A method and apparatus for predicting thermal fusion disasters based on multi-source data

Abstract

This invention discloses a method and apparatus for predicting thermal thaw disasters based on multi-source data, comprising the following steps: Step 1: Collecting meteorological data, geological data, and permafrost monitoring data, and preprocessing the collected data; Step 2: Constructing features for prediction, including: using meteorological data to calculate the temperature change trend in permafrost areas; and using geological and permafrost monitoring data to analyze changes in the permafrost layer; Step 3: Establishing a model using geothermal equations to simulate the thermal response of permafrost, including the following equations: heat conduction equation; thawing equation; and soil moisture equation; Step 4: Integrating information from different data sources into the model to comprehensively assess the risk of thermal thaw disasters in permafrost areas; Step 5: Based on the established model, making real-time or future thermal thaw disaster predictions. This method has high reliability and practicality, and helps to improve the prediction and management capabilities of thermal thaw disasters in permafrost areas.

Description

Technical Field

[0001] This invention relates to the field of thermal melt disaster prediction technology, and in particular to a method and apparatus for thermal melt disaster prediction based on multi-source data. Background Technology

[0002] Thermal thawing in permafrost regions can have a series of severe consequences, with far-reaching impacts on the environment, ecosystems, and human societies. For example, permafrost thawing can lead to soil instability, potentially causing land subsidence and foundation collapse, damaging buildings, roads, and other infrastructure and increasing maintenance and repair costs. Global warming causing sea-level rise and thawing can lead to coastline retreat, exacerbating coastal erosion and impacting coastal communities, fisheries, and tourism. Permafrost ecosystems are highly dependent on cold conditions. Thawing and rising temperatures can lead to habitat loss, altered plant species, and changes in wildlife migration patterns, significantly impacting ecological balance. The thawing of permafrost releases large amounts of greenhouse gases, such as methane, which could accelerate global warming, creating a climate feedback effect that leads to further greenhouse gas releases and exacerbates climate change. Permafrost regions store vast amounts of freshwater resources; thawing or alteration of these resources could impact water supply, particularly in areas with insufficient rainfall. Indigenous communities in permafrost regions may rely on traditional lifestyles and resources; climate change could threaten their cultural and economic livelihoods. Therefore, forecasting thermal thawing disasters in permafrost regions is crucial. Summary of the Invention

[0003] To address the above problems, this invention provides a method and apparatus for predicting thermal fusion disasters based on multi-source data.

[0004] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0005] On one hand, this invention discloses a method for predicting thermal melt disasters based on multi-source data, comprising the following steps:

[0006] Step 1: Collect meteorological data, geological data, and permafrost monitoring data, and preprocess the collected data.

[0007] Step 2: Construct features for prediction, including:

[0008] Using meteorological data, calculate the temperature change trend in permafrost regions;

[0009] Analyze changes in the permafrost layer using geological and permafrost monitoring data;

[0010] Step 3: Use geothermal equations to build a model to simulate the thermal response of permafrost, including the following equations:

[0011] Heat conduction equation: describes the process of heat conduction in soil;

[0012] The thermal melting equation describes the melting and condensation process of ice.

[0013] Soil moisture equation: Considering the impact of changes in soil moisture on heat conduction;

[0014] Step 4: Integrate information from different data sources into the model to comprehensively assess the risk of thermal thaw disasters in permafrost regions;

[0015] Step 5: Based on the established model, make real-time or future predictions of thermal melt disasters.

[0016] Furthermore: Step 2 includes:

[0017] Step 2.1 Temperature Change Trend Analysis:

[0018] Linear regression is used to fit a temperature time series model. Assuming that temperature observations T = [T1, T2, ..., Tn] correspond to time points t = [t1, t2, ..., tn], where n is the number of observations, the linear regression model is expressed as:

[0019] T i =β0+β1t i +∈ i

[0020] Where Ti is the observed temperature value, β0 and β1 are regression coefficients, and εi is the error term;

[0021] Step 2.2 Changes in the permafrost layer:

[0022] Changes in permafrost depth: Using underground temperature monitoring data, changes in permafrost depth are calculated through the permafrost heat conduction equation, as shown below:

[0023]

[0024] Where T is temperature, z is depth, k is the thermal conductivity of the soil, ρ is soil density, and c is soil specific heat capacity.

