A natural disaster monitoring information processing method and system

Abstract

This invention relates to the technical field of natural disaster monitoring information processing, and provides a method and system for natural disaster monitoring information processing. The method includes: receiving deep displacement data and accompanying physical information corresponding to the physical causes of the deep displacement data; performing trend judgment on the deep displacement data to determine whether the deep displacement data exhibits a specific growth trend, obtaining a trend judgment result; when the trend judgment result indicates that the deep displacement data exhibits a specific growth trend, determining the physical causes of the deep displacement data based on the accompanying physical information, obtaining a confidence level label for the deep displacement data; if the confidence level label indicates that the confidence level is lower than a preset lower limit, then limiting or eliminating the influence of the deep displacement data on risk assessment or parameter adjustment. This invention improves the accuracy and timeliness of natural disaster monitoring and early warning.

Description

Technical Field

[0001] This invention relates to the technical field of natural disaster monitoring information processing, and specifically to a method and system for natural disaster monitoring information processing. Background Technology

[0002] In the field of natural disaster monitoring, advanced monitoring information processing methods and systems are typically deployed to effectively prevent geological disasters. These systems utilize multi-source data fusion mechanisms to comprehensively analyze readings from various sensors, such as surface displacement gauges, deep displacement gauges, and rain gauges, to assess geological stability and calculate a comprehensive risk index. During the initial system configuration, deep displacement data is often assigned a higher importance weight due to its indicative nature of landslide precursors.

[0003] However, existing methods for processing natural disaster monitoring information have a technical problem. In practical applications, the external environment of the system deployment area may undergo non-catastrophic, slow but continuous changes. For example, incomplete seepage prevention in small irrigation canals may cause water to slowly infiltrate deep into hillside areas over a long period, gradually altering local soil physical properties and triggering extremely slow creep and subsidence of the soil around the anchor points of deep displacement gauges. The deep displacement data generated by this creep and subsidence is small, continuous, and stable, and is usually judged by the system as valid normal monitoring data. During the system's periodic offline optimization and recalibration, because this "false precursor" deep displacement data continuously exhibits a smooth and coherent growth trend, it shows extremely high statistical significance in the learning dataset. This leads the learning mechanism to assign it an overwhelming weight, while the weights of other real disaster signals are significantly reduced. As a result, when a real disaster occurs, such as a short-term heavy rainfall triggering a surface debris flow on a hillside, the real emergency signal is "diluted" after weighted calculation, and the final output comprehensive risk index is far lower than the actual warning threshold. This prevented the system from issuing timely and effective warnings, thus missing the best opportunity for disaster prevention and mitigation. This phenomenon indicates that when faced with slow, non-catastrophic changes in the external environment, the data fusion weights of existing methods may "drift," causing the system to fall into a "perception blind spot" of real risks due to "logical self-consistency."

[0004] To address the aforementioned issues, existing technologies urgently need improvement. Summary of the Invention

[0005] This application discloses a method and system for processing natural disaster monitoring information, which aims to solve the technical problem that when existing natural disaster monitoring methods face slow, non-hazardous changes in the external environment, the data fusion weights may "drift," causing the system to fall into a "perception blind spot" of real risks due to "logical self-consistency," and thus fail to issue timely and effective early warnings.

[0006] The technical solution of this application is as follows:

[0007] Firstly, this application discloses a method for processing natural disaster monitoring information. A method for processing natural disaster monitoring information includes:

[0008] Receive deep displacement data and the accompanying physical information corresponding to the physical causes of the deep displacement data;

[0009] Perform trend judgment on deep displacement data to determine whether the deep displacement data shows a specific growth trend and obtain the trend judgment result;

[0010] When the trend judgment result indicates that the deep displacement data shows a specific growth trend, the physical cause of the deep displacement data is determined based on the accompanying physical information, and the credibility label of the deep displacement data is obtained.

[0011] If the confidence level is lower than the preset lower limit, then the influence of deep displacement data on risk assessment or parameter adjustment is limited or excluded.

[0012] This application introduces a mechanism for judging the credibility of deep displacement data through its technical solution. Specifically, when the data exhibits a specific growth trend, credibility is marked by analyzing its physical causes. This effectively identifies and processes "false precursor" data caused by non-catastrophic slow changes, avoiding its interference with risk assessment and early warning. This overcomes the problem of data weight "drift" in existing technologies, significantly improving the accuracy and timeliness of early warnings.

[0013] Furthermore, when the trend judgment result indicates that the deep displacement data exhibits a specific growth trend, the steps for determining the physical cause of the deep displacement data based on the accompanying physical information and obtaining the reliability label of the deep displacement data include:

[0014] When the accompanying physical information does not clearly indicate the physical cause, and the regional geostress monitoring data and regional microseismic activity monitoring data do not show any abnormalities, the electrochemical response data of the anchor point interface of the deep displacement gauge is obtained.

[0015] Based on the electrochemical response data, analyze whether microscopic chemical reactions related to the growth trend occur at the anchoring point interface;

[0016] If a microchemical reaction related to the growth trend occurs at the anchor point interface, the deep displacement data is determined to be caused by the microchemical reaction, and the deep displacement data is marked with a confidence level lower than the preset lower limit.

[0017] This application provides a refined method for identifying the causes of abnormal growth in deep displacement data when conventional accompanying information is unclear. By introducing electrochemical response data analysis to examine the microscopic chemical reactions at the anchorage point interface, it is possible to accurately distinguish non-catastrophic displacements caused by equipment or environmental factors. This allows for more precise labeling of data reliability, avoids misjudgments, and further enhances the intelligence and reliability of data processing.

[0018] More specifically, the steps for analyzing whether microscopic chemical reactions related to the growth trend occur at the anchoring point interface based on electrochemical response data include:

[0019] Extract interfacial impedance parameters and / or redox peak parameters from electrochemical response data;

[0020] Determine whether the interface impedance parameter shows a continuous decreasing trend, and / or whether the redox peak parameter shows a continuous increasing trend or a peak potential shift trend, and obtain the parameter judgment result;

[0021] If the parameter judgment result is yes, then it is determined that a microscopic chemical reaction related to the growth trend has occurred at the anchoring point interface.

[0022] Through this technical solution, this application clarifies the specific quantitative indicators and criteria for judging microscopic chemical reactions using electrochemical response data, namely, the changing trends of interfacial impedance parameters and / or redox peak parameters. This makes the judgment of microscopic chemical reactions at the anchoring point interface more objective, accurate, and operable, providing a solid scientific basis for the credibility marking of deep displacement data.

[0023] In some preferred embodiments, steps to limit or eliminate the influence of deep displacement data on risk assessment or parameter adjustment include:

[0024] Obtain the confidence and causation labels corresponding to the deep displacement data;

[0025] The risk level of deep displacement data is determined based on the confidence level and causal markers.

[0026] Select the corresponding impact limitation strategy based on the risk level;

[0027] In accordance with the impact limitation strategy, the impact of deep displacement data on risk assessment or parameter adjustment is limited or eliminated.

