Ground sensing network-based geological disaster real-time monitoring method and system

Abstract

The invention discloses a geological disaster real-time monitoring method and system based on a ground sensing network, and belongs to the technical field of geophysical exploration, and the method comprises the steps: obtaining an acceleration parameter and a moisture content parameter of each sensor node; according to the change rate of the acceleration parameter and the moisture content parameter, identifying an environment sudden change event, generating an environment trigger signal to activate an emergency sampling mode of a target node and an adjacent node, and constructing a dynamic monitoring cluster; requesting a plurality of nodes in the dynamic monitoring cluster to synchronously measure similar parameters, and comparing measurement results to generate a compressed alarm packet; adjusting weight factors of different monitoring parameters by using the compressed alarm packet and pre-acquired real-time environment parameters, and generating an environment calibration risk index; and when the environment calibration risk index exceeds a preset risk threshold value, outputting a geological disaster early warning instruction. According to the method, multi-parameter cooperative triggering, dynamic cluster response, weight adaptive evaluation and compression transmission technologies are adopted, and the monitoring precision, timeliness and environmental adaptability can be comprehensively improved.

Description

Technical Field

[0001] The present invention relates to the field of geophysical exploration technology, and in particular to a real-time monitoring method and system for geological disasters based on a ground sensor network. Background Art

[0002] Real-time geological disaster monitoring is a key branch of geophysical exploration, primarily focused on providing automated early warnings for natural disasters such as landslides, rockfalls, and debris flows. Traditional technologies rely on satellite remote sensing, manual inspections, and fixed sensor networks, which struggle to meet real-time requirements in complex terrain.

[0003] Existing technologies generally use fixed sensor nodes for periodic data collection, triggering alarms based on preset thresholds. Some solutions incorporate rainfall data for auxiliary judgment, but data transmission relies on wired networks or conventional wireless communication protocols. A few systems attempt multi-node collaborative sampling, but these systems employ a static management model with fixed groups.

[0004] However, the single threshold trigger mechanism is prone to false alarms and missed alarms when environmental parameters change dynamically; the fixed frequency sampling mode cannot capture the instantaneous mutation characteristics of geological disasters; the lack of spatial consistency verification of multi-node data leads to misjudgment of local interference; and the unoptimized transmission protocol leads to the loss of key data when the network is congested in remote mountainous areas. Summary of the Invention

[0005] To solve the above problems, the present invention provides a real-time monitoring method and system for geological disasters based on a ground sensor network, which adopts multi-parameter collaborative triggering, dynamic cluster response, weighted adaptive evaluation and compressed transmission technology, and can comprehensively improve monitoring accuracy, timeliness and environmental adaptability.

[0006] The above objectives can be achieved through the following solutions:

[0007] A real-time monitoring method and system for geological disasters based on a ground sensor network, comprising acquiring real-time monitoring data from each sensor node, the real-time monitoring data including acceleration parameters and water content parameters; identifying environmental mutation events based on the change rates of the acceleration parameters and water content parameters, and generating an environmental trigger signal; activating the emergency sampling mode of the target node and adjacent nodes based on the environmental trigger signal, and constructing a dynamic monitoring cluster; requesting multiple nodes in the dynamic monitoring cluster to synchronously measure similar parameters, comparing the measurement results to generate a compressed alarm package; using the compressed alarm package and pre-acquired real-time environmental parameters to adjust weight factors of different monitoring parameters, and generating an environmental calibration risk index; and outputting a geological disaster warning instruction when the environmental calibration risk index exceeds a preset risk threshold.

[0008] Optionally, generating the environmental trigger signal includes: calculating the instantaneous change rate of the acceleration parameter, and when the instantaneous change rate exceeds a preset acceleration mutation threshold, determining it as a first type of trigger event; calculating the minute-level increase of the moisture content parameter, and when the minute-level increase exceeds a preset moisture content mutation threshold, determining it as a second type of trigger event; and combining the first type of trigger event and the second type of trigger event to generate the environmental trigger signal.

[0009] Optionally, the construction of a dynamic monitoring cluster includes: sending the environmental trigger signal to nodes within a preset neighborhood range via short-range wireless broadcast; the node receiving the environmental trigger signal switches to a millisecond sampling frequency and returns a ready confirmation signal to the source node; constructing a temporary communication group including a source node and a response node based on the ready confirmation signal, and generating a status identifier of the dynamic monitoring cluster.

[0010] Optionally, the generation of a compressed alarm package based on the comparison measurement results includes: extracting a list of available nodes from the status identifier of the dynamic monitoring cluster; sending a synchronous measurement instruction to at least three nodes in the available node list; receiving synchronous measurement data returned by each node, and calculating the discreteness of similar parameters; when the discreteness is less than a preset tolerance threshold, generating the compressed alarm package.

[0011] Optionally, when the discreteness is less than the preset tolerance threshold, generating the compressed alarm package includes: when the discreteness is less than the preset tolerance threshold, extracting the acceleration peak, water content change gradient and inclination angle parameters from the synchronous measurement data to obtain a basic data set; associating the basic data set with the geographic location code of the dynamic monitoring cluster, and encapsulating it to obtain a binary data packet including a data header, a payload and a check code; and obtaining a compressed alarm package based on the binary data packet.