[0025] Soil moisture variation: Using soil moisture monitoring data, the spatiotemporal variation of soil moisture is calculated through the soil moisture conduction equation, as shown below:

[0026]

[0027] Where θ is soil moisture, t is time, D is soil moisture diffusion coefficient, and S is water source / sink.

[0028] Furthermore: Step 3 includes:

[0029] Heat conduction equation:

[0030] The heat conduction equation is used to describe the heat conduction process in soil, and its one-dimensional form is as follows:

[0031]

[0032] Where T is temperature, t is time, x is depth, and α is thermal conductivity;

[0033] Thermal melting equation:

[0034] The thermal melting equation is used to describe the melting and condensation process of ice, and its one-dimensional form is as follows:

[0035]

[0036] Where ρi is the density of ice, and r is the distance;

[0037] Soil moisture equation:

[0038] The soil moisture equation, in one-dimensional form, is as follows:

[0039]

[0040] Where θ is soil moisture, t is time, D is soil moisture diffusion coefficient, and S is water source / sink;

[0041] The above equations are coupled together to simulate the thermal response and thawing process throughout the permafrost region.

[0042] Furthermore, step 4 includes:

[0043] Data fusion: Integrating information from different data sources to ensure that the data has a consistent temporal and spatial reference frame;

[0044] Spatial interpolation: For unevenly distributed geographic data, interpolation methods are used to generate a complete spatial data layer in order to match the model;

[0045] Time synchronization: Ensure that timestamps from different data sources are synchronized so that consistent time analysis can be performed in the model.

[0046] To assess the accuracy of the model, it is compared with ground-based observation data. The model's error index is calculated using the following formula:

[0047]

[0048] Where Oi is the observed value, Pi is the model prediction value, and n is the number of data points.

[0049] Furthermore: Step 5 includes:

[0050] Prediction process:

[0051] Time stepping: Using a numerical model, time is stepped to simulate the thermal melting process over a future period of time;

[0052] Initial conditions: Set appropriate initial conditions in the model, including soil temperature and the initial state of frozen soil;

[0053] Model solution: Numerical methods are used to solve the model equations to simulate the spatiotemporal changes of thermal thawing, freezing, and soil moisture.

[0054] Prediction results: Obtain the model's output, including soil temperature and permafrost status at future time points;

[0055] Real-time monitoring:

[0056] Meteorological data: Real-time acquisition of meteorological data, including temperature, precipitation, and wind speed, which will be used to drive the model and provide real-time input;

[0057] Monitoring data: Real-time data on frozen soil and soil moisture are obtained using on-site monitoring stations or satellite remote sensing.

[0058] A thermal melt disaster prediction device based on multi-source data, comprising:

[0059] Data collection and preprocessing module: collects meteorological data, geological data, and permafrost monitoring data, and preprocesses the collected data.

[0060] Feature Engineering Builder: Constructs features for prediction, including:

[0061] Using meteorological data, calculate the temperature change trend in permafrost regions;

[0062] Analyze changes in the permafrost layer using geological and permafrost monitoring data;

[0063] Model building module: Uses geothermal equations to build a model to simulate the thermal response of permafrost, including the following equations:

[0064] Heat conduction equation: describes the process of heat conduction in soil;

[0065] The thermal melting equation describes the melting and condensation process of ice.

[0066] Soil moisture equation: Considering the impact of changes in soil moisture on heat conduction;

[0067] Data integration and model validation module: Integrates information from different data sources into the model to comprehensively assess the risk of thermal thaw disasters in permafrost regions;

[0068] Prediction and Monitoring Module: Based on the established model, it makes real-time or future predictions of thermal melt disasters.