[0028] This application provides a risk level assessment and tiered restriction strategy based on data reliability and causes. This enables the system to adopt differentiated processing methods according to the actual reliability and potential impact of the data, avoiding a blanket exclusion of data. Thus, while ensuring the accuracy of risk assessment, it maximizes the use of effective information, improving the system's flexibility and the sophistication of decision-making.

[0029] Furthermore, the steps for determining the risk level of deep displacement data based on confidence and causal markers include:

[0030] Based on the confidence level marker and the cause marker, the corresponding physical parameters are identified;

[0031] Analyze the numerical values ​​or trends of physical parameters;

[0032] Determine whether the value or trend of a physical parameter exceeds a preset threshold, and obtain the physical parameter judgment result;

[0033] If the physical parameter judgment result is yes, then adjust and determine the risk level corresponding to the deep displacement data.

[0034] This application provides a method for dynamically adjusting risk levels by identifying physical parameters related to credibility and causal markers and judging their values ​​or trends. This makes risk level determination more objective and quantifiable, enabling real-time adjustments based on actual monitoring data, thus making risk assessment more adaptable and accurate.

[0035] Based on the above, the steps to determine whether the value or trend of the physical parameter exceeds a preset threshold and obtain the physical parameter judgment result include:

[0036] To obtain data on regional environmental changes or geological background evolution;

[0037] Calculate the adaptive bias of the preset threshold based on regional environmental change data or geological background evolution data;

[0038] Determine whether the adaptive deviation exceeds the predetermined range, and obtain the deviation judgment result;

[0039] If the deviation judgment result is yes, then adjust the preset threshold;

[0040] Based on the adjusted preset threshold, determine whether the value or trend of the physical parameter exceeds the adjusted preset threshold.

[0041] Through this technical solution, this application introduces an adaptive adjustment mechanism for preset thresholds. By considering regional environmental changes or geological background evolution data, the thresholds are dynamically calculated and adjusted, making the risk assessment thresholds more consistent with the current situation. This avoids misjudgments or omissions caused by fixed thresholds, significantly improving the system's environmental adaptability and the robustness of early warning.

[0042] Preferably, the step of calculating the adaptive bias of the preset threshold includes:

[0043] Analyze the correlation between regional environmental change data or geological background evolution data and historical threshold applicability data to obtain correlation information;

[0044] Based on the associated information, construct a mapping relationship for the associated information;

[0045] Based on current regional environmental change data or geological background evolution data and mapping relationships, the adaptive bias of the preset threshold is calculated.

[0046] This application provides a method for calculating threshold adaptability deviation based on historical data and correlation analysis. This transforms threshold adjustment from a simple empirical judgment into an intelligent adjustment based on data-driven principles and historical patterns, thereby improving the accuracy and scientific rigor of threshold adjustment and further optimizing the precision of risk assessment.

[0047] Furthermore, the steps to analyze the correlation between regional environmental change data or geological background evolution data and historical threshold applicability data to obtain correlation information include:

[0048] Data quality processing is performed on regional environmental change data or geological background evolution data. Data quality processing includes imputing missing data based on the spatiotemporal correlation of the data, adaptive filtering of noisy data, and feature alignment and normalization of heterogeneous data.

[0049] Based on regional environmental change data or geological background evolution data that has undergone data quality processing, combined with historical threshold applicability data, multivariate statistical analysis and nonlinear regression analysis are performed to identify the correlations between data and obtain correlation information.

[0050] This application emphasizes the importance of data quality processing and advanced analytical methods in establishing data correlations through its technical solutions. By performing interpolation, filtering, alignment, and normalization on the raw data, and combining this with multivariate statistics and nonlinear regression analysis, the deep correlations between environmental and geological data and historical thresholds can be more accurately and comprehensively uncovered, providing a more reliable basis for adaptive threshold adjustment.

[0051] In one implementation, the step of constructing a mapping relationship for the association information based on the association information includes:

[0052] Obtain preset physical association rules or experience lookup tables;

[0053] Based on preset physical association rules or experience lookup tables, regional environmental change data or geological background evolution data are mapped to a preset threshold of adaptive deviation to construct a mapping relationship of associated information.

[0054] This application provides a practical method for constructing a mapping relationship of related information through a technical solution, namely, by using preset physical association rules or an experience lookup table. This makes the calculation process of threshold adaptive bias more efficient and controllable, and can be based on either theoretical models or long-term practical experience, thereby improving the system's deployability and practicality.

[0055] Secondly, this application also discloses a natural disaster monitoring information processing system for performing natural disaster monitoring information processing, including:

[0056] The displacement data receiving module is used to receive deep displacement data and the accompanying physical information corresponding to the physical causes of the deep displacement data.

[0057] The displacement trend judgment module is used to judge the trend of deep displacement data, determine whether the deep displacement data shows a specific growth trend, and obtain the trend judgment result.

[0058] The credibility labeling module is used to determine the physical cause of deep displacement data based on the accompanying physical information when the trend judgment result indicates that the deep displacement data shows a specific growth trend, and to obtain the credibility label of the deep displacement data.

[0059] The restriction effect execution module is used to limit or exclude the impact of deep displacement data on risk assessment or parameter adjustment if the confidence level indicated by the confidence level flag is lower than a preset lower limit.

[0060] This application provides a system entity capable of implementing the aforementioned natural disaster monitoring information processing method. Through modular design, the system integrates key functions such as data reception, trend assessment, credibility marking, and impact limitation, providing hardware and software support for practical applications. This makes the implementation of the aforementioned method possible, effectively solving the problems of data misjudgment and early warning failure in existing technologies.

[0061] Beneficial effects

[0062] This application discloses a natural disaster monitoring information processing method that proposes an innovative solution to the problem in existing technologies where deep displacement data may generate "false precursors" due to slow, non-catastrophic changes, leading to an imbalance in the system's weighting of real disaster signals and thus affecting the accuracy of early warnings. This method introduces a mechanism for judging the credibility of deep displacement data. Upon receiving deep displacement data and determining that it exhibits a specific growth trend, it no longer simply regards it as a disaster precursor. Instead, it further determines the true physical cause of the deep displacement data based on the accompanying physical information corresponding to its physical origin, and obtains a credibility marker for the data accordingly. When the credibility marker indicates that the data credibility is below a preset lower limit, the system can intelligently limit or eliminate the impact of the deep displacement data on risk assessment or parameter adjustment.

