Road network climate change adaptability evaluation method and device

Abstract

The invention is applicable to the field of road network climatic change assessment, in particular to a road network climatic change adaptability assessment method and device, in the assessment method, precipitation time sequence data is obtained based on processing of original precipitation data of a target road network area; eMD decomposition is carried out on the precipitation time series data, an obtained modal function and a residual error are used as inputs of a pre-constructed LSTM prediction model, the precipitation is predicted through a multi-input single-output sliding window, and a prediction result comprises spatial distribution of precipitation in a target highway network area; and analyzing the adaptability of the road network infrastructure based on the predicted precipitation, wherein the evaluation indexes of the adaptability comprise the weather resistance and drainage capacity of the physical structure of the road network infrastructure. According to the method, the adaptability of the road network infrastructure is analyzed based on the predicted precipitation, the potential weakness of the road network infrastructure can be identified and improved, and the operation safety and the operation efficiency of the road network under the extreme weather, especially the extreme rainfall condition are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of highway network climate change assessment, and in particular relates to a method and device for assessing the adaptability of highway network climate change. Background Art

[0002] Climate change is a major global challenge facing mankind. Its main manifestations are rising global temperatures, frequent extreme weather and natural disasters, which have a significant impact on human life, the natural environment and building facilities.

[0003] As the frequency of extreme weather events increases, the highway network, as an important transportation infrastructure and key point of economic activities, faces severe challenges of climate change, especially extreme rainfall changes. In order to effectively respond to these challenges, a scientific assessment of the climate change adaptability of the highway network is needed.

[0004] At present, multiple influencing factors are evaluated in the assessment method of climate change adaptability of highway networks; among them, assessing the impact of precipitation on the adaptability of highway networks to climate change is one of the important indicators.

[0005] In recent years, with the rapid development of computing technology and the advent of the big data era, deep learning, as an advanced machine learning technology, has demonstrated its outstanding performance in many fields and has attracted much attention due to its advantages in precipitation prediction. However, the prediction accuracy of related technologies in precipitation prediction models is poor, which will further affect the assessment results of the climate change adaptability of the highway network. Summary of the invention

[0006] The purpose of the embodiments of the present invention is to provide a method and device for evaluating the adaptability of a highway network to climate change, aiming to solve the technical problem that the prediction accuracy of the precipitation prediction model in the related technology is poor, which will further affect the evaluation results of the adaptability of the highway network to climate change.

[0007] To achieve the above object, the present invention provides the following technical solutions.

[0008] An embodiment of the present invention provides a method for evaluating the adaptability of a highway network to climate change, the method comprising the following steps: Based on the processing of the original precipitation data in the target highway network area, precipitation time series data is obtained; Perform EMD decomposition on precipitation time series data, use the obtained modal function and residual as input of a pre-constructed LSTM prediction model, and predict precipitation through a multi-input single-output sliding window. The prediction results include the spatial distribution of precipitation in the target highway network area. The adaptability of the road network infrastructure is analyzed based on the predicted precipitation, and the evaluation indicators of the adaptability include the weather resistance and drainage capacity of the physical structure of the road network infrastructure.

[0009] Furthermore, the step of obtaining precipitation time series data based on processing the original precipitation data of the target highway network area includes: normalizing and sequence segmenting the original precipitation data, taking a monthly precipitation time series data segment as a time step, and forming a continuous multiple months into an input sequence, and obtaining precipitation time series data within the target time period expressed as: ;in, Represents the precipitation in the nth month.

[0010] Furthermore, the step of performing EMD decomposition on the precipitation time series data includes: The precipitation time series data is decomposed into a finite number of intrinsic mode functions and residual terms, expressed as: ; In the formula, Represents the original precipitation time series data; It is represented as the i-th eigenmode function obtained by decomposition; represents the residual term; t represents time; n represents the number of decompositions.