[0012] Optionally, the use of the compressed alarm package and pre-acquired real-time environmental parameters to adjust the weight factors of different monitoring parameters to generate an environmental calibration risk index includes: parsing the compressed alarm package to obtain a set of monitoring parameters; synchronously obtaining current rainfall intensity, soil temperature and historical displacement to generate real-time environmental parameters; matching the parameter weight distribution scheme in a preset weight mapping table according to the real-time environmental parameters to obtain an adjusted weight factor; and calculating the environmental calibration risk index using the adjusted weight factor and the compressed alarm package.

[0013] Optionally, the parameter weight distribution scheme includes: when the current rainfall intensity exceeds a preset rainfall threshold, increasing the weight of the moisture content parameter to the first level; when the historical displacement exceeds a preset displacement threshold, reducing the weight of the acceleration parameter to the second level; and proportionally scaling the weight of the inclination angle parameter according to the degree of deviation between the soil layer temperature and the preset temperature range.

[0014] Optionally, the method further includes: obtaining historical warning records; performing trend matching on the environmental calibration risk index and the historical warning records; generating an environmental feedback coefficient based on trend matching; and correcting the acceleration mutation threshold and the moisture content mutation threshold based on the environmental feedback coefficient.

[0015] Optionally, the method further includes: calculating the mean of the separation divergence to obtain an average dispersion; when the average dispersion is continuously lower than a preset optimization threshold, correcting the acceleration mutation threshold and the moisture content mutation threshold; when the average dispersion exceeds a preset alarm threshold, reconstructing the dynamic monitoring cluster.

[0016] Based on the same inventive concept, the present invention also provides a real-time monitoring system for geological disasters based on a ground sensor network, the system comprising: a data acquisition module for acquiring real-time monitoring data of each sensor node, the real-time monitoring data including acceleration parameters and water content parameters; an environmental trigger module for identifying environmental mutation events according to the change rate of the acceleration parameters and water content parameters, and generating an environmental trigger signal; a cluster control module for activating the emergency sampling mode of the target node and adjacent nodes based on the environmental trigger signal, and constructing a dynamic monitoring cluster; an alarm prediction module for requesting multiple nodes in the dynamic monitoring cluster to synchronously measure similar parameters, compare the measurement results and generate a compressed alarm package; a risk calculation module for adjusting the weight factors of different monitoring parameters using the compressed alarm package and pre-acquired real-time environmental parameters to generate an environmental calibration risk index; and an early warning decision module for outputting a geological disaster early warning instruction when the environmental calibration risk index exceeds a preset risk threshold.

[0017] Compared with the prior art, the present invention has the following advantages:

[0018] 1. The present invention significantly improves the sensitivity of early identification of potential geological disasters through a coordinated triggering mechanism based on the change rate of the two parameters of acceleration and water content, effectively overcoming the risk of underreporting in complex geological environments caused by single parameter monitoring.

[0019] 2. This invention adopts a dynamic monitoring cluster construction strategy to achieve on-demand activation of node resources in disaster focus areas, significantly reducing the energy consumption of the global sensor network while ensuring millimeter-level sampling accuracy;

[0020] 3. The risk assessment model of the present invention dynamically adjusts the weight factors based on environmental parameters, eliminating the impact of background interference such as rainfall and temperature on the accuracy of early warning, and enhancing adaptability under different geological and climatic conditions;

[0021] 4. The present invention ensures efficient and reliable transmission of key alarm information in harsh communication environments through multi-node data consistency verification and compression transmission mechanism, thereby gaining valuable time windows for emergency response.

[0022] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0024] Figure 1 The present invention is a flowchart of a method for real-time monitoring of geological disasters based on a ground sensor network according to an embodiment of the present invention.

[0025] Figure 2 The present invention is a structural diagram of a real-time geological disaster monitoring system based on a ground sensor network. DETAILED DESCRIPTION

[0026] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0027] Reference Figure 1 An embodiment of the present invention proposes a real-time monitoring method for geological disasters based on a ground sensor network, which adopts multi-parameter collaborative triggering, dynamic cluster response, weight adaptive evaluation and compression transmission technology to comprehensively improve monitoring accuracy, timeliness and environmental adaptability.

[0028] The method of this embodiment specifically includes:

[0029] Acquiring real-time monitoring data of each sensor node, wherein the real-time monitoring data includes acceleration parameters and moisture content parameters;

[0030] Identify environmental mutation events based on the change rates of the acceleration parameter and the moisture content parameter, and generate an environmental trigger signal;

[0031] Activate the emergency sampling mode of the target node and adjacent nodes based on the environmental trigger signal to build a dynamic monitoring cluster;

[0032] Requesting multiple nodes in the dynamic monitoring cluster to synchronously measure similar parameters, and comparing the measurement results to generate a compressed alarm packet;

[0033] Using the compressed alarm package and pre-acquired real-time environmental parameters, weight factors of different monitoring parameters are adjusted to generate an environmental calibration risk index;

[0034] When the environmental calibration risk index exceeds a preset risk threshold, a geological disaster warning instruction is output.

