-parameter feature;

[0115] Step S44: Make intelligent disaster risk decisions on key disaster information based on the disaster risk intelligent decision-making model to obtain a disaster risk decision report.

[0116] In the embodiment of the present invention, the disaster risk is determined by collecting and analyzing the real-time data of the multi-parameter characteristics of the underground space, and using the disaster risk degree data. The process uses data fusion technology, combines the real-time data and historical data of the sensor, and uses an algorithm model (such as a decision tree, a support vector machine or a neural network) to perform risk assessment on various characteristics of the underground space, thereby generating real-time data of the underground space disaster risk. This real-time data not only reflects the current risk status, but also provides a basis for subsequent decision-making. A disaster risk intelligent decision-making model is constructed based on real-time risk data, and machine learning or deep learning algorithms are applied. The establishment of this model involves feature selection, model training and verification processes, and historical disaster data and real-time risk data are used for training, so that the model can effectively identify and predict the pattern of disaster occurrence. In this way, the generated disaster risk intelligent decision-making model can self-learn and adapt to the changing environment, and improve the accuracy and real-time performance of its decision-making. The multi-parameter characteristics of the underground space are used as input, and the disaster risk intelligent decision-making model is used to perform multi-parameter prediction. This process requires the application of time series analysis and data interpolation technology to ensure the timeliness and continuity of the input data. At the same time, the generated disaster intelligent risk parameter feature prediction data needs to be timestamped so that the original underground space multi-parameter features can be updated and replaced in real time to maintain the accuracy and relevance of the data. Through this process, the model can dynamically respond to environmental changes and potential risks, improving its effectiveness in practical applications. Based on the constructed risk intelligent decision-making model, intelligent decisions are made on key disaster information to generate a disaster risk decision report. At this point, the model comprehensively considers multiple factors, evaluates the relationship between different parameters, and uses weighted decision-making or fuzzy logic methods to form decision recommendations for specific risk scenarios.

[0117] Preferably, step S42 includes the following steps:

[0118] Step S421: using the underground space multi-parameter features as the data input layer of the model to synchronize the data timestamps and generate an underground space model input layer;

[0119] Step S422: constructing a model feature extraction layer for the real-time data of underground space disaster risk to generate an underground space model feature extraction layer;

[0120] Step S423: Use the disaster status perception data as training data to train the underground space model input layer and the underground space model feature extraction layer, thereby generating a disaster risk intelligent decision-making model.

[0121] In the embodiment of the present invention, the multi-parameter features of the underground space are used as the data input layer of the model for timestamp synchronization, which involves data synchronization technology. The core of this process is to ensure the timeliness and consistency of the data. A time series database or a real-time data stream processing platform is used to align the underground space feature data collected at different times to generate a standardized underground space model input layer. Timestamp synchronization not only ensures the consistency of the data in the time dimension, but also provides reliable basic data for subsequent model training. The feature extraction layer of the real-time data of underground space disaster risk is constructed, and the key technologies include feature engineering and data dimensionality reduction. Feature engineering aims to improve the prediction performance of the model by selecting and extracting features related to disaster risks, such as underground space humidity, groundwater level and rock and soil mechanical properties. Common feature extraction techniques include principal component analysis (PCA), linear discriminant analysis (LDA), etc., which can effectively reduce the dimension of the data, extract the most informative features, and form the underground space model feature extraction layer. This layer not only helps to reduce the complexity of the model, but also improves the computational efficiency and accuracy of the model when processing high-dimensional data. The disaster state perception data is used as training data to train the model layer number of the underground space model input layer and feature extraction layer. This process mainly relies on machine learning algorithms, including the multi-layer perceptron (MLP) in deep learning, which is used to establish deep nonlinear mapping relationships. The training process is divided into forward propagation and back propagation. By adjusting the model parameters to minimize the loss function, the model can accurately predict disaster risks.

