A three-dimensional displacement sensor and its real-time transmission method

Abstract

The present invention relates to the technical field of three-dimensional displacement sensors, and discloses a three-dimensional displacement sensor and a real-time transmission method thereof. The three-dimensional displacement sensor includes a three-dimensional displacement measurement unit, and an edge computing module connected to the three-dimensional displacement measurement unit. The edge computing module includes: a data preprocessing unit, a trend analysis and prediction unit, a dynamic displacement threshold adjustment unit, an intelligent decision-making and control unit, and a data transmission unit. By introducing an edge computing module at the sensor device end, the present invention efficiently processes and optimizes the transmission of the collected data, solves many problems in the process of traditional sensor data processing and transmission, and improves the overall performance and application flexibility of the system; at the same time, the dynamic displacement threshold adjustment unit can dynamically adjust the displacement warning threshold according to the on-site environmental conditions, and through intelligent analysis of the displacement prediction value and the dynamic displacement warning threshold, early warning can be realized, increasing the safety of mining activities.

Description

Technical Field

[0001] The present invention relates to the technical field of three-dimensional displacement sensors, and specifically to a three-dimensional displacement sensor and a real-time transmission method thereof. Background Art

[0002] In mining activities, three-dimensional displacement sensors play many important roles, such as displacement monitoring, stability assessment, safety warning, etc. Traditional three-dimensional displacement sensors often simply transmit the collected raw data directly to the cloud server. Due to the lack of effective local processing capabilities, a large amount of unprocessed raw data not only occupies a large amount of transmission bandwidth, but also brings a heavy computational burden to the cloud server. At the same time, the mining working environment is complex, and geological factors, mining depth factors, etc. will cause changes in the safety warning thresholds of three-dimensional displacement sensors. Existing three-dimensional displacement sensors cannot dynamically adjust the warning thresholds according to the actual working environment of the mine, and there are still potential safety hazards of untimely warnings.

[0003] To solve the above problems, we propose a three-dimensional displacement sensor and a real-time transmission method thereof. Summary of the Invention

[0004] The purpose of the present invention is to provide a three-dimensional displacement sensor and a real-time transmission method thereof, which can dynamically adjust the warning threshold according to the actual working environment of the mine and improve the safety of mining activities.

[0005] To achieve the above purpose, the present invention provides the following technical solution: A real-time transmission method for a three-dimensional displacement sensor includes the following steps:

[0006] The edge computing module obtains the displacement time series data of the monitoring point through data transmission with the three-dimensional displacement measurement unit;

[0007] The edge computing module performs stationarity tests and differencing processing on the displacement time series data to determine the order of the ARIMA model;

[0008] Build a displacement trend prediction model based on the ARIMA model in the edge computing module to predict the future displacement of the monitoring point and obtain the displacement prediction value;

[0009] The edge computing module transmits data to the cloud server through the data transmission unit, and the cloud server sends the geological comprehensive information, rainfall information, wind speed information, and mining depth information of each monitoring point to the edge computing modules arranged at each monitoring point at regular intervals;

[0010] Dynamically adjust the displacement warning threshold based on the new geological comprehensive information, rainfall information, wind speed information, and mining depth information to obtain the dynamically adjusted displacement warning threshold after comprehensive adjustment;

[0011] Compare the displacement prediction value with the dynamic displacement threshold, and trigger an early warning when the prediction value is equal to or exceeds the dynamic displacement threshold;

[0012] The edge computing module sends the displacement data real-time monitored by the three-dimensional displacement measurement unit, the prediction value of the displacement trend prediction model, and the early warning information to the cloud server through the data transmission unit.

[0013] Preferably, the method for performing stationarity test and differencing on the displacement time series data to determine the order of the ARIMA model is as follows:

[0014] Judge whether the displacement time series data has stationarity characteristics;

[0015] If the displacement time series data is non-stationary, perform differencing processing to make the data stationary;

[0016] Determine the appropriate order combination of the ARIMA model by observing the autocorrelation function and partial autocorrelation function.

