Remote monitoring method and system for intelligently sensing situation, vibration and orientation

Abstract

The invention relates to the technical field of information monitoring, in particular to a remote monitoring method and system for intelligently sensing situation, vibration and orientation. Identifying and acquiring a dynamic construction behavior mode, constructing a space-time dependent topological graph of the dynamic construction behavior mode by monitoring a traceability pattern and an orientation tracking community, and inferring state potential energy of the space-time dependent topological graph according to evolution of the vibration sensing signal and the horizontal sensing data so as to determine a global construction sensing situation of the target monitoring area; and performing phase space evolution trajectory evaluation on the global construction perception situation through an evaluation phase space of a spatial evolution evaluation criterion to obtain a stable node pattern, and introducing a Lyapunov function to analyze whether the divergence degree of the stable node pattern interferes with the buried facility or not so as to control remote monitoring equipment to send out a monitoring early warning signal. According to the invention, all-around real-time monitoring can be carried out on construction behaviors around the buried facility, early warning of potential risks is realized through active perception, and the safety and operation efficiency of the buried facility are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of information monitoring, and in particular to a remote monitoring method and system for intelligently sensing situation, vibration and orientation. Background Art

[0002] With the acceleration of urbanization, the installation of underground infrastructure, including electrical cables, optical cables, and pipelines, is becoming increasingly dense. However, these buried facilities face the risk of damage from construction, natural disasters, and environmental changes, posing safety risks to key sectors such as power, communications, and water supply. Therefore, real-time monitoring technology for underground facilities has become a crucial tool for ensuring the safe operation of this infrastructure. However, existing remote monitoring equipment or systems for buried facilities primarily rely on manual inspections or single-sensor monitoring. Consequently, they possess only a single sensing function and are unable to comprehensively monitor the development of conditions, vibrations, and location caused by construction in real time, significantly reducing the threat posed by surrounding construction activities to the buried system. Furthermore, most existing remote monitoring equipment or systems rely on single-point data collection and lack multi-dimensional data fusion and intelligent analysis capabilities. This lacks efficient computational methods for situation evolution, vibration source tracing, and vibration location tracking, making it difficult to accurately identify construction activities and environmental changes, impacting the accuracy, continuity, and reliability of monitoring. Summary of the Invention

[0003] The present invention overcomes the deficiencies of the prior art and provides a remote monitoring method and system for intelligently sensing situation, vibration and orientation.

[0004] In order to achieve the above object, the technical solution adopted by the present invention is: A first aspect of the present invention provides a remote monitoring method for intelligently sensing situation, vibration, and orientation, comprising the following steps: S102: Acquire horizontal and azimuth sensing data of the target monitoring area, perform quaternion accumulation on the gravity direction of the remote monitoring device using the horizontal and azimuth sensing data, and obtain an azimuth tracking group of the azimuth sensing data; S104: Calculate the actual vibration beam of each vibration sensing signal in the target monitoring area by accumulating the time delay, and search for the kernel density window of the vibration sensing signal in the azimuth tracking community based on the actual vibration beam to obtain the monitoring and tracing pattern of the vibration sensing signal; S106: Identify and obtain dynamic construction behavior patterns, construct a spatiotemporal dependency topology map of the dynamic construction behavior patterns by monitoring the traceability pattern and the orientation tracking community, and infer the state potential of the spatiotemporal dependency topology map based on the evolution of the vibration sensing signal and the horizontal sensing data to determine the global construction sensing situation of the target monitoring area; S108: The phase space evolution trajectory of the global construction perception situation is evaluated through the evaluation phase space of the spatial evolution evaluation criterion to obtain a stable node pattern. The Lyapunov function is introduced to analyze whether the divergence degree of the stable node pattern interferes with the buried facilities, so as to control the remote monitoring equipment to send out monitoring warning signals.

[0005] More specifically, the step S102 includes the following steps: Monitoring a target monitoring area through a remote monitoring device to obtain a plurality of horizontal sensing data and azimuth sensing data of the target monitoring area, and obtaining a starting horizontal posture of the remote monitoring device when starting monitoring by identifying the horizontal sensing data, and obtaining a starting horizontal amplitude of the starting horizontal posture; Based on the initial horizontal amplitude, a native horizontal quaternion is constructed when the remote monitoring device is placed on the target monitoring area. The pitch and roll angle states of the remote monitoring device are estimated using the orientation sensing data to obtain the real-time gravity direction of the remote monitoring device in the monitoring state. Obtaining a gravity knowledge graph based on a big data network, and simultaneously obtaining a monitoring principle specification for a remote monitoring device, identifying the horizontal quaternion in the gravity knowledge graph based on the monitoring principle specification, and outputting the gravity direction under the prerequisite that the remote monitoring device achieves the horizontal quaternion, which is defined as the native gravity direction; A quaternion difference algorithm is introduced to calculate the error of the real-time gravity direction compared to the original gravity direction, to obtain the gravity direction overturning error, to calculate the gradient vector of the gravity direction overturning error with respect to the real-time gravity direction, and to obtain the monitoring time step of the orientation sensing data; Taking the gravity direction overturning error as the target elimination function, a quaternion accumulation domain of the target monitoring area is constructed, and the native horizontal quaternion is integrated using the orientation perception data along the gradient vector at the monitoring time step to generate an accumulated horizontal quaternion; If the gravity direction overturning error is 0, the accumulated horizontal quaternion is used to replace the original horizontal quaternion to generate a quaternion integral matrix. The azimuth tracking community of each azimuth sensing data located in the target monitoring area is determined according to the quaternion integral matrix.

[0006] More specifically, the step S104 includes the following steps: Acquire several vibration sensing signals from the target monitoring area, introduce a short-time Fourier transform algorithm to calculate these vibration sensing signals, obtain the vibration signal spectra of different signal sources, calculate the cross-correlation function between the corresponding vibration signal spectra of each signal source, and determine the actual time delay difference between the corresponding vibration signal spectra of each signal source based on the cross-correlation function; Taking the azimuth tracking cluster as the signal termination direction distribution, the vibration signal monitoring weight of the remote monitoring system is assigned based on the actual time delay difference to obtain the signal source array weight. According to the signal source array weight, the time delay of the signal termination direction distribution of sensors monitored by different signal sources is accumulated to obtain the actual vibration beam of each vibration sensing signal in the azimuth tracking cluster. Based on the actual vibration beam, a kernel density window is preset for each vibration sensing signal. A Gaussian kernel function is used to weight all the azimuth points in the neighborhood around each azimuth point in the azimuth tracking community to obtain several weighted means. Based on the weighted means, the azimuth point is iterated to the mean position of the azimuth points in the neighborhood to determine the maximum density point for each azimuth point. The kernel density window of each vibration perception signal is adjusted according to the maximum density point to obtain a new kernel density window. The covariance matrix of the iterative search window is updated based on the new kernel density window until the preset iteration frequency is reached, and finally the monitoring and tracing pattern of the vibration perception signal is obtained.

