Method for evaluating vulnerability of concurrent composite disaster system

Abstract

The invention discloses a method for evaluating the vulnerability of a concurrent composite disaster system, and the method comprises the steps: constructing a composite disaster system, and obtaining the vulnerability of the concurrent composite disaster system through the vulnerability of the composite disaster system; the method comprises the following steps: constructing a collapse-landslide vulnerability evaluation index system, performing sensitivity evaluation on collapse disasters and landslide disasters in a concurrent composite disaster system through a deep learning method according to the vulnerability of the concurrent composite disaster system to obtain collapse-landslide sensitivity, and calculating the collapse disasters and the landslide disasters by adopting an entropy weight method to obtain the collapse-landslide sensitivity. The collapse-landslide exposure degree and the collapse-landslide adaptability are obtained; and in combination with the collapse-landslide sensitivity, performing weighted comprehensive evaluation on the collapse-landslide exposure degree and the collapse-landslide adaptability through a TOPSIS method to obtain a vulnerability evaluation result of the concurrent composite disaster system in the collapse-landslide scene. According to the method, the evaluation index weight can be dynamically adjusted to adapt to complex disasters, and the regional evaluation precision and efficiency are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of disaster assessment, and in particular to a concurrent composite disaster system vulnerability assessment method. Background Art

[0002] In recent years, due to the impact of climate change and human activities, natural disasters have occurred frequently, and extreme weather and natural disasters have gradually become important factors causing heavy casualties and property losses. Among them, climate disasters such as storm surges, saltwater tides, floods, and waterlogging, as well as derivative geological disasters such as riverbank collapse, landslides, and ground collapse, often cause more serious losses through mutual influence and risk transmission to form complex compound disasters. Therefore, it is of great significance for disaster prevention and mitigation to explore the superposition mechanism of compound disaster vulnerability and evaluate the vulnerability of compound disaster systems.

[0003] In the field of vulnerability assessment, the focus was initially on the study of a single type of disaster, including the impact of natural disasters on communities, cities, and buildings. At present, existing literature has evaluated the vulnerability of heavy rains in Liaoning Province, China using a random forest model based on the exposure-sensitivity-adaptability framework. In addition, based on the regional disaster system theory, the social vulnerability and exposure of the population and economy in typhoon disasters were comprehensively considered to assess the typhoon risk in Guangdong Province. In addition, the hazard factors of storm surge disasters, the vulnerability of disaster-bearing bodies, and the stability of the disaster-prone environment were deeply explored, and the response process of storm surge disasters was analyzed. However, this single disaster type assessment failed to fully consider the complexity of the combined effects of multiple disasters. With the in-depth discussion of the concept of vulnerability, scholars began to realize the importance of the combined impact of multiple disasters to systematic analysis. For example, through an in-depth study of the mechanism of the typhoon-flood disaster chain, a vulnerability curve model of house collapse in typhoon disasters in five provinces along the southeast coast was fitted based on historical disaster data. Based on the regional disaster system, the economic loss vulnerability in rainfall-landslides was assessed using random forests, decision trees, and XGBoost algorithms. Indicators were selected from four aspects: danger of extreme precipitation-geological disasters, exposure, vulnerability, and disaster prevention and mitigation capabilities, and the risks in Shuicheng County, Guizhou Province were assessed.

[0004] In summary, in the context of specific compound disasters, multi-factor comprehensive assessment is crucial for vulnerability analysis. However, the existing vulnerability assessment for compound disasters often directly constructs a vulnerability assessment index system covering multiple disasters in compound disasters, which fails to fully consider the superposition (reduction) effect of vulnerability in different dimensions in compound disasters. Summary of the invention

[0005] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0006] In view of the above existing problems, the present invention is proposed. Therefore, the present invention provides a concurrent composite disaster system vulnerability assessment method to solve the problems raised in the background technology.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions: a concurrent composite disaster system vulnerability assessment method, comprising:

[0008] Obtain disaster data, build a composite disaster system, and consider the superposition effect of composite disasters in the composite disaster system. Through the composite disaster system vulnerability, the vulnerability of the concurrent composite disaster system is obtained.