[0035] Specifically, by collecting the acceleration and water content parameters of sensor nodes in real time, capturing their rate of change to identify sudden geological environmental events, generating environmental trigger signals to activate the emergency sampling mode of nodes in the target area, and constructing dynamic monitoring clusters to achieve spatial collaborative measurement. Based on the discreteness analysis of synchronized data from multiple nodes within the cluster, compressed alarm packages are generated. The weight factors of each monitoring indicator are dynamically adjusted based on real-time environmental parameters, and ultimately, through weighted calculation, an environmental calibration risk index is generated to enable early warning decisions. This significantly improves the sensitivity and reliability of geological hazard monitoring, effectively overcoming the vulnerability of traditional single-point monitoring to local interference. The dynamic clustering mechanism significantly reduces system energy consumption while ensuring data timeliness. The adaptive weight adjustment function eliminates the influence of environmental background noise, enhancing the accuracy of assessments under different geological conditions. The compressed transmission mechanism optimizes network resource utilization, making it particularly suitable for long-term and stable monitoring in remote areas.

[0036] Optionally, generating an environment trigger signal includes:

[0037] Calculating the instantaneous rate of change of the acceleration parameter, and determining it as a first type trigger event when the instantaneous rate of change exceeds a preset acceleration mutation threshold;

[0038] Calculating the minute-level increase in the moisture content parameter, and determining it as a second-type trigger event when the minute-level increase exceeds a preset moisture content mutation threshold;

[0039] The first type of trigger event and the second type of trigger event are combined to generate the environmental trigger signal.

[0040] Specifically, generating an environmental trigger signal includes the following steps: first, obtaining the acceleration parameters and water content parameters monitored in real time by the sensor node; calculating the instantaneous change rate of the acceleration parameter, which is defined as the difference between the current acceleration value and the acceleration value of the previous sampling point divided by the sampling time interval. ,have:

[0041] ,

[0042] in Indicates the current acceleration parameter value, which is obtained by real-time measurement of the acceleration sensor. Indicates the acceleration parameter value at the previous sampling moment, Indicates a preset fixed sampling time interval, such as 10 milliseconds; when the instantaneous rate of change exceeds the preset acceleration mutation threshold, it is determined to be a first-class trigger event, which is preset based on historical geological disaster data experiments; secondly, the minute-level increase in the water content parameter is calculated, and the increase is defined as the absolute difference between the current water content value and the water content value 60 seconds ago. ,have:

[0043] ,

[0044] in Indicates the current water content parameter value, which is obtained by real-time measurement of the water content sensor. Indicates the moisture content parameter value 60 seconds ago; when the minute-level increase exceeds the preset moisture content mutation threshold, it is determined to be a second-type trigger event, which is preset based on soil property experiments; finally, the first-type trigger event and the second-type trigger event are combined, and an environmental trigger signal is generated when any event occurs; by capturing the rapid changes in physical parameters in real time to identify potential geological disaster events, the sensitivity and reliability of event detection are improved, delayed response is avoided, and sensor energy consumption is optimized.

[0045] Optionally, the building of a dynamic monitoring cluster includes:

[0046] Sending the environmental trigger signal to nodes within a preset neighborhood range via short-range wireless broadcast;

[0047] The node receiving the environmental trigger signal switches to a millisecond sampling frequency and returns a ready confirmation signal to the source node;

[0048] A temporary communication group including a source node and a response node is constructed according to the ready confirmation signal, and a status identifier of the dynamic monitoring cluster group is generated.

[0049] Specifically, the source node first sends the generated environmental trigger signal to all nodes within a preset neighborhood range through a short-range wireless broadcast protocol. This neighborhood range is determined by the maximum effective transmission distance of the wireless communication module, for example, set to a radius of 50 meters. Second, the node that receives the environmental trigger signal immediately switches to a millisecond sampling frequency, which is set to 1000 samples per second by the microcontroller configuration register. At the same time, each node returns a ready confirmation signal containing its own node ID to the source node through the same wireless channel. Finally, the source node collects all ready confirmation signals within a preset waiting time window, establishes a temporary communication group member list based on the node ID contained in the signal, and generates a status identification code representing the activation status of the dynamic monitoring cluster. The status identification code is stored in the non-volatile memory of the source node. Through the event trigger mechanism, dynamic aggregation of monitoring resources is achieved, and a high-precision monitoring network is constructed on demand, ensuring data timeliness while significantly reducing the overall energy consumption of the system.

[0050] For example, in a debris flow monitoring scenario, a source node located in a valley detects a sudden change in moisture content and generates an environmental trigger signal, which is broadcast to five nodes within 50 meters of the surrounding area through the Zigbee protocol. After receiving the signal, each node switches the acceleration sensor sampling frequency from the conventional 10Hz to 1000Hz, and sends a ready confirmation packet containing a unique ID to the source node within 200 milliseconds. The source node receives all five responses within a 300-millisecond waiting period, generates a member list containing six node IDs and a status identification code 0xAA55, and achieves rapid self-organization of monitoring resources in disaster-focused areas, ensuring that key locations obtain ultra-high-frequency data collection capabilities, while avoiding ineffective power consumption of nodes in non-related areas, greatly improving the emergency response efficiency and energy utilization of the monitoring system.