[0063] Through the aforementioned technical solution, this application can effectively identify and filter out deep displacement data with "false precursor" characteristics caused by non-catastrophic factors (such as slow creep subsidence of anchor points due to seepage in small irrigation canals). This avoids assigning excessive weight to such data during system learning and calibration, thereby preventing the "dilution" or masking of real disaster signals. Therefore, when a real disaster occurs, the system can conduct risk assessments based on more accurate and reliable data, and issue timely and effective early warnings. This overcomes the shortcomings of existing methods that fall into "perception blind spots" due to "logical self-consistency," significantly improving the accuracy and timeliness of natural disaster monitoring and early warning, and providing more reliable technical support for disaster prevention and mitigation. Attached Figure Description

[0064] Figure 1 This is a flowchart of a natural disaster monitoring information processing method according to one embodiment of the present invention;

[0065] Figure 2 This is one of the flowcharts of a natural disaster monitoring information processing method according to another embodiment of the present invention;

[0066] Figure 3 This is a second flowchart of a natural disaster monitoring information processing method according to another embodiment of the present invention;

[0067] Figure 4 This is a third flowchart of a natural disaster monitoring information processing method according to another embodiment of the present invention;

[0068] Figure 5 This is a fourth flowchart of a natural disaster monitoring information processing method according to another embodiment of the present invention;

[0069] Figure 6 This is the fifth flowchart of a natural disaster monitoring information processing method according to another embodiment of the present invention;

[0070] Figure 7 This is a system block diagram of a natural disaster monitoring information processing system according to another embodiment of the present invention;

[0071] Explanation of reference numerals in the attached figures:

[0072] 1. Natural disaster monitoring and information processing system; 11. Displacement data receiving module; 12. Displacement trend judgment module; 13. Credibility marking module; 14. Limitation of impact execution module. Detailed Implementation

[0073] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0074] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0075] In the field of natural disaster monitoring, traditional information processing methods suffer from data fusion weight "drift" when faced with slow, non-hazardous changes in the external environment. For example, non-hazardous factors may cause deep displacement data to show a continuous upward trend, which the system may misjudge as a valid precursor signal, thus assigning it excessive weight during offline optimization and recalibration. This dilutes the weight of real disaster signals, causing the system to fail to issue timely and effective warnings, missing the optimal opportunity for disaster prevention and mitigation.

[0076] In response, this application proposes a method for processing natural disaster monitoring information, combining... Figure 1 As shown, it includes:

[0077] S1 receives deep displacement data and the accompanying physical information corresponding to the physical causes of the deep displacement data;

[0078] S2, perform trend judgment on deep displacement data to determine whether the deep displacement data shows a specific growth trend and obtain the trend judgment result;

[0079] S3, When the trend judgment result indicates that the deep displacement data shows a specific growth trend, the physical cause of the deep displacement data is judged based on the accompanying physical information to obtain the credibility label of the deep displacement data.

[0080] S4. If the confidence level indicator indicates that the confidence level is lower than the preset lower limit, then the influence of deep displacement data on risk assessment or parameter adjustment is limited or excluded.

[0081] To facilitate a clearer understanding of the technical solutions in this application, the key terms and implementation environment involved are explained below. Deep displacement data refers to data acquired through sensors such as deep displacement gauges, reflecting changes in displacement of the deep geological structure. It is typically used to monitor deformation within geological bodies and is an important basis for judging precursors to geological disasters such as landslides and collapses. Accompanying physical information refers to other physical quantities acquired simultaneously with or in conjunction with deep displacement data, reflecting the physical causes of the deep displacement data. Examples include regional geostress monitoring data, regional microseismic activity monitoring data, groundwater level data, soil moisture content data, and temperature data. This information helps to comprehensively determine the true cause of deep displacement changes. Trend judgment refers to analyzing the patterns of deep displacement data over time to determine whether it exhibits a specific growth pattern, such as linear growth, accelerated growth, or periodic growth. A specific growth trend refers to a pre-defined displacement growth pattern that may indicate abnormal changes in the geological body, such as continuous, non-linear, or accelerated displacement growth. A credibility marker is an identifier assigned to deep displacement data after assessing its reliability or validity. This marker indicates whether the data truly reflects precursors to geological disasters or whether it has been affected by non-hazardous factors. A preset lower limit is a threshold used to judge data credibility; when the data credibility falls below this lower limit, it indicates that the data may be unreliable or lack indicative significance. Risk assessment is the process of comprehensively judging the probability, potential impact, and severity of geological disasters based on monitoring data. Parameter adjustment refers to the dynamic correction of warning thresholds, model parameters, or data weights within the monitoring system based on monitoring data and risk assessment results to optimize system performance and accuracy. The natural disaster monitoring information processing method of this application is typically deployed at monitoring stations or data processing centers in geologically hazardous areas. This environment typically includes various geological sensors, data acquisition units, data transmission networks, and computing servers or cloud computing platforms for data processing and analysis. After being received, the data undergoes preprocessing, analysis, and judgment, ultimately guiding risk assessment and dynamic adjustment of system parameters.

[0082] Specifically, the various technical features of this application can be implemented in the following ways.

[0083] Regarding the reception of deep displacement data and the accompanying physical information corresponding to the physical causes of deep displacement data, deep displacement data can be acquired in real time using deep displacement gauges deployed deep within the strata and transmitted to a data processing center via wired or wireless networks. Accompanying physical information can be acquired synchronously by other sensors and received together with the deep displacement data after time-stamping. Alternatively, deep displacement data can be imported in batches from historical databases, which may have already undergone preliminary cleaning and formatting. Accompanying physical information can be obtained from different data sources and received via data interfaces or file transfers. During the reception process, data verification mechanisms can be employed to ensure the integrity and accuracy of the data.

[0084] To determine the trend of deep displacement data and whether it exhibits a specific growth trend, linear regression analysis can be used. This involves calculating the slope of the displacement data over time and comparing it to a preset threshold to determine if a sustained growth trend exists. For example, if the slope consistently exceeds a certain positive value, a specific growth trend is considered to be present. As a preferred implementation, moving averages or exponential smoothing can be used to smooth the deep displacement data. Then, by comparing the current data point with the average or smoothed value of the previous time period, it can be determined whether an accelerating growth trend is present. For example, if the smoothed displacement value shows an increase for several consecutive periods, and the rate of increase gradually increases, it is determined to exhibit a specific growth trend. Furthermore, time series analysis models, such as autoregressive integral moving average models or long short-term memory networks, can be used to model and predict deep displacement data. By comparing the deviation between the actual displacement data and the model's predicted values, and combining this with statistical significance tests, it can be determined whether an abnormal, sustained growth trend exists.

[0085] When the trend assessment indicates that deep displacement data exhibits a specific growth trend, the system determines the physical causes of the deep displacement data based on accompanying physical information to obtain a confidence level label. When deep displacement data shows a specific growth trend, the system can first check whether there are clear hazard indicators in the accompanying physical information, such as short-term heavy rainfall, significant regional ground stress anomalies, or high-frequency microseismic activity. If such clear indicators exist, the deep displacement data is marked as high confidence. If no clear indicators exist, further analysis is required. As a preferred implementation, a rule-based expert system can be constructed. When deep displacement data shows a specific growth trend, the system will infer the possible causes of deep displacement based on preset physical association rules and accompanying physical information. For example, if the increase in deep displacement occurs simultaneously with a continuous, slow rise in groundwater levels, and there are no other hazard signals, it may be inferred to be non-hazardous creep and marked as low confidence. Furthermore, machine learning classification models, such as support vector machines or decision trees, can also be used. This model learns the relationship between deep displacement data, accompanying physical information, and the actual physical causes in historical data. When new deep displacement data shows a specific growth trend, the model inputs its corresponding accompanying physical information, outputs the probability distribution of its physical causes, and assigns a confidence level label accordingly. For example, if the model determines that the probability of non-catastrophic creep is high, it is labeled as having a confidence level below a preset lower limit.