[0009] A collapse-landslide vulnerability assessment index system is constructed. According to the vulnerability of the concurrent composite disaster system, the collapse disaster and landslide disaster in the concurrent composite disaster system are evaluated for sensitivity by a deep learning method to obtain collapse-landslide sensitivity. At the same time, the collapse disaster and landslide disaster are calculated by an entropy weight method to obtain collapse-landslide exposure and collapse-landslide adaptability, respectively.

[0010] Combined with the collapse-landslide sensitivity, the collapse-landslide exposure and collapse-landslide adaptability were weighted and comprehensively evaluated using the TOPSIS method to obtain the vulnerability assessment results of the concurrent composite disaster system under the collapse-landslide scenario.

[0011] As a preferred solution of the concurrent composite disaster system vulnerability assessment method of the present invention, the step of obtaining disaster data and constructing a composite disaster system includes:

[0012] Set the compound disaster system as concurrent compound disaster C n , the concurrent compound disaster C n It is expressed as:

[0013]

[0014] Among them, A1, A2, …, A n They are independent disasters that occur simultaneously.

[0015] As a preferred solution of the concurrent composite disaster system vulnerability assessment method of the present invention, the method of considering the superposition effect of composite disasters in the composite disaster system includes:

[0016] Set composite disaster C n The superposition effect coefficient of disaster losses is gC n , gC n Indicates compound disaster C n The disaster loss L n Relative to disasters A1, A2, …, A n The cumulative effect of the sum of disaster losses.

[0017] As a preferred solution of the concurrent composite disaster system vulnerability assessment method of the present invention, the composite disaster system vulnerability includes:

[0018] It is obtained based on the calculation results of sensitivity, exposure and adaptability of the composite disaster system.

[0019] As a preferred solution of the concurrent compound disaster system vulnerability assessment method of the present invention, the vulnerability of the concurrent compound disaster system includes:

[0020] According to the regional disaster theory, the vulnerability of the concurrent composite disaster system is calculated and expressed as:

[0021]

[0022] Among them, the exposure degree E Cn , Sensitivity S Cn , Adaptability Cn The results were normalized and The weights of exposure, sensitivity and adaptability are represented by W e , W s , W f express.

[0023] As a preferred solution of the concurrent composite disaster system vulnerability assessment method of the present invention, the sensitivity assessment of collapse disasters and landslide disasters in the concurrent composite disaster system is performed by a deep learning method, including:

[0024] Based on the AlexNet-2 model, the data of the input layer is converted into a two-dimensional array, and the evaluation index is extracted from the disaster data. After the evaluation index is superimposed, it is input into the CNN model to form a multi-band 2D image;

[0025] The 2D image is the input image, and the image of each pixel after superposition is the original data. The layers of each evaluation index are overlapped by using the layerstacking function of the ENVI software, and divided into 12×12 multi-band grid images with the disaster point as the center, and then input into the CNN model for forecasting.

[0026] As a preferred solution of the concurrent composite disaster system vulnerability assessment method of the present invention, the AlexNet-2 model includes:

[0027] Place two neurons in the output layer classifier to indicate whether a disaster has occurred;

[0028] The historical disaster data set is randomly divided into three parts, 60% of which are used as training set learning data and 20% as validation set data for learning. The trained model is recorded and saved, and the remaining 20% ​​of the disaster point data is used as test set data.

[0029] As a preferred solution of the concurrent composite disaster system vulnerability assessment method of the present invention, the entropy weight method is used to calculate the collapse disaster and landslide disaster, including:

[0030] The entropy weight method calculation process is as follows:

[0031] Construct a judgment matrix R with m evaluation objects and n evaluation indicators x;

[0032] Convert all indicators into very large indicators, and use the reciprocal method to make the very small indicators positive;

[0033] Standardize each indicator and obtain the standardized matrix for the non-existent negative data;

[0034] By normalizing the judgment matrix R, we get the judgment matrix B;

[0035] According to the definition of entropy, there are m evaluation objects and n evaluation indicators, and the entropy of the evaluation indicators is determined;

[0036] Calculate the entropy weight W of the evaluation index.

