Artificial intelligence-based climate risk prevention and control method and device

Abstract

The application discloses an artificial intelligence-based climate risk prevention and control method and device. The method comprises the following steps: case learning and simulation generation: after the case library is cleaned and standardized, the case technical features are extracted, the feature vectors are generated, and the key features are obtained through dimension reduction processing; a simulation case library is generated based on the key features; technical requirements are extracted and verified: the case technical features are classified into four categories of risk assessment, early warning, prevention and control, and resilience improvement; a technical requirement matrix is constructed to form the association between the climate risk types and the technical requirements and to verify the association; a technical system is constructed: all technical requirements are clustered; the association between different technical requirements is mined; a multi-level technical system covering risk assessment, early warning, prevention and control and resilience improvement is generated; and the multi-level technical system is used for climate risk prevention and control. The application solves the technical problem that the climate risk prevention and control is not comprehensive in the related art.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of climate risk prevention and control, in particular, to a climate risk prevention and control method and device based on artificial intelligence. BACKGROUND

[0002] This section is intended to provide background or context to the content recited in the claims or specification and is not admitted to be prior art merely by inclusion in this section.

[0003] Global climate change has significantly increased the frequency and intensity of extreme weather events, posing unprecedented threats to waterway transportation infrastructure such as ports, channels, bridges, etc. These infrastructures are important components of regional economy and transportation networks, but at the same time, they are vulnerable to climate risks such as storm surges, floods, heavy rains, extreme high or low temperatures, etc. These climate events can cause damage to facilities, disrupt operations, and result in significant economic losses, posing serious challenges to the planning, construction and operation of infrastructure. Therefore, climate risk prevention and control of waterway transportation infrastructure has become a key area of global concern.

[0004] In order to effectively respond to complex and changing climate risks, it is urgent to establish a comprehensive technical system covering the whole process from risk assessment, early warning, prevention and control to resilience improvement. However, the existing technology still has the following outstanding problems in practical application:

[0005] 1) Technology is scattered and lacks systematic integration: current research and practice of climate risk prevention and control technology focuses on single technology or specific application scenarios, for example, risk assessment for floods or early warning technology for storm surges is relatively mature, but it has not formed a systematic technical framework, which is difficult to fully cover the needs of waterway transportation infrastructure in response to various climate risks;

[0006] 2) Lack of technology demand classification, difficult to fully respond to complex climate events: the demand of waterway transportation infrastructure in different regions and climate scenarios is significantly different, different risk assessment, early warning and prevention and control technologies are needed for typhoon, flood, rainstorm or extreme temperature, however, the existing method is difficult to fully identify and classify these complex and diverse technology needs, resulting in blind spots in technology development and application;

[0007] 3) Lack of intelligent methods with strong prediction ability: existing risk management methods rely mainly on experts' experience summary or limited case analysis, lacking intelligent analysis tools based on large-scale data. This method is not only inefficient, but also has obvious shortcomings in prediction ability and response effect when dealing with new or complex climate risks.

[0008] In view of the above problems, no effective solution has been proposed so far. SUMMARY

[0009] The embodiment of the application provides a climate risk prevention and control method and device based on artificial intelligence, and the scheme is based on machine learning, simulation case generation and clustering analysis technology, and a climate risk prevention and control technology system for waterway transportation infrastructure is constructed. The scheme learns existing climate risk prevention and control technology cases, extracts and classifies technical requirements, and finally generates a systematic technical framework, providing scientific support for climate risk prevention and control of waterway transportation infrastructure, to at least solve the technical problem that climate risk prevention and control is not comprehensive in related technologies.

[0010] According to an aspect of an embodiment of the application, a climate risk prevention and control method based on artificial intelligence is provided, including: case learning and simulation generation: data cleaning and standardization processing are performed on a case library of climate risk prevention and control technology, case technical features are extracted from the processed case library by using natural language processing technology, a case technical feature vector of the case technical features is generated by using a machine learning model, and key features are obtained by dimension reduction processing on the case technical feature vector; a generative adversarial network or a variational autoencoder is used to generate a simulation case library based on the key features, thereby expanding the coverage range of case data; extracting and verifying technical requirements: the case technical features in the simulation case library are classified into four categories of risk assessment, early warning, prevention and control, and resilience improvement based on a classification model; a technology demand matrix is constructed by combining an unsupervised learning algorithm, forming the association between climate risk types and technical requirements; the technical requirements in the technology demand matrix are verified; constructing a technology system: clustering all technical requirements based on a clustering analysis algorithm to form clustering analysis results; using association rule mining technology to analyze the association between different technical requirements to form association analysis results; generating a multi-level technology system covering risk assessment, early warning, prevention and control and resilience improvement according to the clustering analysis results and the association analysis results; using the multi-level technology system for climate risk prevention and control.

[0011] Optionally, the case technical features are extracted from the processed case library by using natural language processing technology, including one of the following: the case technical features are extracted from the processed case library by using a TF-IDF algorithm; the case technical features are extracted from the processed case library by using a deep learning model BERT based on a Transformer architecture; and the case technical features are extracted from the processed case library by using other deep learning models based on a non-Transformer architecture.