[0011] Furthermore, the pre-constructed LSTM prediction model adopts mean square error MSE as the loss function, and adds a penalty term for prediction errors of high precipitation values; Adam is selected as the optimizer, and the initial learning rate is set to 0.001 to balance the convergence speed and training stability; an early stopping strategy is implemented during training, and the patience value is set to 30 training cycles to prevent overfitting of the model training, the number of neural units in the hidden layer is set to 150, and a dropout layer is added with a dropout rate set to 0.3.

[0012] Furthermore, in the training of the pre-constructed LSTM prediction model, the particle swarm optimization algorithm PSO is used to optimize the initial parameters of the LSTM prediction model; Wherein, in the particle swarm optimization algorithm, the space dimension is set to D and the number of particles is N; The position of the i-th particle is ; The velocity of the i-th particle is ; The optimal position searched by the i-th particle is expressed as: ; All particles search for the global optimal position as: ; The formula for updating the position and velocity of the jth component of the ith particle is expressed as: ; ; In the formula, k represents the number of iterations; represents the inertia weight; represents the learning factor; Represents a random number.

[0013] Furthermore, in the particle swarm optimization algorithm PSO, a linear decreasing weight strategy balancing algorithm is used to optimize the inertia weight to improve the global search capability and local search capability of the algorithm during the search process; The linear decreasing weight strategy is expressed as: ; In the formula, represents the maximum number of iterations; g represents the current number of iterations; represents the initial inertia weight; Represents the inertia weight when the iteration reaches the maximum number.

[0014] Furthermore, in order to enhance the ability of particles to converge to the global optimal point, the particle swarm optimization algorithm PSO uses an asymmetric learning factor calculation formula to optimize the learning factor: ; ; In the formula, Indicates the maximum number of iterations; Respectively represent the final value and initial value of the individual learning factor of the particle; Respectively represent the termination value and initial value of the particle group learning factor; Among them, the termination value of the particle individual learning factor is 0.5; the initial value of the particle individual learning factor is 2.5; the termination value of the particle group learning factor is 2.25; and the initial value of the particle group learning factor is 1.

[0015] Furthermore, in the step of predicting precipitation through a multi-input single-output sliding window, the sliding window is set to 12.

[0016] Furthermore, the step of analyzing the adaptability of the highway network infrastructure based on the predicted precipitation includes: Assess the weather resistance of road network infrastructure based on the predicted temporal and spatial distribution of precipitation; Evaluate the drainage of the road network infrastructure based on the road network infrastructure data, road network drainage structure parameters and predicted precipitation; The weather resistance assessment results and drainage assessment results were integrated to obtain the climate change adaptability assessment results of the highway network, including: ; In the formula, V represents the adaptability of the road network to climate change; E represents the weather resistance of the road network infrastructure; A represents the drainage performance of the road network infrastructure.

[0017] Another embodiment of the present invention provides a device for evaluating the adaptability of a highway network to climate change, the device comprising: A data processing module, used for obtaining precipitation time series data based on processing of original precipitation data in the target highway network area; The climate prediction module is used to perform EMD decomposition on precipitation time series data, use the obtained modal function and residual as the input of the pre-constructed LSTM prediction model, and predict precipitation through a multi-input single-output sliding window; The adaptability assessment module is used to analyze the adaptability of the highway network infrastructure based on the predicted precipitation, and the adaptability assessment indicators include the weather resistance and drainage capacity of the physical structure of the highway network infrastructure.