[0037] As a preferred solution of the concurrent composite disaster system vulnerability assessment method of the present invention, the TOPSIS method is used to perform a weighted comprehensive evaluation on the collapse-landslide exposure and collapse-landslide adaptability, including:

[0038] There are m targets and n characteristic quantities in the set indicator evaluation system. The calculation process using the TOPSIS method is as follows:

[0039] Each feature x ij (i=1,2,…,m;j=1,2,…,n) and the original data of positive and negative reference points form the initial judgment matrix X, and normalize it to obtain the matrix Y;

[0040] Combined with the weights of each evaluation index obtained by the entropy weight method, the weighted judgment matrix Z is obtained. Based on Z, the positive and negative ideal values ​​of the evaluated system are obtained. and By Euclidean distance and Calculate the relative closeness between each element in Z and the positive and negative ideal values, and convert the closeness T i Normalize and get a comprehensive score.

[0041] Compared with the prior art, the invention has the following beneficial effects:

[0042] 1. By combining the deep learning CNN model and the entropy weight method, the present invention can perform sensitivity analysis and weight allocation on different disaster types in a concurrent composite disaster system, significantly improving the accuracy of vulnerability assessment. In particular, in the assessment of composite disasters such as collapse and landslide, a multi-dimensional assessment index system is adopted to make the vulnerability assessment results more comprehensive and accurate.

[0043] 2. Compared with the traditional single disaster type assessment method, by constructing a composite disaster system, the superposition effect and complex joint action of multiple disasters are taken into account, which can more truly reflect the vulnerability of the disaster system. At the same time, based on the TOPSIS method, a weighted comprehensive evaluation is performed, so that the assessment results can reflect the relative importance of various disasters in different scenarios;

[0044] 3. When evaluating complex disasters such as collapse and landslides, the entropy weight method and deep learning technology can be used to adjust the weights and adaptability of various evaluation indicators in real time, which can not only reflect the current state of the disaster system, but also make dynamic adjustments according to environmental changes, thereby improving the system's ability to respond to complex disaster situations;

[0045] 4. Construct a composite disaster system model, which is applicable to the scenario where multiple disasters occur simultaneously. It can be applied not only to collapse and landslide scenarios, but also to composite scenario assessments of other disaster types, such as combined disasters of floods, typhoons and earthquakes. Its versatility and adaptability greatly expand the scope of application of the solution of the present invention;

[0046] 5. By introducing the CNN model and TOPSIS method, automatic analysis and evaluation of disaster vulnerability can be achieved, which reduces manual intervention and improves evaluation efficiency. At the same time, the standardized data processing and analysis process is adopted to facilitate disaster management departments to quickly obtain and process evaluation results, thereby achieving rapid response and decision-making support for disaster risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0048] Figure 1 This is an overall flow chart of a concurrent composite disaster system vulnerability assessment method according to an embodiment of the present invention;

[0049] Figure 2 This is an example diagram of the interaction mechanism of compound disasters in a concurrent compound disaster system vulnerability assessment method according to an embodiment of the present invention;

[0050] Figure 3 A flow chart of collapse-landslide vulnerability assessment of a concurrent composite disaster system vulnerability assessment method according to an embodiment of the present invention;

[0051] Figure 4 A schematic diagram of factor layer superposition of a concurrent composite disaster system vulnerability assessment method according to an embodiment of the present invention;

[0052] Figure 5 A collapse-landslide vulnerability zoning map of the Guangdong-Hong Kong-Macao Greater Bay Area according to a concurrent composite disaster system vulnerability assessment method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0053] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0054] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0055] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0056] The present invention is described in detail with reference to schematic diagrams. When describing the embodiments of the present invention, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0057] At the same time, in the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "upper, lower, inner and outer" are based on the directions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0058] In the present invention, unless otherwise clearly specified and limited, the terms "install, connect, connect" should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0059] Example 1

[0060] Reference Figures 1 to 4 , which is the first embodiment of the present invention, and provides a concurrent composite disaster system vulnerability assessment method, comprising:

[0061] S1. Obtain disaster data, build a composite disaster system, and consider the superposition effect of composite disasters in the composite disaster system. Through the composite disaster system vulnerability, the vulnerability of the concurrent composite disaster system is obtained;