[0012] Optionally, the simulation case library is generated by using a generative adversarial network or a variational autoencoder, including: the simulation case library covering multiple climate events, multiple technical scenarios and multiple technical types is generated by using the generative adversarial network or the variational autoencoder.

[0013] Optionally, the cases in the simulation case library are classified into four categories of risk assessment, early warning, prevention and control, and resilience enhancement based on a classification model, including one of the following: the case technical features in the simulation case library are classified into four categories of risk assessment, early warning, prevention and control, and resilience enhancement based on a support vector machine (SVM) algorithm; the case technical features in the simulation case library are classified into four categories of risk assessment, early warning, prevention and control, and resilience enhancement based on a random forest algorithm; and the case technical features in the simulation case library are classified into four categories of risk assessment, early warning, prevention and control, and resilience enhancement based on a deep learning model.

[0014] Optionally, a technology demand matrix is constructed in combination with an unsupervised learning algorithm, including: a K-means clustering algorithm or a hierarchical clustering algorithm is used to construct the technology demand matrix, wherein the elements in the technology demand matrix represent the degree of association between climate risk types and technology demands.

[0015] Optionally, when generating a multi-level technology system covering risk assessment, early warning, prevention and control, and resilience enhancement based on the clustering analysis result and the association analysis result, the output multi-level technology system includes: a technology classification table, the elements in the technology classification table are classified by climate risk type and technology type; a technology demand priority table, the technology demand priority table is used to output priorities for sorting based on the importance and frequency of classified demands; and a systematic technology framework, the systematic technology framework forms a technology system covering risk assessment, early warning, prevention and control, and resilience enhancement.

[0016] Optionally, the artificial intelligence-based climate risk prevention and control method dynamically adjusts the priority and classification structure of technology demands during operation to adapt to changing climate scenarios and regional characteristics.

[0017] According to another aspect of the embodiments of the present application, a climate risk prevention and control device based on artificial intelligence is also provided, comprising: a learning simulation module, configured to case learning and simulation generation: performing data cleaning and standardization processing on a case library of climate risk prevention and control technology, extracting case technical features from the processed case library by using natural language processing technology, generating a case technical feature vector of the case technical features by using a machine learning model, and performing dimension reduction processing on the case technical feature vector to obtain key features; generating a simulation case library based on the key features by using a generative adversarial network or a variational autoencoder, so as to expand the coverage range of case data; a verification module, configured to extract and verify technical requirements: classifying case technical features in the simulation case library into four categories of risk assessment, early warning, prevention and control, and resilience improvement based on a classification model; constructing a technical requirement matrix by combining an unsupervised learning algorithm, forming an association between climate risk types and technical requirements; verifying the technical requirements in the technical requirement matrix; a construction module, configured to construct a technical system: clustering all technical requirements based on a clustering analysis algorithm to form a clustering analysis result; analyzing the association between different technical requirements by using an association rule mining technology to form an association analysis result; generating a multi-level technical system covering risk assessment, early warning, prevention and control, and resilience improvement according to the clustering analysis result and the association analysis result; and an evaluation module, configured to use the multi-level technical system for climate risk prevention and control.

[0018] According to another aspect of the embodiments of the present application, a computer readable storage medium is also provided, the storage medium comprising a stored program, the program being executed to perform the above method.

[0019] According to another aspect of the embodiments of the present application, an electronic device is also provided, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor executing the above method by the computer program.

[0020] According to an aspect of the present application, a computer program product or computer program is provided, the computer program product or computer program comprising computer instructions stored in a computer readable storage medium. A processor of a computer device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions, so that the computer device executes the steps of any embodiment of the above method.

[0021] In the embodiments of the present application, case learning and simulation generation: data cleaning and standardization processing are performed on the case library of climate risk prevention and control technology, natural language processing technology is used to extract case technology features from the processed case library, a machine learning model is used to generate a case technology feature vector of the case technology features, and key features are obtained by dimensionality reduction processing on the case technology feature vector; a generative adversarial network or a variational autoencoder is used to generate a simulation case library based on the key features, thereby expanding the coverage range of the case data; extracting and verifying technical requirements: the case technology features in the simulation case library are classified into four categories of risk assessment, early warning, prevention and control, and resilience improvement based on a classification model; a technology demand matrix is constructed by combining an unsupervised learning algorithm, forming the association between climate risk types and technical requirements; the technical requirements in the technology demand matrix are verified; constructing a technology system: clustering all technical requirements based on a clustering analysis algorithm to form clustering analysis results; using association rule mining technology to analyze the association between different technical requirements to form association analysis results; generating a multi-level technology system covering risk assessment, early warning, prevention and control, and resilience improvement according to the clustering analysis results and the association analysis results; using the multi-level technology system for climate risk prevention and control. Thus, the technical problem of insufficient comprehensive climate risk prevention and control in the related art is solved. BRIEF DESCRIPTION OF DRAWINGS

[0022] The drawings described herein are used to provide further understanding of the present application, and form a part of the present application. The illustrative embodiments of the present application and their descriptions serve to explain the present application, and do not constitute an improper limitation on the present application. In the drawings:

[0023] Figure 1 is a flowchart of an optional artificial intelligence-based climate risk prevention and control method according to an embodiment of the present application;

[0024] Figure 2 is a schematic diagram of an optional artificial intelligence-based climate risk prevention and control device according to an embodiment of the present application;

[0025] Figure 3 is a structural block diagram of a terminal according to an embodiment of the present application. DETAILED DESCRIPTION

[0026] In order for those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should be within the scope of protection of the present application.