[0007] More specifically, the step S106 includes the following steps: Obtain a construction behavior characteristic network that matches the target monitoring area, dynamically reconstruct and identify the neuron structure of the construction behavior characteristic network using vibration perception signals and horizontal perception data, and obtain the dynamic construction behavior pattern generated by vibration perception signals and horizontal perception data in the target monitoring area; A spatiotemporal dependency topology map of dynamic construction behavior patterns is constructed using the monitoring traceability pattern as a spatiotemporal node and the orientation tracking community as a spatiotemporal boundary, and the flash frequency of each dynamic construction behavior pattern as the temporal perception step length progresses is obtained; Based on the flash frequency, the potential function of each spatiotemporal node and the spatiotemporal boundary potential energy of the dependency relationship between each spatiotemporal node and its neighboring spatiotemporal nodes are preset. The overall energy of the spatiotemporal dependency topology graph is calculated by the potential function of each spatiotemporal node and the spatiotemporal boundary potential energy between the corresponding neighboring spatiotemporal nodes to obtain the global spatiotemporal potential energy. Set a start-stop time period and obtain the number of evolutions of each vibration perception signal and horizontal perception data generated between adjacent dynamic construction behavior patterns within the start-stop-stop time period. Calculate the new message after the spatiotemporal node receives the perception messages from all neighboring spatiotemporal nodes based on the number of evolutions, and obtain the perception evolution iteration information. The belief network algorithm is introduced to infer and estimate the state of space-time nodes based on the iterative information of perceptual evolution, and the marginal distribution of each space-time node is obtained. During the inference process, the current global space-time potential energy is extracted; If the current global space-time potential energy is greater than the minimum global space-time potential energy, the state of the space-time nodes will continue to be inferred until it approaches the minimum global space-time potential energy, and the node combination trend of the space-time dependency topology graph will be generated. The global construction perception situation of the target monitoring area will be determined based on the node combination trend.

[0008] More specifically, the method of obtaining a construction behavior characteristic network that matches the target monitoring area, dynamically reconstructing and identifying the neuron structure of the construction behavior characteristic network using vibration sensing signals and horizontal sensing data, and obtaining a dynamic construction behavior pattern generated by vibration sensing signals and horizontal sensing data in the target monitoring area specifically includes the following steps: Obtaining geographic information of the target monitoring area, and based on this information, obtaining a construction behavior feature network that matches the construction output vibration perception signals and horizontal perception data of the target monitoring area in the big data network. Extracting the predetermined weight vector of each construction behavior feature neuron based on the construction behavior feature network; Obtain the time-series perception step size and current perception state of the remote monitoring device, construct a combined solution space tree for the current perception state, and recursively combine each vibration perception signal with each horizontal perception data in the combined solution space tree using the time-series perception step size as a discrete constraint to generate several vibration-horizontal monitoring samples. Calculate the response distance between each vibration-level monitoring sample and each established weight vector to obtain N response distances. Only the construction behavior feature neurons corresponding to the established weight vector with the minimum response distance are extracted and defined as the best matching unit. Construct the neighborhood space of the construction behavior feature network, extract the matching position of the best matching unit in the construction behavior feature network, mark it as the best matching position, preset the neighborhood planning threshold based on the best matching position, and include the surrounding neurons of the best matching unit in the neighborhood space. If the current neighborhood planning value reaches the neighborhood planning threshold, stop the inclusion planning and obtain the neighborhood neuron set; The neighborhood function of the neighborhood neuron set is obtained, and the weight vector of each neighborhood neuron in the neighborhood neuron set is recalculated according to the best matching unit and the neighborhood function, and the construction behavior feature network structure is re-identified to generate a dynamic construction behavior pattern of vibration perception signals and horizontal perception data in the target monitoring area.

[0009] More specifically, the step S108 includes the following steps: Obtain a schematic diagram of the target monitoring area, build a multidimensional spatial model of the target monitoring area based on the schematic diagram, and obtain spatial evolution evaluation criteria for different construction situations based on the big data network; An iterative mapping trend equation for the global construction perception situation is constructed. An evaluation phase space for the construction situation is constructed based on the spatial evolution evaluation criterion. The Runge-Kutta algorithm is introduced to solve the iterative mapping trend equation by integrating the vibration perception signals, horizontal perception data, and azimuth perception data in the evaluation phase space. By solving the problem, multiple phase space evolution trajectories of the global construction perception situation are obtained. Based on these multiple phase space evolution trajectories, a situation evolution evaluation phase diagram is drawn. The stable node pattern corresponding to each vibration perception signal, horizontal perception data, and azimuth perception data generated under the premise of evaluating the global construction perception situation is extracted and evaluated through the situation evolution evaluation phase diagram. The Lyapunov function is introduced and the abnormal divergence point is preset according to the maximum evaluation criterion of the spatial evolution evaluation criterion. The divergence index of each stable node in the stable node pattern relative to the abnormal divergence point is calculated by the Lyapunov function to obtain multiple Lyapunov indices. If the Lyapunov exponent is positive, the stable node corresponding to the positive Lyapunov exponent is marked as a divergent stable point; if the Lyapunov exponent is negative, it is marked as a divergent saddle point; Based on the Lyapunov exponent, a divergence critical threshold is preset, and all divergence stable points and all divergence saddle point planning spaces are integrated in the multidimensional space model until the divergence critical threshold is reached, thereby generating a divergence space of the global construction perception situation; Obtain a distribution diagram of buried facilities and construct a safe distribution space for the buried facilities based on the distribution diagram. If there is an interference between the divergent space and the safe distribution space, mark the construction behavior in the interference area as a dangerous monitoring action and control the remote monitoring equipment to issue a monitoring warning signal.

