A method for analyzing the evolution of surface runoff in mountainous areas based on multivariate stereo information

Abstract

The present application relates to a method for analyzing the causes of surface runoff evolution in mountainous areas based on multivariate stereo information. The method includes: using artificial intelligence technology based on deep learning to perform fine-grained time series analysis on the temperature, precipitation, vegetation coverage, soil moisture and groundwater level data of the basin's historical period, respectively mining the dynamic evolution characteristics of each factor in the local time domain, and performing time series aggregation processing based on feature significance on the local time series feature representation of each factor to obtain the global time domain core feature representation of each factor, and then based on the global embedded joint features of each factor, intelligently identify the core causes of the evolution of surface runoff in mountainous areas. In this way, a multi-dimensional dynamic analysis of the evolution process of surface runoff in mountainous areas can be achieved, improving the prediction accuracy and reliability.

Description

Technical Field

[0001] The present application relates to the technical field of surface runoff evolution in mountainous areas, and in particular, to a method for analyzing the causes of surface runoff evolution in mountainous areas based on multivariate stereoscopic information. Background Art

[0002] Surface runoff in mountainous areas refers to the water flowing along the slope formed on the surface after atmospheric precipitation. It is an important component of runoff at the outlet section of the basin and plays a key role in flood formation, water resource allocation and ecological environment maintenance. However, the evolution of surface runoff in mountainous areas is affected by multiple factors such as topography, landform, climate, vegetation and soil. The interaction between these factors makes the runoff evolution process highly complex and uncertain.

[0003] Traditional methods for studying the evolution of surface runoff in mountainous areas mainly rely on empirical models and statistical analysis, which usually make it difficult to fully consider various influencing factors. When dealing with complex and changeable natural factors, they may not be able to fully consider the complex interactions between multiple factors, and ignore the nonlinear relationships and collinearity problems between factors, making it difficult to achieve a multi-dimensional dynamic analysis of runoff evolution.

[0004] Therefore, a method for analyzing the evolution causes of surface runoff in mountainous areas based on multivariate stereoscopic information is expected. Summary of the invention

[0005] This summary is provided to introduce concepts in a brief form that will be described in detail in the detailed description below. This summary is not intended to identify key features or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

[0006] In a first aspect, the present application provides a method for analyzing the evolution causes of surface runoff in mountainous areas based on multivariate stereoscopic information, the method comprising:

[0007] Obtain historical data of temperature, precipitation, vegetation coverage, soil moisture and groundwater level;

[0008] The historical data of the temperature value, the historical data of the precipitation, the historical data of the vegetation coverage rate, the historical data of the soil moisture and the historical data of the groundwater level are respectively subjected to time series feature extraction based on the local time scale to obtain a sequence of local time series implicit correlation feature vectors of the temperature, a sequence of local time series implicit correlation feature vectors of the precipitation, a sequence of local time series implicit correlation feature vectors of the vegetation coverage rate, a sequence of local time series implicit correlation feature vectors of the soil moisture and a sequence of local time series implicit correlation feature vectors of the groundwater level;

[0009] The sequence of the implicitly associated feature vectors of the local time series of the temperature, the sequence of the implicitly associated feature vectors of the local time series of the precipitation, the sequence of the implicitly associated feature vectors of the local time series of the vegetation coverage, the sequence of the implicitly associated feature vectors of the local time series of the soil moisture, and the sequence of the implicitly associated feature vectors of the local time series of the groundwater level are respectively subjected to information aggregation based on feature significance to obtain the temperature time series associated aggregation feature vector, the precipitation time series associated aggregation feature vector, the vegetation coverage time series associated aggregation feature vector, the soil moisture time series associated aggregation feature vector, and the groundwater level time series associated aggregation feature vector;

[0010] Performing multivariate information joint encoding on the temperature time series associated aggregation feature vector, the precipitation time series associated aggregation feature vector, the vegetation coverage time series associated aggregation feature vector, the soil moisture time series associated aggregation feature vector, and the groundwater level time series associated aggregation feature vector to obtain a multivariate stereo information fusion representation vector;

[0011] Based on the multivariate stereo information fusion representation vector, the analytical results of the causes of the evolution of surface runoff in mountainous areas are determined.

[0012] Optionally, the historical data of the temperature value, the historical data of the precipitation, the historical data of the vegetation coverage, the historical data of the soil moisture and the historical data of the groundwater level are respectively subjected to time series feature extraction based on the local time scale to obtain a sequence of local time series implicit correlation feature vectors of the temperature, a sequence of local time series implicit correlation feature vectors of the precipitation, a sequence of local time series implicit correlation feature vectors of the vegetation coverage, a sequence of local time series implicit correlation feature vectors of the soil moisture and a sequence of local time series implicit correlation feature vectors of the groundwater level, including: extracting the time series feature vectors of the temperature value based on the local time scale The historical data, the historical data of precipitation, the historical data of vegetation coverage, the historical data of soil moisture and the historical data of groundwater level are respectively segmented according to a predetermined time scale and input into a sequence encoder based on a Bi-LSTM model to obtain a sequence of local time series implicit correlation feature vectors of temperature, a sequence of local time series implicit correlation feature vectors of precipitation, a sequence of local time series implicit correlation feature vectors of vegetation coverage, a sequence of local time series implicit correlation feature vectors of soil moisture and a sequence of local time series implicit correlation feature vectors of groundwater level.

[0013] Optionally, the sequence of the implicitly associated feature vectors of the local time series of the temperature, the sequence of the implicitly associated feature vectors of the local time series of the precipitation, the sequence of the implicitly associated feature vectors of the local time series of the vegetation coverage, the sequence of the implicitly associated feature vectors of the local time series of the soil moisture, and the sequence of the implicitly associated feature vectors of the local time series of the groundwater level are respectively subjected to information aggregation based on feature significance to obtain the temperature time series associated aggregated feature vector, the precipitation time series associated aggregated feature vector, the vegetation coverage time series associated aggregated feature vector, the soil moisture time series associated aggregated feature vector, and the groundwater level time series associated aggregated feature vector, including: The sequence of local time series implicit association feature vectors, the sequence of local time series implicit association feature vectors of precipitation, the sequence of local time series implicit association feature vectors of vegetation coverage, the sequence of local time series implicit association feature vectors of soil moisture and the sequence of local time series implicit association feature vectors of groundwater level are respectively input into the feature aggregation network based on local feature significance transfer attenuation annotation to obtain the temperature time series associated aggregation feature vector, the precipitation time series associated aggregation feature vector, the vegetation coverage time series associated aggregation feature vector, the soil moisture time series associated aggregation feature vector and the groundwater level time series associated aggregation feature vector.