[0027] It should be noted that the terms "first", "second", and the like in the description and in the claims of the present application and the above-described accompanying drawings are used to distinguish similar objects, and do not necessarily have to be used to describe a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product, or apparatus including a series of steps or units does not have to be limited to only those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to such a process, method, product, or apparatus.

[0028] According to an aspect of an embodiment of the present application, a method embodiment of an artificial intelligence-based climate risk prevention and control method is provided. By integrating simulation case generation, classification model, clustering analysis, and association rule mining technology, a data-driven intelligent technology system is constructed. This method can be applied to the planning, construction, and operation and maintenance stages of waterway transportation infrastructure, can dynamically adjust the priority and classification structure of technical requirements to adapt to changing climate scenarios and regional characteristics, and provides technical support for climate risk prevention and control.

[0029] Figure 1 is a flowchart of an optional artificial intelligence-based climate risk prevention and control method according to an embodiment of the present application, as shown in Figure 1 The method can include the following steps:

[0030] Step S102, case learning and simulation generation: data cleaning and standardization processing are performed on the case library of climate risk prevention and control technology, natural language processing technology is used to extract case technical features from the processed case library, a machine learning model is used to generate a case technical feature vector of the case technical features, and key features are obtained by dimensionality reduction processing on the case technical feature vector; a generative adversarial network (GAN) or a variational autoencoder (VAE) is used to generate a simulation case library based on the key features, thereby expanding the coverage of the case data.

[0031] The natural language processing technology described above includes but is not limited to TF-IDF, BERT, or other deep learning models, which are used to extract technical types, application scenarios, applicable conditions, and effect evaluations in the cases; the simulation cases generated by the generative adversarial network (GAN) or the variational autoencoder (VAE) cover various technical scenarios and technical types of different climate events (including but not limited to floods, storm surges, heavy rainfall, and extreme high temperatures).

[0032] Step S104, extracting and verifying technical requirements: based on the classification model, the technical features of the simulation case library are classified into four categories: risk assessment, early warning, prevention and control, and resilience improvement; combined with unsupervised learning algorithm to construct technical requirement matrix, form the association between climate risk type and technical requirement; verify the technical requirements in the technical requirement matrix.

[0033] The above classification model is a support vector machine (SVM), random forest or deep learning model, which is used to classify the technical features in the simulation case library into four categories: risk assessment, early warning, prevention and control, and resilience improvement; The construction of the above technical requirement matrix is based on unsupervised learning algorithm, including but not limited to K-means clustering and hierarchical clustering, and the matrix elements represent the association degree between climate risk type and technical requirement.

[0034] Step S106, constructing a technical system: based on clustering analysis algorithm, all technical requirements are clustered to form clustering analysis results; use association rule mining technology to analyze the association between different technical requirements to form association analysis results; generate a multi-level technical system covering risk assessment, early warning, prevention and control, and resilience improvement based on clustering analysis results and association analysis results.

[0035] The above technical system construction includes: integrating similar technical requirements based on clustering algorithm to form a preliminary technical system framework; use association rule mining technology (such as Apriori algorithm) to analyze the logical relationship between technologies; build a multi-level technical system, including first-level classification (risk assessment, early warning, prevention and control, resilience improvement) and second-level technical subdivision (specific technical type). The final output of the climate risk prevention and control technical system includes: technical classification table: output specific classification results according to climate risk type and technical type; technical requirement priority table: output priority order based on the importance and frequency of classification requirements; Systematic technology framework: technical system covering risk assessment, early warning, prevention and control and resilience improvement.

[0036] Step S108, using a multi-level technical system for climate risk prevention and control.

[0037] In view of the technical problems of the related art mentioned in the background art that the climate risk prevention and control is not comprehensive enough, the present application proposes a systematic technical system generation scheme based on artificial intelligence and machine learning, which uses existing climate risk prevention and control technology cases, combined with simulation case generation and clustering analysis technology, to construct a climate risk prevention and control technical system covering risk assessment, early warning, prevention and control, and resilience improvement, thereby solving the problem. The specific solution includes the following three core stages:

[0038] 1) Case study and simulation generation: Preprocess and learn features from existing climate risk prevention and control technology cases using machine learning techniques to generate a simulation case library, providing a data foundation for subsequent technology requirement extraction and cluster analysis; 2) Technology requirement extraction and classification: Extract technology requirements under complex climate events from the simulation case library and verify them to ensure the comprehensiveness and accuracy of the classification; 3) Technology system construction: Integrate the extracted technology requirements through cluster analysis and association analysis to generate a systematic climate risk prevention and control technology system for waterway transportation infrastructure.