[0010] The second aspect of the present invention provides a remote monitoring system that can intelligently perceive situation, vibration and orientation. The remote monitoring system includes a memory and a processor. The memory stores a remote monitoring method program that can intelligently perceive situation, vibration and orientation. When the remote monitoring method program is executed by the processor, any one of the steps of the remote monitoring method is implemented.

[0011] The present invention solves the technical defects existing in the background technology, and the beneficial technical effects of the present invention are: The horizontal and azimuth sensing data of the target monitoring area are obtained, and the gravity direction of the remote monitoring equipment is accumulated by the horizontal and azimuth sensing data to obtain the azimuth tracking community of the azimuth sensing data; the actual vibration beam of each vibration sensing signal in the target monitoring area is calculated by time-delay accumulation, and the kernel density window of the vibration sensing signal located in the azimuth tracking community is searched and replaced based on the actual vibration beam to obtain the monitoring traceability pattern of the vibration sensing signal; the dynamic construction behavior pattern is identified and obtained, and the spatiotemporal dependency topology of the dynamic construction behavior pattern is constructed by the monitoring traceability pattern and the azimuth tracking community. The state potential energy of the spatiotemporal dependency topology is inferred according to the evolution of the vibration sensing signal and the horizontal sensing data to determine the global construction perception situation of the target monitoring area; the phase space evolution trajectory of the global construction perception situation is evaluated through the evaluation phase space of the spatial evolution evaluation criterion to obtain a stable node pattern, and the Lyapunov function is introduced to analyze whether the divergence degree of the stable node pattern interferes with the buried facilities, so as to control the remote monitoring equipment to send out monitoring warning signals. The present invention can conduct all-weather, all-round, real-time monitoring of the construction behavior of buried facilities and their surrounding environment, and realize early warning of potential risks by actively sensing the characteristics of surrounding construction behavior, thereby improving the safety and operation efficiency of buried facilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] 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 only some embodiments of the present invention. For ordinary technicians in this field, without paying any creative work, they can also obtain drawings of other embodiments based on these drawings.

[0013] Figure 1 A first method flow chart of a remote monitoring method for intelligently sensing situation, vibration and orientation is shown; Figure 2 A second method flow chart of a remote monitoring method for intelligently sensing situation, vibration, and orientation is shown; Figure 3 The system framework diagram of a remote monitoring system with intelligent perception of situation, vibration and orientation is shown. DETAILED DESCRIPTION

[0014] In order to more clearly understand the above-mentioned objects, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that, in the absence of conflict, the embodiments of the present application and the features therein can be combined with each other.

[0015] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the scope of protection of the present invention is not limited to the specific embodiments disclosed below.

[0016] The first aspect of the present invention provides a remote monitoring method for intelligently sensing situation, vibration and orientation, such as Figure 1 As shown, the following steps are included: S102: Acquire horizontal and azimuth sensing data of the target monitoring area, perform quaternion accumulation on the gravity direction of the remote monitoring device using the horizontal and azimuth sensing data, and obtain an azimuth tracking group of the azimuth sensing data; S104: Calculate the actual vibration beam of each vibration sensing signal in the target monitoring area by accumulating the time delay, and search for the kernel density window of the vibration sensing signal in the azimuth tracking community based on the actual vibration beam to obtain the monitoring and tracing pattern of the vibration sensing signal; S106: Identify and obtain dynamic construction behavior patterns, construct a spatiotemporal dependency topology map of the dynamic construction behavior patterns by monitoring the traceability pattern and the orientation tracking community, and infer the state potential of the spatiotemporal dependency topology map based on the evolution of the vibration sensing signal and the horizontal sensing data to determine the global construction sensing situation of the target monitoring area; S108: The phase space evolution trajectory of the global construction perception situation is evaluated through the evaluation phase space of the spatial evolution evaluation criterion to obtain a stable node pattern. The Lyapunov function is introduced to analyze whether the divergence degree of the stable node pattern interferes with the buried facilities, so as to control the remote monitoring equipment to send out monitoring warning signals.

[0017] More specifically, the step S102 includes the following steps: Monitoring a target monitoring area through a remote monitoring device to obtain a plurality of horizontal sensing data and azimuth sensing data of the target monitoring area, and obtaining a starting horizontal posture of the remote monitoring device when starting monitoring by identifying the horizontal sensing data, and obtaining a starting horizontal amplitude of the starting horizontal posture; Based on the initial horizontal amplitude, a native horizontal quaternion is constructed when the remote monitoring device is placed on the target monitoring area. The pitch and roll angle states of the remote monitoring device are estimated using the orientation sensing data to obtain the real-time gravity direction of the remote monitoring device in the monitoring state. Obtaining a gravity knowledge graph based on a big data network, and simultaneously obtaining a monitoring principle specification for a remote monitoring device, identifying the horizontal quaternion in the gravity knowledge graph based on the monitoring principle specification, and outputting the gravity direction under the prerequisite that the remote monitoring device achieves the horizontal quaternion, which is defined as the native gravity direction; A quaternion difference algorithm is introduced to calculate the error of the real-time gravity direction compared to the original gravity direction, to obtain the gravity direction overturning error, to calculate the gradient vector of the gravity direction overturning error with respect to the real-time gravity direction, and to obtain the monitoring time step of the orientation sensing data; Taking the gravity direction overturning error as the target elimination function, a quaternion accumulation domain of the target monitoring area is constructed, and the native horizontal quaternion is integrated using the orientation perception data along the gradient vector at the monitoring time step to generate an accumulated horizontal quaternion; If the gravity direction overturning error is 0, the accumulated horizontal quaternion is used to replace the original horizontal quaternion to generate a quaternion integral matrix. The azimuth tracking community of each azimuth sensing data located in the target monitoring area is determined according to the quaternion integral matrix.

[0018] It should be noted that horizontal sensing data includes the tilt sensor angle and the horizontal axis rotation angle. Azimuth sensing data includes the gyroscope sensor angular velocity, the device attitude angle, and the magnetometer measurement value. Remote monitoring equipment can perform azimuth sensing of vibrations in the surrounding environment of buried facilities, enabling accurate acquisition of the overall situation of surrounding construction activities. When the azimuth of construction vibrations changes, it will cause the ground to sink and collapse, and these conditions will cause changes in the horizontal sensing data of the remote monitoring equipment. Therefore, horizontal sensing data and azimuth sensing data are closely related. Horizontal sensing data is an important basis for tracking azimuth changes in vibrations. However, current remote monitoring equipment has poor processing performance for its own horizontal sensing, making it difficult to perceive the specific direction of the vibration through horizontal changes. This leads to large errors in the azimuth tracking of construction vibrations and inaccurate azimuth monitoring results. In this regard, this method can use the orientation perception data to perform quaternion accumulation on the gravity direction change error of the remote monitoring device's own horizontal perception data, so that the remote monitoring device can quickly track the specific distribution of its orientation when the vibration occurs based on the horizontal perception data, thereby visualizing the tracking trajectory of the orientation perception data, reducing the positioning error of the vibration orientation, and improving the orientation monitoring performance and accuracy of the remote monitoring device.