[0014] Optionally, the sequence of the local time series implicitly associated feature vectors of the air temperature, the sequence of the local time series implicitly associated feature vectors of the precipitation, the sequence of the local time series implicitly associated feature vectors of the vegetation coverage, the sequence of the local time series implicitly associated feature vectors of the soil moisture and the sequence of the local time series implicitly associated feature vectors of the groundwater level are respectively input into a feature aggregation network based on local feature significance transfer attenuation annotation to obtain the air temperature time series associated aggregation feature vector, the precipitation time series associated aggregation feature vector, the vegetation coverage time series associated aggregation feature vector, the soil moisture time series associated aggregation feature vector and the groundwater level time series associated aggregation feature vector, including: calculating the feature significance description factor of each local time series implicitly associated feature vector of the air temperature in the sequence of the local time series implicitly associated feature vector of the air temperature, wherein the feature significance description factor is related to the mean and variance of each local time series implicitly associated feature vector of the air temperature; using the local time series implicitly associated feature vector of the air temperature as the feature aggregation network; and obtaining the feature significance description factor of each local time series implicitly associated feature vector of the air temperature. The last temperature local time series implicit association feature vector in the sequence of feature vectors is used as the current temperature local time series implicit association feature vector; based on the distance span between each temperature local time series implicit association feature vector in the sequence of temperature local time series implicit association feature vectors and the current temperature local time series implicit association feature vector, the feature significance attenuation factor of each temperature local time series implicit association feature vector is constructed; based on the feature significance attenuation factor of each temperature local time series implicit association feature vector, the feature significance description factor of each temperature local time series implicit association feature vector is modulated to obtain a sequence of feature significance attenuation description factors; the sequence of feature significance attenuation description factors is input into the gated mask module to obtain a sequence of feature significance attenuation weight factors; with the sequence of feature significance attenuation weight factors as the sequence of weights, the weighted sum of the sequence of temperature local time series implicit association feature vectors is calculated to obtain the temperature time series associated aggregated feature vector.

[0015] Optionally, the characteristic significance description factor of each air temperature local time series implicit correlation feature vector in the sequence of the air temperature local time series implicit correlation feature vector is calculated, including: calculating the expected value of the fourth power of the difference between each eigenvalue in the air temperature local time series implicit correlation feature vector and its characteristic mean, and dividing the expected value by the square of the characteristic variance of the air temperature local time series implicit correlation feature vector to obtain the characteristic significance description factor.

[0016] Optionally, based on the distance span between each temperature local time series implicitly associated feature vector in the sequence of temperature local time series implicitly associated feature vectors and the current temperature local time series implicitly associated feature vector, a feature significance attenuation factor of each temperature local time series implicitly associated feature vector is constructed, including: calculating the difference between the maximum eigenvalue of the current temperature local time series implicitly associated feature vector and the maximum eigenvalue of each temperature local time series implicitly associated feature vector in the sequence of temperature local time series implicitly associated feature vectors, and then dividing the difference by the number of eigenvectors separated therefrom to obtain the feature significance attenuation factor of each temperature local time series implicitly associated feature vector.

[0017] Optionally, based on the feature significance attenuation factor of each air temperature local time series implicitly associated feature vector, the feature significance description factor of each air temperature local time series implicitly associated feature vector is modulated to obtain a sequence of feature significance attenuation description factors, including: calculating the product between the feature significance attenuation factor of each air temperature local time series implicitly associated feature vector and its feature significance description factor to obtain the sequence of feature significance attenuation description factors.

[0018] Optionally, the temperature time series associated aggregated feature vector, the precipitation time series associated aggregated feature vector, the vegetation coverage rate time series associated aggregated feature vector, the soil moisture time series associated aggregated feature vector and the groundwater level time series associated aggregated feature vector are subjected to multivariate information joint encoding to obtain a multivariate stereo information fusion representation vector, including: inputting the temperature time series associated aggregated feature vector, the precipitation time series associated aggregated feature vector, the vegetation coverage rate time series associated aggregated feature vector, the soil moisture time series associated aggregated feature vector and the groundwater level time series associated aggregated feature vector into a multivariate stereo information joint encoder based on an embedding layer to obtain the multivariate stereo information fusion representation vector.

[0019] Optionally, based on the multivariate stereo information fusion representation vector, an analysis result of the cause of evolution of surface runoff in mountainous areas is determined, including: inputting the multivariate stereo information fusion representation vector into a classifier-based analysis module of the cause of evolution of surface runoff in mountainous areas to obtain the analysis result, and the analysis result is used to represent the type label of the core cause of evolution of surface runoff in mountainous areas.

[0020] The above technical solution uses deep learning-based artificial intelligence technology to conduct fine-grained time series analysis on the temperature, precipitation, vegetation coverage, soil moisture and groundwater level data of the basin in the historical period, respectively mining the dynamic evolution characteristics of each factor in the local time domain, and performing time series aggregation based on feature significance on the local time series feature representation of each factor to obtain the global time domain core feature representation of each factor, and then based on the global embedded joint features of each factor, intelligently identify the core causes of the evolution of surface runoff in mountainous areas. In this way, a multi-dimensional dynamic analysis of the evolution process of surface runoff in mountainous areas can be achieved, improving the prediction accuracy and reliability.

[0021] Other features and advantages of the present application will be described in detail in the subsequent specific implementation section. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The above and other features, advantages and aspects of the embodiments of the present application will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale. In the drawings:

[0023] Figure 1 It is a flow chart of a method for analyzing the evolution causes of surface runoff in mountainous areas based on multivariate stereo information according to an exemplary embodiment.

[0024] Figure 2 It is a schematic diagram of the evolution of meteorological and hydrological underlying surface elements in a watershed according to an exemplary embodiment.

[0025] Figure 3 It is a block diagram of a system for analyzing the evolution causes of surface runoff in mountainous areas based on multivariate stereo information according to an exemplary embodiment.

[0026] Figure 4 It is a block diagram of an electronic device according to an exemplary embodiment.

[0027] Figure 5 It is an application scenario diagram of a method for analyzing the evolution causes of surface runoff in mountainous areas based on multivariate stereo information according to an exemplary embodiment. DETAILED DESCRIPTION

[0028] The embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be construed as being limited to the embodiments described herein. Instead, these embodiments are provided to provide a more thorough and complete understanding of the present application. It should be understood that the drawings and embodiments of the present application are only for exemplary purposes and are not intended to limit the scope of protection of the present application.