[0122] Preferably, step S44 includes the following steps:

[0123] Step S441: predicting the real-time data of underground space disaster risk based on the intelligent decision-making model for disaster risk, and generating underground space disaster prediction results, wherein the underground space disaster prediction results include underground space low-risk disaster data, underground space medium-risk disaster data, and underground space high-risk disaster data;

[0124] Step S442: When the underground space disaster prediction result is underground space low risk disaster data, threshold classification analysis is performed on the disaster risk level threshold based on the underground space data, an underground space low risk threshold data analysis report is generated, and a green label is performed, thereby completing the underground space low risk disaster strategy;

[0125] Step S443: When the underground space disaster prediction result is underground space medium risk disaster data, the underground space humidity data is monitored to generate medium risk underground space humidity analysis data; the groundwater level data is marked with abnormal increase, and abnormal underground space water level data is found during the generation; the medium risk underground space humidity analysis data and the underground space water level abnormal data are predicted by exponential smoothing method to generate an underground space medium risk prediction report, thereby completing the underground space medium risk disaster strategy;

[0126] Step S444: When the underground space disaster prediction result is underground space high-risk disaster data, the underground space moisture comparison position data is marked as an emergency outlier to generate underground space high-risk underground space humidity outlier data; the underground space geotechnical perception data is identified as key points, and positioned in combination with GIS to generate underground space high-risk key point data; the underground space high-risk underground space humidity outlier data and underground space high-risk key point data are reported to generate an underground space high-risk disaster report, and the report is sent using the cloud platform, thereby formulating an underground space high-risk disaster strategy;

[0127] Step S445: Merge the underground space low risk threshold data analysis report, the underground space medium risk prediction report and the underground space high risk disaster report to generate a disaster risk decision report.

[0128] In an embodiment of the present invention, the real-time data of disaster risk in underground space is predicted by using a disaster risk intelligent decision-making model. This process relies on machine learning or deep learning algorithms, especially regression analysis and classification models. By analyzing the characteristic variables in the real-time data, such as underground space humidity, groundwater level, etc., low, medium and high risk disaster data of underground space are generated. This multi-level risk classification enables risk management to formulate corresponding response strategies for different risk levels. When the prediction result is low risk, the classification analysis of the threshold of disaster risk level based on underground space data is a key technical means. Specifically, statistical methods such as cluster analysis and decision tree algorithm are used to determine the threshold of risk level and generate a low risk threshold data analysis report. This report not only helps to clarify the low risk state, but also provides intuitive visual instructions through green label marking, promotes rapid decision-making and response, and thus forms a low risk disaster strategy for underground space. Medium risk underground space humidity analysis data is generated by real-time monitoring of underground space humidity data, and abnormal increase of groundwater level is marked. The technical means used here include anomaly detection algorithm and time series analysis, and the data is processed in combination with exponential smoothing method to improve the accuracy of risk prediction. Finally, the generated medium-risk forecast report provides the necessary basis for decision makers and facilitates the formation of medium-risk disaster strategies. The processing of high-risk data is also critical. Emergency outlier marking is performed on the underground space moisture comparison position data, and threshold-based detection technology and GIS (geographic information system) technology are used to identify key points to generate high-risk key point data.

[0129] In this specification, a system for underground space disaster perception and risk intelligent decision-making analysis is provided, which is used to execute the above-mentioned underground space disaster perception and risk intelligent decision-making analysis method. The underground space disaster perception and risk intelligent decision-making analysis system:

[0130] Data acquisition module: Use sensors to collect data from underground space and generate underground space data; pre-process underground space data to obtain standard underground space data;

[0131] Feature extraction module: extract multi-parameter features from standard underground space data to obtain multi-parameter features of underground space; identify key disaster information of underground space based on multi-parameter features of underground space to obtain key disaster information;

[0132] Disaster perception analysis module: conduct disaster status perception analysis on key disaster information to obtain disaster status perception data; conduct disaster risk level threshold assessment on disaster status perception data to generate disaster risk level threshold; calculate the risk level of disaster status perception data based on disaster risk level data to obtain disaster risk level data;