[0069] The technological advancements achieved by this invention compared to existing technologies are as follows:

[0070] This method integrates multiple data sources, including meteorological, geological, and permafrost monitoring data, to provide comprehensive information, contributing to a more holistic understanding of the thermal thaw disaster risks in permafrost regions. It employs a physical modeling approach, using physical equations to simulate heat conduction, thawing, and freezing processes. This enhances the interpretability of the predictions and facilitates a deeper understanding of the mechanisms of thermal thaw disasters. Despite using physical modeling, this method fully leverages data-driven approaches, such as monitoring data and real-time meteorological data, to update and validate the model, improving its accuracy and robustness. By combining real-time meteorological and monitoring data, this method enables real-time prediction and monitoring of thermal thaw disasters in permafrost regions, allowing decision-makers to take timely countermeasures. The specific implementation of this method can be customized and improved according to different regions and needs. It can be flexibly adjusted based on available data and regional characteristics. This method integrates knowledge from multiple fields, including meteorology, geology, earth science, and numerical modeling, providing scientific support for decision-makers to better understand and manage thermal thaw disaster risks. By predicting and monitoring thermal melt disasters in advance, this approach helps communities and governments take timely measures to mitigate potential disaster impacts and improve societal response capabilities.

[0071] In summary, this method has high reliability and practicality, and helps to improve the ability to predict and manage thermal thawing disasters in permafrost regions. Attached Figure Description

[0072] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.

[0073] In the attached diagram:

[0074] Figure 1 This is a flowchart of the present invention. Detailed Implementation

[0075] The following specific embodiments are combined with each other. Concepts or processes that are the same or similar may not be described again in some embodiments. The embodiments of the present invention will now be described with reference to the accompanying drawings.

[0076] Example 1

[0077] like Figure 1 As shown, this invention discloses a method and apparatus for predicting thermal calving disasters based on multi-source data, comprising the following steps:

[0078] Step 1: Data Collection and Preprocessing

[0079] First, it is necessary to collect data from multiple sources, including meteorological data, geological data, and permafrost monitoring data, which will provide information on climate, topography, soil conditions, and ice layer changes.

[0080] Step 2: Feature Engineering

[0081] This step involves constructing features for prediction, which includes the following aspects:

[0082] Temperature change trend analysis: Using meteorological data, calculate the temperature change trend in permafrost regions. This can be achieved by fitting a temperature time series model, such as a linear trend or a nonlinear trend.

[0083] Changes in the permafrost layer: Analyze changes in the permafrost layer using geological and permafrost monitoring data. This can be achieved by measuring parameters such as soil temperature, humidity, and density, using appropriate geological and permafrost physics equations.

[0084] Step 3: Model Building

[0085] In this step, a physical modeling approach is used, employing geothermal equations to simulate the thermal response of permafrost, including the following equations:

[0086] Heat conduction equation: describes the process of heat conduction in soil.

[0087] Thermal melting equation: describes the melting and condensation process of ice.

[0088] Soil moisture equation: Considering the changes in soil moisture, which have a significant impact on heat conduction.

[0089] Step 4: Data Integration and Model Validation

[0090] Information from different data sources is integrated into the model to comprehensively assess the risk of thermal thaw disasters in permafrost regions. Model validation is a crucial step, using ground-based observation data to verify the model's accuracy and robustness.

[0091] Step 5: Prediction and Monitoring

[0092] Based on the established model, real-time or future thermal calving disaster predictions can be made. This can be done by combining real-time meteorological data and monitoring data. The prediction results can be presented in the form of maps, visualization charts or numerical indicators to help decision-makers take appropriate measures.

[0093] This method combines physical modeling and multi-source data analysis, enabling a more comprehensive understanding and prediction of thermal thaw disasters in permafrost regions.