[0062] Furthermore, the composite disaster system is set as concurrent composite disaster C n , concurrent compound disaster C n It is expressed as:

[0063]

[0064] Among them, A1, A2, …, A n These are independent disasters that occur simultaneously;

[0065] Specifically, the interaction mechanism between compound disasters can be referred to Figure 2 ;

[0066] Furthermore, let us assume that the composite disaster C n The superposition effect coefficient of disaster losses is g Cn , g Cn Indicates compound disaster C n The disaster loss L n Relative to disasters A1, A2, …, A n The superposition effect of the sum of the disaster losses is expressed as follows:

[0067]

[0068] Furthermore, the vulnerability of the composite disaster system is obtained based on the calculation results of the composite disaster system sensitivity, exposure, and adaptability, as shown below:

[0069] V Cn =f(E Cn ,S Cn ,F Cn )

[0070] Specifically, sensitivity, also known as susceptibility, refers to the sensitivity of the comprehensive external environment of the area threatened by compound disasters to disasters or damages. Under the same disaster-causing factors, the higher the sensitivity (susceptibility), the greater the probability of compound disasters, and the more serious the impact of the disasters.

[0071] Specifically, exposure refers to the composition and distribution of human social systems under the influence of natural disasters, including the number, composition, value and distribution of disaster-prone objects such as population, property and engineering facilities. Only disaster-prone objects exposed to natural disasters are likely to suffer losses. Exposure is the direct cause of disaster risk. It reflects the degree to which the system encounters disasters or dangers and determines the size of the potential losses of the system under the influence of disaster events.

[0072] Specifically, adaptability refers to the system's ability to respond to disaster events and recover from disaster losses, which mainly depends on social wealth, technology, education, information, skills, infrastructure, stability and management capabilities. For example, measures such as setting up disaster shelters and building embankments have improved the region's adaptability to disasters, which is conducive to reducing losses caused by disasters.

[0073] Furthermore, according to the regional disaster theory, the vulnerability of the concurrent composite disaster system is calculated and expressed as:

[0074]

[0075] Among them, the exposure E Cn , Sensitivity S Cn , Adaptability Cn The results were normalized and The weights of exposure, sensitivity and adaptability are represented by W e , W s , W f express;

[0076] Furthermore, the sensitivity, exposure and adaptability of concurrent composite disaster systems are theoretically derived;

[0077] Furthermore, the sensitivity derivation process of concurrent composite disaster system is as follows:

[0078] Use S Cn Indicates compound disaster C n Sensitivity (probability of occurrence), S Aj Indicates disaster A j sensitivity;

[0079] Assume that in a concurrent compound disaster, disaster A j The occurrence of disasters is independent of each other. j The sensitivity of the disaster is not affected by the occurrence of other disasters in the complex disaster;

[0080] Compound disaster C n The occurrence of disasters A1, A2, …, A n Therefore, the compound disaster C n The sensitivity of disasters A1, A2, …, A n Sensitivity A1 ,S A2 ,…,S An It is deduced that:

[0081]

[0082] For the special case of two disasters constituting a composite disaster, the sensitivity of the composite disaster C2 is expressed as:

[0083] S C2 =S A1 ×S A2

[0084] Among them, S A1 , S A2 They represent the sensitivity to disasters A1 and A2 respectively;

[0085] Furthermore, the exposure derivation process of concurrent composite disaster system is as follows:

[0086] Use E Cn Indicates compound disaster C n Exposure (potential disaster losses), E Aj Indicates disaster A j exposure;

[0087] Compound disaster C n The exposure degree can be expressed by disasters A1, A2, …, A n Exposure S A1 ,S A2 ,…,S An It is deduced that:

[0088]

[0089] For the special case of two disasters forming a composite disaster, the exposure of the composite disaster C2 is expressed as:

[0090] E C2 =g C2 (E A1 +E A2 )

[0091] Among them, E A1 、E A2 They represent the exposure to single hazards A1 and A2 respectively;

[0092] Furthermore, the fitness derivation process of the concurrent composite disaster system is as follows:

[0093] Use F Cn Indicates that the region faces a compound disaster C n Adaptability (coping ability), with F Aj Indicates that the area is facing disaster A j adaptability;