[0051] Optionally, the generating a compressed alarm package by comparing the measurement results includes:

[0052] Extracting a list of available nodes from the status identifier of the dynamic monitoring cluster;

[0053] Sending a synchronous measurement instruction to at least three nodes in the available node list;

[0054] Receive the synchronous measurement data returned by each node and calculate the dispersion of similar parameters;

[0055] When the dispersion is less than a preset tolerance threshold, the compressed alarm packet is generated.

[0056] Specifically, first, the available node list is extracted from the status identifier of the dynamic monitoring cluster, which contains the information of all member nodes of the established temporary communication group; secondly, a synchronous measurement instruction with a GPS timestamp is sent to at least three nodes in the available node list to ensure that all nodes collect similar monitoring parameters under the same time reference; then, the synchronous measurement data packet returned by each node is received, which contains the node ID, measurement time and parameter value; then the dispersion of the same type of parameters is calculated, and the dispersion is calculated using the standard deviation formula. ,have:

[0057] ,

[0058] in Indicates the number of nodes participating in the measurement, which is determined by the list of available nodes. Represents the measurement parameter value of the i-th node, obtained from the synchronous measurement data packet, It represents the arithmetic mean of all node measurement values. Finally, when the discreteness calculation result is less than the preset tolerance threshold, the compressed alarm package generation program is triggered. Through spatial multi-node data consistency verification, single-point errors are eliminated, which greatly improves the credibility of monitoring data and the accuracy of alarms, while reducing network transmission load.

[0059] Optionally, when the discreteness is less than the preset tolerance threshold, generating the compressed alarm package includes:

[0060] When the discreteness is less than a preset tolerance threshold, the acceleration peak value, the water content change gradient and the inclination angle parameters are extracted from the synchronous measurement data to obtain a basic data set;

[0061] Associating the basic data set with the geographic location code of the dynamic monitoring cluster, and encapsulating it to obtain a binary data packet including a data header, a payload, and a check code;

[0062] A compressed alarm packet is obtained based on the binary data packet.

[0063] Specifically, when the calculated discreteness is less than the preset tolerance threshold, three key parameters are first extracted from the synchronous measurement data. The acceleration peak is defined as the maximum value of the acceleration measurements of all nodes, which is directly obtained from the synchronous measurement data packet returned by each node. The water content change gradient is defined as the ratio of the water content change during the event duration to the time. For the water content change gradient, we have:

[0064] ,

[0065] in Indicates the current moisture content value, obtained from the latest measurement data, Indicates the initial moisture content value when the environmental trigger signal is generated, obtained from the historical storage record. represents the time interval from the generation of the environmental trigger signal to the current moment, which is calculated by the system clock; the tilt angle parameter is obtained through the MEMS sensor; these three parameters constitute the basic data set; the basic data set is then associated with the geographic location code of the dynamic monitoring cluster, which is generated by the average GPS coordinates of the nodes in the cluster; finally, the associated data is encapsulated into a binary data packet, the data header contains the protocol version and data type identifier, the payload stores the basic data set and geographic location code, and the check code is generated using the CRC32 algorithm; this binary data packet is the compressed alarm packet; the core characteristics of the disaster are refined and the transmission format is standardized, which greatly reduces the data transmission volume while retaining key warning information, enhancing the system's adaptability in low-bandwidth environments.

[0066] Optionally, the adjusting weight factors of different monitoring parameters using the compressed alarm package and pre-acquired real-time environmental parameters to generate an environmental calibration risk index includes:

[0067] Parsing the compressed alarm packet to obtain a monitoring parameter set; synchronously obtaining current rainfall intensity, soil temperature and historical displacement to generate real-time environmental parameters;

[0068] Matching a parameter weight distribution scheme in a preset weight mapping table according to the real-time environmental parameters to obtain an adjusted weight factor;

[0069] The environmental calibration risk index is calculated using the adjusted weight factor and the compressed alarm package.

[0070] Specifically, the generated compressed alarm package is first parsed to extract the monitoring parameter set contained therein, which includes at least the acceleration peak, moisture content gradient, and tilt angle parameters. Simultaneously, the current rainfall intensity parameter is obtained through the weather station interface, the soil layer temperature parameter is obtained through the temperature sensor array, and the historical displacement parameter is obtained through the displacement monitoring database. The three constitute the real-time environmental parameters. Then, the real-time environmental parameters are input into the preset weight mapping table for matching, as shown in Table 1.

[0071] Table 1 Weight mapping table

[0072] As shown in the table, the rainfall intensity threshold is set to 30 mm / h, the historical displacement threshold is set to 10 mm, the soil temperature deviation from the preset range is set to 10-20 ° C, and the weight is scaled proportionally by 0.02 / ° C according to the degree of deviation; the environmental parameters are matched in the order of rainfall intensity, historical displacement, and soil temperature, and the higher order is given priority in case of conflict; the default baseline weights are acceleration 0.5, moisture content 0.4, and tilt angle 0.5. For example, if the rainfall intensity is 35 mm / h, which is greater than 30 mm / h, the moisture content weight is increased to 0.6, the historical displacement of 8 mm does not exceed the threshold, the acceleration weight remains 0.5, the soil temperature is 12 ° C, the preset median is 15 ° C, and the deviation is -3 ° C. The calculated tilt angle weight is 0.5 × (1-3 × 0.02) = 0.47. The parameter weight distribution scheme is obtained based on the matching results, and the adjusted weight factor set is output; finally, the weighted summation formula is applied to calculate the environmental calibration risk index. For the environmental calibration risk index ,have:

[0073] ,

[0074] in represents the adjusted acceleration weight factor, obtained from the weight distribution scheme, represents the adjusted moisture content weight factor, represents the adjusted time weight factor, Indicates the normalized peak acceleration value, obtained by compressing the alarm packet data and dividing it by the preset reference value of 10m / s². It represents the normalized water content gradient, which is obtained by dividing the original gradient value by the reference value 0.1% / s. It represents the normalized tilt angle parameter, which is obtained by dividing the tilt angle value by the maximum tilt angle at which the disaster occurs; the normalization process ensures that all parameters are dimensionless values; the risk assessment model is optimized by dynamically integrating environmental background information, eliminating the assessment bias caused by environmental interference, and improving the spatiotemporal adaptability of geological disaster warning.

[0075] Optionally, the parameter weight distribution scheme includes:

[0076] When the current rainfall intensity exceeds the preset rainfall threshold, the weight of the moisture content parameter is increased to the first level;

[0077] When the historical displacement exceeds the preset displacement threshold, the weight of the acceleration parameter is reduced to the second level;

[0078] The weight of the tilt angle parameter is scaled proportionally according to the degree of deviation of the soil temperature from the preset temperature range.

[0079] Specifically, when the current rainfall intensity parameter value obtained through the weather station interface exceeds the preset rainfall threshold, the system automatically increases the weight factor of the moisture content parameter to the first level, the weight value of this level is fixed at 0.6, and the preset rainfall threshold is set to 30 mm per hour according to the regional historical meteorological data; when the historical displacement parameter value obtained through the displacement monitoring database exceeds the preset displacement threshold, the weight factor of the acceleration parameter is automatically reduced to the second level, the weight value of this level is fixed at 0.3, and the preset displacement threshold is set to 10 mm according to the geological survey report; for the weight adjustment of the inclination angle parameter, first calculate the deviation of the soil layer temperature parameter value from the median of the preset temperature range, and for the deviation ,have:

[0080] ,

[0081] in Indicates the median of the preset temperature range, which is determined to be 15 degrees Celsius by local geological characteristics. Represents the soil temperature parameter value obtained in real time; then calculate the weight scaling ratio :

[0082] ,

[0083] in The preset proportional coefficient is 0.02 per degree Celsius; finally, the baseline weight of the tilt angle parameter 0.5 is multiplied by the scaling ratio to obtain the adjusted weight; all adjustment operations are updated in real time in the weight mapping table; a dynamic association between environmental factors and monitoring parameter sensitivity is established to eliminate the impact of seasonal environmental interference on the assessment model and enhance the applicability of the risk index under different geological and climatic conditions.

[0084] Optionally, the method further includes:

[0085] Obtain historical warning records;

[0086] Performing trend matching on the environmental calibration risk index and the historical warning records;

[0087] According to the trend matching demerits, the environmental feedback coefficient is generated;

[0088] The acceleration mutation threshold and the moisture content mutation threshold are modified according to the environmental feedback coefficient.

[0089] Specifically, the method also includes the operation of adaptively adjusting the trigger threshold: first, the historical warning record is obtained from the system database, which contains the event type, timestamp and index value corresponding to each time the environmental calibration risk index exceeds the limit within at least three months; then the currently generated environmental calibration risk index is trend matched with the historical records. The matching process includes calculating the change rate of the current index within the historical time window of the same period, and for the change rate ,have:

[0090] ,

[0091] in Represents the current environmental calibration risk index, generated by the aforementioned method, It represents the average risk index within the three-day sliding window of the same period in history, extracted from historical warning records; based on the change rate The absolute value of the generated environmental feedback coefficient ,when When the coefficient is taken , otherwise take 1; finally use this coefficient to modify the acceleration mutation threshold and moisture content mutation threshold, for the new threshold ,have:

[0092] ,

[0093] in is the original preset threshold, is the environmental feedback coefficient, and the corrected threshold is stored in the system configuration register. The trigger sensitivity is optimized through historical data feedback, dynamically adapting to the long-term evolution trend of the geological environment, and avoiding false triggering or missed triggering problems caused by environmental baseline drift.

[0094] Optionally, the method further includes:

[0095] Calculating the mean of the separation divergence to obtain the average dispersion;

[0096] When the average dispersion is continuously lower than a preset optimization threshold, modifying the acceleration mutation threshold and the moisture content mutation threshold;

[0097] When the average dispersion exceeds a preset alarm threshold, the dynamic monitoring cluster is reconstructed.