[0039] As an optional embodiment, the embodiments of this application are described in detail below with reference to specific technical solutions:

[0040] The algorithm flow for generating the AI-based climate risk prevention and control technology system for waterway transportation infrastructure in this application is as follows:

[0041] Input: A case study library of climate risk prevention and control technologies, including technology cases related to risk assessment, risk warning, prevention and control measures, and resilience enhancement.

[0042] 1) Case study and simulation generation

[0043] 1.1) Case Preprocessing

[0044] The input case library is cleaned and standardized to remove redundant and invalid data; Natural Language Processing (NLP) technology is used to perform structured analysis on the case content to extract key technical features, such as technology type, application scenario, applicable conditions, and effect evaluation.

[0045] 1.2) Feature Learning

[0046] The technical features of the case are represented using machine learning models (such as TF-IDF, BERT or other deep learning models) to generate technical feature vectors.

[0047] The technical features of the case study are subjected to dimensionality reduction to extract key features for subsequent analysis. The formula is as follows:

[0048]

[0049] V i : No. i The feature vector of each case represents the core technical features of that case; C i : No. i The text content or description of each case; f Text feature extraction function, which uses natural language processing (NLP) or machine learning models (such as TF-IDF, BERT) to extract key features of a case.

[0050] 1.3) Case Simulation Generation

[0051] Use Generative Adversarial Networks (GAN) or Variational Autoencoder (VAE) to expand case data and generate simulation case library.

[0052] Ensure that simulation cases cover more scenarios and technology types, and improve case coverage. The formula is as follows:

[0053]

[0054] E: Expectation operator; G : Generator, used to generate new simulation cases; D : Discriminator, used to distinguish real cases and simulation cases; P data (C) : Distribution of real cases; P z (z) : Distribution of random noise used to generate simulation cases; G(z) : Simulation cases generated by the generator; Goal: Through the use of Generative Adversarial Networks (GAN), generate simulation cases that cover more scenarios and expand the case library.

[0055] 2) Technology Requirement Extraction and Verification

[0056] 2.1) Case Classification

[0057] Use a classification model (such as Support Vector Machine SVM) to classify case features, and the classification objective function is:

[0058]

[0059] w : Weight vector of the classification model; C : Regularization parameter, controls the complexity of the model; ξ i : Slack variable, allows a certain classification error, n is the number of training samples.

[0060] Satisfy the constraints:

[0061]

[0062] yi : Classification label of the sample; b : Bias term of the classification model.

[0063] 2.2) Technology Requirement Matrix Construction

[0064] The extracted technology requirements are represented by a matrix:

[0065]

[0066] t ij : No. i Climate risk and the first j The degree of relevance of a technology requirement is usually determined by the results of case analysis.

[0067] 2.3) Requirements Verification

[0068] Verify technical requirements using real-world case studies, employing a requirement coverage metric:

[0069]

[0070] i Climate risk type code, i = 1, 2,..., N, N Total number of climate risk types; j Technical requirement number, j= 1, 2,..., M, M Total number of technology requirements; t ij The degree of correlation between the i-th climate risk and the j-th technology requirement is usually determined by the results of case analysis. δ ij : Coverage indicator variable, when the j-th technological requirement can effectively address the i-th climate risk δ ij = 1 ,otherwise δ ij =0。

[0071] 3) Construction of the technical system

[0072] 3.1) Technology Clustering

[0073] Cluster analysis is performed on the technology requirements, with the objective function being:

[0074]

[0075] K The number of clusters; n Number of cases; V i (k) : Belongs to the k The feature vector of the class; μ k : No. k The center vector of the class.

[0076] 3.2) Inter-technology correlation analysis

[0077] To analyze the correlation between technical requirements, support and confidence formulas are used:

[0078] Support:

[0079] A and B represent technical requirements A and B, and the formula above indicates the frequency at which technical requirements A and B are applied simultaneously.

[0080] Confidence level:

[0081] This indicates the possibility that when requirement A is adopted, requirement B will also be adopted.

[0082] 3.3) Allocation of Technical Weights

[0083] Calculate technology weights based on demand frequency:

[0084]

[0085] W j : No. j The weight of each technology requirement; f ij : No. i In the case of the first one, the first j The frequency of occurrence of each technical requirement; n Total number of cases.

[0086] 3.4) Multi-layered framework of the technical system

[0087] The technology system is generated according to a multi-level framework, in which the primary and secondary classifications are determined by clustering results and weight priority ranking, ultimately forming a multi-dimensional technology system.

[0088] The final output technology system can be represented by the following matrix:

[0089]

[0090] in: s ij : indicates the first i The first type of climate risk corresponding to j The priority of each technology is filled in after being sorted by weight; rows represent climate risk types; columns represent technology categories (risk assessment, early warning, prevention and control, resilience enhancement).