[0018] By implementing the method and device for evaluating the adaptability of highway network to climate change of the present invention, the following effects can be achieved: First, the present invention performs EMD decomposition on precipitation time series data, uses the obtained modal function and residual as the input of a pre-constructed LSTM prediction model, and predicts precipitation through a multi-input single-output sliding window, thereby obtaining more accurate precipitation prediction data, wherein EMD decomposition can adaptively decompose signals and is more suitable for LSTM prediction model processing, thereby improving the accuracy of prediction; Second, in the training of the pre-constructed LSTM prediction model, the present invention optimizes and solves the initial parameters of the LSTM prediction model through the particle swarm optimization algorithm PSO, thereby improving the generalization ability of the LSTM prediction model, so that it can maintain a high prediction accuracy; Third, the present invention analyzes the adaptability of the highway network infrastructure based on the predicted precipitation, which helps to identify and improve the potential weaknesses of the highway network infrastructure and improve the operational safety and efficiency of the highway network under extreme weather conditions, especially extreme rainfall conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention.

[0020] Figure 1 A flowchart of a method for evaluating the adaptability of a highway network to climate change according to the present invention; Figure 2 A sub-flow chart of a method for assessing the adaptability of a highway network to climate change according to the present invention; Figure 3 The present invention is a structural block diagram of a device for evaluating the adaptability of a highway network to climate change. DETAILED DESCRIPTION

[0021] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0022] The terms "including" and "having" and any variations thereof in the specification and claims of this application are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device comprising a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or devices.

[0023] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0024] As the frequency of extreme weather events increases, the highway network, as an important transportation infrastructure and key point of economic activities, faces severe climate change, especially the challenge of extreme rainfall changes. In order to effectively respond to these challenges, it is necessary to scientifically evaluate the adaptability of the highway network to climate change. At present, in the assessment method of the adaptability of the highway network to climate change, multiple influencing factors are evaluated. Among them, the impact of precipitation on the adaptability of the highway network to climate change is one of the important indicators. The prediction accuracy of the relevant technology in the prediction model of precipitation is poor, which will further affect the assessment results of the adaptability of the highway network to climate change.

[0025] Firstly, the present invention performs EMD decomposition on precipitation time series data, uses the obtained modal function and residual as the input of a pre-constructed LSTM prediction model, and predicts precipitation through a multi-input single-output sliding window, so as to obtain more accurate precipitation prediction data, wherein the EMD decomposition can adaptively decompose the signal and is more suitable for LSTM prediction model processing, thereby improving the accuracy of the prediction; secondly, in the training of the pre-constructed LSTM prediction model, the present invention optimizes and solves the initial parameters of the LSTM prediction model through the particle swarm optimization algorithm PSO, thereby improving the generalization ability of the LSTM prediction model and enabling it to maintain a high prediction accuracy; finally, the present invention analyzes the adaptability of the highway network infrastructure based on the predicted precipitation, which helps to identify and improve the potential weaknesses of the highway network infrastructure and improve the operation safety and efficiency of the highway network under extreme weather conditions, especially extreme rainfall conditions.

[0026] The specific implementation of the method and device for evaluating the adaptability of a highway network to climate change of the present invention is described in detail below in conjunction with specific embodiments.

[0027] Please refer to Figure 1 In a preferred embodiment of the present disclosure, a method for assessing the adaptability of a highway network to climate change comprises the following steps: S101: obtaining precipitation time series data based on processing the original precipitation data of the target road network area; wherein, in the embodiment of the present disclosure, the step of obtaining precipitation time series data based on processing the original precipitation data of the target road network area includes: The original precipitation data is normalized and sequenced, and the monthly precipitation time series data segment is taken as a time step. Multiple consecutive months are combined into an input sequence. The precipitation time series data within the target time period is expressed as: ,in, Represents the precipitation in the nth month.

[0028] Please continue to refer to Figure 1 The evaluation method of the embodiment of the present disclosure further includes the following steps: S102: performing EMD decomposition on the precipitation time series data, using the obtained modal function and residual as the input of the pre-constructed LSTM prediction model, and predicting the precipitation through a multi-input single-output sliding window, wherein the prediction result includes the spatial distribution of precipitation in the target highway network area; Among them, in the disclosed embodiment, the present invention performs EMD decomposition on precipitation time series data, uses the obtained modal function and residual as the input of the pre-constructed LSTM prediction model, and predicts the precipitation through a multi-input single-output sliding window, which can obtain more accurate precipitation prediction data, wherein EMD decomposition can adaptively decompose the signal, which is more suitable for LSTM prediction model processing, thereby improving the accuracy of the prediction; S103: Analyze the adaptability of the highway network infrastructure based on the predicted precipitation, wherein the adaptability evaluation index includes the weather resistance and drainage capacity of the physical structure of the highway network infrastructure.