[0094] Assume that the area faces disaster A j Adaptability F Aj Indicated as regional response disaster A j Capacity Aj and regional disasters j Emergency demand R Aj Ratio between:

[0095] F Aj =P Aj / R Aj

[0096] Similarly, the region faces a complex disaster C n The adaptability is expressed as the regional response to compound disasters C n Capacity Cn and regional compound disasters n Emergency demand R Cn Ratio between:

[0097] F Cn =P Cn / P Cn

[0098] Assume that the regional ability to cope with various disasters is consistent and universal, that is: P Aj =P Cn =P, regional composite disaster C n Emergency demand R Cn The superposition (reduction) effect coefficient of the disaster loss is consistent with the superposition (reduction) effect coefficient of the disaster loss. n Adaptability F A1 ,FA2 ,…,F An It is deduced that:

[0099]

[0100] For the special case of two disasters constituting a composite disaster, the adaptability of the region to the composite disaster C2 is expressed as:

[0101] F C2 =F A1 ×F A2 / (F A1 +F A2 ) C2

[0102] Among them, F A1 、F A2 They represent the adaptability to disasters A1 and A2 respectively;

[0103] S2. Construct a collapse-landslide vulnerability assessment index system. According to the vulnerability of the concurrent composite disaster system, the collapse disaster and landslide disaster in the concurrent composite disaster system are evaluated by deep learning method to obtain collapse-landslide sensitivity. At the same time, the entropy weight method is used to calculate the collapse disaster and landslide disaster to obtain collapse-landslide exposure and collapse-landslide adaptability respectively.

[0104] Specifically, the collapse-landslide vulnerability assessment index system is constructed as a reference Figure 3 ;

[0105] It should be explained that the Convolutional Neural Network simulates the structure and function of biological neural networks and is one of the most efficient methods in the current field of image recognition. Unlike traditional neural networks, the input data of CNN is usually multi-dimensional image data, such as two-dimensional image data and image data of multiple dimensions such as RGB color channels. CNN can automatically extract features from the original data, thus avoiding the repeated manual feature extraction work. At the same time, the hierarchical structure of CNN makes it easier to process and explain images intuitively.

[0106] Furthermore, the deep learning method uses the CNN evaluation method as follows:

[0107] Evaluation factor superposition: Based on the AlexNet-2 model, the data of the input layer is converted into a two-dimensional array, and the evaluation index is extracted from the disaster data. After the evaluation index is superimposed, it is input into the CNN model to form a multi-band 2D image, such as Figure 4 As shown;

[0108] It should be noted that when processing disaster data, the neighboring areas of the pixels need to be taken into account, so the data of the input layer must be converted into a two-dimensional array;

[0109] Selection of evaluation unit: The 2D image is used as the input image, and the image of each pixel after superposition is used as the original data. The layer stacking function of ENVI software is used to overlap the layers of each evaluation index, and the disaster point is taken as the center to split it into a 12×12 multi-band raster map, which is then input into the CNN model for forecasting;

[0110] It should be noted that the disaster sensitivity assessment model is established using convolutional neural networks. Due to the large number of network layers, there are certain requirements on the matrix size when inputting data, so regular grid elements are used for evaluation;

[0111] AlexNet-2 model construction: Place two neurons in the output layer classifier to indicate whether a disaster has occurred; randomly divide the historical disaster data set into three parts, take 60% of them as training set learning data and 20% as validation set data for learning, and record and save the trained model, and the remaining 20% ​​of the disaster point data is used as the test set data;

[0112] It should be noted that the CNN-based sensitivity assessment is a binary classification problem for landslide and collapse hazards;

[0113] Furthermore, the entropy weight method is used to calculate the collapse and landslide hazards, and the process is as follows:

[0114] Construct a judgment matrix R with m evaluation objects and n evaluation indicators x:

[0115] R=(x ij ), (i=1,2,…,n; j=1,2,…,m)

[0116] Convert all indicators into very large indicators, and use the reciprocal method to make the very small indicators positive:

[0117]

[0118] Standardize each indicator and get the standardized matrix for the non-existent negative data:

[0119]

[0120] By normalizing the judgment matrix R, we get the judgment matrix B:

[0121]

[0122] Among them, x max 、xmin Respectively represent the maximum and minimum values ​​of satisfaction of each goal;

[0123] According to the definition of entropy, there are m evaluation objects and n evaluation indicators, and the entropy of the evaluation indicators is determined as follows:

[0124]

[0125]

[0126] Calculate the entropy weight W of the evaluation index:

[0127] W=(ω i ) 1×n

[0128]

[0129] Among them, ω is the weight;

[0130] S3. Combined with collapse-landslide sensitivity, the collapse-landslide exposure and collapse-landslide adaptability were weighted and comprehensively evaluated by TOPSIS method to obtain the vulnerability assessment results of concurrent composite disaster system under collapse-landslide scenario;

[0131] Furthermore, it is assumed that there are m targets and n characteristic quantities in the indicator evaluation system, and the calculation process using the TOPSIS method is as follows:

[0132] Each feature x ij (i=1,2,…,m;j=1,2,…,n) and the original data of positive and negative reference points form the initial judgment matrix X, and normalize it to obtain the matrix Y;

[0133] Combined with the weights of each evaluation index obtained by the entropy weight method, the weighted judgment matrix Z is obtained. Based on Z, the positive and negative ideal values ​​of the evaluated system are obtained. and By Euclidean distance and Calculate the relative closeness between each element in Z and the positive and negative ideal values, and convert the closeness T i Normalize and get the comprehensive score:

[0134]

[0135] It should be explained that the TOPSI method, also known as the approximate ideal solution method, is a general comprehensive evaluation method suitable for the comparative selection of multiple feature quantities and multiple targets. That is, the Euclidean distance between each feature quantity and the positive and negative ideal values ​​is used to obtain a degree of closeness, and then the state interval is divided according to the size of this degree of closeness, so as to evaluate the advantages and disadvantages of each feature quantity.

[0136] Example 2

[0137] Reference Figure 5 , which is the second embodiment of the present invention, and which provides a concurrent composite disaster system vulnerability assessment method, comprising: verifying the collapse-landslide sensitivity assessment involving CNN in the scheme of the present invention through simulation experiments;

[0138] First, the collapse-landslide vulnerability assessment index system constructed in the scheme of the present invention is described as follows:

[0139] The Guangdong-Hong Kong-Macao Greater Bay Area is located in southern China, facing the South China Sea to the south and Southeast Asia across the sea. It has outstanding location advantages. It is not only an important starting point of the Maritime Silk Road, but also an important hub for shipping in the Pacific and Indian Oceans. It is one of the regions with the highest economic development level, urbanization development level and population in China. It is also a region where natural disasters such as heavy rains, storm surges, salt tides, floods, and derivative geological disasters such as collapses and landslides frequently occur and trigger each other, forming a complex disaster network. Therefore, the region is very likely to form major urban public events, triggering chain reactions in a large area, and exposing nearby facilities and residents to safety risks. Therefore, the present invention selects the Guangdong-Hong Kong-Macao Greater Bay Area as the research area to conduct collapse-landslide vulnerability assessment;

[0140] The research data is mainly divided into three dimensions. Among them, the exposure and adaptability index data such as per capita disposable income and number of medical beds are mainly derived from the "Statistical Yearbook 2021", and the exposure and adaptability indicators such as population and gender ratio are mainly derived from the "Population Census 2021". The sensitivity index data such as river network distribution, elevation, fault zone distribution, and disaster point distribution data are mainly derived from platforms such as the Resource and Environmental Science and Data Center of the Chinese Academy of Sciences and the website of the China Meteorological Administration. The sensitivity index data such as surface undulation, distance from the fault zone, and river network density are calculated based on the existing data using the statistical analysis function of ArcGIS;

[0141] The collapse-landslide vulnerability assessment index system was constructed from three dimensions: sensitivity, exposure, and adaptability, as shown in Table 1;

[0142] Table 1 Collapse-landslide vulnerability evaluation index system

[0143]

[0144]