[0091] The final output is a set of climate risk prevention and control technology system for waterway transportation infrastructure, which includes: a technology classification table: outputting specific technology classifications according to risk type and technology type; a technology demand priority table: providing a priority ranking of technology demands based on the importance and frequency of the classified demands; and a systematic technology framework: generating a multi-level technology system covering risk assessment, early warning, prevention and control, and resilience enhancement.

[0092] In a specific implementation case, the construction of a climate risk prevention and control technology system for a coastal port terminal is taken as an example:

[0093] 1) Basic data and scenarios

[0094] Port basic situation: Location: Southeast coast; Main functions: container and bulk cargo transportation; Key facilities: terminal, yard, handling equipment, storage facilities; Main climate risks: typhoon, storm surge, heavy rain, extreme high temperature.

[0095] Historical case data: Collect climate risk prevention and control cases of the port and 20 similar ports in the surrounding area in the past 10 years, including 150 specific technology application cases, covering risk assessment, early warning, prevention and control measures, and resilience improvement.

[0096] 2) Specific implementation steps

[0097] 2.1) Case learning and simulation generation

[0098] Generate simulation cases: Original cases: 150; Supplement cases generated by GAN: 300; Total case library size: 450.

[0099] 2.2) Technology demand extraction and classification

[0100] The identified key technology demand matrix is shown in Table 1 below:

[0101] Table 1

[0102]

[0103] 2.3) Technology system construction

[0104] Technology clustering results: Monitoring and early warning technology cluster (meteorological monitoring system, hydrological monitoring system, equipment state monitoring system), protection facility technology cluster (wind and moisture protection facility, drainage and waterlogging prevention facility, corrosion prevention facility), emergency disposal technology cluster (emergency response system, equipment protection scheme, personnel evacuation scheme).

[0105] 3) Implementation effect

[0106] Application effect of the generated technology system:

[0107] Risk assessment accuracy rate improved (typhoon risk prediction accuracy rate: 92%, heavy rain impact assessment accuracy rate: 88%); early warning time advanced (typhoon early warning time increased by 6 hours, heavy rain early warning time increased by 4 hours); prevention and control effect improved (equipment loss reduced by 45%, operation interruption time shortened by 35%).

[0108] 4) Technology verification

[0109] The generated technical system is verified in the field, and two typhoons and one heavy rain event landing in 2023 are selected as verification cases:

[0110] Typhoon "Blue" event verification: early warning lead time: 8 hours; protection measures in place rate: 95%; facility damage rate: reduced by 52%.

[0111] This implementation case demonstrates the feasibility and effectiveness of the patent technology in practical application. Through the systematic technical system, the climate risk prevention and control capability of the port terminal is significantly improved, realizing the double improvement of economic benefit and social benefit.

[0112] It should be noted that for the foregoing method embodiments, in order to simply describe, they are all expressed as a series of action combinations, but those skilled in the art should know that the present application is not limited by the order of the described actions, because according to the present application, certain steps can be performed in other order or simultaneously. Secondly, those skilled in the art should know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily necessary for the present application.

[0113] From the above description of the embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be realized by means of software and the necessary general hardware platform, of course, it can also be realized by hardware, but in many cases the former is a better implementation. Based on such understanding, the technical solutions of the present application can be embodied in the form of software product in essence or say the part that contributes to the prior art, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disc), including a plurality of instructions to make a terminal device (which can be a mobile phone, computer, server, or network device, etc.) execute the method described in each embodiment of the present application.

[0114] According to another aspect of the embodiments of the present application, a kind of based on artificial intelligence's climate risk prevention and control device based on artificial intelligence's climate risk prevention and control method described above is provided. Figure 2 It is a kind of optional based on artificial intelligence's climate risk prevention and control device according to the embodiments of the present application, as shown in Figure 2 The device can include:

[0115] The learning simulation module 21 is configured to learn and simulate generation. The case library of climate risk prevention and control technologies is subjected to data cleaning and standardization processing. Natural language processing technology is used to extract case technical features from the processed case library. A machine learning model is used to generate a case technical feature vector of the case technical features, and the case technical feature vector is subjected to dimension reduction processing to obtain key features. A generative adversarial network or a variational autoencoder is used to generate a simulation case library based on the key features, thereby expanding the coverage of the case data.

[0116] The verification module 23 is configured to extract and verify technical requirements. The case technical features in the simulation case library are classified into four categories of risk assessment, early warning, prevention and control, and resilience improvement based on a classification model. A technical requirement matrix is constructed by combining an unsupervised learning algorithm, forming an association between climate risk types and technical requirements. The technical requirements in the technical requirement matrix are verified.

[0117] The construction module 25 is configured to construct a technical system. All technical requirements are clustered based on a clustering analysis algorithm to form a clustering analysis result. The association between different technical requirements is analyzed using association rule mining technology to form an association analysis result. A multi-level technical system covering risk assessment, early warning, prevention and control, and resilience improvement is generated according to the clustering analysis result and the association analysis result.

[0118] The evaluation module 27 is configured to use the multi-level technical system for climate risk prevention and control.

[0119] Optionally, the natural language processing technology is used to extract case technical features from the processed case library, including one of the following: using a TF-IDF algorithm to extract case technical features from the processed case library; using a deep learning model BERT based on a Transformer architecture to extract case technical features from the processed case library; and using other deep learning models based on a non-Transformer architecture to extract case technical features from the processed case library.