[0029] Furthermore, in the embodiment of the present disclosure, the step of performing EMD decomposition on the precipitation time series data includes: The precipitation time series data is decomposed into a finite number of intrinsic mode functions and residual terms, expressed as: ; In the formula, Represents the original precipitation time series data; It is represented as the i-th eigenmode function obtained by decomposition; represents the residual term; t represents time; n represents the number of decompositions.

[0030] Furthermore, in an embodiment of the present invention, the pre-constructed LSTM prediction model adopts mean square error MSE as the loss function, and adds a penalty term for incorrect prediction of high precipitation values; the optimizer selects Adam, and the initial learning rate is set to 0.001 to balance the convergence speed and training stability; the early stopping strategy is implemented during training, and the patience value is set to 30 training cycles to prevent overfitting of the model training, the number of neural units in the hidden layer is set to 150, and a dropout layer is added, and the dropout rate is set to 0.3.

[0031] Further, in the disclosed embodiment, the pre-constructed LSTM prediction model includes an input gate, a forget gate and an output gate, which are used to extract feature representations of input data; wherein: The input gate is represented as: ; The forget gate is expressed as: ; The output gate is represented as: ; In the formula, represents the activation function; represents the input gate; represents the output gate; represents the forget gate; Indicates the hidden layer state at the previous moment; represents the model input data at time t, that is, the mode function and residual obtained by EMD decomposition of precipitation time series data at time t; Represent the weight matrix and bias of the input gate respectively; Represent the weight matrix and bias of the forget gate respectively; Represent the weight matrix and bias of the output gate respectively.

[0032] Furthermore, in the training of the pre-constructed LSTM prediction model in the embodiment of the present invention, the initial parameters of the LSTM prediction model are optimized and solved by the particle swarm optimization algorithm PSO; Therefore, in the training of the pre-constructed LSTM prediction model, the present invention optimizes the initial parameters of the LSTM prediction model through the particle swarm optimization algorithm PSO, thereby improving the generalization ability of the LSTM prediction model and enabling it to maintain a high prediction accuracy.

[0033] Preferably, in the disclosed embodiment, in the particle swarm optimization algorithm, the space dimension is set to D and the number of particles is N; The position of the i-th particle is ; The velocity of the i-th particle is ; The optimal position searched by the i-th particle is expressed as: ; All particles search for the global optimal position as: ; The formula for updating the position and velocity of the jth component of the ith particle is expressed as: ; ; In the formula, k represents the number of iterations; represents the inertia weight; represents the learning factor; Represents a random number.

[0034] As a preferred embodiment of the present disclosure, in the particle swarm optimization algorithm PSO, a linear decreasing weight strategy balancing algorithm is used to optimize the inertia weight to improve the global search capability and local search capability of the algorithm during the search process; wherein the linear decreasing weight strategy provided in the embodiment of the present disclosure is expressed as: ; In the formula, represents the maximum number of iterations; g represents the current number of iterations; represents the initial inertia weight; Represents the inertia weight when the iteration reaches the maximum number.

[0035] As a preferred embodiment of the present disclosure, in order to enhance the ability of particles to converge to the global optimal point, the particle swarm optimization algorithm PSO uses an asymmetric learning factor calculation formula to optimize the learning factor: ; ; In the formula, Indicates the maximum number of iterations; Respectively represent the final value and initial value of the individual learning factor of the particle; Respectively represent the termination value and initial value of the particle group learning factor; Among them, as a preferred embodiment, the termination value of the particle individual learning factor is 0.5; the initial value of the particle individual learning factor is 2.5; the termination value of the particle group learning factor is 2.25; and the initial value of the particle group learning factor is 1.