[0098] Specifically, first calculate the arithmetic mean of the discreteness. ,have:

[0099] ,

[0100] in Indicates the number of synchronous measurements performed within the preset statistical period, recorded by the system clock. Indicates the dispersion value calculated by the jth synchronous measurement, extracted from the historical measurement database; when the average dispersion of five consecutive statistical periods is When the acceleration mutation threshold value is continuously lower than the preset optimization threshold value, the acceleration mutation threshold value and the moisture content mutation threshold value are modified proportionally. The modified acceleration mutation threshold value and moisture content mutation threshold ,have:

[0101] , ,

[0102] in is the original acceleration mutation threshold, is the original moisture content mutation threshold, which is read from the system configuration register. The preset correction coefficient is 0.2; when the single discrete When the preset alarm threshold is exceeded, a reconstruction instruction is immediately sent to the cluster control module to trigger the reconstruction process of the dynamic monitoring cluster. The trigger mechanism is reversely optimized using data consistency indicators to dynamically balance monitoring sensitivity and system stability to prevent threshold rigidity problems in long-term low-discrete environments.

[0103] Based on the same inventive concept, Figure 2 As shown, the present invention also provides a real-time geological disaster monitoring system based on a ground sensor network, the system comprising:

[0104] A data acquisition module is used to obtain real-time monitoring data of each sensor node, wherein the real-time monitoring data includes acceleration parameters and water content parameters;

[0105] An environmental trigger module, configured to identify environmental mutation events based on the change rates of the acceleration parameter and the moisture content parameter, and generate an environmental trigger signal;

[0106] A cluster control module is used to activate the emergency sampling mode of the target node and adjacent nodes based on the environmental trigger signal to build a dynamic monitoring cluster;

[0107] An alarm prediction module is used to request multiple nodes in the dynamic monitoring cluster to synchronously measure similar parameters, compare the measurement results and generate a compressed alarm package;

[0108] a risk calculation module for adjusting weight factors of different monitoring parameters using the compressed alarm package and pre-acquired real-time environmental parameters to generate an environmental calibration risk index;

[0109] The early warning decision module is used to output a geological disaster early warning instruction when the environmental calibration risk index exceeds a preset risk threshold.

[0110] Example 1

[0111] To verify the feasibility and advancement of this invention in practical applications, a monitoring and early warning demonstration project was conducted in a high-risk landslide area in a mountainous area of ​​Yunnan. This area has complex geological structures and loose soils. During the annual rainy season (May to September), continuous heavy rainfall frequently causes landslides, mudslides, and other geological disasters, seriously threatening the lives and property of residents in the villages at the foot of the mountain, as well as traffic safety on a provincial highway.

[0112] Traditional monitoring methods rely primarily on regular manual inspections and single-threshold alarms based on rain gauges. These methods suffer from issues such as delayed response, high false alarm rates, and an inability to capture the physical and mechanical changes that often precede disasters. Transmitting large amounts of raw data is particularly challenging in mountainous areas with poor communication signals. This demonstration project aims to utilize the methods and systems of this invention to achieve all-weather, high-precision, low-power, real-time monitoring and intelligent early warning of landslides.

[0113] In this implementation, the project team deployed 30 sensor nodes along the selected landslide surface, integrating triaxial accelerometers, soil moisture sensors, GPS positioning modules, and low-power wireless communication modules. Under normal conditions, these nodes sample data at a low frequency of once every 10 minutes to maximize energy conservation. The system's backend was deployed at the local emergency management department's monitoring center.

[0114] On July 20, 2024, the area experienced continuous heavy rain. At 3:15 a.m., sensor node 8, located in the center-front portion of the landslide, detected a minute-by-minute increase in soil moisture of 6.5%. This exceeded the preset 5% moisture content mutation threshold, identifying it as a Category II trigger event. Almost simultaneously, the instantaneous rate of change of the acceleration parameter at the same sensor node also exceeded the preset threshold, identifying it as a Category I trigger event. With both conditions met, the system immediately generated an environmental trigger signal. The cluster control module of node 8 (the source node) immediately broadcast the environmental trigger signal via short-range wireless to its 50-meter neighborhood. Nearby nodes 6, 7, 9, and 12 responded swiftly within an average of 180 milliseconds of receiving the signal, switching their sampling frequency from the standard 0.0017 Hz (once every 10 minutes) to an emergency sampling mode of 100 Hz. They also sent a ready confirmation signal containing their own ID to node 8. Based on these signals, the system successfully built a dynamic monitoring cluster with node 8 as the core and a total of 5 nodes, and generated a unique status identifier for the cluster.

[0115] The alarm prediction module immediately sent synchronized measurement commands to the five nodes within the dynamic monitoring cluster. Each node simultaneously collected acceleration parameters and returned data. The backend system calculated that the dispersion of the acceleration readings returned by these five nodes was 0.08g, far below the preset tolerance threshold of 0.5g. This indicated that the detected vibration was a regional, real physical event, rather than a false alarm caused by a single sensor failure or local interference (such as animal contact). After confirming data consistency, the system extracted peak acceleration, water content gradient, and tilt angle parameters from the synchronized measurement data as a basic dataset. This dataset was then associated with the dynamic monitoring cluster's geolocation code (calculated by averaging the GPS coordinates of each node) and encapsulated into a binary data packet consisting of a header, payload, and CRC checksum. This compressed alarm packet, a mere 72 bytes, was efficiently transmitted to the monitoring center via the telecommunications network.