[0094] Specifically, step 1 includes:

[0095] Data collection:

[0096] Meteorological data collection: Obtaining meteorological data from meteorological stations, satellites, or meteorological models, including temperature, humidity, precipitation, wind speed, etc., which are usually provided in time series form.

[0097] Geological data collection: Collecting geological information, including soil type, topography, groundwater level, etc. This data can come from geological exploration, geological maps or geological databases.

[0098] Permafrost monitoring data: Acquire monitoring data about permafrost, including temperature, thickness, and movement speed. This data usually comes from dedicated monitoring stations or satellite remote sensing.

[0099] Data preprocessing:

[0100] Data cleaning: Remove outliers and erroneous data. The cleaned data is represented as Dc.

[0101] Missing value imputation: For data with missing values, interpolation methods can be used to impute the missing values. The linear interpolation formula is as follows:

[0102] If there are missing values ​​between xi and x{i+1}, the missing value to be filled is x. missing :

[0103]

[0104] Where t is the time corresponding to the missing value, and ti and t{i+1} are adjacent time points in the time series.

[0105] Data format conversion: Converting data from different data sources into the same units and coordinate system for subsequent analysis.

[0106] Data merging: Integrating data from different data sources into a single dataset for comprehensive analysis. Geographic Information System (GIS) tools can be used to handle spatial data merging.

[0107] The steps above can help prepare the dataset for subsequent analysis, ensuring the quality and consistency of the data. This prepared data will be used to build models and predict thermal melt disasters.

[0108] Specifically, step 2 includes:

[0109] Step 2.1 Temperature Change Trend Analysis:

[0110] To analyze temperature change trends, linear regression can be used to fit a temperature time series model. Assuming a series of temperature observations T = [T1, T2, ..., Tn] corresponding to time points t = [t1, t2, ..., tn], where n is the number of observations, the linear regression model can be expressed as:

[0111] T i =β0+β1t i +∈ i

[0112] Where Ti is the observed temperature value, β0 and β1 are regression coefficients, and εi is the error term. By fitting this model, the value of β can be estimated, which represents the trend of temperature change over time. If β is significantly non-zero, it indicates that there is a temperature change trend.

[0113] Step 2.2 Changes in the permafrost layer:

[0114] The analysis of permafrost layer changes includes the following characteristic engineering aspects:

[0115] Changes in permafrost depth: Using underground temperature monitoring data, changes in permafrost depth can be calculated. This can be achieved using the permafrost heat conduction equation, as shown below:

[0116]

[0117] Where T is temperature, z is depth, k is the thermal conductivity of the soil, ρ is soil density, and c is the specific heat capacity of the soil. By integrating this equation, the variation in the depth of permafrost can be estimated.

[0118] Soil moisture variation: Using soil moisture monitoring data, the spatiotemporal variation of soil moisture can be calculated. This can be achieved using the soil moisture conduction equation, as shown below:

[0119]

[0120] Where θ is soil moisture, t is time, D is the soil moisture diffusion coefficient, and S is the water source / sink ratio. By numerically solving this equation, the trend of soil moisture variation can be estimated.

[0121] The above steps can be used to analyze and extract features related to thermal thaw disasters in permafrost regions. These features will be used to build predictive models to assess potential thermal thaw disaster risks.

[0122] Specifically, step 3 includes:

[0123] 1. Heat conduction equation:

[0124] The heat conduction equation describes the process of heat conduction in soil. Its one-dimensional form is as follows:

[0125]

[0126] Where T is temperature, t is time, x is depth, and α is thermal conductivity. This equation can be solved numerically using the finite difference or finite element method.

[0127] 2. Thermal melting equation:

[0128] The thermal melting equation describes the melting and condensation process of ice. Its one-dimensional form is as follows:

[0129]

[0130] Where ρi is the density of ice and r is the distance. This equation takes into account the heat conduction and phase transition processes of ice. Numerical solutions can be obtained using implicit or explicit numerical methods.