[0120] Optionally, when the generative adversarial network or the variational autoencoder is used to generate the simulation case library, the generative adversarial network or the variational autoencoder is used to generate a simulation case library covering multiple climate events, multiple technical scenarios, and multiple technical types.

[0121] Optionally, the technical features of the cases in the simulation case library are classified into four categories of risk assessment, early warning, prevention and control, and resilience improvement based on a classification model, including one of the following: the technical features of the cases in the simulation case library are classified into four categories of risk assessment, early warning, prevention and control, and resilience improvement based on a support vector machine (SVM) algorithm; the technical features of the cases in the simulation case library are classified into four categories of risk assessment, early warning, prevention and control, and resilience improvement based on a random forest algorithm; and the technical features of the cases in the simulation case library are classified into four categories of risk assessment, early warning, prevention and control, and resilience improvement based on a deep learning model.

[0122] Optionally, when the technology demand matrix is constructed in combination with an unsupervised learning algorithm: a K-means clustering algorithm or a hierarchical clustering algorithm is used to construct the technology demand matrix, wherein an element in the technology demand matrix represents the degree of association between a climate risk type and a technology demand.

[0123] Optionally, when the multi-level technology system covering risk assessment, early warning, prevention and control, and resilience improvement is generated based on the clustering analysis result and the association analysis result, the output multi-level technology system includes: a technology classification table, elements in the technology classification table are classified by climate risk type and technology type; a technology demand priority table, the technology demand priority table is used to output priorities for sorting based on the importance and frequency of classified demands; and a systematic technology framework, the systematic technology framework forms a technology system covering risk assessment, early warning, prevention and control, and resilience improvement.

[0124] Optionally, the artificial intelligence-based climate risk prevention and control method dynamically adjusts the priority and classification structure of technology demands during operation to adapt to changing climate scenarios and regional characteristics.

[0125] In view of the problem of insufficient comprehensive climate risk prevention and control in the related art mentioned in the background, the present application proposes a systematic technology system generation scheme based on artificial intelligence and machine learning. The scheme uses existing climate risk prevention and control technology cases, combines simulation case generation and clustering analysis technology, and constructs a climate risk prevention and control technology system covering risk assessment, early warning, prevention and control, and resilience improvement, thereby solving the problem.

[0126] According to another aspect of an embodiment of the present application, a server or terminal for implementing the above-mentioned artificial intelligence-based climate risk prevention and control method is also provided.

[0127] Figure 3 is a structural block diagram of a terminal according to an embodiment of the present application, as shown in Figure 3 The terminal can include one or more (only one is shown in the figure) processors 301, a memory 303, and a transmission device 305, as shown in Figure 3As shown, the terminal can further include an input / output device 307.

[0128] The memory 303 can be used to store software programs and modules, such as program instructions / modules corresponding to the climate risk prevention and control method based on artificial intelligence and the device in the embodiments of the present application. The processor 301 executes various functions and data processing by running the software programs and modules stored in the memory 303, that is, implements the climate risk prevention and control method based on artificial intelligence. The memory 303 can include a high-speed random access memory, and can further include a non-volatile memory, such as one or more magnetic storage devices, flash memories, or other non-volatile solid-state memories. In some examples, the memory 303 can further include a memory remotely arranged with respect to the processor 301, and the remote memory can be connected to the terminal through a network. Examples of the network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0129] The transmission device 305 described above is used to receive or send data via a network, and can also be used for data transmission between the processor and the memory. Specific examples of the network can include wired networks and wireless networks. In one example, the transmission device 305 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices and routers through a network cable to communicate with the Internet or a local area network. In one example, the transmission device 305 is a radio frequency (Radio Frequency, RF) module, which is used to communicate with the Internet in a wireless manner.

[0130] Specifically, the memory 303 is used to store application programs.

[0131] The processor 301 can call the application programs stored in the memory 303 through the transmission device 305 to perform the following steps:

[0132] Case learning and simulation generation: data cleaning and standardization processing are performed on a case library of climate risk prevention and control technologies, natural language processing technology is used to extract case technology features from the processed case library, a machine learning model is used to generate a case technology feature vector of the case technology features, and key features are obtained by dimension reduction processing on the case technology feature vector; a generative adversarial network or a variational autoencoder is used to generate a simulation case library based on the key features, thereby expanding the coverage range of the case data; extracting and verifying technical requirements: the case technology features in the simulation case library are classified into four categories of risk assessment, early warning, prevention and control, and resilience improvement based on a classification model; a technology demand matrix is constructed by combining unsupervised learning algorithms, forming the association between climate risk types and technical requirements; the technical requirements in the technology demand matrix are verified; constructing a technology system: clustering all technical requirements based on clustering analysis algorithms to form clustering analysis results; using association rule mining technology to analyze the association between different technical requirements to form association analysis results; generating a multi-level technology system covering risk assessment, early warning, prevention and control, and resilience improvement according to the clustering analysis results and the association analysis results; using the multi-level technology system for climate risk prevention and control.