[0086] Regarding limiting or excluding the influence of deep displacement data on risk assessment or parameter adjustment if the confidence level label indicates that the confidence level is below a preset lower limit, when the confidence level label of deep displacement data indicates that its confidence level is below the preset lower limit, the weight of the deep displacement data in the calculation of the comprehensive risk index can be reduced during the risk assessment process. For example, its weight can be reduced from 0.4 to 0.1, thereby reducing its contribution to the final risk assessment result. As a preferred implementation, deep displacement data with a confidence level below the preset lower limit can be completely excluded from the dataset used for system parameter adjustment. This means that this data will not participate in the subsequent machine learning model training or optimization process, avoiding its negative impact on the system's adaptive capability. In addition, deep displacement data with a confidence level below the preset lower limit can be specially labeled and stored in isolation. When conducting risk assessment, the system will prioritize the use of high-confidence data and only refer to these low-confidence data when necessary, but will apply strict correction factors or manual review processes to them.

[0087] Optional, combined Figure 2 As shown in Figure S3, when the trend judgment result indicates that the deep displacement data exhibits a specific growth trend, the steps for determining the physical cause of the deep displacement data based on the accompanying physical information and obtaining the confidence label of the deep displacement data include:

[0088] S31. When the accompanying physical information does not clearly indicate the physical cause, and the regional geostress monitoring data and regional microseismic activity monitoring data do not show any abnormalities, acquire the electrochemical response data of the anchor point interface of the deep displacement gauge.

[0089] S32, based on electrochemical response data, analyze whether microscopic chemical reactions related to the growth trend occur at the anchoring point interface;

[0090] S33, if a microchemical reaction related to the growth trend occurs at the anchor point interface, it is determined that the deep displacement data is caused by the microchemical reaction, and the deep displacement data is marked with a confidence level lower than the preset lower limit.

[0091] In deep displacement gauges, the anchoring interface refers to the fixed point where the gauge connects to the surrounding soil or structure. Its stability directly affects the accuracy of the displacement data. Electrochemical response data can be understood as real-time monitoring of the electrochemical properties of the interface material, such as impedance, potential, and current, by deploying an electrochemical sensor at the anchoring interface. Changes in these parameters reflect whether microscopic chemical reactions such as corrosion, dissolution, or material fatigue occur at the anchoring interface. Further, after acquiring the electrochemical response data, it is necessary to analyze whether microscopic chemical reactions related to the growth trend occur at the anchoring interface. These microscopic chemical reactions refer to chemical processes that may lead to anchoring loosening, failure, or sensor degradation, such as electrochemical corrosion of metal anchors or aging and decomposition of polymer encapsulation materials. These reactions may cause the measurement reference of the deep displacement gauge to drift, resulting in displacement data that does not reflect the true geological displacement growth trend. Therefore, if the analysis results indicate that microscopic chemical reactions related to the growth trend of deep displacement data have indeed occurred at the anchoring interface, it can be determined that the deep displacement data is not caused by actual geological displacement but by microscopic chemical reactions at the anchoring interface. In this scenario, to ensure the accuracy of subsequent risk assessments or parameter adjustments, deep displacement data will be marked with a confidence level below a preset lower limit. This preset lower limit can be set according to the actual application scenario and the requirements for data reliability; for example, data below a certain threshold may be considered unreliable.

[0092] In some preferred embodiments, a specific example is given below. Suppose that in a geological disaster monitoring area, a deep displacement gauge continuously monitors its deep displacement data, showing a slow but persistent upward trend. However, concurrently monitored physical information such as rainfall, temperature, and groundwater level are all within normal ranges, and regional stress monitoring data and regional microseismic activity monitoring data do not show any abnormal fluctuations. In this situation, relying solely on traditional macroscopic accompanying physical information would make it difficult to determine whether the increase in deep displacement data is due to genuine geological creep or caused by other factors.

[0093] At this point, the proposed solution is activated. The system first determines that the accompanying physical information does not clearly indicate the physical cause, and that neither the regional geostress monitoring data nor the regional microseismic activity monitoring data show any abnormalities. Subsequently, the system acquires the electrochemical response data of the anchoring point interface of the deep displacement gauge. By analyzing this electrochemical response data, for example, the interfacial impedance parameter and redox peak parameter are extracted. If the interfacial impedance parameter shows a continuous decreasing trend, and / or the redox peak parameter shows a continuous increasing trend or a peak potential shift trend, this indicates that the anchoring point interface may be undergoing microscopic chemical reactions such as corrosion or material degradation.

[0094] Based on this analysis, the system determined that the deep displacement data was not caused by actual geological displacement, but rather by a microscopic chemical reaction at the anchoring point interface. Therefore, the deep displacement data was marked with a confidence level below a preset lower limit. In this way, the abnormal data was effectively identified and its impact on risk assessment or parameter adjustment was limited, avoiding misjudgments caused by sensor malfunctions and ensuring the accuracy and reliability of the monitoring system.

[0095] Optional, combined Figure 3 As shown, S32, based on electrochemical response data, analyzes whether microscopic chemical reactions related to the growth trend occur at the anchoring point interface, including the following steps:

[0096] S321, Extract interfacial impedance parameters and / or redox peak parameters from electrochemical response data;

[0097] S322, determine whether the interface impedance parameter shows a continuous decreasing trend, and / or whether the redox peak parameter shows a continuous increasing trend or a peak potential shift trend, and obtain the parameter judgment result;

[0098] S323, if the parameter judgment result is yes, then it is determined that a microscopic chemical reaction related to the growth trend has occurred at the anchoring point interface.

[0099] Specifically, electrochemical response data can be acquired by installing an electrochemical sensor at the anchoring point interface of the deep displacement gauge. This sensor can monitor the electrochemical state at the anchoring point interface in real time or periodically. Interface impedance parameters refer to parameters such as interface resistance and capacitance measured by electrochemical impedance spectroscopy (EIS), reflecting charge transfer resistance, diffusion resistance, and double-layer characteristics at the interface. Redox peak parameters are typically obtained using techniques such as cyclic voltammetry (CV) or linear sweep voltammetry (LSV), including oxidation peak current, reduction peak current, and peak potential. These parameters are closely related to the type, rate, and reversibility of redox reactions occurring at the interface. In practical applications, when microscopic chemical reactions occur at the anchoring point interface, such as corrosion, dissolution, or the formation of new compounds, its electrochemical characteristics change significantly. For example, corrosion typically leads to a decrease in interface impedance because the formation of corrosion products or metal dissolution increases the conductivity of the interface or alters its structure. Simultaneously, certain redox reactions can increase the redox peak current or shift the peak potential, indicating the formation of new electrochemically active substances or changes in the reaction pathway. Therefore, by monitoring the changing trends of these parameters, it is possible to effectively determine whether a microscopic chemical reaction related to the deep displacement growth trend is occurring at the anchorage interface.

[0100] Optional, combined Figure 4 As shown, the steps in S4 to limit or exclude the influence of deep displacement data on risk assessment or parameter adjustment include:

[0101] S41, Obtain the confidence label and causation label corresponding to the deep displacement data;

[0102] S42, Determine the risk level of deep displacement data based on confidence and causal markers;

[0103] S43. Select the corresponding impact limitation strategy based on the risk level;

[0104] S44. In accordance with the influence limitation strategy, limit or eliminate the influence of deep displacement data on risk assessment or parameter adjustment.