[0131] 3. Soil moisture equation:

[0132] The soil moisture equation considers changes in soil moisture, which has a significant impact on heat conduction. Its one-dimensional form is as follows:

[0133]

[0134] Where θ is soil moisture, t is time, D is the soil moisture diffusion coefficient, and S is the water source / sink. This equation describes the flow and changes of moisture in the soil, and can be solved numerically using the finite element method or the finite difference method.

[0135] 4. Model Coupling:

[0136] The above equations can be coupled together to simulate the thermal response and thawing process of the entire permafrost region. The initial and boundary conditions of the model need to be set according to the actual situation.

[0137] 5. Numerical solution:

[0138] The numerical solution of the model uses numerical analysis methods that discretize the above partial differential equations into a form that can be processed by a computer and allow for time and space steps on a discrete grid.

[0139] Specifically, step 4 includes:

[0140] Data integration:

[0141] Data fusion: Integrating information from different data sources to ensure the data has a consistent temporal and spatial reference frame. Geographic Information System (GIS) tools can be used to handle the integration of spatial data.

[0142] Spatial interpolation: For unevenly distributed geographic data, kriging interpolation can be used to generate a complete spatial data layer to match the model.

[0143] Time synchronization: Ensure that timestamps from different data sources are synchronized so that consistent time analysis can be performed in the model.

[0144] Model validation:

[0145] Ground observation data: Comparison with ground observation data. To assess the model's accuracy, compare model predictions with actual observation data. The model's error metric can be calculated using the following formula:

[0146]

[0147] Where Oi is the observed value, Pi is the model prediction value, and n is the number of data points.

[0148] The above steps can be used to evaluate the accuracy, reliability, and robustness of the model, ensuring that it can reliably predict the risk of thermal thaw disasters in permafrost regions, and also help to determine the applicability and improvement directions of the model.

[0149] Specifically, step 5 includes:

[0150] Prediction process:

[0151] Time stepping: Using a numerical model, time is stepped to simulate the thermal melting process over a future period of time.

[0152] Initial conditions: Set appropriate initial conditions in the model, including soil temperature, initial state of frozen soil, etc.

[0153] Model solution: Numerical methods are used to solve the model equations to simulate the spatiotemporal changes of thermal thawing, freezing, and soil moisture.

[0154] Prediction results: Obtain the model's output, including soil temperature and permafrost conditions at future time points.

[0155] Real-time monitoring:

[0156] Real-time monitoring includes the following aspects:

[0157] Meteorological data: Real-time acquisition of meteorological data, including temperature, precipitation, wind speed, etc. This data will be used to drive the model and provide real-time input.

[0158] Monitoring data: Real-time data on frozen soil and soil moisture are obtained using on-site monitoring stations or satellite remote sensing.

[0159] Model update:

[0160] As new real-time data becomes available, the model can be updated periodically to reflect the latest developments. This can be achieved by rerunning the model and incorporating the new observations.

[0161] Example 2

[0162] This embodiment discloses a thermal melt disaster prediction device based on multi-source data, including:

[0163] Data collection and preprocessing module: collects meteorological data, geological data, and permafrost monitoring data, and preprocesses the collected data.

[0164] Feature Engineering Builder: Constructs features for prediction, including:

[0165] Using meteorological data, calculate the temperature change trend in permafrost regions;

[0166] Analyze changes in the permafrost layer using geological and permafrost monitoring data;

[0167] Model building module: Uses geothermal equations to build a model to simulate the thermal response of permafrost, including the following equations:

[0168] Heat conduction equation: describes the process of heat conduction in soil;

[0169] The thermal melting equation describes the melting and condensation process of ice.

[0170] Soil moisture equation: Considering the impact of changes in soil moisture on heat conduction;

[0171] Data integration and model validation module: Integrates information from different data sources into the model to comprehensively assess the risk of thermal thaw disasters in permafrost regions;

[0172] Prediction and Monitoring Module: Based on the established model, it makes real-time or future predictions of thermal melt disasters.