[0029] It should be understood that the various steps described in the method implementation of the present application can be performed in different orders and / or performed in parallel. In addition, the method implementation may include additional steps and / or omit the steps shown. The scope of the present application is not limited in this respect.

[0030] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0031] It should be noted that the concepts such as "first" and "second" mentioned in this application are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0032] It should be noted that the modifications of "one" and "plurality" mentioned in the present application are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0033] The names of the messages or information exchanged between multiple devices in the embodiments of the present application are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0034] The specific implementation of the present application is described in detail below with reference to the accompanying drawings.

[0035] Figure 1 is a flow chart of a method for analyzing the evolution causes of surface runoff in mountainous areas based on multivariate stereo information according to an exemplary embodiment. Figure 1 As shown, the method includes:

[0036] Step S101, obtaining historical data of temperature values, historical data of precipitation, historical data of vegetation coverage, historical data of soil moisture and historical data of groundwater level;

[0037] Step S102, performing time series feature extraction based on local time scale on the historical data of the temperature value, the historical data of the precipitation, the historical data of the vegetation coverage rate, the historical data of the soil moisture and the historical data of the groundwater level respectively to obtain a sequence of local time series implicit correlation feature vectors of the temperature, a sequence of local time series implicit correlation feature vectors of the precipitation, a sequence of local time series implicit correlation feature vectors of the vegetation coverage rate, a sequence of local time series implicit correlation feature vectors of the soil moisture and a sequence of local time series implicit correlation feature vectors of the groundwater level;

[0038] Step S103, respectively performing information aggregation based on feature significance on the sequence of the local time series implicitly associated feature vectors of the temperature, the sequence of the local time series implicitly associated feature vectors of the precipitation, the sequence of the local time series implicitly associated feature vectors of the vegetation coverage, the sequence of the local time series implicitly associated feature vectors of the soil moisture, and the sequence of the local time series implicitly associated feature vectors of the groundwater level to obtain the temperature time series associated aggregated feature vector, the precipitation time series associated aggregated feature vector, the vegetation coverage time series associated aggregated feature vector, the soil moisture time series associated aggregated feature vector, and the groundwater level time series associated aggregated feature vector;

[0039] Step S104, performing multivariate information joint encoding on the temperature time series associated aggregation feature vector, the precipitation time series associated aggregation feature vector, the vegetation coverage time series associated aggregation feature vector, the soil moisture time series associated aggregation feature vector, and the groundwater level time series associated aggregation feature vector to obtain a multivariate stereo information fusion representation vector;

[0040] Step S105: Determine the analytical result of the evolution cause of surface runoff in the mountainous area based on the multivariate stereo information fusion representation vector.

[0041] In response to the above technical problems, the technical concept of this application is to use artificial intelligence technology based on deep learning to conduct fine-grained time series analysis on the temperature, precipitation, vegetation coverage, soil moisture and groundwater level data of the basin in the historical period, respectively dig out the dynamic evolution characteristics of each factor in the local time domain, and perform time series aggregation processing based on feature significance on the local time series feature representation of each factor to obtain the global time domain core feature representation of each factor, and then intelligently identify the core causes of the evolution of surface runoff in mountainous areas based on the global embedded joint features of each factor. In this way, a multi-dimensional dynamic analysis of the evolution process of surface runoff in mountainous areas can be achieved, and the prediction accuracy and reliability can be improved.

[0042] In one embodiment of the present application, Figure 2As shown in the figure, the evolution of the meteorological and hydrological underlying surface elements in the basin includes meteorological elements, underlying surface elements and hydrological elements. Among them, meteorological elements include climate warming and precipitation homogenization; underlying surface elements include vegetation greening; hydrological elements include soil drying and groundwater cavitation, which together form the natural runoff changes.

[0043] In the technical solution of the present application, firstly, historical data of temperature, historical data of precipitation, historical data of vegetation coverage, historical data of soil moisture and historical data of groundwater level are obtained. It should be understood that the evolution of surface runoff in mountainous areas is jointly affected by a variety of environmental factors, which are interrelated and interact with each other. Specifically, temperature affects the transpiration of vegetation, thereby affecting soil moisture and groundwater recharge; precipitation is one of the main sources of surface runoff and directly affects the amount of surface runoff; vegetation regulates surface runoff by intercepting precipitation, slowing surface runoff, increasing soil moisture infiltration and reducing evaporation. Areas with high vegetation coverage usually have better soil and water conservation capabilities and can reduce the formation of surface runoff; soil moisture is a key factor in the conversion between surface runoff and groundwater. Wet soil will increase water infiltration and reduce surface runoff, while dry soil may cause more surface water to directly form runoff; and the height of the groundwater level reflects the contribution of groundwater to surface runoff. Rising groundwater levels may lead to an increase in surface water bodies, while a decrease in groundwater levels may reduce the recharge of surface water bodies. These environmental factors are the core elements in the surface water cycle and directly affect the generation, confluence and distribution of surface runoff. Therefore, in the technical solution of the present application, a multivariate three-dimensional information input is constructed using the temperature, precipitation, vegetation coverage, soil moisture and groundwater level data of the basin's historical period, and a deep analysis of the causes of the evolution of surface runoff in mountainous areas is achieved by comprehensively considering various environmental factors that affect the evolution of surface runoff in mountainous areas.

[0044] Considering that the time span of the historical data of the above-mentioned environmental factors is large, directly processing the data may lead to an increase in computational complexity and the concealment of local key information. Therefore, in the technical solution of the present application, in order to capture the short-term change trend of each environmental factor in a finer granularity, the historical data of the temperature value, the historical data of the precipitation, the historical data of the vegetation coverage, the historical data of the soil moisture and the historical data of the groundwater level are first segmented at a predetermined time scale, and the long time series data is divided into multiple shorter subsequences, and then the sequence encoder based on the Bi-LSTM model is used to perform time series encoding on each segmented subsequence data, and the bidirectional learning ability of the Bi-LSTM model is used, while considering the forward and backward time dependencies in the sequence data, so as to effectively capture the dynamic evolution information of each environmental factor in the local time domain, and generate a sequence of implicitly associated feature vectors of the local time series of temperature, a sequence of implicitly associated feature vectors of the local time series of precipitation, a sequence of implicitly associated feature vectors of the local time series of vegetation coverage, a sequence of implicitly associated feature vectors of the local time series of soil moisture and a sequence of implicitly associated feature vectors of the local time series of groundwater level.