[0105] The acquisition of credibility and causal tags for deep displacement data refers to the preliminary processing performed by the system after receiving the data to obtain a reliability assessment result (credibility tag) and the potential causes leading to its specific growth trend (causal tag). The credibility tag indicates whether the deep displacement data is reliable; for example, a credibility level below a preset lower limit may indicate data anomalies or interference from non-geological factors. The causal tag identifies the physical causes of the growth trend in the deep displacement data, such as geological tectonic activity, groundwater changes, temperature effects, or other non-geological factors (e.g., equipment failure, environmental disturbance). These tags are typically generated after the deep displacement data is received and preliminarily assessed.

[0106] Furthermore, the risk level of the deep displacement data is determined based on the reliability and causal markers. This step aims to comprehensively consider the reliability of the data and its physical causes to quantitatively assess the potential risks that the deep displacement data may cause. For example, if the deep displacement data is marked as low reliability and the causal marker points to non-geological factors, its risk level may be rated as low, indicating that its reference value for actual geological hazard risk assessment is limited. Conversely, if the data has high reliability and the causal marker points to potential precursors of geological hazards, its risk level may be rated as high. The determination of the risk level can be based on pre-defined rules, expert knowledge systems, or machine learning models, mapping the reliability and causal markers to different risk levels, such as "high risk," "medium risk," "low risk," or "invalid data."

[0107] Based on this, appropriate impact mitigation strategies are selected according to the risk level. Different risk levels require different handling methods to avoid unreliable or misleading data negatively impacting subsequent risk assessments and parameter adjustments. For example, deep displacement data assessed as "invalid data" can be completely excluded; for "low-risk" data, its weight can be limited or it can be used only as a reference; for "medium-risk" data, further verification or correction may be required before use; and for "high-risk" data, even if its reliability is questionable, it may need to be handled with caution to prevent underreporting of true risks. Impact mitigation strategies can include data removal, data weight reduction, data correction, data isolation, or use only for auxiliary reference, among other methods.

[0108] Ultimately, the impact of deep displacement data on risk assessment or parameter adjustment is limited or excluded according to the influence restriction strategy. This means that, based on the previously determined risk level and selected strategy, the system will intervene in the role of deep displacement data in subsequent stages such as risk assessment model input, early warning threshold adjustment, and geological parameter calibration. For example, if the strategy is "data removal," the deep displacement data will not be included in the risk assessment calculation; if the strategy is "data deweighting," the data will be assigned a lower weight in the calculation; if the strategy is "data isolation," the data may be stored and analyzed separately, without being mixed with mainstream data. This aims to ensure the accuracy of risk assessment and the effectiveness of parameter adjustment, avoiding misjudgments or improper operations due to erroneous or unreliable data.

[0109] Optionally, the step of determining the risk level of deep displacement data based on confidence and causal labels may include the following operations:

[0110] Based on the confidence level marker and the cause marker, the corresponding physical parameters are identified;

[0111] Analyze the numerical values ​​or trends of physical parameters;

[0112] Determine whether the value or trend of a physical parameter exceeds a preset threshold, and obtain the physical parameter judgment result;

[0113] If the physical parameter judgment result is yes, then adjust and determine the risk level corresponding to the deep displacement data.

[0114] The identification of corresponding physical parameters refers to the system's ability to intelligently filter out the physical parameters most relevant to the current monitoring context and potential risk type, based on the credibility markers of deep displacement data and the identified physical causal markers. For example, if the deep displacement data is marked as being caused by groundwater activity, physical parameters such as pore water pressure and groundwater level changes may be identified; if the causal markers point to geological tectonic activity, parameters such as in-situ stress and microseismic activity frequency may be identified. The selection of these physical parameters aims to provide a deeper physical insight into deep displacement phenomena.

[0115] Furthermore, analyzing the numerical values ​​or trends of physical parameters refers to continuously monitoring the identified physical parameters and conducting in-depth analysis of their current values ​​and trends over time. This can include assessing the absolute values, rates of change, accelerations, and correlations with other relevant parameters. For example, a sustained increase in pore water pressure or a rapid accumulation of geostress may indicate potential risks.

[0116] Therefore, determining whether the value or trend of a physical parameter exceeds a preset threshold involves comparing the analyzed physical parameter value or trend with a pre-set safety threshold. This preset threshold is determined based on historical data, geological models, or expert experience and is used to define the boundary between normal fluctuations and abnormal changes. When the value or trend of a physical parameter breaks through these thresholds, the physical parameter judgment result is positive, indicating a potential escalation of risk.

[0117] If the physical parameter assessment result is positive, the risk level corresponding to the deep displacement data is adjusted and determined. This means that once the physical parameters show anomalies, even if the reliability of the deep displacement data itself is low or the cause is unclear, the system will adjust or more accurately determine the risk level indicated by the deep displacement data based on these more physically meaningful parameter changes to reflect the actual risk situation.

[0118] In some embodiments of this application, when determining the risk level of deep displacement data based on confidence and causal markers, it is necessary to determine whether the values ​​or trends of physical parameters exceed preset thresholds. However, in practical applications, the occurrence and evolution of natural disasters are often influenced by various dynamic factors such as regional environmental changes and geological background evolution. If the preset threshold remains fixed, it may not accurately reflect the current actual geological environment, leading to a decrease in the accuracy of risk assessment, and even potential misjudgments or omissions. Therefore, this application further proposes an optimized method for determining whether physical parameters exceed thresholds, aiming to improve the accuracy and adaptability of risk assessment by dynamically adjusting the preset threshold.

[0119] Optionally, the steps for determining whether the value or trend of a physical parameter exceeds a preset threshold and obtaining the physical parameter determination result include:

[0120] To obtain data on regional environmental changes or geological background evolution;

[0121] Calculate the adaptive bias of the preset threshold based on regional environmental change data or geological background evolution data;

[0122] Determine whether the adaptive deviation exceeds the predetermined range, and obtain the deviation judgment result;

[0123] If the deviation judgment result is yes, then adjust the preset threshold;

[0124] Based on the adjusted preset threshold, determine whether the value or trend of the physical parameter exceeds the adjusted preset threshold.

[0125] Specifically, acquiring regional environmental change data or geological background evolution data refers to collecting dynamic information on external environmental factors and the internal state of geological bodies related to the monitoring area. Regional environmental change data may include, but is not limited to, meteorological and hydrological data such as rainfall, temperature, groundwater level, river flow, and soil moisture, as well as data on human activities such as engineering construction and reservoir water storage and release. Geological background evolution data may include data on changes in geostress, microseismic activity, surface deformation, and changes in groundwater chemical composition, reflecting long-term or short-term changes in the internal state of geological bodies. This data can be acquired through various means, including various sensors deployed within the monitoring area, remote sensing technology, historical data records, and geological exploration reports. The aim is to provide comprehensive background information for subsequent threshold adaptive adjustments.