[0133] Optionally, specific examples in this embodiment can refer to the examples described in the above embodiments, which will not be repeated here.

[0134] Those skilled in the art can understand that, Figure 3 The structure shown is only schematic, and the terminal can be a smart phone (such as an Android phone, an iOS phone, etc.), a tablet computer, a palm computer, a mobile Internet device (MID), a PAD, and the like. Figure 3 It does not limit the structure of the above-mentioned electronic device. For example, the terminal can further include more or less components (such as a network interface, a display device, etc.) than those shown in the above-mentioned electronic device, or have a different configuration from that shown in the above-mentioned electronic device. Figure 3 Figure 3 The structure shown is only schematic, and the terminal can be a smart phone (such as an Android phone, an iOS phone, etc.), a tablet computer, a palm computer, a mobile Internet device (MID), a PAD, and the like. Figure 3 It does not limit the structure of the above-mentioned electronic device. For example, the terminal can further include more or less components (such as a network interface, a display device, etc.) than those shown in the above-mentioned electronic device, or have a different configuration from that shown in the above-mentioned electronic device.

[0135] Those skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be instructed by a program to complete the related hardware of the terminal device, and the program can be stored in a computer readable storage medium, which can include a flash disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0136] Embodiments of the present application also provide a storage medium. Optionally, in the present embodiment, the above-mentioned storage medium can be used to execute the program code of the artificial intelligence-based climate risk prevention and control method.

[0137] Optionally, in the embodiment, the storage medium can be located on at least one of the plurality of network devices in the network shown in the above embodiment.

[0138] Optionally, in the embodiment, the storage medium is configured to store program code for performing the following steps:

[0139] Case learning and simulation generation: data cleaning and standardization processing are performed on a case library of climate risk prevention and control technologies, natural language processing technology is used to extract case technology features from the processed case library, a machine learning model is used to generate a case technology feature vector of the case technology features, and key features are obtained by dimensionality reduction processing on the case technology feature vector; a generative adversarial network or a variational autoencoder is used to generate a simulation case library based on the key features, thereby expanding the coverage range of the case data; extracting and verifying technical requirements: the case technology features in the simulation case library are classified into four categories of risk assessment, early warning, prevention and control, and resilience improvement based on a classification model; a technology demand matrix is constructed by combining unsupervised learning algorithms, forming the association between climate risk types and technical requirements; the technical requirements in the technology demand matrix are verified; constructing a technology system: clustering all technical requirements based on a clustering analysis algorithm to form clustering analysis results; using association rule mining technology to analyze the association between different technical requirements to form association analysis results; generating a multi-level technology system covering risk assessment, early warning, prevention and control, and resilience improvement according to the clustering analysis results and the association analysis results; using the multi-level technology system for climate risk prevention and control.

[0140] Optionally, specific examples in the embodiment can refer to the examples described in the above embodiments, which will not be described here again.

[0141] Optionally, in the embodiment, the storage medium can include but is not limited to a U disk, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and various media that can store program codes.

[0142] The serial numbers of the embodiments of the present application are only for description, and do not represent the advantages and disadvantages of the embodiments.

[0143] The integrated units in the above embodiments, if implemented in the form of software function units and sold or used as independent products, can be stored in the above computer-readable storage medium. Based on such understanding, the technical solutions of the present application, essentially or in other words, the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a number of instructions to make one or more computer devices (which can be personal computers, servers or network devices, etc.) execute all or part of the steps of the methods described in the various embodiments of the present application.

[0144] In the above embodiments of the present application, the description of each embodiment has its own focus, and the parts not described in detail in a certain embodiment can be referred to the relevant description of other embodiments.

[0145] In the several embodiments provided by the present application, it should be understood that the disclosed client can be implemented in other ways. Of course, the above device embodiment is only illustrative, for example, the division of the units is only a logical function division, and there can be another division manner in actual implementation, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interface, unit or module, and can be electrical or other forms.

[0146] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, i.e., they can be located in one place or distributed on multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiment according to actual needs.

[0147] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software function unit.

[0148] The above is only the preferred embodiment of the present application, and it should be pointed out that for those skilled in the art, without departing from the principle of the present application, some improvements and refinements can be made, and these improvements and refinements should be regarded as the protection scope of the present application.