[0045] In one embodiment of the present application, the historical data of the temperature value, the historical data of the precipitation, the historical data of the vegetation coverage, the historical data of the soil moisture and the historical data of the groundwater level are respectively subjected to time series feature extraction based on the local time scale to obtain a sequence of local time series implicit correlation feature vectors of the temperature, a sequence of local time series implicit correlation feature vectors of the precipitation, a sequence of local time series implicit correlation feature vectors of the vegetation coverage, a sequence of local time series implicit correlation feature vectors of the soil moisture and a sequence of local time series implicit correlation feature vectors of the groundwater level, including: extracting the time series feature vectors based on the local time scale The historical data of air temperature values, the historical data of precipitation, the historical data of vegetation coverage, the historical data of soil moisture and the historical data of groundwater level are respectively segmented according to a predetermined time scale and input into a sequence encoder based on a Bi-LSTM model to obtain a sequence of local time series implicit correlation feature vectors of air temperature, a sequence of local time series implicit correlation feature vectors of precipitation, a sequence of local time series implicit correlation feature vectors of vegetation coverage, a sequence of local time series implicit correlation feature vectors of soil moisture and a sequence of local time series implicit correlation feature vectors of groundwater level.

[0046] Next, in order to fully understand the dynamic evolution characteristics of various environmental factors on the global time scale, the local time series feature representations of various environmental factors are further time-series aggregated to form a global time series feature representation. In particular, considering that in the process of surface runoff evolution in mountainous areas, environmental factors in different periods may have different impact intensities on the evolution of surface runoff in mountainous areas. Based on this, the present application introduces a feature aggregation network based on local feature significance transfer attenuation annotation to aggregate the sequence of the local time series implicitly associated feature vectors of the temperature, the sequence of the local time series implicitly associated feature vectors of the precipitation, the sequence of the local time series implicitly associated feature vectors of the vegetation coverage, the sequence of the local time series implicitly associated feature vectors of the soil moisture, and the sequence of the local time series implicitly associated feature vectors of the groundwater level, respectively, and dynamically adjust the feature weights of different time steps to highlight the key time series features, weaken the non-key time series features, and improve the significance and relevance of the feature representation. Specifically, the feature aggregation network first calculates the feature significance descriptor of each local feature vector based on the mean and variance of each local feature vector in the input sequence, providing an important priority indicator for subsequent feature selection and aggregation. Here, the mean reflects the central position of the feature value, while the variance represents the degree of dispersion of the feature value, and the two together determine the significance of the feature. Next, a feature significance attenuation factor is constructed based on the distance span between each local feature vector and the current local feature vector (the end of the time series), so that feature vectors that are closer to the current environmental state in the feature space receive a higher weight, thereby strengthening the relevance of local features. Furthermore, the feature significance attenuation factor of each local feature vector is applied to its feature significance descriptor to modulate it, so that it more accurately reflects the relative importance of each local feature in the overall sequence. Then, the modulated descriptive factors are further screened using activation functions and gating mechanisms, and the input sequence is weighted summed with the screened descriptive factors as weights to integrate the characteristic information of each local time domain in the input sequence and generate the temperature time series associated aggregation feature vector, precipitation time series associated aggregation feature vector, vegetation coverage time series associated aggregation feature vector, soil moisture time series associated aggregation feature vector and groundwater level time series associated aggregation feature vector.

[0047] In one embodiment of the present application, the sequence of the local time series implicitly associated feature vectors of the temperature, the sequence of the local time series implicitly associated feature vectors of the precipitation, the sequence of the local time series implicitly associated feature vectors of the vegetation coverage, the sequence of the local time series implicitly associated feature vectors of the soil moisture, and the sequence of the local time series implicitly associated feature vectors of the groundwater level are respectively subjected to information aggregation based on feature significance to obtain the temperature time series associated aggregated feature vector, the precipitation time series associated aggregated feature vector, the vegetation coverage time series associated aggregated feature vector, the soil moisture time series associated aggregated feature vector, and the groundwater level time series associated aggregated feature vector, including: The sequence of the local time series implicitly associated feature vectors of the temperature, the sequence of the local time series implicitly associated feature vectors of the precipitation, the sequence of the local time series implicitly associated feature vectors of the vegetation coverage rate, the sequence of the local time series implicitly associated feature vectors of the soil moisture, and the sequence of the local time series implicitly associated feature vectors of the groundwater level are respectively input into a feature aggregation network based on local feature significance transfer attenuation annotation to obtain the temperature time series associated aggregation feature vector, the precipitation time series associated aggregation feature vector, the vegetation coverage rate time series associated aggregation feature vector, the soil moisture time series associated aggregation feature vector, and the groundwater level time series associated aggregation feature vector.

[0048] Furthermore, in one embodiment of the present application, the sequence of the local time series implicitly associated feature vectors of the temperature, the sequence of the local time series implicitly associated feature vectors of the precipitation, the sequence of the local time series implicitly associated feature vectors of the vegetation coverage, the sequence of the local time series implicitly associated feature vectors of the soil moisture, and the sequence of the local time series implicitly associated feature vectors of the groundwater level are respectively input into a feature aggregation network based on local feature significance transfer attenuation annotation to obtain the temperature time series associated aggregation feature vector, the precipitation time series associated aggregation feature vector, the vegetation coverage time series associated aggregation feature vector, the soil moisture time series associated aggregation feature vector, and the groundwater level time series associated aggregation feature vector, including: calculating the feature significance description factor of each local time series implicitly associated feature vector of the temperature in the sequence of the local time series implicitly associated feature vector of the temperature, wherein the feature significance description factor is related to the mean and variance of each local time series implicitly associated feature vector of the temperature; using the local time series implicitly associated feature vector of the temperature as the feature aggregation network; and calculating the feature significance description factor of each local time series implicitly associated feature vector of the temperature as the feature aggregation network. The last temperature local time series implicit association feature vector in the sequence of the temperature local time series implicit association feature vectors is taken as the current temperature local time series implicit association feature vector; based on the distance span between each temperature local time series implicit association feature vector in the sequence of the temperature local time series implicit association feature vectors and the current temperature local time series implicit association feature vector, the feature significance attenuation factor of each temperature local time series implicit association feature vector is constructed; based on the feature significance attenuation factor of each temperature local time series implicit association feature vector, the feature significance description factor of each temperature local time series implicit association feature vector is modulated to obtain a sequence of feature significance attenuation description factors; the sequence of feature significance attenuation description factors is input into the gated mask module to obtain a sequence of feature significance attenuation weight factors; with the sequence of feature significance attenuation weight factors as the sequence of weights, the weighted sum of the sequence of the temperature local time series implicit association feature vectors is calculated to obtain the temperature time series association aggregation feature vector.