[0036] Preferably, in the embodiment of the present invention, in the step of predicting precipitation through a multi-input single-output sliding window, since the precipitation prediction law presents an annual periodicity, the sliding window is set to 12.

[0037] Please refer to Figure 2 In the disclosed embodiment, the step of analyzing the adaptability of the highway network infrastructure based on the predicted precipitation includes the following steps: S201: Assess the weather resistance of road network infrastructure based on the temporal variation and spatial distribution of predicted precipitation; In this embodiment, the daily maximum precipitation data in the historical scenes are arranged in ascending order, and the extreme precipitation threshold is determined based on the statistical results; then, the historical daily maximum predicted precipitation data are compared with the extreme precipitation threshold, the number of days when the precipitation exceeds the threshold is counted, and the roadbed of the highway network is evaluated when the precipitation exceeds the threshold; wherein, the deformation of the roadbed of the highway network is used as a measurement indicator, and the roadbed deformation result obtained when extreme precipitation occurs is calculated. Compared with the roadbed deformation results without extreme precipitation By comparison, the difference in deformation is obtained: ;in, It represents the difference in roadbed deformation under extreme precipitation conditions; k represents the climate change scenarios of precipitation in different periods; the deformation results are further divided into four levels, with amplitudes of 1 to 4 points from small to large.

[0038] S202: Evaluate the drainage performance of the road network infrastructure based on the road network infrastructure data, road network drainage structure parameters and predicted precipitation; In this embodiment, the collected highway network infrastructure data is integrated, and the impact of precipitation on highway infrastructure in future scenarios is simulated. The number of serious waterlogging on the road surface under the influence of precipitation is used as a measurement indicator to evaluate the drainage of the highway network infrastructure. The evaluation results are divided into four levels, with amplitudes of 1 to 4 points from small to large.

[0039] S203: Integrate the weather resistance assessment results and the drainage assessment results to obtain the road network climate change adaptability assessment results, including: ; In the formula, V represents the adaptability of the road network to climate change; E represents the weather resistance of the road network infrastructure; A represents the drainage performance of the road network infrastructure.

[0040] This embodiment takes a highway network infrastructure as an example to illustrate the adaptability of the highway network infrastructure under climate change, especially extreme precipitation conditions. First, the highway network infrastructure data is investigated, including collecting the history of the highway network; extracting the daily maximum precipitation data; then comparing the historical daily maximum predicted precipitation data with the precipitation threshold, and counting the number of days when the precipitation exceeds the threshold; and evaluating the highway network roadbed when the precipitation exceeds the threshold; wherein, the deformation of the highway network roadbed is used as a measurement indicator, and the roadbed deformation result obtained when extreme precipitation occurs is used as the measurement indicator. Compared with the roadbed deformation results without extreme precipitation Compare and get the difference of deformation, the difference is between 0-N, where 1 point is assigned to the days between [0, N / 4], 2 points to the days between [N / 4, N / 2], 3 points to the days between [N / 2, 3N / 4], and 4 points to the days between [3N / 4, N], and finally get the weather resistance evaluation result; Furthermore, based on the statistical results, the impact of precipitation on highway infrastructure in future scenarios was simulated, and the evaluation results were divided into four levels, with amplitudes of 1 to 4 points from small to large, using the number of serious waterlogging on the road surface under the influence of precipitation as a measurement indicator; Finally, the weather resistance evaluation results and drainage evaluation results are integrated according to the formula The climate change adaptability assessment results of the highway network are calculated.

[0041] In summary, the present invention analyzes the adaptability of the highway network infrastructure based on the predicted precipitation, which helps to identify and improve the potential weaknesses of the highway network infrastructure and improve the operational safety and efficiency of the highway network under extreme weather conditions, especially extreme rainfall conditions.