[0173] The modules described above are used to implement the functions in Embodiment 1.

[0174] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or substitute some of the technical features. Any modifications, substitutions, or improvements made within the spirit and principles of the present invention should be included within the scope of protection of the claims of the present invention.

Claims

1. A method for predicting thermal fusion disasters based on multi-source data, characterized in that, Includes the following steps: Step 1: Collect meteorological data, geological data, and permafrost monitoring data, and preprocess the collected data; Step 2: Construct features for prediction, including: Using meteorological data, calculate the temperature change trend in permafrost regions; Analyze changes in the permafrost layer using geological and permafrost monitoring data; Step 3: Use geothermal equations to build a model to simulate the thermal response of permafrost, including the following equations: Heat conduction equation: describes the process of heat conduction in soil; The thermal melting equation describes the melting and condensation process of ice. Soil moisture equation: Considering the impact of changes in soil moisture on heat conduction; Step 4: Integrate information from different data sources into the model to comprehensively assess the risk of thermal thaw disasters in permafrost regions; Step 5: Based on the established model, make real-time or future predictions of thermal melt disasters; Step 2 includes: Step 2.1 Temperature Change Trend Analysis: Linear regression is used to fit a temperature time series model. Assuming that temperature observations T = [T1, T2, ..., Tn] correspond to time points t = [t1, t2, ..., tn], where n is the number of observations, the linear regression model is expressed as: Where Ti is the observed temperature value, β0 and β1 are regression coefficients, and εi is the error term; Step 2.2 Changes in the permafrost layer: Changes in permafrost depth: Using underground temperature monitoring data, changes in permafrost depth are calculated through the permafrost heat conduction equation, as shown below: Where T is temperature, z is depth, k is the thermal conductivity of the soil, ρ is soil density, and c is soil specific heat capacity. Soil moisture variation: Using soil moisture monitoring data, the spatiotemporal variation of soil moisture is calculated through the soil moisture conduction equation, as shown below: Where θ is soil moisture, t is time, D is soil moisture diffusion coefficient, and S is water source / sink; Step 3 includes: Heat conduction equation: The heat conduction equation is used to describe the heat conduction process in soil, and its one-dimensional form is as follows: Where T is temperature, t is time, x is depth, and α is thermal conductivity; Thermal melting equation: The thermal melting equation is used to describe the melting and condensation process of ice, and its one-dimensional form is as follows: Where ρi is the density of ice, and r is the distance; Soil moisture equation: The soil moisture equation, in one-dimensional form, is as follows: Where θ is soil moisture, t is time, D is soil moisture diffusion coefficient, and S is water source / sink; The above equations are coupled together to simulate the thermal response and thawing process throughout the permafrost region; Step 4 includes: Data fusion: Integrating information from different data sources to ensure that the data has a consistent temporal and spatial reference frame; Spatial interpolation: For unevenly distributed geographic data, interpolation methods are used to generate a complete spatial data layer in order to match the model; Time synchronization: Ensure that timestamps from different data sources are synchronized so that consistent time analysis can be performed in the model; To assess the accuracy of the model, it is compared with ground-based observation data. The model's error index is calculated using the following formula: Where Oi is the observed value, Pi is the model prediction value, and n is the number of data points; Step 5 includes: Prediction process: Time stepping: Using a numerical model, time is stepped to simulate the thermal melting process over a future period of time; Initial conditions: Set appropriate initial conditions in the model, including soil temperature and the initial state of frozen soil; Model solution: Numerical methods are used to solve the model equations to simulate the spatiotemporal changes of thermal thawing, freezing, and soil moisture. Prediction results: Obtain the model's output, including soil temperature and permafrost status at future time points; Real-time monitoring: Meteorological data: Real-time acquisition of meteorological data, including temperature, precipitation, and wind speed, which will be used to drive the model and provide real-time input; Monitoring data: Real-time data on frozen soil and soil moisture are obtained using on-site monitoring stations or satellite remote sensing.

Images