[0049] Furthermore, in one embodiment of the present application, the characteristic significance description factor of each air temperature local time series implicit correlation feature vector in the sequence of the air temperature local time series implicit correlation feature vector is calculated, including: calculating the expected value of the fourth power of the difference between each eigenvalue in the air temperature local time series implicit correlation feature vector and its characteristic mean, and dividing the expected value by the square of the characteristic variance of the air temperature local time series implicit correlation feature vector to obtain the characteristic significance description factor.

[0050] Furthermore, in one embodiment of the present application, based on the distance span between each temperature local time series implicitly associated feature vector in the sequence of the temperature local time series implicitly associated feature vector and the current temperature local time series implicitly associated feature vector, a feature significance attenuation factor of each temperature local time series implicitly associated feature vector is constructed, including: calculating the difference between the maximum eigenvalue of the current temperature local time series implicitly associated feature vector and the maximum eigenvalue of each temperature local time series implicitly associated feature vector in the sequence of the temperature local time series implicitly associated feature vector, and then dividing the difference by the number of eigenvectors separated therefrom to obtain the feature significance attenuation factor of each temperature local time series implicitly associated feature vector.

[0051] Furthermore, in one embodiment of the present application, the feature significance description factor of each air temperature local time series implicitly associated feature vector is modulated based on the feature significance attenuation factor of each air temperature local time series implicitly associated feature vector to obtain a sequence of feature significance attenuation description factors, including: calculating the product between the feature significance attenuation factor of each air temperature local time series implicitly associated feature vector and its feature significance description factor to obtain the sequence of feature significance attenuation description factors.

[0052] Specifically, the sequence of the implicit correlation feature vectors of the local temperature time series is processed by the following feature significance attenuation guided fusion formula to obtain the temperature time series associated aggregated feature vector, wherein the feature significance attenuation guided fusion formula is:

[0053]

[0054]

[0055]

[0056]

[0057]

[0058]

[0059]

[0060] in, represents the sequence of implicitly associated feature vectors of the local time series of temperature, The value of is the number of eigenvectors in the sequence of implicitly associated eigenvectors of the local time series of temperature. , , , and They represent the first, second, and third in the sequence of implicit correlation feature vectors of the local temperature time series. , and the current local temperature time series implicitly associated feature vector, represents the maximum eigenvalue of the eigenvector, Indicates the The first implicit correlation feature vector of the local temperature time series eigenvalues, and Respectively represent the The characteristic mean and square of the characteristic variance of the implicit correlation characteristic vector of the local temperature time series, Indicates the The fourth-order central moment of the implicit correlation eigenvector of the local temperature time series, that is, The expected value of the fourth power of the difference between each eigenvalue and its eigenmean in the implicit correlation eigenvector of the local temperature time series, Indicates the The characteristic significance attenuation factor of the implicit correlation eigenvector of the local temperature time series, Indicates The feature significance decay factor is described by Indicates Normalized feature significance decay description factor, Indicates feature significance decay weight factor, is the preset threshold, Indicates mask processing, Represents the temperature time series associated aggregation feature vector.

[0061] Then, in order to better capture the complex interactive relationship between environmental factors and consider the comprehensive impact of various environmental factors on the evolution of surface runoff in mountainous areas, the present application further adopts a multivariate stereo information joint encoder based on an embedding layer to jointly encode the temperature time series correlation aggregation feature vector, the precipitation time series correlation aggregation feature vector, the vegetation coverage time series correlation aggregation feature vector, the soil moisture time series correlation aggregation feature vector and the groundwater level time series correlation aggregation feature vector. Through the embedding layer, the characteristics of various environmental elements from different sources and dimensions are mapped into a unified representation space to capture the intrinsic connection and mutual influence between environmental factors, realize the integration of multivariate stereo information, and generate a multivariate stereo information fusion representation vector, thereby more comprehensively reflecting the complex driving mechanism of the evolution of surface runoff in mountainous areas.

[0062] In one embodiment of the present application, multi-information joint encoding is performed on the air temperature time series associated aggregated feature vector, the precipitation time series associated aggregated feature vector, the vegetation coverage rate time series associated aggregated feature vector, the soil moisture time series associated aggregated feature vector and the groundwater level time series associated aggregated feature vector to obtain a multi-information fusion representation vector, including: inputting the air temperature time series associated aggregated feature vector, the precipitation time series associated aggregated feature vector, the vegetation coverage rate time series associated aggregated feature vector, the soil moisture time series associated aggregated feature vector and the groundwater level time series associated aggregated feature vector into a multi-information joint encoder based on an embedding layer to obtain the multi-information fusion representation vector.

[0063] Finally, the multivariate stereo information fusion representation vector is input into the classifier-based mountain surface runoff evolution cause analysis module to obtain the analysis result, and the analysis result is used to represent the type label of the core cause of the evolution of mountain surface runoff. In the technical solution of the present application, the classifier-based mountain surface runoff evolution cause analysis module can output the corresponding core cause type label of surface runoff evolution in real time, such as climate and meteorology, vegetation coverage, soil state, etc., by learning and mining the deep structure and characteristic patterns in the multivariate stereo information fusion representation vector, so that corresponding water resources management strategies and ecological protection measures can be formulated according to the analysis results. For example, if the analysis results show that climate and meteorology are the main causes, it may be necessary to predict future climate change trends in order to formulate adaptive water resources allocation plans; if vegetation coverage is the key factor, it may be necessary to implement vegetation restoration projects to enhance the water conservation capacity of the basin.