[0116] After receiving and parsing the compressed alert package, the monitoring center's risk calculation module simultaneously acquired real-time external environmental parameters: regional weather station data indicated a current rainfall intensity of 42 mm / h, and a historical database indicated minimal displacement in the area over the past month. According to a pre-set weight mapping table, since the current rainfall intensity exceeded the 30 mm / h rainfall threshold, the system automatically increased the weight of the moisture content parameter to the first level (adjusting the weight factor to 0.6). Using this adjusted weight factor, combined with the monitoring parameter set within the compressed alert package, the system calculated an environmental calibration risk index of 86 (out of 100). This index exceeded the pre-set risk threshold of 80. The early warning decision module determined that the situation had reached Level 2 (orange) warning criteria and immediately issued a geological disaster warning directive to the village's emergency liaison and highway management section via text message, voice call, and dedicated app. It also recommended the immediate activation of the emergency plan and the temporary relocation of personnel in the danger zone.

[0117] In order to quantify the effect of the present invention, the data of the demonstration project running for half a year were compared with the system of the adjacent area using the traditional monitoring method (fixed high-frequency sampling and single rainfall threshold), as shown in Tables 2 and 3.

[0118] Table 2 Geological disaster monitoring events and system response data

[0119] time Trigger Node Event Type Cluster members Risk Index Early warning decision-making Final Result 2024-06-12 04:50 No. 15 Moisture content exceeds threshold 14, 15, 18 45 No warning No disaster occurred 2024-07-20 03:15 No. 8 Acceleration + moisture content 6, 7, 8, 9, 12 86 Level 2 warning Small landslide 2024-08-05 11:22 No. 21 Acceleration exceeds threshold 20, 21, 22, 25 68 Internal Focus No disaster occurred 2024-08-28 22:04 No. 7 Acceleration + moisture content 6, 7, 8, 9 92 Level 1 warning Medium-sized landslide

[0120] Table 3 System performance comparison

[0121] Performance indicators System of the present invention Traditional single threshold system Early warning accuracy 95% 68% False negative rate < 2% About 15% False positive rate < 5% About 30% Average warning response time < 3 minutes > 45 minutes Average daily power consumption of a single node 45mAh 210mAh (fixed high frequency) Single alarm data transmission volume < 100 bytes > 10 KB (raw waveform)

[0122] As can be seen from Table 2, the system of the present invention can effectively distinguish events of different risk levels. High-level warnings are triggered only when multiple key indicators work together and the risk index exceeds the threshold, avoiding interference caused by fluctuations in a single parameter. The comparative data in Table 3 shows that the present system significantly improves warning accuracy and significantly reduces false alarm and missed alarm rates. The average warning response time has been shortened from nearly an hour to less than three minutes, creating an extremely valuable window of opportunity for disaster prevention and mitigation. The weighted adaptive assessment model dynamically adjusts its judgment logic based on environmental factors such as real-time rainfall, enhancing the system's adaptability to varying meteorological conditions. Furthermore, the dynamic cluster's "activation on demand" mode and data compression technology reduce the system's total energy consumption and data transmission volume by over 80%, making it ideal for long-term, stable deployment in remote mountainous areas with limited power and communications.

[0123] To sum up, this embodiment fully demonstrates that the method and system proposed in the present invention have great advantages of high precision, high timeliness, high adaptability and low energy consumption in actual geological disaster monitoring, can effectively improve the early warning capability of geological disasters, and has extremely high promotion and application value.

[0124] It should be noted that the formulas appearing above can translate physical quantities of different properties into unitless standard values ​​or superimposable parameters of the same dimension through the principle of dimensional consistency and mathematical standardization means (such as normalization, dimensionless parameter conversion or unit system unification), thereby eliminating the interference of different dimensions on the operation logic, so that the formulas have mathematical operation rationality and objective law adaptability while retaining the distribution characteristics of the original data. It is a conventional technical means and will not be elaborated here. The electrical connection between the above-mentioned units does not necessarily mean a direct connection of the circuit. The indirect connection method can be applied to the embodiments of the present invention as long as the purpose of the present invention is achieved. The above is only an exemplary embodiment of the present invention and the scope of the present invention cannot be limited thereto.

[0125] That is, any equivalent changes and modifications made according to the teachings of the present invention are still within the scope of the present invention. Those skilled in the art will readily conceive of other embodiments of the present invention after considering the disclosure of the specification and practical truths. This application is intended to cover any variations, uses, or adaptations of the present invention that follow the general principles of the present invention and include common knowledge or customary technical means in the art not described herein.

Claims

1. A real-time monitoring method for geological disasters based on a ground sensor network, characterized in that: The method comprises: Acquiring real-time monitoring data of each sensor node, wherein the real-time monitoring data includes acceleration parameters and moisture content parameters; Identify environmental mutation events based on the change rates of the acceleration parameter and the moisture content parameter, and generate an environmental trigger signal; Activate the emergency sampling mode of the target node and adjacent nodes based on the environmental trigger signal to build a dynamic monitoring cluster; Requesting multiple nodes in the dynamic monitoring cluster to synchronously measure similar parameters, and comparing the measurement results to generate a compressed alarm packet; Using the compressed alarm package and pre-acquired real-time environmental parameters, weight factors of different monitoring parameters are adjusted to generate an environmental calibration risk index; When the environmental calibration risk index exceeds a preset risk threshold, a geological disaster warning instruction is output.