[0019] It should be noted that this method obtains the initial horizontal posture of the current remote monitoring device when it starts monitoring through horizontal sensing data, and uses quaternions to define the horizontal amplitude of the initial horizontal posture. Quaternions are a mathematical tool used to represent the rotational change of gravity in three-dimensional space, ensuring that the subsequent horizontal filtering process of orientation perception starts from a known state of the device. Then, the pitch angle state and roll angle state of the remote monitoring device are calculated through data such as acceleration and magnetometer measurement values ​​contained in the orientation sensing data to determine its real-time gravity direction based on the current horizontal sensing data. From the known horizontal sensing data and orientation sensing data, it can be seen that the horizontal degree of the remote monitoring device has been tilted from the original gravity direction to the real-time gravity direction under the continuous advancement of the construction vibration direction. Therefore, the gravity tilt change of the remote monitoring device itself when the horizontal change occurs is calculated based on the actual gravity direction and the native gravity direction expressed by the quaternion, that is, the gravity direction overturning error. The gravity rollover error reveals the trajectory changes in orientation-sensing data, such as gyroscope sensor angular velocity, device attitude angle, and magnetometer measurement values, when each horizontal sensing data, such as the tilt sensor angle and horizontal axis rotation angle, undergoes temporal changes. Therefore, this method corrects this gravity error by integrating the native horizontal quaternion using orientation-sensing data along the gradient vector with respect to the real-time gravity direction. This allows the integrated quaternion to accurately reflect the specific direction of device vibration monitoring, ensuring that the remote monitoring device maintains accurate vibration azimuth estimation and tracking during its own horizontal changes. If the gravity rollover error is 0, it indicates that the device's own gravity direction has returned to its original gravity direction. During the restoration process, the device tracks the vibration position changes of the orientation-sensing data. Therefore, this method replaces the native horizontal quaternion with the accumulated horizontal quaternion to generate a quaternion integral matrix. This quaternion integral matrix represents the final cluster of all tracked vibration azimuth change locations, achieving accurate tracking of orientation-sensing data using horizontal sensing data.

[0020] More specifically, the step S104 includes the following steps: Acquire several vibration sensing signals from the target monitoring area, introduce a short-time Fourier transform algorithm to calculate these vibration sensing signals, obtain the vibration signal spectra of different signal sources, calculate the cross-correlation function between the corresponding vibration signal spectra of each signal source, and determine the actual time delay difference between the corresponding vibration signal spectra of each signal source based on the cross-correlation function; Using the azimuth tracking cluster as the signal termination direction distribution, the vibration signal monitoring weight of the remote monitoring device is assigned based on the actual time delay difference to obtain the signal source array weight. Based on the signal source array weight, the time delay of the signal termination direction distribution of sensors monitored by different signal sources is accumulated to obtain the actual vibration beam of each vibration sensing signal in the azimuth tracking cluster. Based on the actual vibration beam, a kernel density window is preset for each vibration sensing signal. A Gaussian kernel function is used to weight all the azimuth points in the neighborhood around each azimuth point in the azimuth tracking community to obtain several weighted means. Based on the weighted means, the azimuth point is iterated to the mean position of the azimuth points in the neighborhood to determine the maximum density point for each azimuth point. The kernel density window of each vibration perception signal is adjusted according to the maximum density point to obtain a new kernel density window. The covariance matrix of the iterative search window is updated based on the new kernel density window until the preset iteration frequency is reached, and finally the monitoring and tracing pattern of the vibration perception signal is obtained.

[0021] It should be noted that vibration data is usually processed in the form of signal source acquisition. However, since there may be a certain time delay in the perception of vibration signals in different directions in time sequence, this makes it difficult for remote monitoring equipment to distinguish the directional areas to which each vibration signal belongs when centrally processing the perceived vibration signals. This causes the equipment to have inaccurate perception of the vibration signal source, which can easily lead to confusion in the subsequent vibration signal data processing. This greatly reduces the accuracy of the vibration perception monitoring results. Therefore, it is crucial to trace the specific perception direction of the vibration signal. Therefore, this method adjusts the kernel density window of each vibration perception signal through the maximum density point of each azimuth point in the azimuth tracking community, so as to achieve accurate tracking of the azimuth sources of different vibration signals. Since the perception of vibration signals has time delays at different distances, the resulting vibration beams are different. Therefore, the time delay difference of each vibration signal received by the remote monitoring equipment can be calculated as a reference for accurate tracing. To this end, this method first extracts the characteristics of each vibration signal, that is, the vibration signal spectrum of different signal sources. Since the perception of these vibration signals maintains a certain temporal correlation for beamforming, the cross-correlation function between the corresponding vibration signal spectra of each signal source is used. This cross-correlation function is the embodiment of the actual time delay difference between the vibration signals; weight allocation can significantly enhance the target signal and improve the quality of the signal beam. Therefore, the vibration signal monitoring weight of the remote monitoring equipment is allocated according to the actual time delay difference. Then, according to the signal source array weight, the time delay of the signal termination direction distribution of the vibration sensors monitoring different signal sources is accumulated, so that the target signal is spatially enhanced and the interference signal is suppressed, so as to synthesize a directional signal or receive a signal from a specific direction, thereby enhancing the ability to distinguish different signal sources and ensuring the accuracy of the vibration beam.

[0022] It should be noted that after obtaining the actual vibration beam of the vibration signal, the specific direction of the source can be traced based on the time delay feedback of the beam. This method uses the actual vibration beam to preset the kernel density window of each vibration sensing signal. The kernel density window will be dynamically adjusted as the distance of the neighborhood below the beam delay difference reflects the change. This makes the distance tracking of each vibration signal based on the vibration delay difference more reliable, and improves the accuracy of the remote monitoring equipment for tracing the direction of the vibration signal. Among them, the azimuth point is moved to a new position (mean point). With each iteration, the azimuth point will move closer to the high-density area, making it close to the center of data density. This can further accurately mark the affiliated direction of each vibration signal according to the actual vibration beam, effectively ensuring the correct location of the vibration signal in the direction tracking community.