[0126] Furthermore, based on regional environmental change data or geological background evolution data, the adaptive bias of the preset threshold is calculated. Adaptive bias refers to the difference between the preset threshold and the actual required threshold under the current regional environmental and geological background. This calculation can be based on historical data, statistical models, machine learning algorithms, or expert rules. For example, a mapping model can be established by analyzing the relationship between historical environmental changes and threshold effectiveness. When new environmental data is input, the model can output a suggested threshold adjustment amount, i.e., the adaptive bias.

[0127] Next, it is determined whether the adaptive bias exceeds a predetermined range. The predetermined range refers to the allowable adjustment magnitude of the threshold; for example, a percentage range (e.g., ±10%) or an absolute value range can be set. This step aims to avoid over-adjusting the threshold and ensure the rationality and stability of the adjustment. If the bias is too large, it may indicate abnormally drastic changes in the current environment or a problem with the calculation model, which may require manual intervention or deeper analysis. If the bias determination result is yes, meaning the adaptive bias exceeds the predetermined range, the preset threshold is adjusted. The adjustment method can be simple addition or subtraction operations, or a more complex non-linear adjustment, to make the threshold more closely reflect the current actual situation.

[0128] Finally, based on the adjusted preset threshold, it is determined whether the values ​​or trends of the physical parameters exceed the adjusted preset threshold. After the threshold is dynamically adjusted, the physical parameters (such as displacement rate, acceleration, etc.) corresponding to the deep displacement data are compared with this new, more adaptive threshold. This step ensures that the risk assessment is based on a dynamic threshold that matches the current environmental and geological conditions, thereby improving the accuracy and reliability of the judgment.

[0129] In some preferred embodiments, a specific example is given below. Suppose that in a landslide monitoring area, a deep displacement gauge continuously detects a small increase in displacement. According to a conventional fixed preset threshold, this displacement data may not have reached the critical value that triggers an early warning. However, the solution of this application further acquires regional environmental change data for the area, such as monitoring recent continuous heavy rainfall, leading to a significant increase in soil moisture content; simultaneously, geological background evolution data, such as regional microseismic activity monitoring, shows that internal stress is accumulating in the strata. The system calculates an adaptive deviation of the preset displacement threshold based on this rainfall and microseismic activity data. For example, through historical data analysis, it is known that under conditions of continuous heavy rainfall and stress accumulation, the critical displacement threshold for landslides will significantly decrease. If the calculated adaptive deviation indicates that the current threshold should be lowered, and the deviation is within a predetermined range, the system will adjust the original preset threshold. At this time, even if the deep displacement data has not yet reached the original fixed threshold, it may have already exceeded the adjusted, lower adaptive threshold. As a result, the system can promptly determine the increased risk level corresponding to the deep displacement data and trigger corresponding early warnings or take restrictive measures, thereby effectively avoiding the underestimation of potential risks caused by environmental changes and significantly improving the timeliness and accuracy of early warnings.

[0130] Optionally, the step of calculating the adaptive bias of the preset threshold includes:

[0131] Analyze the correlation between regional environmental change data or geological background evolution data and historical threshold applicability data to obtain correlation information;

[0132] Based on the associated information, construct a mapping relationship for the associated information;

[0133] Based on current regional environmental change data or geological background evolution data and mapping relationships, the adaptive bias of the preset threshold is calculated.

[0134] This study analyzes the correlation between regional environmental change data or geological background evolution data and historical threshold applicability data, aiming to identify key factors influencing threshold applicability and their interrelationships. Historical threshold applicability data can include cases where preset thresholds have proven effective or ineffective under different environmental or geological conditions in the past, along with their corresponding environmental or geological parameters. Analysis of this data can reveal the patterns of threshold performance under different contexts.

[0135] Furthermore, based on the obtained correlation information, a mapping relationship can be constructed. This mapping relationship can be a mathematical model, a lookup table, a rule set, or a machine learning model. Its purpose is to transform current regional environmental change data or geological background evolution data into predictions or calculations of adaptive biases to preset thresholds through identified correlations. For example, if the correlation information indicates that an increase in a certain environmental parameter will lead to a downward adjustment of the threshold, then the mapping relationship will reflect this adjustment rule.

[0136] Finally, based on the currently acquired regional environmental change data or geological background evolution data, and combined with the established mapping relationship, the adaptability deviation of the preset threshold can be calculated. This deviation quantifies the difference between the preset threshold and the actual needs under the current environmental or geological conditions, providing a quantitative basis for subsequent threshold adjustments.

[0137] In some of the embodiments described above in this application, it is proposed to obtain correlation information by analyzing the correlation between regional environmental change data or geological background evolution data and historical threshold applicability data. However, in practical applications, the original regional environmental change data or geological background evolution data may have problems such as missing data, noise, or heterogeneity. If correlation analysis is performed directly, the accuracy of the correlation information may be insufficient, thereby affecting the calculation accuracy of the preset threshold adaptability deviation.

[0138] Optional, combined Figure 5 As shown, the steps to analyze the correlation between regional environmental change data or geological background evolution data and historical threshold applicability data to obtain correlation information include:

[0139] A1. Perform data quality processing on regional environmental change data or geological background evolution data. Data quality processing includes imputing missing data based on the spatiotemporal correlation of the data, adaptive filtering of noisy data, and feature alignment and normalization of heterogeneous data.

[0140] A2. Based on regional environmental change data or geological background evolution data that has undergone data quality processing, combined with historical threshold applicability data, multivariate statistical analysis and nonlinear regression analysis are performed to identify the correlation between data and obtain correlation information.

[0141] Specifically, data quality processing refers to preprocessing raw regional environmental change data or geological background evolution data to improve data integrity, accuracy, and consistency. This includes imputation of missing data based on spatiotemporal correlation. When gaps exist in the monitoring data, the missing values ​​are estimated and filled using the data's temporal and spatial continuity or correlation. Methods such as linear interpolation, spline interpolation, kriging interpolation, or machine learning-based methods can be used for imputation. The aim is to ensure the integrity of the data sequence and avoid impacting subsequent analysis due to data discontinuity. Adaptive filtering of noisy data refers to processing random interference, sensor errors, or environmental noise that may occur during monitoring. Algorithms that dynamically adjust filtering parameters based on data characteristics can be used. Methods such as Kalman filtering, wavelet denoising, or adaptive median filtering can be employed. The goal is to remove invalid information from the data and improve the signal-to-noise ratio. Feature alignment and normalization of heterogeneous data refers to making data comparable by unifying the data representation and units when the data comes from different sensors, different measurement standards, or different data formats. For example, feature alignment can be performed by using min-max normalization, Z-score normalization, or principal component analysis (PCA). The purpose is to eliminate the heterogeneity between data and ensure the effectiveness of joint analysis of different types of data.

[0142] After data quality processing, multivariate statistical analysis and nonlinear regression analysis are performed based on the processed regional environmental change data or geological background evolution data, combined with historical threshold applicability data. Multivariate statistical analysis considers the relationships between multiple variables simultaneously, employing methods such as correlation analysis, cluster analysis, factor analysis, or principal component analysis. Its purpose is to reveal potential correlation patterns between multiple environmental or geological factors and historical threshold applicability. Nonlinear regression analysis establishes mathematical models of nonlinear relationships between variables, using methods such as multinomial regression, support vector regression (SVR), neural networks, or Gaussian process regression. Its purpose is to capture complex nonlinear dependencies between data, thereby more accurately quantifying the impact of environmental or geological changes on threshold applicability.