2. A method for real-time monitoring of geological disasters based on a ground sensor network according to claim 1, characterized in that: The generating of the environment trigger signal comprises: Calculating the instantaneous rate of change of the acceleration parameter, and determining it as a first type trigger event when the instantaneous rate of change exceeds a preset acceleration mutation threshold; Calculating the minute-level increase in the moisture content parameter, and determining it as a second-type trigger event when the minute-level increase exceeds a preset moisture content mutation threshold; The first type of trigger event and the second type of trigger event are combined to generate the environmental trigger signal.

3. The method for real-time monitoring of geological disasters based on a ground sensor network according to claim 2, characterized in that: The construction of a dynamic monitoring cluster includes: Sending the environmental trigger signal to nodes within a preset neighborhood range via short-range wireless broadcast; The node receiving the environmental trigger signal switches to a millisecond sampling frequency and returns a ready confirmation signal to the source node; A temporary communication group including a source node and a response node is constructed according to the ready confirmation signal, and a status identifier of the dynamic monitoring cluster group is generated.

4. A method for real-time monitoring of geological disasters based on a ground sensor network according to claim 3, characterized in that: The generating of a compressed alarm packet by comparing the measurement results includes: Extracting a list of available nodes from the status identifier of the dynamic monitoring cluster; Sending a synchronous measurement instruction to at least three nodes in the available node list; Receive the synchronous measurement data returned by each node and calculate the dispersion of similar parameters; When the dispersion is less than a preset tolerance threshold, the compressed alarm packet is generated.

5. The method for real-time monitoring of geological disasters based on a ground sensor network according to claim 4, characterized in that: When the discreteness is less than the preset tolerance threshold, generating the compressed alarm packet includes: When the discreteness is less than a preset tolerance threshold, the acceleration peak value, the water content change gradient and the inclination angle parameters are extracted from the synchronous measurement data to obtain a basic data set; Associating the basic data set with the geographic location code of the dynamic monitoring cluster, and encapsulating it to obtain a binary data packet including a data header, a payload, and a check code; A compressed alarm packet is obtained based on the binary data packet.

6. The method for real-time monitoring of geological disasters based on a ground sensor network according to claim 4, characterized in that: The step of adjusting weight factors of different monitoring parameters using the compressed alarm package and pre-acquired real-time environmental parameters to generate an environmental calibration risk index includes: Parsing the compressed alarm packet to obtain a monitoring parameter set; Synchronously obtain current rainfall intensity, soil temperature and historical displacement to generate real-time environmental parameters; Matching a parameter weight distribution scheme in a preset weight mapping table according to the real-time environmental parameters to obtain an adjusted weight factor; The environmental calibration risk index is calculated using the adjusted weight factor and the compressed alarm package.

7. The method for real-time monitoring of geological disasters based on a ground sensor network according to claim 6, characterized in that: The parameter weight distribution scheme includes: When the current rainfall intensity exceeds the preset rainfall threshold, the weight of the moisture content parameter is increased to the first level; When the historical displacement exceeds the preset displacement threshold, the weight of the acceleration parameter is reduced to the second level; The weight of the tilt angle parameter is scaled proportionally according to the degree of deviation of the soil temperature from the preset temperature range.

8. The method for real-time monitoring of geological disasters based on a ground sensor network according to claim 2, characterized in that: The method further comprises: Obtain historical warning records; Performing trend matching on the environmental calibration risk index and the historical warning records; According to the trend matching demerits, the environmental feedback coefficient is generated; The acceleration mutation threshold and the moisture content mutation threshold are modified according to the environmental feedback coefficient.

9. The method for real-time monitoring of geological disasters based on a ground sensor network according to claim 4, characterized in that: The method further comprises: Calculating the mean of the separation divergence to obtain the average dispersion; When the average dispersion is continuously lower than a preset optimization threshold, modifying the acceleration mutation threshold and the moisture content mutation threshold; When the average dispersion exceeds a preset alarm threshold, the dynamic monitoring cluster is reconstructed.

10. A real-time monitoring system for geological disasters based on a ground sensor network, applied to the real-time monitoring method for geological disasters based on a ground sensor network according to any one of claims 1 to 9, characterized in that: The system comprises: A data acquisition module is used to obtain real-time monitoring data of each sensor node, wherein the real-time monitoring data includes acceleration parameters and water content parameters; An environmental trigger module, configured to identify environmental mutation events based on the change rates of the acceleration parameter and the moisture content parameter, and generate an environmental trigger signal; A cluster control module is used to activate the emergency sampling mode of the target node and adjacent nodes based on the environmental trigger signal to build a dynamic monitoring cluster; An alarm prediction module is used to request multiple nodes in the dynamic monitoring cluster to synchronously measure similar parameters, compare the measurement results and generate a compressed alarm package; a risk calculation module for adjusting weight factors of different monitoring parameters using the compressed alarm package and pre-acquired real-time environmental parameters to generate an environmental calibration risk index; The early warning decision module is used to output a geological disaster early warning instruction when the environmental calibration risk index exceeds a preset risk threshold.

Images