[0023] More specifically, the step S106 is as follows: Figure 2 As shown, the specific steps include: S202: Obtain a construction behavior characteristic network that matches the target monitoring area, dynamically reconstruct and identify the neuron structure of the construction behavior characteristic network using the vibration sensing signal and the horizontal sensing data, and obtain a dynamic construction behavior pattern generated by the vibration sensing signal and the horizontal sensing data in the target monitoring area; S204: constructing a spatiotemporal dependency topology map of the dynamic construction behavior pattern using the monitoring traceability pattern as a spatiotemporal node and the orientation tracking community as a spatiotemporal boundary, and obtaining the flash frequency of each dynamic construction behavior pattern as the temporal perception step progresses; S206: Preset the potential function of each spatiotemporal node based on the flash frequency and the spatiotemporal boundary potential energy of the dependency relationship between each spatiotemporal node and its neighboring spatiotemporal nodes, and calculate the overall energy of the spatiotemporal dependency topology graph through the potential function of each spatiotemporal node and the spatiotemporal boundary potential energy between the corresponding neighboring spatiotemporal nodes to obtain the global spatiotemporal potential energy; S208: Setting a start-end-end time period, obtaining the number of evolutions of each vibration perception signal and horizontal perception data generated between adjacent dynamic construction behavior patterns within the start-end-end time period, calculating the new message after the spatiotemporal node receives the perception messages from all neighboring spatiotemporal nodes based on the number of evolutions, and obtaining perception evolution iteration information; S210: Introducing the belief network algorithm, based on the perceptual evolution iterative information, the state of the spatiotemporal nodes is inferred and estimated to obtain the marginal distribution of each spatiotemporal node. During the inference process, the current global spatiotemporal potential energy is extracted; S212: If the current global space-time potential energy is greater than the minimum global space-time potential energy, continue to infer the state of the space-time nodes until it approaches the minimum global space-time potential energy, generate the node combination trend of the space-time dependency topology graph, and determine the global construction perception situation of the target monitoring area based on the node combination trend.

[0024] It should be noted that the calculated monitoring traceability pattern and orientation tracking community can reflect the specific trends of the construction environment surrounding the remote monitoring equipment. As vibration, horizontal, and orientation sensing data are continuously acquired, the construction situation and status also change accordingly. Therefore, in order to accurately monitor the global construction situation within the target monitoring area, it is necessary to utilize these sensing data to dynamically analyze the evolution of construction behavior. To this end, this method first identifies vibration sensing signals and horizontal sensing data through a construction behavior feature network to quickly identify dynamic construction behavior patterns within the target monitoring area. These dynamic construction behavior patterns include, but are not limited to, road roller compaction, pile foundation drilling, and public facility demolition. Because these construction behavior patterns are executed synchronously in time and space, their evolution follows certain spatiotemporal characteristics. Therefore, a spatiotemporal dependency topology map of dynamic construction behavior patterns is constructed, using the monitoring traceability pattern as spatiotemporal nodes and the orientation tracking community as spatiotemporal boundaries. This spatiotemporal dependency topology map reveals the relationships between construction behavior variables in time and space, thereby identifying directly related and conditionally independent construction behavior variables, making subsequent evolution more convenient and accurate. The flash frequency represents the average rate at which each dynamic construction behavior pattern intermittently occurs over the time-series perception step. This frequency can illustrate the correlation and random probability of the trend evolution between dynamic construction behavior patterns, consistent with the evolutionary laws of construction behavior. Therefore, based on this flash frequency, we can define the potential function for each spatiotemporal node and the spatiotemporal boundary potential energy of the dependency relationship between each spatiotemporal node and its neighboring spatiotemporal nodes. The potential function defines the strength of the relationship between construction behavior variables and quantifies their interactions. Because the state of each spatiotemporal node in the situation evolution depends only on the state of its neighboring nodes, the spatiotemporal boundary potential energy represents the degree of state trend dependence between pairs of neighboring spatiotemporal nodes.

[0025] It should be noted that the traditional situation awareness method requires inferring each construction behavior pattern one by one, which increases the number of calculation steps and the amount of computation. In order to improve the efficiency of situation evolution, this method calculates a global space-time potential energy through the potential functions of all space-time nodes and the boundary potential energy that expresses the dependency relationship between space-time nodes. This global space-time potential energy can reflect the state evolution trend of all construction behavior patterns in the space-time dependency topology diagram, thereby reducing the amount of computation. Monitoring different sensory data is essentially the evolutionary process of a spatiotemporal node with construction behavior characteristics receiving state messages from all neighboring spatiotemporal nodes. Therefore, the sensory evolution information between each neighboring spatiotemporal node can be iterated according to the number of evolutions of each vibration sensory signal and horizontal sensory data generated between adjacent dynamic construction behavior patterns, ensuring that the construction behavior model of the spatiotemporal node obtains the complete contextual information of the construction behavior models of the neighboring spatiotemporal nodes, so as to approach the true state evolution distribution inference. If the current global spatiotemporal potential energy is greater than the minimum global spatiotemporal potential energy, it means that the inferred state does not conform to the response of the sensory data. Therefore, it is necessary to make the current global spatiotemporal potential energy approach the node combination trend generated by the minimum global spatiotemporal potential energy in order to more accurately reflect the global construction perception situation of the target monitoring area in the current spatiotemporal space. Through this method, the spatiotemporal state evolution of dynamic construction behavior patterns can be analyzed using sensory data, so that the construction situation that conforms to the sensory data expression can be quickly and accurately monitored. At the same time, compared with traditional methods, it can achieve global monitoring perception of the situation of multiple construction behavior patterns, improve the efficiency of situation monitoring, and reduce the error of the results.