[0042] In addition, by predicting and analyzing future extreme rainfall conditions, the present invention can predict possible conditions in the highway network in advance and take a series of preventive measures to reduce the damage caused by extreme climate conditions to highway network infrastructure and highway operations and reduce the adverse effects of extreme weather.

[0043] Please refer to Figure 3 In another embodiment of the present disclosure, a road network climate change adaptability assessment device is provided, the assessment device comprising a data processing module 301, a climate prediction module 302 and an adaptability assessment module 303; wherein: The data processing module 301 is used to obtain precipitation time series data based on processing of original precipitation data in the target highway network area; The climate prediction module 302 is used to perform EMD decomposition on the precipitation time series data, use the obtained modal function and residual as the input of the pre-constructed LSTM prediction model, and predict the precipitation through a multi-input single-output sliding window; The adaptability evaluation module 303 is used to analyze the adaptability of the highway network infrastructure based on the predicted precipitation. The adaptability evaluation index includes the weather resistance and drainage capacity of the physical structure of the highway network infrastructure.

[0044] It should be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art may understand and implement it without paying any creative effort.

[0045] Those skilled in the art can clearly understand that for the sake of convenience and brevity, the specific working process of the device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0046] Another preferred embodiment of the present invention provides a computer device, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor implements a method for assessing the adaptability of a highway network to climate change as described in any one of the above embodiments when executing the computer program, wherein the assessment method comprises the following steps: obtaining precipitation time series data based on processing of original precipitation data of a target highway network area; performing EMD decomposition on the precipitation time series data, using the obtained modal function and residual as input of a pre-constructed LSTM prediction model, predicting precipitation through a multi-input single-output sliding window, wherein the prediction result includes the spatial distribution of precipitation in the target highway network area; analyzing the adaptability of the highway network infrastructure based on the predicted precipitation, wherein the evaluation indicators of the adaptability include the weather resistance and drainage capacity of the physical structure of the highway network infrastructure.

[0047] The computer device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc.; the identification device may include but is not limited to a processor and a memory.

[0048] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the terminal device, and uses various interfaces and lines to connect various parts of the entire terminal device.

[0049] The memory can be used to store the computer program, and the processor realizes various functions of the terminal device by running or executing the computer program stored in the memory and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required for a function, etc.; the data storage area can store data created according to the use of the mobile phone, etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0050] Another preferred embodiment of the present invention provides a storage medium, which is a computer-readable storage medium. The computer program is stored in the computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-mentioned method embodiments can be implemented.

[0051] The computer program includes computer program code, which may be in source code form, object code form, executable file or some intermediate form, etc. The computer readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc.

[0052] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for assessing the adaptability of a highway network to climate change, characterized in that: The assessment method includes the following steps: Based on the processing of the original precipitation data in the target highway network area, precipitation time series data is obtained; Perform EMD decomposition on precipitation time series data, use the obtained modal function and residual as input of a pre-constructed LSTM prediction model, and predict precipitation through a multi-input single-output sliding window. The prediction results include the spatial distribution of precipitation in the target highway network area. The adaptability of the road network infrastructure is analyzed based on the predicted precipitation, and the evaluation indicators of the adaptability include the weather resistance and drainage capacity of the physical structure of the road network infrastructure.

2. A method for evaluating the adaptability of a highway network to climate change according to claim 1, characterized in that: The step of obtaining precipitation time series data based on processing the original precipitation data of the target highway network area includes: normalizing and sequence segmenting the original precipitation data, taking a monthly precipitation time series data segment as a time step, and forming a continuous multiple months into an input sequence, and obtaining precipitation time series data within the target time period represented as: ,in, Represents the precipitation in the nth month.