[0143] In some preferred embodiments, a specific example is given below. Assume this application is applied to natural disaster monitoring in a landslide-prone area. Regional environmental change data for this area includes rainfall, temperature, and soil moisture, while geological background evolution data includes surface deformation and groundwater level changes. Historical threshold applicability data includes environmental and geological parameter thresholds and their applicability assessments at the time of past landslide events in the area. When analyzing the correlation between regional environmental change data or geological background evolution data and historical threshold applicability data, the raw data are first processed for data quality. For example, if rainfall data is missing, it can be interpolated based on data from adjacent monitoring points and time series patterns; if surface deformation sensor data is affected by environmental electromagnetic interference and generates noise, an adaptive Kalman filter algorithm can be used to remove the noise; if geological background evolution data comes from different types of sensors (such as GNSS displacement meters and InSAR satellite imagery), feature alignment and normalization are required to eliminate scale and format differences between different data sources. After data quality processing, this high-quality data, combined with historical threshold applicability data, is used for multivariate statistical analysis. For example, principal component analysis (PCA) identifies the main environmental and geological factors influencing landslide occurrence. Then, nonlinear regression analysis, such as support vector regression (SVR) or neural network models, establishes a complex nonlinear relationship model between these factors and historical threshold applicability, thereby accurately identifying the correlations between data and obtaining correlation information used to calculate the adaptive bias of preset thresholds. This allows for a more accurate assessment of the risk level under current environmental and geological conditions, and corresponding adjustments to the warning thresholds.

[0144] Optional, combined Figure 6 As shown, the steps for constructing the mapping relationship of association information based on the association information include:

[0145] B1, retrieve the preset physical association rules or experience lookup table;

[0146] B2, based on preset physical association rules or experience lookup tables, maps regional environmental change data or geological background evolution data to a preset threshold adaptive deviation, and constructs a mapping relationship of associated information.

[0147] The preset physical association rules can be understood as mathematical models or logical rules pre-established based on physical principles, geomechanical models, or engineering experience. These rules describe the quantitative or qualitative relationship between regional environmental change data or geological background evolution data and the adaptive deviation of preset thresholds. For example, these rules can define how the deep displacement monitoring threshold should be adjusted under specific temperature, humidity, or formation pressure changes. The experience lookup table is a data table constructed through historical data analysis, expert experience summaries, or simulation experiment results. It contains the corresponding adaptive deviation values ​​of preset thresholds under different combinations of regional environmental change data or geological background evolution data. When receiving current regional environmental change data or geological background evolution data, the system can quickly obtain the corresponding adaptive deviation by looking up the table. Mapping regional environmental change data or geological background evolution data to the adaptive deviation of preset thresholds refers to the process of transforming or associating input data (regional environmental change data or geological background evolution data) with output data (adaptive deviation of preset thresholds) through the aforementioned rules or lookup tables. Its purpose is to dynamically and accurately adjust monitoring thresholds in complex and changing environments to better reflect actual conditions, thereby improving the accuracy of risk assessment.

[0148] This application also discloses a natural disaster monitoring information processing system for performing natural disaster monitoring information processing, combined with... Figure 7 As shown, the natural disaster monitoring information processing system 1 includes:

[0149] The displacement data receiving module 11 is used to receive deep displacement data and the accompanying physical information corresponding to the physical causes of the deep displacement data;

[0150] The displacement trend judgment module 12 is used to judge the trend of deep displacement data, determine whether the deep displacement data shows a specific growth trend, and obtain the trend judgment result.

[0151] The credibility labeling module 13 is used to determine the physical cause of the deep displacement data based on the accompanying physical information when the trend judgment result indicates that the deep displacement data shows a specific growth trend, and to obtain the credibility label of the deep displacement data.

[0152] The restriction effect execution module 14 is used to restrict or exclude the influence of deep displacement data on risk assessment or parameter adjustment if the confidence level indicated by the confidence level flag is lower than the preset lower limit.

[0153] The natural disaster monitoring and information processing system proposed in this application aims to address the problem that traditional methods may experience data fusion weight "drift" when facing slow, non-hazardous changes in the external environment, leading to a "perception blind spot" in the system due to "logical self-consistency." This system, through a modular design, achieves intelligent processing and reliability management of deep displacement data. Specifically, the displacement data receiving module is responsible for comprehensive data acquisition, the displacement trend judgment module performs preliminary data screening, the reliability labeling module analyzes the physical causes of the data in depth and assigns reliability labels, and finally, the influence limitation execution module intelligently adjusts the data based on the impact of the reliability labels. Therefore, this system can effectively avoid the interference of "false precursor" data caused by non-hazardous factors on the accuracy of the system's early warning, significantly improving the reliability and timeliness of natural disaster monitoring and early warning.

[0154] Specifically, the various technical features of this application can be implemented in the following ways.

[0155] Regarding the displacement data receiving module, this module can be configured to connect to sensors such as deep displacement gauges via wired or wireless communication interfaces to receive deep displacement data in real time. Simultaneously, this module can also receive accompanying physical information corresponding to the physical causes of deep displacement data from other monitoring devices or databases via network interfaces or data buses, such as regional geostress monitoring data, regional microseismic activity monitoring data, and groundwater level data. As one implementation, the displacement data receiving module can be a standalone hardware unit, such as a data acquisition card or embedded processor, integrating the necessary communication protocol stack and data caching mechanism. Alternatively, the module can be a software service deployed on a server or cloud computing platform, receiving data streams from different data sources through application programming interfaces (APIs) or message queues.

[0156] Regarding the displacement trend judgment module, this module is configured to perform trend analysis on the deep displacement data received by the displacement data receiving module. The specific method for judging the trend of deep displacement data has already been described in the above embodiments and will not be repeated here. It should be emphasized that the displacement trend judgment module can be a dedicated signal processing unit, such as a digital signal processor (DSP) or a field-programmable gate array (FPGA), for efficiently executing complex trend analysis algorithms. Alternatively, this module can also be a software component running on a general-purpose processor, implementing the trend judgment function by calling statistical analysis libraries or machine learning frameworks.

[0157] Regarding the credibility labeling module, this module is configured to determine the physical cause of the deep displacement data and obtain a credibility label for the deep displacement data based on the accompanying physical information when the trend judgment result of the displacement trend judgment module indicates that the deep displacement data exhibits a specific growth trend. The specific method for determining the physical cause of deep displacement data and obtaining the credibility label has already been described in the above embodiments and will not be repeated here. It is important to emphasize that the credibility labeling module can be a rule-based reasoning expert system, which internally stores a series of preset physical association rules for inferring the cause of displacement based on the accompanying physical information. As a preferred implementation, this module can be an inference engine integrating a machine learning model. This model learns the complex relationship between deep displacement data, accompanying physical information, and the true physical cause through offline training, thereby enabling real-time classification and credibility assessment of new data.