[0026] More specifically, the method of obtaining a construction behavior characteristic network that matches the target monitoring area, dynamically reconstructing and identifying the neuron structure of the construction behavior characteristic network using vibration sensing signals and horizontal sensing data, and obtaining a dynamic construction behavior pattern generated by vibration sensing signals and horizontal sensing data in the target monitoring area specifically includes the following steps: Obtaining geographic information of the target monitoring area, and based on this information, obtaining a construction behavior feature network that matches the construction output vibration perception signals and horizontal perception data of the target monitoring area in the big data network, and extracting the predetermined weight vector of each construction behavior feature neuron based on the construction behavior feature network; Obtain the time-series perception step size and current perception state of the remote monitoring device, construct a combined solution space tree for the current perception state, and recursively combine each vibration perception signal with each horizontal perception data in the combined solution space tree using the time-series perception step size as a discrete constraint to generate several vibration-horizontal monitoring samples. Calculate the response distance between each vibration-level monitoring sample and each established weight vector to obtain N response distances. Only the construction behavior feature neurons corresponding to the established weight vector with the minimum response distance are extracted and defined as the best matching unit. Construct the neighborhood space of the construction behavior feature network, extract the matching position of the best matching unit in the construction behavior feature network, mark it as the best matching position, preset the neighborhood planning threshold based on the best matching position, and include the surrounding neurons of the best matching unit in the neighborhood space. If the current neighborhood planning value reaches the neighborhood planning threshold, stop the inclusion planning and obtain the neighborhood neuron set; The neighborhood function of the neighborhood neuron set is obtained, and the weight vector of each neighborhood neuron in the neighborhood neuron set is recalculated according to the best matching unit and the neighborhood function, and the construction behavior feature network structure is re-identified to generate a dynamic construction behavior pattern of vibration perception signals and horizontal perception data in the target monitoring area.

[0027] It should be noted that for the identification of dynamic construction behavior models, this method uses a construction behavior feature network derived from a big data network that matches the construction output of the target monitoring area as the basis for feature identification. Each neuron in the construction behavior feature network represents a feature classification, and each feature classification has a threshold for pattern recognition, namely a weight vector. Therefore, it is necessary to extract a predetermined weight vector for each construction behavior feature neuron. Because construction behavior is an ongoing state, the construction behavior state perceived by remote monitoring equipment is inconsistent at each time step. For example, within a certain time step, the drilling rate of a ground drill changes from low to high speed, forming a transient dynamic construction behavior pattern. These changes are reflected in the perception data within different time steps. Therefore, it is necessary to recursively combine the vibration perception signal and horizontal perception data monitored within the time step to ensure that all possible dynamic construction behavior patterns are detected based on the perception data and generate multiple vibration-horizontal monitoring samples. Next, the response distance between each vibration-level monitoring sample and each predetermined weight vector is calculated. This response distance expresses the matching rate between each construction behavior feature and each neuron. The neuron that best represents the current vibration-level monitoring sample input, i.e., the best matching unit, is then found. By adjusting the weights of the best matching unit and its neighboring neurons, the construction behavior feature neurons are brought closer to the input sample, thereby adjusting the mapping of the entire construction behavior feature network. The adjustment of the weight vector of each neighboring neuron in this neighborhood neuron set is a dynamic adjustment process, which enables it to conform to the dynamic perception and recognition of construction behavior changes under the continuous advancement of the time series step. The final adjusted construction behavior feature network structure is the dynamic construction behavior pattern of vibration perception signals and level perception data generated in the target monitoring area. This method can adjust the weights of network neurons to identify and analyze the perception data generated in the area, thereby accurately and efficiently identifying the dynamic construction behavior patterns appearing in the area, providing a reliable inference basis for subsequent situation evolution calculations, and making the remote monitoring equipment more accurate in perceiving and monitoring real-time construction situations.

[0028] More specifically, the step S108 includes the following steps: Obtain a schematic diagram of the target monitoring area, build a multidimensional spatial model of the target monitoring area based on the schematic diagram, and obtain spatial evolution evaluation criteria for different construction situations based on the big data network; An iterative mapping trend equation for the global construction perception situation is constructed. An evaluation phase space for the construction situation is constructed based on the spatial evolution evaluation criterion. The Runge-Kutta algorithm is introduced to solve the iterative mapping trend equation by integrating the vibration perception signals, horizontal perception data, and azimuth perception data in the evaluation phase space. By solving the problem, multiple phase space evolution trajectories of the global construction perception situation are obtained. Based on these multiple phase space evolution trajectories, a situation evolution evaluation phase diagram is drawn. The stable node pattern corresponding to each vibration perception signal, horizontal perception data, and azimuth perception data generated under the premise of evaluating the global construction perception situation is extracted and evaluated through the situation evolution evaluation phase diagram. The Lyapunov function is introduced and the abnormal divergence point is preset according to the maximum evaluation criterion of the spatial evolution evaluation criterion. The divergence index of each stable node in the stable node pattern relative to the abnormal divergence point is calculated by the Lyapunov function to obtain multiple Lyapunov indices. If the Lyapunov exponent is positive, the stable node corresponding to the positive Lyapunov exponent is marked as a divergent stable point; if the Lyapunov exponent is negative, it is marked as a divergent saddle point; Based on the Lyapunov exponent, a divergence critical threshold is preset, and all divergence stable points and all divergence saddle point planning spaces are integrated in the multidimensional space model until the divergence critical threshold is reached, thereby generating a divergence space of the global construction perception situation; Obtain a distribution diagram of buried facilities and construct a safe distribution space for the buried facilities based on the distribution diagram. If there is an interference between the divergent space and the safe distribution space, mark the construction behavior in the interference area as a dangerous monitoring action and control the remote monitoring equipment to issue a monitoring warning signal.

[0029] It should be noted that as the construction behavior situation continues to develop in time and space, it may cause damage to the cables, optical fibers and pipelines buried underground, which may lead to local system paralysis of buried facilities and affect normal use. Therefore, it is crucial for remote monitoring equipment to accurately judge whether the development of construction behavior situation affects buried facilities. To this end, this method constructs an evaluation phase space by obtaining spatial evolution evaluation criteria for different construction situations to evaluate the trajectory trend of situation development. Then, the global construction perception situation is integrated and solved with existing perception data in the evaluation phase space to obtain the development quality of the global construction perception situation in different local areas. This method visualizes this by drawing a situation evolution evaluation phase diagram, which can be used to obtain the evolution degree corresponding to each vibration perception signal, horizontal perception data, and azimuth perception data generated under the premise of evaluating the global construction perception situation, that is, the stable node pattern. This stable node pattern is an important evaluation criterion for judging whether it affects the safety of buried facilities. If a stable node in the stable node pattern is unstable, it means that the continued outward development of the construction behavior situation in the local area represented by the stable node may interfere with the buried facilities. Therefore, this method calculates the divergence index of each stable node relative to the abnormal divergence point through the Lyapunov function to obtain multiple Lyapunov indices. The Lyapunov index represents the divergence stability of the construction behavior situation in the local area. If the Lyapunov index is positive, it indicates that the construction behavior trend in the local area to which the stable node belongs within the target monitoring area has developed relatively small, showing positive feedback. Therefore, the situation is relatively stable and will not affect underground buried facilities. If the Lyapunov index is negative, it means that the construction behavior trend in the local area to which the stable node belongs has developed beyond the normal range, gradually deviating from positive feedback and tending towards negative feedback, indicating that the construction behavior trend in the stable node area has begun to develop unstablely.