3. A method for evaluating the adaptability of a highway network to climate change according to claim 2, characterized in that: The step of performing EMD decomposition on precipitation time series data includes: The precipitation time series data is decomposed into a finite number of intrinsic mode functions and residual terms, expressed as: ; In the formula, Represents the original precipitation time series data; It is represented as the i-th eigenmode function obtained by decomposition; represents the residual term; t represents time; n represents the number of decompositions.

4. A method for evaluating the adaptability of a highway network to climate change according to claim 3, characterized in that: The pre-constructed LSTM prediction model uses mean square error (MSE) as the loss function, and adds a penalty term for incorrect prediction of high precipitation values; Adam is selected as the optimizer, and the initial learning rate is set to 0.001; an early stopping strategy is implemented during training, and the patience value is set to 30 training cycles; the number of neural units in the hidden layer is set to 150, and a dropout layer is added, and the dropout rate is set to 0.

3.

5. A method for evaluating the adaptability of a highway network to climate change according to claim 4, characterized in that: In the training of the pre-constructed LSTM prediction model, the particle swarm optimization algorithm PSO is used to optimize the initial parameters of the LSTM prediction model; Wherein, in the particle swarm optimization algorithm, the space dimension is set to D and the number of particles is N; The position of the i-th particle is ; The velocity of the i-th particle is ; The optimal position searched by the i-th particle is expressed as: ; All particles search for the global optimal position as: ; The formula for updating the position and velocity of the jth component of the ith particle is expressed as: ; ; In the formula, k represents the number of iterations; represents the inertia weight; represents the learning factor; Represents a random number.

6. A method for evaluating the adaptability of a highway network to climate change according to claim 5, characterized in that: In the particle swarm optimization algorithm PSO, a linear decreasing weight strategy balancing algorithm is used to optimize the inertia weight to improve the global search ability and local search ability of the algorithm during the search process; The linear decreasing weight strategy is expressed as: ; In the formula, represents the maximum number of iterations; g represents the current number of iterations; represents the initial inertia weight; Represents the inertia weight when the iteration reaches the maximum number.

7. A method for evaluating the adaptability of a highway network to climate change according to claim 6, characterized in that: In the particle swarm optimization algorithm PSO, an asymmetric learning factor calculation formula is used to optimize the learning factor: ; ; In the formula, Indicates the maximum number of iterations; Respectively represent the final value and initial value of the individual learning factor of the particle; They represent the termination value and initial value of the particle group learning factor respectively.

8. A method for assessing the adaptability of a highway network to climate change according to claim 7, characterized in that: In the step of predicting precipitation through a multi-input single-output sliding window, the sliding window is set to 12.

9. A method for evaluating the adaptability of a highway network to climate change according to claim 8, characterized in that: The step of analyzing the adaptability of the road network infrastructure based on the predicted precipitation includes: Assess the weather resistance of road network infrastructure based on the predicted temporal and spatial distribution of precipitation; Evaluate the drainage of the road network infrastructure based on the road network infrastructure data, road network drainage structure parameters and predicted precipitation; The weather resistance assessment results and drainage assessment results were integrated to obtain the climate change adaptability assessment results of the highway network, including: ; In the formula, V represents the adaptability of the road network to climate change; E represents the weather resistance of the road network infrastructure; A represents the drainage performance of the road network infrastructure.

10. An evaluation device for implementing a method for evaluating the adaptability of a highway network to climate change as claimed in any one of claims 1 to 9, characterized in that: The evaluation device comprises: A data processing module, used for obtaining precipitation time series data based on processing of original precipitation data in the target highway network area; The climate prediction module is used to perform EMD decomposition on precipitation time series data, use the obtained modal function and residual as the input of the pre-constructed LSTM prediction model, and predict precipitation through a multi-input single-output sliding window; The adaptability assessment module is used to analyze the adaptability of the highway network infrastructure based on the predicted precipitation, and the adaptability assessment indicators include the weather resistance and drainage capacity of the physical structure of the highway network infrastructure.

Images