[0158] Regarding the impact restriction execution module, this module is configured to restrict or exclude the influence of deep displacement data on risk assessment or parameter adjustment if the confidence mark obtained by the confidence mark module indicates a confidence level lower than a preset lower limit. The specific methods for restricting or excluding the influence of deep displacement data on risk assessment or parameter adjustment have already been described in the above embodiments and will not be repeated here. It is important to emphasize that the impact restriction execution module can be a strategy management unit that dynamically adjusts the weight of deep displacement data in the risk assessment model based on the confidence mark, or removes it from the dataset used for system parameter adjustment. For example, this module can be a software agent that intercepts and filters data before it enters the risk assessment or parameter adjustment process, ensuring that only high-confidence data is used for critical decisions. In some implementations, this module can also trigger a manual review process, reminding operators to manually intervene and confirm when the data confidence level is questionable.

[0159] The above are merely embodiments of this application and are not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for processing natural disaster monitoring information, characterized in that, include: Receive deep displacement data and the accompanying physical information corresponding to the physical causes of the deep displacement data; The deep displacement data is subjected to trend judgment to determine whether the deep displacement data shows a specific growth trend, and the trend judgment result is obtained; When the trend judgment result indicates that the deep displacement data shows a specific growth trend, the physical cause of the deep displacement data is determined based on the accompanying physical information, and the credibility label of the deep displacement data is obtained. If the confidence level indicator indicates that the confidence level is lower than the preset lower limit, then the influence of the deep displacement data on risk assessment or parameter adjustment is limited or eliminated. The step of determining the physical cause of the deep displacement data and obtaining a credibility marker for the deep displacement data based on the accompanying physical information when the trend judgment result indicates that the deep displacement data exhibits a specific growth trend includes: When the accompanying physical information does not clearly indicate the physical cause, and the regional geostress monitoring data and regional microseismic activity monitoring data do not show any abnormalities, the electrochemical response data of the anchor point interface of the deep displacement gauge is obtained. Based on the electrochemical response data, analyze whether microscopic chemical reactions related to the growth trend occur at the anchoring point interface; If a microchemical reaction related to the growth trend occurs at the anchor point interface, it is determined that the deep displacement data is caused by the microchemical reaction, and the deep displacement data is marked with a confidence level lower than a preset lower limit.

2. The method for processing natural disaster monitoring information according to claim 1, characterized in that, The step of analyzing whether a microscopic chemical reaction related to the growth trend occurs at the anchoring point interface based on the electrochemical response data includes: Extract the interfacial impedance parameters and / or redox peak parameters from the electrochemical response data; Determine whether the interface impedance parameter shows a continuous decreasing trend, and / or whether the redox peak parameter shows a continuous increasing trend or a peak potential shift trend, to obtain the parameter determination result; If the parameter determination result is yes, then it is determined that a microscopic chemical reaction related to the growth trend has occurred at the anchoring point interface.

3. The method for processing natural disaster monitoring information according to claim 1, characterized in that, The steps for limiting or eliminating the impact of the deep displacement data on risk assessment or parameter adjustment include: Obtain the confidence marker and causal marker corresponding to the deep displacement data; The risk level of the deep displacement data is determined based on the confidence level and cause markers. Based on the stated risk level, select the corresponding impact limitation strategy; According to the aforementioned impact limitation strategy, the impact of the deep displacement data on risk assessment or parameter adjustment is limited or eliminated.

4. The method for processing natural disaster monitoring information according to claim 3, characterized in that, The step of determining the risk level of the deep displacement data based on the confidence marker and the causal marker includes: Based on the confidence marker and the cause marker, the corresponding physical parameters are identified; Analyze the numerical values ​​or trends of the physical parameters; Determine whether the value or trend of the physical parameter exceeds a preset threshold, and obtain the physical parameter determination result; If the physical parameter judgment result is yes, then adjust and determine the risk level corresponding to the deep displacement data.

5. The method for processing natural disaster monitoring information according to claim 4, characterized in that, The step of determining whether the value or trend of the physical parameter exceeds a preset threshold and obtaining the physical parameter determination result includes: To obtain data on regional environmental changes or geological background evolution; Based on the regional environmental change data or geological background evolution data, calculate the adaptive deviation of the preset threshold; Determine whether the adaptive deviation exceeds a predetermined range, and obtain the deviation determination result; If the deviation judgment result is yes, then the preset threshold is adjusted; Based on the adjusted preset threshold, it is determined whether the value or trend of the physical parameter exceeds the adjusted preset threshold.

6. The method for processing natural disaster monitoring information according to claim 5, characterized in that, The step of calculating the adaptive bias of the preset threshold includes: The correlation between the regional environmental change data or geological background evolution data and historical threshold applicability data is analyzed to obtain correlation information; Based on the associated information, a mapping relationship for the associated information is constructed; Based on the current regional environmental change data or geological background evolution data and the mapping relationship, the adaptive deviation of the preset threshold is calculated.

7. A method for processing natural disaster monitoring information according to claim 6, characterized in that, The step of analyzing the correlation between the regional environmental change data or geological background evolution data and historical threshold applicability data to obtain correlation information includes: Data quality processing is performed on the regional environmental change data or geological background evolution data; the data quality processing includes imputing missing data according to the spatiotemporal correlation of the data, adaptive filtering of noisy data, and feature alignment and normalization of heterogeneous data. Based on the regional environmental change data or geological background evolution data that has undergone data quality processing, combined with historical threshold applicability data, multivariate statistical analysis and nonlinear regression analysis are performed to identify the correlations between the data and obtain correlation information.

8. A method for processing natural disaster monitoring information according to claim 6, characterized in that, The step of constructing the mapping relationship of the association information based on the association information includes: Obtain preset physical association rules or experience lookup tables; Based on preset physical association rules or experience lookup tables, the regional environmental change data or geological background evolution data are mapped to an adaptive deviation of a preset threshold to construct a mapping relationship of associated information.

9. A natural disaster monitoring information processing system, used to perform natural disaster monitoring information processing, characterized in that, include: The displacement data receiving module is used to receive deep displacement data and the accompanying physical information corresponding to the physical cause of the deep displacement data; The displacement trend judgment module is used to judge the trend of the deep displacement data, determine whether the deep displacement data shows a specific growth trend, and obtain the trend judgment result. The credibility labeling module is used to determine the physical cause of the deep displacement data based on the accompanying physical information when the trend judgment result indicates that the deep displacement data shows a specific growth trend, and to obtain the credibility label of the deep displacement data. The limiting effect execution module is used to limit or exclude the impact of the deep displacement data on risk assessment or parameter adjustment if the confidence level indicated by the confidence level flag is lower than a preset lower limit. The credibility marking module is also used to acquire the electrochemical response data of the anchor point interface of the deep displacement gauge when the accompanying physical information does not clearly indicate the physical cause and the regional geostress monitoring data and regional microseismic activity monitoring data do not show any abnormalities. Based on the electrochemical response data, analyze whether microscopic chemical reactions related to the growth trend occur at the anchoring point interface; If a microchemical reaction related to the growth trend occurs at the anchor point interface, it is determined that the deep displacement data is caused by the microchemical reaction, and the deep displacement data is marked with a confidence level lower than a preset lower limit.

Images