[0145] Among them, sensitivity is mainly related to the local natural geographical environment. Therefore, the present invention considers the natural geographical conditions that may breed disasters, and determines the sensitivity index based on the five criteria of hydrology, geology, topography, vegetation, and land use; among them, surface roughness and land use indicators are landslide sensitivity indicators, and the remaining indicators are sensitivity indicators related to landslide and collapse disasters; exposure reflects the potential losses that may be caused by the human social system under the action of natural disasters, so the present invention determines the exposure index based on the three criteria of population, economy, and society; adaptability reflects the ability to respond to disaster events and the ability to recover from disaster losses, so the present invention determines the adaptability evaluation index based on the two criteria of disaster prevention ability and adaptability; exposure and adaptability indicators should theoretically vary depending on the type of disaster, but the present invention considers the quantifiability and availability of indicator data, and selects the same indicators for the two disasters;

[0146] Secondly, the CNN-based collapse-landslide susceptibility assessment simulation experiment is as follows:

[0147] In the Guangdong-Hong Kong-Macao Greater Bay Area, the collected disaster point data include 939 landslides and 1330 collapses as of 2019. For the selection of non-disaster points, GIS technology is used to randomly generate the same number of non-disaster points 5km away from the known disaster points. On this basis, the data are divided into 60%, 20%, and 20% as training sets, validation sets, and test sets, respectively. Then, a large number of experiments are carried out in the training process of the convolutional neural network, and finally the best parameter combination of this experiment is obtained as shown in Table 2. The final calculated accuracy of the collapse test set is 86.75%, and the accuracy of the landslide test set is 89.1%;

[0148] Table 2 CNN model parameter settings

[0149] parameter Landslide hazard parameter values Collapse hazard parameter values Input Image 12×12×14 12×12×12 Convolution Kernel 3×3 3×3 Max Pooling 2×2 2×2 Loss Function Cross-entropy Cross-entropy Activation Function RELU RELU Optimizer SGD SGD Iterations 80 65 Iteration batch 32 32 Learning Rate 0.01 0.01

[0150] The results of the collapse-landslide vulnerability assessment of the Guangdong-Hong Kong-Macao Greater Bay Area are obtained through Table 2; Figure 5 , the collapse-landslide high vulnerability areas and sub-high vulnerability areas in the Guangdong-Hong Kong-Macao Greater Bay Area mainly present the following characteristics: concentrated in the central areas of the Bay Area, such as Yuexiu District and Tianhe District, and the central and southern areas of the Bay Area, such as Nanshan District, Futian District, and Luohu District; scattered in the northern area of ​​the Bay Area, as well as the central and eastern regions; low vulnerability areas and sub-low vulnerability areas are concentrated in the southwestern areas of the Bay Area, such as Enping City, Taishan City, Kaiping City, Xinhui City, and the central and central-western regions, such as Dinghu District, Gaoming District, Shunde District, Sanshui District, Nanhai District, Panyu District, and Nansha District; among them, the average vulnerability levels of Yuexiu District, Futian District, and Tianhe District are among the top ten in the Bay Area, and are at a high vulnerability level, and it is necessary to give priority to strengthening the disaster resistance of infrastructure and strengthening disaster management measures;

[0151] Finally, through the analysis of the scheme of the present invention, the collapse-landslide system vulnerability assessment results can be accurately obtained, which provides a reference value for the disaster prevention and mitigation capacity building in specific areas.

[0152] Those skilled in the art will appreciate that the embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the application can adopt the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program codes. The scheme in the embodiments of the present application can be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.

[0153] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0154] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0155] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0156] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0157] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A concurrent composite disaster system vulnerability assessment method, characterized in that: include: Obtain disaster data, build a composite disaster system, and consider the superposition effect of composite disasters in the composite disaster system. Through the composite disaster system vulnerability, the vulnerability of the concurrent composite disaster system is obtained. A collapse-landslide vulnerability assessment index system is constructed. According to the vulnerability of the concurrent composite disaster system, the collapse disaster and landslide disaster in the concurrent composite disaster system are evaluated for sensitivity by a deep learning method to obtain collapse-landslide sensitivity. At the same time, the collapse disaster and landslide disaster are calculated by an entropy weight method to obtain collapse-landslide exposure and collapse-landslide adaptability, respectively. Combined with the collapse-landslide sensitivity, the collapse-landslide exposure and collapse-landslide adaptability were weighted and comprehensively evaluated using the TOPSIS method to obtain the vulnerability assessment results of the concurrent composite disaster system under the collapse-landslide scenario.