[0017] Preferably, the formula of the displacement trend prediction model based on the ARIMA model is as follows:

[0018] ;

[0019] Among them, p is the autoregressive order, q is the moving average order, is the predicted value of the i-th monitoring point at time, represents the number of future time steps, is the mean of the time series, is the autoregressive coefficient, is the moving average coefficient, is the index variable, is the actual displacement of the i-th monitoring point at time, is the white noise sequence.

[0020] Preferably, the method for dynamically adjusting the displacement early warning threshold based on the new geological comprehensive information, rainfall information, wind speed information, and mining depth information to obtain the dynamically adjusted displacement early warning threshold after comprehensive adjustment is as follows:

[0021] Calculate the threshold adjustment coefficients of the geological comprehensive information, rainfall information, wind speed information, and mining depth information respectively;

[0022] The threshold adjustment coefficient of the geological comprehensive information is: ;

[0023] Among them, represents the geological comprehensive influence factor around the i-th monitoring point, is the geological condition influence strength coefficient;

[0024] The threshold adjustment coefficient of rainfall information is: ;

[0025] in, It is a weight coefficient determined based on experience and is used to adjust the degree of influence of different rainfall intervals on threshold adjustment. is the displacement exceeding threshold frequency during rainfall, is the frequency of displacement exceeding the threshold in normal weather;

[0026] The threshold adjustment coefficient of wind speed information is: ;

[0027] in, It is a coefficient determined according to the accuracy requirements and stability characteristics of the monitoring facilities and is used to adjust the degree of wind speed influence. To monitor the shaking amplitude of the facility, is the threshold of the allowable shaking amplitude; the threshold adjustment coefficient of the mining depth information is: ;

[0028] in, It is the reference value of the depth threshold initially set. It is a depth value predetermined based on the geological conditions of the mine and engineering experience. Increase the amount for mining depth;

[0029] The weights of the threshold adjustment coefficients of geological comprehensive information, rainfall information, wind speed information, and mining depth information are calculated by the hierarchical analysis method, respectively: ;

[0030] Calculate the comprehensive threshold adjustment factor: ;

[0031] Calculate the synthetically adjusted dynamic displacement threshold: ;in, is the initial displacement warning threshold.

[0032] Preferably, the method of calculating the weights of the threshold adjustment coefficients of the geological comprehensive information, rainfall information, wind speed information, and mining depth information respectively by using the hierarchical analysis method is as follows:

[0033] The problem of determining the weight of the threshold adjustment coefficient of the mine displacement monitoring and early warning system is divided into the target layer, the criterion layer and the scheme layer. The main task of the target layer is to reasonably determine the weight of the threshold adjustment coefficient. The criterion layer includes the threshold adjustment coefficient corresponding to the geological comprehensive information, rainfall information, wind speed information, and mining depth information. The scheme layer is a specific weight allocation scheme.

[0034] Construct a judgment matrix by comparing the relative importance among the factors in the criterion layer;

[0035] Perform eigenvalue decomposition on the judgment matrix, calculate the maximum eigenvalue and its corresponding eigenvector, and normalize the eigenvector to obtain the weight vector of the threshold adjustment coefficients for each factor.

[0036] A three-dimensional displacement sensor, comprising:

[0037] A three-dimensional displacement measurement unit for collecting displacement time series data of a monitoring point;

[0038] An edge computing module connected to the three-dimensional displacement measurement unit, the edge computing module comprising:

[0039] A data preprocessing unit that receives the displacement time series data from the three-dimensional displacement measurement unit and performs filtering on it to remove noise interference;

[0040] A trend analysis and prediction unit for predicting the displacement value of the monitoring point and outputting a displacement prediction value;

[0041] A dynamic displacement threshold adjustment unit for adjusting the dynamic displacement warning threshold according to the geological comprehensive information, rainfall information, wind speed information, and mining depth information of the monitoring point sent by the cloud server to the edge computing module;

[0042] An intelligent decision-making and control unit compares the displacement prediction value with the dynamic displacement warning threshold, and triggers an alarm signal or controls the action of relevant devices when the displacement prediction value is equal to or greater than the dynamic displacement warning threshold;

[0043] A data transmission unit connected to the edge computing module for transmitting the data processed by the edge computing module to the cloud server. The data transmission unit adopts an optimized transmission protocol to reduce protocol overhead, and combines a reliable transmission mechanism and a real-time transmission protocol to ensure stable and fast data transmission.