Claims

1. A climate risk prevention and control method based on artificial intelligence, characterized in that, include: Case study and simulation generation: The case library of climate risk prevention and control technologies is cleaned and standardized. Natural language processing technology is used to extract case technology features from the processed case library. A machine learning model is used to generate case technology feature vectors of the case technology features, and the case technology feature vectors are dimensionality reduced to obtain key features. A simulation case library is generated based on the key features using generative adversarial networks or variational autoencoders, thereby expanding the coverage of case data. The simulation cases generated by the generative adversarial networks or variational autoencoders cover a variety of technical scenarios and technical types for different climate events. Extracting and verifying technical requirements: Based on a classification model, the technical features of the cases in the simulation case library are classified into four categories: risk assessment, early warning, prevention and control, and resilience enhancement; a technical requirement matrix is ​​constructed by combining unsupervised learning algorithms to form the correlation between climate risk types and technical requirements; Verify the technical requirements in the aforementioned technical requirements matrix; Building a technical system: Clustering all technical requirements based on clustering analysis algorithms to generate clustering analysis results; The association rule mining technique is used to analyze the correlation between different technical requirements and generate association analysis results; Based on the clustering analysis results and the association analysis results, a multi-level technical system covering risk assessment, early warning, prevention and control, and resilience enhancement is generated; Climate risk prevention and control are carried out using the aforementioned multi-level technology system; Among these features, generative adversarial networks (GANs) are used to generate simulation cases covering a wider range of scenarios. ， E : indicates the expectation operator ，G : Represents a generator used to generate new simulation cases. D： The discriminator is used to distinguish between real-world and simulated cases. P data (C) Distribution of real-world cases P z (z) The distribution of random noise is used to generate simulation cases. G (z) Simulation cases generated by the generator.

2. The method according to claim 1, characterized in that, Natural language processing techniques are used to extract case technical features from the processed case database, including one of the following: The TF-IDF algorithm is used to extract case technical features from the processed case library; The BERT deep learning model, based on the Transformer architecture, extracts case technical features from the processed case library. Other deep learning models based on non-Transformer architectures extract case technical features from the processed case library.

3. The method according to claim 1, characterized in that, Based on the classification model, the technical features of the cases in the simulation case library are classified into four categories: risk assessment, early warning, prevention and control, and resilience enhancement, including one of the following: Based on the Support Vector Machine (SVM) algorithm, the technical features of the cases in the simulation case library are classified into four categories: risk assessment, early warning, prevention and control, and resilience enhancement. Based on the random forest algorithm, the technical features of the cases in the simulation case library are classified into four categories: risk assessment, early warning, prevention and control, and resilience enhancement. Based on a deep learning model, the technical features of the cases in the simulation case library are classified into four categories: risk assessment, early warning, prevention and control, and resilience enhancement.

4. The method according to claim 1, characterized in that, A technology requirements matrix is ​​constructed by combining unsupervised learning algorithms, including: The technology demand matrix is ​​constructed using K-means clustering or hierarchical clustering algorithms, wherein the elements in the technology demand matrix represent the degree of correlation between climate risk types and technology demands.

5. The method according to claim 1, characterized in that, When generating a multi-level technical system covering risk assessment, early warning, prevention and control, and resilience enhancement based on the clustering analysis results and the association analysis results, the output multi-level technical system includes: A technology classification table, wherein the elements in the technology classification table are classified according to climate risk type and technology type; A technology requirement priority table, which is used to sort requirements based on their importance and frequency. A systematic technical framework, which forms a technical system covering risk assessment, early warning, prevention and control, and resilience enhancement.

6. The method according to any one of claims 1 to 5, characterized in that, The AI-based climate risk prevention and control method dynamically adjusts the priority and classification structure of technical requirements during operation to adapt to changing climate scenarios and regional characteristics.

7. A climate risk prevention and control device based on artificial intelligence, characterized in that, include: The learning simulation module is used for case learning and simulation generation: It performs data cleaning and standardization on a case library of climate risk prevention and control technologies; extracts case technology features from the processed case library using natural language processing technology; generates case technology feature vectors using machine learning models; and performs dimensionality reduction on these feature vectors to obtain key features. Based on these key features, it uses generative adversarial networks (GANs) or variational autoencoders (VAEs) to generate a simulation case library, thereby expanding the coverage of case data. The simulation cases generated by the GANs or VAEs cover various technical scenarios and types of different climate events. The verification module is used to extract and verify technical requirements: based on a classification model, the technical features of the cases in the simulation case library are classified into four categories: risk assessment, early warning, prevention and control, and resilience enhancement; and a technical requirement matrix is ​​constructed by combining an unsupervised learning algorithm to form the correlation between climate risk types and technical requirements. Verify the technical requirements in the aforementioned technical requirements matrix; The building module is used to construct the technology system: it clusters all technical requirements based on clustering analysis algorithms to generate clustering analysis results; The association rule mining technique is used to analyze the correlation between different technical requirements and generate association analysis results; Based on the clustering analysis results and the association analysis results, a multi-level technical system covering risk assessment, early warning, prevention and control, and resilience enhancement is generated; An assessment module is used to conduct climate risk prevention and control using the aforementioned multi-level technology system; Among these features, generative adversarial networks (GANs) are used to generate simulation cases covering a wider range of scenarios. ， E : indicates the expectation operator ，G : Represents a generator used to generate new simulation cases. D： The discriminator is used to distinguish between real-world and simulated cases. P data (C) Distribution of real-world cases P z (z) The distribution of random noise is used to generate simulation cases. G (z) Simulation cases generated by the generator.

8. A computer-readable storage medium, characterized in that, The storage medium includes a stored program, wherein the program executes the method described in any one of claims 1 to 6 when it is run.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, The processor executes the method described in any one of claims 1 to 6 via the computer program.

Images