[0064] Here, the temperature time series associated aggregation feature vector, the precipitation time series associated aggregation feature vector, the vegetation coverage rate time series associated aggregation feature vector, the soil moisture time series associated aggregation feature vector and the groundwater level time series associated aggregation feature vector respectively represent the time series significant aggregation features of temperature value, precipitation, vegetation coverage rate, soil moisture and groundwater level. Considering that the significant differences in the source domain time series distribution of the temperature value, precipitation, vegetation coverage rate, soil moisture and groundwater level will cause the time series distribution of the temperature time series associated aggregation feature vector, the precipitation time series associated aggregation feature vector, the vegetation coverage rate time series associated aggregation feature vector, the soil moisture time series associated aggregation feature vector and the groundwater level time series associated aggregation feature vector in the feature domain to be significantly different, this causes the fine-grained embedding aggregation offset of different modal time series features when performing joint encoding of multivariate stereo information based on the embedding layer, resulting in the obtained multivariate stereo information fusion representation vector having a complex spatial structure with significant aggregation of multivariate information. Therefore, it is expected to improve its classification regression convergence and generalization effect under complex aggregation spatial structure.

[0065] Preferably, in an example of the present application, inputting the multivariate stereo information fusion representation vector into a classifier-based mountain surface runoff evolution factor analysis module to obtain an analysis result includes:

[0066] Calculating the sum of the absolute values ​​of each eigenvalue of the multivariate stereo information fusion representation vector to obtain a first multivariate stereo information fusion spatial structure value, and calculating the square root of the sum of the squares of each eigenvalue of the multivariate stereo information fusion representation vector to obtain a second multivariate stereo information fusion spatial structure value;

[0067] Multiplying each eigenvalue of the multivariate stereo information fusion representation vector by the first multivariate stereo information fusion spatial structure value and the second multivariate stereo information fusion spatial structure value respectively to obtain a first multivariate stereo information fusion structure reference value and a second multivariate stereo information fusion structure reference value corresponding to each eigenvalue;

[0068] Multiplying each eigenvalue of the multivariate stereo information fusion representation vector by the length of the multivariate stereo information fusion representation vector and the square root of the length to obtain a first multivariate stereo information fusion scale transformation value and a second multivariate stereo information fusion scale transformation value corresponding to each eigenvalue;

[0069] Dividing the first multivariate stereo information fusion structure reference value by the difference between the first multivariate stereo information fusion spatial structure value and the first multivariate stereo information fusion scale transformation value to obtain a first multivariate stereo information fusion transformation adjustment value;

[0070] Dividing the second multivariate stereo information fusion structure reference value by the difference between the second multivariate stereo information fusion spatial structure value and the second multivariate stereo information fusion scale transformation value to obtain a second multivariate stereo information fusion transformation adjustment value;

[0071] Calculating a weighted sum of the first multivariate stereo information fusion transformation adjustment value and the second multivariate stereo information fusion transformation adjustment value to obtain each eigenvalue of the optimized multivariate stereo information fusion representation vector;

[0072] The optimized multivariate stereo information fusion representation vector is input into a classifier-based mountain surface runoff evolution cause analysis module to obtain an analysis result.

[0073] Here, the multivariate stereo information fusion representation vector is recorded as The optimization expression is:

[0074]

[0075]

[0076]

[0077]

[0078]

[0079]

[0080]

[0081]

[0082] in, represents the first eigenvalues, represents the length of the multivariate stereo information fusion representation vector, represents the first multivariate stereo information fusion spatial structure value, represents the second multivariate stereo information fusion spatial structure value, It means point multiplication by position. It means adding by position. represents the first The first multivariate stereo information fusion transformation adjustment value corresponding to the eigenvalue, is a first multivariate stereo information fusion transformation adjustment vector composed of the first multivariate stereo information fusion transformation adjustment values, represents the first The second multivariate stereo information fusion transformation adjustment value corresponding to the eigenvalue, is a second multivariate stereo information fusion transformation adjustment vector composed of each of the second multivariate stereo information fusion transformation adjustment values, is the weight hyperparameter, Represents the optimized multivariate stereo information fusion representation vector.

[0083] Therefore, in the preferred example, for the spatial structure information of the feature set of the multivariate stereo information fusion representation vector in the high-dimensional space, the scale-based frame transformation of each eigenvalue of the multivariate stereo information fusion representation vector is performed by taking the class-norm spatial structured representation of the multivariate stereo information fusion representation vector as a reference window, and the spatial structure-based frame attention weight adjustment of each eigenvalue of the multivariate stereo information fusion representation vector is implemented to ensure the invariance of the spatial transformation (translation, scaling and rotation) of the multivariate stereo information fusion representation vector under the interaction of feature space, thereby improving the convergence and generalization effect of classification regression of the feature set of the multivariate stereo information fusion representation vector under complex spatial structure representation, and improving the accuracy of the analysis result obtained by transforming the surface runoff in mountainous area into the cause analysis module based on the classifier.

[0084] In summary, the above scheme uses artificial intelligence technology based on deep learning to conduct fine-grained time series analysis on the temperature, precipitation, vegetation coverage, soil moisture and groundwater level data of the basin in the historical period, respectively mining the dynamic evolution characteristics of each factor in the local time domain, and performing time series aggregation processing based on feature significance on the local time series feature representation of each factor to obtain the global time domain core feature representation of each factor, and then based on the global embedded joint features of each factor, intelligently identify the core causes of the evolution of surface runoff in mountainous areas. In this way, a multi-dimensional dynamic analysis of the evolution process of surface runoff in mountainous areas can be achieved, improving the prediction accuracy and reliability.

[0085] Figure 3 FIG. 1 is a block diagram of a system for analyzing the evolution of surface runoff in mountainous areas based on multivariate stereo information according to an exemplary embodiment. Figure 3 As shown, the system 200 includes:

[0086] The data acquisition module 201 is used to acquire historical data of temperature values, historical data of precipitation, historical data of vegetation coverage, historical data of soil moisture and historical data of groundwater level;

[0087] The time series feature extraction module 202 is used to extract time series features based on local time scales for the historical data of the temperature value, the historical data of the precipitation, the historical data of the vegetation coverage rate, the historical data of the soil moisture, and the historical data of the groundwater level, respectively, to obtain a sequence of local time series implicit correlation feature vectors of the temperature, a sequence of local time series implicit correlation feature vectors of the precipitation, a sequence of local time series implicit correlation feature vectors of the vegetation coverage rate, a sequence of local time series implicit correlation feature vectors of the soil moisture, and a sequence of local time series implicit correlation feature vectors of the groundwater level;

[0088] The information aggregation module 203 is used to perform information aggregation based on feature significance on the sequence of the local time series implicitly associated feature vectors of the temperature, the sequence of the local time series implicitly associated feature vectors of the precipitation, the sequence of the local time series implicitly associated feature vectors of the vegetation coverage, the sequence of the local time series implicitly associated feature vectors of the soil moisture, and the sequence of the local time series implicitly associated feature vectors of the groundwater level, respectively, to obtain the temperature time series associated aggregation feature vector, the precipitation time series associated aggregation feature vector, the vegetation coverage time series associated aggregation feature vector, the soil moisture time series associated aggregation feature vector, and the groundwater level time series associated aggregation feature vector;

[0089] A multivariate information joint encoding module 204 is used to perform multivariate information joint encoding on the temperature time series associated aggregation feature vector, the precipitation time series associated aggregation feature vector, the vegetation coverage time series associated aggregation feature vector, the soil moisture time series associated aggregation feature vector and the groundwater level time series associated aggregation feature vector to obtain a multivariate stereo information fusion representation vector;

[0090] The analysis result determination module 205 of the evolution cause of surface runoff in mountainous area is used to determine the analysis result of the evolution cause of surface runoff in mountainous area based on the multivariate stereo information fusion representation vector.