[0044] Preferably, the three-dimensional displacement measurement unit is composed of a laser interferometric, capacitive or other principle capable of realizing three-dimensional displacement measurement.

[0045] Preferably, the data preprocessing unit in the edge computing module uses Kalman filtering, mean filtering or other effective filtering algorithms to remove noise, and uses dictionary coding, wavelet transform or other suitable data compression algorithms.

[0046] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0047] 1. By introducing an edge computing module at the sensor device end, the present invention efficiently processes and optimally transmits the collected data, solves many problems in the traditional sensor data processing and transmission process, and improves the overall performance and application flexibility of the system.

[0048] 2. By introducing a trend analysis and prediction unit in the edge computing module, the present invention can predict future displacement values based on the displacement time series data of the monitoring point. At the same time, through the dynamic displacement threshold adjustment unit, the displacement warning threshold can be dynamically adjusted according to the on-site environmental conditions. Through the intelligent analysis of the predicted value of the trend analysis and prediction unit and the dynamic displacement warning threshold of the dynamic displacement threshold adjustment unit, early warning can be realized, increasing the safety of mining activities. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is a schematic flow chart of a real-time transmission method of a three-dimensional displacement sensor according to the present invention;

[0050] Figure 2 It is a schematic diagram of a three-dimensional displacement sensor system according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0051] Next, the technical solutions of the present invention will be clearly and completely described in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0052] Please refer to Figure 1 - Figure 2 , the present invention provides a technical solution: a real-time transmission method of a three-dimensional displacement sensor, including the following steps:

[0053] The edge computing module obtains the displacement time series data of the monitoring point through data transmission with the three-dimensional displacement measurement unit; the displacement time series data refers to the data set of the displacement conditions of the monitoring point at different time points, which can be used to analyze the displacement movement law of the monitoring point, predict future displacement conditions, etc.

[0054] The edge computing module performs stationarity test and differencing processing on the displacement time series data to determine the order of the ARIMA model; the method is as follows:

[0055] Judge whether the displacement time series data has stationarity characteristics;

[0056] If the displacement time series data is not stationary, perform differencing processing to make the data stationary;

[0057] Determine the appropriate order combination of the ARIMA model by observing the autocorrelation function and the partial autocorrelation function.

[0058] The autocorrelation function describes the correlation between a time series and its own lagged values. The partial autocorrelation function, on the other hand, measures the correlation between a time series and its lagged values after controlling for the effects of intermediate lag terms. For an ARIMA model, the values of p and q are initially determined by observing the autocorrelation function and the partial autocorrelation function plots. If the autocorrelation function suddenly truncates (i.e., quickly approaches zero) after a certain lag order, it may indicate the value of q; if the partial autocorrelation function suddenly truncates after a certain lag order, it may indicate the value of p.

[0059] Construct a displacement trend prediction model based on the ARIMA model within the edge computing module to predict the future displacement of the monitoring point and obtain the displacement prediction value;

[0060] The formula for the displacement trend prediction model based on the ARIMA model is as follows:

[0061] ;

[0062] where p is the autoregressive order, q is the moving average order, is the predicted value of the i-th monitoring point at time, represents the number of future time steps, is the mean of the time series;

[0063] are the autoregressive coefficients. In the autoregressive part of the ARIMA model, they are used to measure the degree of influence of past displacement values on the current predicted displacement value. For example, represents the influence weight of the displacement at the previous time step on the current predicted displacement, represents the influence weight of the displacement at the two previous time steps on the current predicted displacement, and so on. These coefficients are estimated from historical displacement data through model training (such as the least squares method, etc.) to reflect the autocorrelation of the displacement data.