[0133] Risk decision-making module: Real-time disaster risk measurement of multi-parameter characteristics of underground space is carried out through disaster risk degree data to obtain real-time data of underground space disaster risk; a disaster risk intelligent decision-making model is constructed based on the real-time data of underground space disaster risk to generate a disaster risk intelligent decision-making model; intelligent disaster risk decision-making is carried out on key disaster information based on the disaster risk intelligent decision-making model to obtain a disaster risk decision-making report, thereby completing the underground space disaster perception and risk intelligent decision-making tasks.

[0134] The beneficial effect of the present invention is that the underground space data generated by using sensor technology for data collection provides basic information for subsequent analysis. This process ensures the comprehensiveness and real-time nature of the data and lays a solid foundation for the feature extraction of underground space. In the feature extraction stage, through multi-parameter analysis of standardized data, key data reflecting important characteristics of the underground environment can be extracted, including underground space humidity, groundwater level and geotechnical properties, which provide a basis for the identification of disaster information. It enables researchers to conduct in-depth analysis of key disaster information and generate accurate risk assessment standards through risk level threshold assessment. This process not only improves the sensitivity to disaster status, but also provides data support for the calculation of subsequent risk levels. By comprehensively analyzing the disaster status perception data, the system can monitor the risk changes of underground space in real time, generate disaster risk level data, and make the decision-making process more scientific and reliable. Through the construction of real-time disaster risk determination and risk intelligent decision-making model, dynamic management of disaster risk is realized. Based on real-time data input, the model can flexibly respond to various potential risks, provide accurate basis for decision-making of key disaster information, and generate corresponding disaster risk decision reports. This series of steps not only improves the accuracy and efficiency of underground space disaster monitoring, but also provides scientific support for the safety management of underground space. Therefore, the present invention significantly improves the information processing efficiency and assessment accuracy in underground space management through systematic data collection and analysis, and effectively solves the problems of information delay and insufficient assessment due to environmental constraints in traditional methods. This innovation improves disaster perception capabilities, optimizes the timeliness and scientificity of risk decision-making, and provides managers with more accurate decision-making support.

[0135] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0136] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for underground space disaster perception and risk intelligent decision-making analysis, characterized in that: The following steps are involved: Step S1: using sensors to collect data from underground space and generate underground space data; Preprocess the underground space data to obtain standard underground space data; Step S2: extracting multi-parameter features from standard underground space data to obtain multi-parameter features of underground space; Identify key disaster information of underground space based on multi-parameter characteristics of underground space to obtain key disaster information; Step S3: Perform disaster status perception analysis on key disaster information to obtain disaster status perception data; Conduct disaster risk level threshold assessment on disaster status perception data to generate disaster risk level threshold; calculate the risk level of disaster status perception data based on the disaster risk level data to obtain disaster risk level data; Step S4: Real-time disaster risk determination is performed on the multi-parameter characteristics of underground space through disaster risk degree data to obtain real-time data on underground space disaster risk; Construct a disaster risk intelligent decision-making model based on the real-time data of underground space disaster risks to generate a disaster risk intelligent decision-making model; Based on the disaster risk intelligent decision-making model, intelligent disaster risk decision-making is carried out on key disaster information to obtain a disaster risk decision-making report, thereby completing the underground space disaster perception and risk intelligent decision-making tasks.

2. The underground space disaster perception and risk intelligent decision-making analysis method according to claim 1 is characterized in that: Step S1 includes the following steps: Step S11: using sensors to collect data from underground space and generate underground space data; Step S12: performing data cleaning on the underground space data to generate underground space cleaning data; Step S13: performing data standardization processing on the underground space cleaning data to generate standard underground space data.