[0030] It should be noted that after identifying stable and unstable nodes, the distribution pattern of these nodes can be integrated and planned within the target monitoring area to form a divergent space that reflects the global construction perception situation. If the divergent space and the safe distribution space are in conflict, it means that the divergent space of a certain local area will seriously affect the safety of buried facilities and easily cause damage. Therefore, the construction behavior in the conflicting area is marked as a dangerous monitoring action, and the remote monitoring equipment is controlled to issue a monitoring warning signal to issue a corresponding warning. This method can be used to conduct real-time judgment and analysis on whether the development of the construction behavior situation perceived by the remote monitoring equipment affects the safety of buried facilities, thereby promptly discovering construction behaviors that threaten the safe operation of the buried system, improving the safety factor of surrounding construction for buried facilities, and having high reliability.

[0031] The second aspect of the present invention provides a remote monitoring system for intelligently sensing situation, vibration and orientation, such as Figure 3As shown, the remote monitoring system includes a memory 31 and a processor 32. The memory 31 stores a remote monitoring method program for intelligently sensing status, vibration and orientation. When the remote monitoring method program is executed by the processor 32, any one of the remote monitoring method steps is implemented.

[0032] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A remote monitoring method for intelligently sensing situation, vibration and orientation, characterized in that: The following steps are involved: S102: Acquire horizontal and azimuth sensing data of the target monitoring area, perform quaternion accumulation on the gravity direction of the remote monitoring device using the horizontal and azimuth sensing data, and obtain an azimuth tracking group of the azimuth sensing data; S104: Calculate the actual vibration beam of each vibration sensing signal in the target monitoring area by accumulating the time delay, and search for the kernel density window of the vibration sensing signal in the azimuth tracking community based on the actual vibration beam to obtain the monitoring and tracing pattern of the vibration sensing signal; S106: Identify and obtain dynamic construction behavior patterns, construct a spatiotemporal dependency topology map of the dynamic construction behavior patterns by monitoring the traceability pattern and the orientation tracking community, and infer the state potential of the spatiotemporal dependency topology map based on the evolution of the vibration sensing signal and the horizontal sensing data to determine the global construction sensing situation of the target monitoring area; S108: The phase space evolution trajectory of the global construction perception situation is evaluated through the evaluation phase space of the spatial evolution evaluation criterion to obtain a stable node pattern. The Lyapunov function is introduced to analyze whether the divergence degree of the stable node pattern interferes with the buried facilities, so as to control the remote monitoring equipment to send out monitoring warning signals.

2. The remote monitoring method for intelligently sensing situation, vibration and orientation according to claim 1, characterized in that: The step S102 specifically includes the following steps: Monitoring a target monitoring area through a remote monitoring device to obtain a plurality of horizontal sensing data and azimuth sensing data of the target monitoring area, and obtaining a starting horizontal posture of the remote monitoring device when starting monitoring by identifying the horizontal sensing data, and obtaining a starting horizontal amplitude of the starting horizontal posture; Based on the initial horizontal amplitude, a native horizontal quaternion is constructed when the remote monitoring device is placed on the target monitoring area. The pitch and roll angle states of the remote monitoring device are estimated using the orientation sensing data to obtain the real-time gravity direction of the remote monitoring device in the monitoring state. Obtaining a gravity knowledge graph based on a big data network, and simultaneously obtaining a monitoring principle specification for a remote monitoring device, identifying the horizontal quaternion in the gravity knowledge graph based on the monitoring principle specification, and outputting the gravity direction under the prerequisite that the remote monitoring device achieves the horizontal quaternion, which is defined as the native gravity direction; A quaternion difference algorithm is introduced to calculate the error of the real-time gravity direction compared to the original gravity direction, to obtain the gravity direction overturning error, to calculate the gradient vector of the gravity direction overturning error with respect to the real-time gravity direction, and to obtain the monitoring time step of the orientation sensing data; Taking the gravity direction overturning error as the target elimination function, a quaternion accumulation domain of the target monitoring area is constructed, and the native horizontal quaternion is integrated using the orientation perception data along the gradient vector at the monitoring time step to generate an accumulated horizontal quaternion; If the gravity direction overturning error is 0, the accumulated horizontal quaternion is used to replace the original horizontal quaternion to generate a quaternion integral matrix. The azimuth tracking community of each azimuth sensing data located in the target monitoring area is determined according to the quaternion integral matrix.

3. The remote monitoring method for intelligently sensing situation, vibration and orientation according to claim 1, characterized in that: The step S104 specifically includes the following steps: Acquire several vibration sensing signals from the target monitoring area, introduce a short-time Fourier transform algorithm to calculate these vibration sensing signals, obtain the vibration signal spectra of different signal sources, calculate the cross-correlation function between the corresponding vibration signal spectra of each signal source, and determine the actual time delay difference between the corresponding vibration signal spectra of each signal source based on the cross-correlation function; Taking the azimuth tracking cluster as the signal termination direction distribution, the vibration signal monitoring weight of the remote monitoring system is assigned based on the actual time delay difference to obtain the signal source array weight. According to the signal source array weight, the time delay of the signal termination direction distribution of sensors monitored by different signal sources is accumulated to obtain the actual vibration beam of each vibration sensing signal in the azimuth tracking cluster. Based on the actual vibration beam, a kernel density window is preset for each vibration sensing signal. A Gaussian kernel function is used to weight all the azimuth points in the neighborhood around each azimuth point in the azimuth tracking community to obtain several weighted means. Based on the weighted means, the azimuth point is iterated to the mean position of the azimuth points in the neighborhood to determine the maximum density point for each azimuth point. The kernel density window of each vibration perception signal is adjusted according to the maximum density point to obtain a new kernel density window. The covariance matrix of the iterative search window is updated based on the new kernel density window until the preset iteration frequency is reached, and finally the monitoring and tracing pattern of the vibration perception signal is obtained.