[0091] Reference below Figure 4 , which shows a schematic diagram of the structure of an electronic device 600 suitable for implementing the embodiment of the present application. The terminal device in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 4 The electronic device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0092] like Figure 4 As shown, the electronic device 600 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage device 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the electronic device 600 are also stored. The processing device 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0093] Typically, the following devices may be connected to the I / O interface 605: an input device 606 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 608 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 609. The communication device 609 may allow the electronic device 600 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 4 The electronic device 600 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0094] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device 609, or installed from the storage device 608, or installed from the ROM 602. When the computer program is executed by the processing device 601, the above-mentioned functions defined in the method of the embodiment of the present application are executed.

[0095] It should be noted that the computer-readable medium mentioned above in the present application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0096] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0097] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0098] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including, but not limited to, object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0099] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0100] The modules involved in the embodiments described in this application can be implemented by software or hardware. The name of the module does not limit the module itself in some cases. For example, the test parameter acquisition module can also be described as a "module for acquiring device test parameters corresponding to the target device".

[0101] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0102] In the context of the present application, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0103] Figure 5 FIG. 1 is an application scenario diagram of a method for analyzing the evolution of surface runoff in mountainous areas based on multivariate stereo information according to an exemplary embodiment. Figure 5 As shown, in this application scenario, first, obtain the historical data of temperature values ​​(for example, Figure 5 C1 shown in the figure), historical data on precipitation (e.g. Figure 5 C2 as shown in the figure), historical data on vegetation cover (e.g. Figure 5 C3 as shown in the figure), historical data on soil moisture (e.g. Figure 5 C4 as shown in the figure) and historical data on groundwater levels (e.g. Figure 5 Then, the acquired historical data of temperature, precipitation, vegetation coverage, soil moisture and groundwater level are input into a server (for example, Figure 5 In S) shown in the figure, the server can process the historical data of temperature values, historical data of precipitation, historical data of vegetation coverage, historical data of soil moisture and historical data of groundwater level based on the analytical algorithm of the evolution causes of surface runoff in mountainous areas based on multivariate stereoscopic information to determine the analytical results of the evolution causes of surface runoff in mountainous areas.

[0104] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in this application (but not limited to) by each other to form a technical solution.

[0105] In addition, although each operation is described in a specific order, this should not be construed as requiring these operations to be performed in the specific order shown or to be performed in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details are included in the above discussion, these should not be construed as limiting the scope of the application. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

[0106] Although the subject matter has been described in language specific to structural features and / or method logic actions, it should be understood that the subject matter defined is not necessarily limited to the specific features or actions described above. On the contrary, the specific features and actions described above are merely exemplary forms of implementation. With respect to the apparatus in the above-mentioned embodiments, the specific manner in which each module performs the operation has been described in detail in the embodiments related to the method, and will not be described in detail here.