3. The underground space disaster perception and risk intelligent decision-making analysis method according to claim 1 is characterized in that: Step S2 includes the following steps: Step S21: extracting multi-parameter features from standard underground space data to obtain multi-parameter features of the underground space; Step S22: performing multi-feature combination processing on the underground space multi-parameter features to generate underground space multi-parameter combination data; Step S23: using the underground space multi-parameter combination data to identify underground space key disaster information based on the underground space multi-parameter characteristics, and obtaining key disaster information, wherein the key disaster information includes underground space humidity data, groundwater level data and rock and soil mechanics property data.

4. The underground space disaster perception and risk intelligent decision-making analysis method according to claim 1 is characterized in that: Step S3 includes the following steps: Step S31: performing disaster status perception analysis on key disaster information to obtain disaster status perception data; Step S32: performing time series analysis on the disaster status perception data to generate a disaster status time series analysis result; performing disaster risk level threshold evaluation on the disaster status time series analysis result to generate a disaster risk level threshold; Step S33: Calculate the risk level of the disaster status perception data according to the disaster risk level data to obtain disaster risk level data.

5. The underground space disaster perception and risk intelligent decision-making analysis method according to claim 4 is characterized in that: Step S31 includes the following steps: Step S311: Acquire historical underground space humidity data; compare the moisture retention capacity of the historical underground space humidity data with the underground space humidity data, and perform underground area spatial analysis in combination with the geographic information system to generate underground space moisture comparison position data; perform underground space disaster perception on the underground space moisture comparison position data to generate underground area underground space moisture perception data; Step S312: Analyze the water level fluctuation amplitude of the groundwater level data to generate groundwater level fluctuation amplitude data; analyze the underground space particle compaction degree of the groundwater level fluctuation amplitude data to generate groundwater level sedimentation particle analysis data; use the underground space moisture comparison position data to perceive the water level disaster state of the groundwater level sedimentation particle analysis data to generate groundwater level disaster perception data; Step S313: Perform triaxial shear test on the geotechnical mechanics property data to generate underground space shear strength data; perform consolidation test processing on the geotechnical mechanics property data to generate underground space porosity data; perform geotechnical consolidation load disaster perception on the underground space shear strength data and underground space porosity data to generate underground space geotechnical perception data; Step S314: Perform multi-level analysis on the underground space moisture perception data, groundwater level disaster perception data and underground space rock and soil perception data in the underground area to generate disaster status perception data.

6. The underground space disaster perception and risk intelligent decision-making analysis method according to claim 4 is characterized in that: Step S33 includes the following steps: Step S331: compare the historical underground space humidity data and the underground space moisture perception data of the underground area to generate the underground space moisture comparison data of the underground area; calculate the underground space humidity risk index for the underground space moisture comparison data of the underground area using the disaster risk level threshold to generate the underground space humidity risk index; Step S332: monitor the water level fluctuation amplitude of the groundwater level disaster sensing data to generate underground space water level monitoring data; use the disaster risk level threshold to perform abnormal disaster fluctuation assessment calculation on the underground space water level monitoring data to generate an underground space water level risk index; Step S333: using the disaster risk level threshold to evaluate and calculate the collapse risk index of the underground space geotechnical sensing data to generate the underground space geotechnical risk index; Step S334: weight the risk index of underground space humidity, the risk index of underground space water level and the risk index of underground space geotechnical risk to generate disaster risk degree data.

7. The underground space disaster perception and risk intelligent decision-making analysis method according to claim 1 is characterized in that: Step S4 includes the following steps: Step S41: Real-time disaster risk determination is performed on the multi-parameter characteristics of underground space through disaster risk degree data to obtain real-time data on underground space disaster risk; Step S42: constructing a disaster risk intelligent decision-making model based on the real-time data of underground space disaster risk to generate a disaster risk intelligent decision-making model; Step S43: using the underground space multi-parameter feature as input to the disaster risk intelligent decision-making model to perform underground disaster multi-parameter prediction, generate disaster intelligent risk parameter feature prediction data, and perform update timestamp processing to replace the underground space multi-parameter feature; Step S44: Make intelligent disaster risk decisions on key disaster information based on the disaster risk intelligent decision-making model to obtain a disaster risk decision report.