2. The concurrent composite disaster system vulnerability assessment method according to claim 1, characterized in that: The method of obtaining disaster data and constructing a composite disaster system includes: Set the compound disaster system as concurrent compound disaster C n , the concurrent compound disaster C n It is expressed as: Among them, A1, A2, …, A n They are independent disasters that occur simultaneously.

3. The concurrent composite disaster system vulnerability assessment method according to claim 2, characterized in that: The superposition effect of compound disasters in the compound disaster system includes: Set composite disaster C n The superposition effect coefficient of disaster losses is g Cn , g Cn Indicates compound disaster C n The disaster loss L n Relative to disasters A1, A2, …, A n The cumulative effect of the sum of disaster losses.

4. The concurrent composite disaster system vulnerability assessment method according to claim 1, characterized in that: The vulnerability of the complex disaster system includes: It is obtained based on the calculation results of sensitivity, exposure and adaptability of the composite disaster system.

5. The concurrent composite disaster system vulnerability assessment method according to claim 1, characterized in that: The vulnerability of the concurrent composite disaster system includes: According to the regional disaster theory, the vulnerability of the concurrent composite disaster system is calculated and expressed as: Among them, the exposure E Cn , Sensitivity S Cn , Adaptability Cn The results were normalized and The weights of exposure, sensitivity and adaptability are represented by W e , W s , W f express.

6. The concurrent composite disaster system vulnerability assessment method according to claim 5, characterized in that: The sensitivity assessment of collapse and landslide hazards in concurrent composite hazard systems is carried out through deep learning methods, including: Based on the AlexNet-2 model, the data of the input layer is converted into a two-dimensional array, and the evaluation index is extracted from the disaster data. After the evaluation index is superimposed, it is input into the CNN model to form a multi-band 2D image; The 2D image is the input image, and the image of each pixel after superposition is the original data. The layers of each evaluation index are overlapped by using the layerstacking function of the ENVI software, and divided into 12×12 multi-band grid images with the disaster point as the center, and then input into the CNN model for forecasting.

7. The concurrent composite disaster system vulnerability assessment method according to claim 6, characterized in that: The AlexNet-2 model includes: Place two neurons in the output layer classifier to indicate whether a disaster has occurred; The historical disaster data set is randomly divided into three parts, 60% of which are used as training set learning data and 20% as validation set data for learning. The trained model is recorded and saved, and the remaining 20% ​​of the disaster point data is used as test set data.

8. The concurrent composite disaster system vulnerability assessment method according to claim 1, characterized in that: The entropy weight method is used to calculate collapse hazards and landslide hazards, including: The entropy weight method calculation process is as follows: Construct a judgment matrix R with m evaluation objects and n evaluation indicators x; Convert all indicators into very large indicators, and use the reciprocal method to make the very small indicators positive; Standardize each indicator and obtain the standardized matrix for the non-existent negative data; By normalizing the judgment matrix R, we get the judgment matrix B; According to the definition of entropy, there are m evaluation objects and n evaluation indicators, and the entropy of the evaluation indicators is determined; Calculate the entropy weight W of the evaluation index.

9. The concurrent composite disaster system vulnerability assessment method according to claim 8, characterized in that: The collapse-landslide exposure and collapse-landslide adaptability were evaluated by weighted comprehensive evaluation using the TOPSIS method, including: There are m targets and n characteristic quantities in the set indicator evaluation system. The calculation process using the TOPSIS method is as follows: Each feature x ij (i=1,2,…,m;j=1,2,…,n) and the original data of positive and negative reference points form the initial judgment matrix X, and normalize it to obtain the matrix Y; Combined with the weights of each evaluation index obtained by the entropy weight method, the weighted judgment matrix Z is obtained. Based on Z, the positive and negative ideal values ​​of the evaluated system are obtained. and By Euclidean distance and Calculate the relative closeness between each element in Z and the positive and negative ideal values, and convert the closeness T i Normalize and get a comprehensive score.

Images