Claims

1. A method for analyzing the evolution of surface runoff in mountainous areas based on multivariate stereo information, characterized in that: include: Obtain historical data of temperature, precipitation, vegetation coverage, soil moisture and groundwater level; The historical data of the temperature value, the historical data of the precipitation, the historical data of the vegetation coverage rate, the historical data of the soil moisture and the historical data of the groundwater level are respectively subjected to time series feature extraction based on the local time scale to obtain a sequence of local time series implicit correlation feature vectors of the temperature, a sequence of local time series implicit correlation feature vectors of the precipitation, a sequence of local time series implicit correlation feature vectors of the vegetation coverage rate, a sequence of local time series implicit correlation feature vectors of the soil moisture and a sequence of local time series implicit correlation feature vectors of the groundwater level; The sequence of the implicitly associated feature vectors of the local time series of the temperature, the sequence of the implicitly associated feature vectors of the local time series of the precipitation, the sequence of the implicitly associated feature vectors of the local time series of the vegetation coverage, the sequence of the implicitly associated feature vectors of the local time series of the soil moisture, and the sequence of the implicitly associated feature vectors of the local time series of the groundwater level are respectively subjected to information aggregation based on feature significance to obtain the temperature time series associated aggregation feature vector, the precipitation time series associated aggregation feature vector, the vegetation coverage time series associated aggregation feature vector, the soil moisture time series associated aggregation feature vector, and the groundwater level time series associated aggregation feature vector; Performing multivariate information joint encoding on the temperature time series associated aggregation feature vector, the precipitation time series associated aggregation feature vector, the vegetation coverage time series associated aggregation feature vector, the soil moisture time series associated aggregation feature vector, and the groundwater level time series associated aggregation feature vector to obtain a multivariate stereo information fusion representation vector; Determine the analytical results of the evolution causes of surface runoff in mountainous areas based on the multivariate stereo information fusion representation vector; The historical data of the temperature value, the historical data of the precipitation, the historical data of the vegetation coverage, the historical data of the soil moisture and the historical data of the groundwater level are respectively subjected to time series feature extraction based on the local time scale to obtain a sequence of local time series implicit correlation feature vectors of the temperature, a sequence of local time series implicit correlation feature vectors of the precipitation, a sequence of local time series implicit correlation feature vectors of the vegetation coverage, a sequence of local time series implicit correlation feature vectors of the soil moisture and a sequence of local time series implicit correlation feature vectors of the groundwater level, including: The historical data of the temperature value, the historical data of the precipitation, the historical data of the vegetation coverage rate, the historical data of the soil moisture and the historical data of the groundwater level are respectively segmented according to a predetermined time scale and input into a sequence encoder based on a Bi-LSTM model to obtain a sequence of the local time series implicit correlation feature vectors of the temperature, a sequence of the local time series implicit correlation feature vectors of the precipitation, a sequence of the local time series implicit correlation feature vectors of the vegetation coverage rate, a sequence of the local time series implicit correlation feature vectors of the soil moisture and a sequence of the local time series implicit correlation feature vectors of the groundwater level; The sequence of the implicitly associated feature vectors of the local time series of the temperature, the sequence of the implicitly associated feature vectors of the local time series of the precipitation, the sequence of the implicitly associated feature vectors of the local time series of the vegetation coverage, the sequence of the implicitly associated feature vectors of the local time series of the soil moisture, and the sequence of the implicitly associated feature vectors of the local time series of the groundwater level are respectively subjected to information aggregation based on feature significance to obtain the temperature time series associated aggregation feature vector, the precipitation time series associated aggregation feature vector, the vegetation coverage time series associated aggregation feature vector, the soil moisture time series associated aggregation feature vector, and the groundwater level time series associated aggregation feature vector, including: Inputting the sequence of the local time series implicitly associated feature vectors of the temperature, the sequence of the local time series implicitly associated feature vectors of the precipitation, the sequence of the local time series implicitly associated feature vectors of the vegetation coverage, the sequence of the local time series implicitly associated feature vectors of the soil moisture, and the sequence of the local time series implicitly associated feature vectors of the groundwater level into a feature aggregation network based on local feature significance transfer attenuation annotation respectively to obtain the temperature time series associated aggregation feature vector, the precipitation time series associated aggregation feature vector, the vegetation coverage time series associated aggregation feature vector, the soil moisture time series associated aggregation feature vector, and the groundwater level time series associated aggregation feature vector; The sequence of the local time series implicitly associated feature vectors of the temperature, the sequence of the local time series implicitly associated feature vectors of the precipitation, the sequence of the local time series implicitly associated feature vectors of the vegetation coverage, the sequence of the local time series implicitly associated feature vectors of the soil moisture, and the sequence of the local time series implicitly associated feature vectors of the groundwater level are respectively input into a feature aggregation network based on local feature significance transfer attenuation annotation to obtain the temperature time series associated aggregation feature vector, the precipitation time series associated aggregation feature vector, the vegetation coverage time series associated aggregation feature vector, the soil moisture time series associated aggregation feature vector, and the groundwater level time series associated aggregation feature vector, including: Calculating the characteristic significance description factor of each temperature local time series implicit correlation feature vector in the sequence of the temperature local time series implicit correlation feature vector, wherein the characteristic significance description factor is related to the mean and variance of each temperature local time series implicit correlation feature vector; Taking the last temperature local time series implicit association feature vector in the sequence of the temperature local time series implicit association feature vectors as the current temperature local time series implicit association feature vector, constructing the feature significance attenuation factor of each temperature local time series implicit association feature vector based on the distance span between each temperature local time series implicit association feature vector in the sequence of the temperature local time series implicit association feature vectors and the current temperature local time series implicit association feature vector; Based on the characteristic significance attenuation factor of each temperature local time series implicit correlation feature vector, the characteristic significance description factor of each temperature local time series implicit correlation feature vector is modulated to obtain a sequence of characteristic significance attenuation description factors; Inputting the sequence of feature significance attenuation description factors into a gated mask module to obtain a sequence of feature significance attenuation weight factors; Taking the sequence of the feature significance attenuation weight factors as the sequence of weights, calculating the weighted sum of the sequence of the temperature local time series implicit correlation feature vectors to obtain the temperature time series correlation aggregation feature vector; The temperature time series associated aggregation feature vector, the precipitation time series associated aggregation feature vector, the vegetation coverage time series associated aggregation feature vector, the soil moisture time series associated aggregation feature vector, and the groundwater level time series associated aggregation feature vector are jointly encoded with multivariate information to obtain a multivariate stereo information fusion representation vector, including: The temperature time series associated aggregated feature vector, the precipitation time series associated aggregated feature vector, the vegetation coverage time series associated aggregated feature vector, the soil moisture time series associated aggregated feature vector and the groundwater level time series associated aggregated feature vector are input into a multivariate stereo information joint encoder based on an embedding layer to obtain the multivariate stereo information fusion representation vector.

2. The method for analyzing the evolution causes of surface runoff in mountainous areas based on multivariate stereoscopic information according to claim 1 is characterized in that: Calculating the characteristic significance description factor of each temperature local time series implicit correlation feature vector in the sequence of temperature local time series implicit correlation feature vectors, including: The expected value of the fourth power of the difference between each eigenvalue in the local time series implicit correlation eigenvector of the air temperature and its eigenmean is calculated, and the expected value is divided by the square of the eigenvariance of the local time series implicit correlation eigenvector of the air temperature to obtain the eigensignificance description factor.

3. The method for analyzing the evolution causes of surface runoff in mountainous areas based on multivariate stereoscopic information according to claim 2 is characterized in that: Based on the distance span between each temperature local time series implicit correlation feature vector in the sequence of the temperature local time series implicit correlation feature vector and the current temperature local time series implicit correlation feature vector, constructing the feature significance attenuation factor of each temperature local time series implicit correlation feature vector, including: Calculate the difference between the maximum eigenvalue of the current air temperature local time series implicit correlation feature vector and the maximum eigenvalue of each air temperature local time series implicit correlation feature vector in the sequence of the air temperature local time series implicit correlation feature vector, and then divide the difference by the number of eigenvectors separated by the two to obtain the feature significance attenuation factor of each air temperature local time series implicit correlation feature vector.

4. The method for analyzing the evolution causes of surface runoff in mountainous areas based on multivariate stereoscopic information according to claim 3 is characterized in that: The characteristic significance description factor of each temperature local time series implicit correlation feature vector is modulated based on the characteristic significance attenuation factor of each temperature local time series implicit correlation feature vector to obtain a sequence of characteristic significance attenuation description factors, including: The product of the characteristic significance attenuation factor of each of the local time series implicit correlation characteristic vectors of the air temperature and its characteristic significance description factor is calculated to obtain a sequence of the characteristic significance attenuation description factors.

5. The method for analyzing the evolution of surface runoff in mountainous areas based on multivariate stereoscopic information according to claim 4 is characterized in that: Based on the multivariate stereo information fusion representation vector, the analytical results of the causes of surface runoff evolution in mountainous areas are determined, including: The multivariate stereo information fusion representation vector is input into a classifier-based mountain surface runoff evolution cause analysis module to obtain the analysis result, and the analysis result is used to represent the type label of the core cause of the mountain surface runoff evolution.

Images