4. The remote monitoring method for intelligently sensing situation, vibration and orientation according to claim 1, characterized in that: The step S106 specifically includes the following steps: Obtain a construction behavior characteristic network that matches the target monitoring area, dynamically reconstruct and identify the neuron structure of the construction behavior characteristic network using vibration perception signals and horizontal perception data, and obtain the dynamic construction behavior pattern generated by vibration perception signals and horizontal perception data in the target monitoring area; A spatiotemporal dependency topology map of dynamic construction behavior patterns is constructed using the monitoring traceability pattern as a spatiotemporal node and the orientation tracking community as a spatiotemporal boundary, and the flash frequency of each dynamic construction behavior pattern as the temporal perception step length progresses is obtained; Based on the flash frequency, the potential function of each spatiotemporal node and the spatiotemporal boundary potential energy of the dependency relationship between each spatiotemporal node and its neighboring spatiotemporal nodes are preset. The overall energy of the spatiotemporal dependency topology graph is calculated by the potential function of each spatiotemporal node and the spatiotemporal boundary potential energy between the corresponding neighboring spatiotemporal nodes to obtain the global spatiotemporal potential energy. Set a start-stop time period and obtain the number of evolutions of each vibration perception signal and horizontal perception data generated between adjacent dynamic construction behavior patterns within the start-stop-stop time period. Calculate the new message after the spatiotemporal node receives the perception messages from all neighboring spatiotemporal nodes based on the number of evolutions, and obtain the perception evolution iteration information. The belief network algorithm is introduced to infer and estimate the state of space-time nodes based on the iterative information of perceptual evolution, and the marginal distribution of each space-time node is obtained. During the inference process, the current global space-time potential energy is extracted; If the current global space-time potential energy is greater than the minimum global space-time potential energy, the state of the space-time nodes will continue to be inferred until it approaches the minimum global space-time potential energy, and the node combination trend of the space-time dependency topology graph will be generated. The global construction perception situation of the target monitoring area will be determined based on the node combination trend.

5. The remote monitoring method for intelligently sensing situation, vibration and orientation according to claim 4, characterized in that: The method of obtaining a construction behavior characteristic network that conforms to the target monitoring area, dynamically reconstructing and identifying the neuron structure of the construction behavior characteristic network using vibration sensing signals and horizontal sensing data, and obtaining a dynamic construction behavior pattern generated by vibration sensing signals and horizontal sensing data in the target monitoring area specifically includes the following steps: Obtaining geographic information of the target monitoring area, and based on this information, obtaining a construction behavior feature network that matches the construction output vibration perception signals and horizontal perception data of the target monitoring area in the big data network, and extracting the predetermined weight vector of each construction behavior feature neuron based on the construction behavior feature network; Obtain the time-series perception step size and current perception state of the remote monitoring device, construct a combined solution space tree for the current perception state, and recursively combine each vibration perception signal with each horizontal perception data in the combined solution space tree using the time-series perception step size as a discrete constraint to generate several vibration-horizontal monitoring samples. Calculate the response distance between each vibration-level monitoring sample and each established weight vector to obtain N response distances. Only the construction behavior feature neurons corresponding to the established weight vector with the minimum response distance are extracted and defined as the best matching unit. Construct the neighborhood space of the construction behavior feature network, extract the matching position of the best matching unit in the construction behavior feature network, mark it as the best matching position, preset the neighborhood planning threshold based on the best matching position, and include the surrounding neurons of the best matching unit in the neighborhood space. If the current neighborhood planning value reaches the neighborhood planning threshold, stop the inclusion planning and obtain the neighborhood neuron set; The neighborhood function of the neighborhood neuron set is obtained, and the weight vector of each neighborhood neuron in the neighborhood neuron set is recalculated according to the best matching unit and the neighborhood function, and the construction behavior feature network structure is re-identified to generate a dynamic construction behavior pattern of vibration perception signals and horizontal perception data in the target monitoring area.

6. The remote monitoring method for intelligently sensing situation, vibration and orientation according to claim 1, characterized in that: The step S108 specifically includes the following steps: Obtain a schematic diagram of the target monitoring area, build a multidimensional spatial model of the target monitoring area based on the schematic diagram, and obtain spatial evolution evaluation criteria for different construction situations based on the big data network; An iterative mapping trend equation for the global construction perception situation is constructed. An evaluation phase space for the construction situation is constructed based on the spatial evolution evaluation criterion. The Runge-Kutta algorithm is introduced to solve the iterative mapping trend equation by integrating the vibration perception signals, horizontal perception data, and azimuth perception data in the evaluation phase space. By solving the problem, multiple phase space evolution trajectories of the global construction perception situation are obtained. Based on these multiple phase space evolution trajectories, a situation evolution evaluation phase diagram is drawn. The stable node pattern corresponding to each vibration perception signal, horizontal perception data, and azimuth perception data generated under the premise of evaluating the global construction perception situation is extracted and evaluated through the situation evolution evaluation phase diagram. The Lyapunov function is introduced and the abnormal divergence point is preset according to the maximum evaluation criterion of the spatial evolution evaluation criterion. The divergence index of each stable node in the stable node pattern relative to the abnormal divergence point is calculated by the Lyapunov function to obtain multiple Lyapunov indices. If the Lyapunov exponent is positive, the stable node corresponding to the positive Lyapunov exponent is marked as a divergent stable point; if the Lyapunov exponent is negative, it is marked as a divergent saddle point; Based on the Lyapunov exponent, a divergence critical threshold is preset, and all divergence stable points and all divergence saddle point planning spaces are integrated in the multidimensional space model until the divergence critical threshold is reached, thereby generating a divergence space of the global construction perception situation; Obtain a distribution diagram of buried facilities and construct a safe distribution space for the buried facilities based on the distribution diagram. If there is an interference between the divergent space and the safe distribution space, mark the construction behavior in the interference area as a dangerous monitoring action and control the remote monitoring equipment to issue a monitoring warning signal.

7. A remote monitoring system with intelligent perception of situation, vibration and orientation, characterized in that: The remote monitoring system includes a memory and a processor. The memory stores a remote monitoring method program that intelligently perceives situation, vibration and orientation. When the remote monitoring method program is executed by the processor, the remote monitoring method steps as described in any one of claims 1 to 6 are implemented.

Images