8. The underground space disaster perception and risk intelligent decision-making analysis method according to claim 7 is characterized in that: Step S42 includes the following steps: Step S421: using the underground space multi-parameter features as the data input layer of the model to synchronize the data timestamps and generate an underground space model input layer; Step S422: constructing a model feature extraction layer for the real-time data of underground space disaster risk to generate an underground space model feature extraction layer; Step S423: Use the disaster status perception data as training data to train the underground space model input layer and the underground space model feature extraction layer, thereby generating a disaster risk intelligent decision-making model.

9. The underground space disaster perception and risk intelligent decision-making analysis method according to claim 7 is characterized in that: Step S44 includes the following steps: Step S441: predicting the real-time data of underground space disaster risk based on the intelligent decision-making model for disaster risk, and generating underground space disaster prediction results, wherein the underground space disaster prediction results include underground space low-risk disaster data, underground space medium-risk disaster data, and underground space high-risk disaster data; Step S442: When the underground space disaster prediction result is underground space low risk disaster data, threshold classification analysis is performed on the disaster risk level threshold based on the underground space data, an underground space low risk threshold data analysis report is generated, and a green label is performed, thereby completing the underground space low risk disaster strategy; Step S443: When the underground space disaster prediction result is underground space medium risk disaster data, the underground space humidity data is monitored to generate medium risk underground space humidity analysis data; the groundwater level data is marked with abnormal increase, and abnormal underground space water level data is found during the generation; the medium risk underground space humidity analysis data and the underground space water level abnormal data are predicted by exponential smoothing method to generate an underground space medium risk prediction report, thereby completing the underground space medium risk disaster strategy; Step S444: When the underground space disaster prediction result is underground space high-risk disaster data, the underground space moisture comparison position data is marked as an emergency outlier to generate underground space high-risk underground space humidity outlier data; the underground space geotechnical perception data is identified as key points, and positioned in combination with GIS to generate underground space high-risk key point data; the underground space high-risk underground space humidity outlier data and underground space high-risk key point data are reported to generate an underground space high-risk disaster report, and the report is sent using the cloud platform, thereby formulating an underground space high-risk disaster strategy; Step S445: Merge the underground space low risk threshold data analysis report, the underground space medium risk prediction report and the underground space high risk disaster report to generate a disaster risk decision report.

10. An underground space disaster perception and risk intelligent decision-making analysis system, characterized in that: Used to execute the underground space disaster perception and risk intelligent decision-making analysis method as claimed in claim 1, the underground space disaster perception and risk intelligent decision-making analysis system comprises: Data acquisition module: Use sensors to collect data from underground space and generate underground space data; pre-process underground space data to obtain standard underground space data; Feature extraction module: extract multi-parameter features from standard underground space data to obtain multi-parameter features of underground space; identify key disaster information of underground space based on multi-parameter features of underground space to obtain key disaster information; Disaster perception analysis module: conduct disaster status perception analysis on key disaster information to obtain disaster status perception data; conduct disaster risk level threshold assessment on disaster status perception data to generate disaster risk level threshold; calculate the risk level of disaster status perception data based on disaster risk level data to obtain disaster risk level data; Risk decision-making module: Real-time disaster risk measurement of multi-parameter characteristics of underground space is carried out through disaster risk degree data to obtain real-time data of underground space disaster risk; a disaster risk intelligent decision-making model is constructed based on the real-time data of underground space disaster risk to generate a disaster risk intelligent decision-making model; intelligent disaster risk decision-making is carried out on key disaster information based on the disaster risk intelligent decision-making model to obtain a disaster risk decision-making report, thereby completing the underground space disaster perception and risk intelligent decision-making tasks.

Images