[0064] are the moving average coefficients. In the moving average part (MA part) of the ARIMA model, they are used to measure the degree of influence of past prediction errors (white noise) on the current predicted displacement value. For example, represents the influence weight of the prediction error at the previous time step on the current predicted displacement, represents the influence weight of the prediction errors at the two previous time steps on the current predicted displacement, and so on. These coefficients are also estimated from historical data through model training.

[0065] is the index variable in the autoregressive part of the formula , is used to traverse the past time steps corresponding to the autoregressive order p. For example, when p = 3, will take 1, 2, and 3 respectively. This means that when the model predicts , it will consider the impact of the displacement deviation from the mean at the past 1, 2, and 3 time steps on the current predicted value. Specifically, multiplied by represents the contribution of the displacement deviation from the mean at the previous time step to the current predicted displacement, multiplied by represents the contribution of the displacement deviation from the mean at the previous two time steps to the current predicted displacement, and so on.

[0066] In the moving average part of the formula , is used to traverse the past time steps corresponding to the moving average order q. For example, when q = 2, will take 1 and 2 respectively. This means that when the model predicts , it will consider the impact of the prediction errors (white noise) at the past 1 and 2 time steps on the current predicted value. Specifically, multiplied by represents the contribution of the prediction error at the previous time step to the current predicted displacement, multiplied by represents the contribution of the prediction error at the previous two time steps to the current predicted displacement.

[0067] is the actual displacement of the i-th monitoring point at time. After subtracting the mean , the deviation of the displacement at this time relative to the average level can be obtained. This deviation is multiplied by the autoregressive coefficient to calculate the contribution of the past displacement deviation to the future predicted displacement.

[0068] is a white noise sequence. White noise is a random process that represents the random fluctuation part in the time series that cannot be explained by the autoregressive and moving average parts in the model. In the ARIMA model, it is assumed that the prediction error (i.e., the difference between the actual displacement and the predicted displacement) follows a white noise distribution. The values of these white noise sequences are obtained through fitting and estimation of historical data during the model training process, and are used to consider the random uncertainty factors in the displacement data.

[0069] The edge computing module transmits data to the cloud server through the data transmission unit. The cloud server sends information about factors that affect the displacement monitoring value to the edge computing modules deployed at each monitoring point at regular intervals, such as every hour. The information includes comprehensive geological information, rainfall information, wind speed information, mining depth information, etc. of each monitoring point.

[0070] Based on the new comprehensive geological information, rainfall information, wind speed information, mining depth information, and other information about factors that affect the displacement monitoring value, the displacement warning threshold is dynamically adjusted to obtain the dynamically adjusted displacement warning threshold after comprehensive adjustment. In this embodiment, taking the comprehensive geological information, rainfall information, wind speed information, and mining depth information as examples, the method is as follows:

[0071] Calculate the threshold adjustment coefficients of the comprehensive geological information, rainfall information, wind speed information, and mining depth information respectively;

[0072] The threshold adjustment coefficient of the comprehensive geological information is:

[0073] , where represents the comprehensive geological influence factor around the i-th monitoring point, and its value range is 0 - 1, which is determined comprehensively according to geological factors such as formation lithology, fault distribution, and degree of joint fissure development. is the influence intensity coefficient of geological conditions, which is also determined according to experience or tests. When the geological conditions around the monitoring point are complex (such as soft lithology, many faults, and well-developed joint fissures), the value is larger, is also larger, so that the adjusted displacement warning threshold increases. This is because in areas with poor geological conditions, the deformability of the rock mass itself is larger, so the displacement threshold needs to be appropriately relaxed.

[0074] The threshold adjustment coefficient of the rainfall information is:

[0075] , where is the weight coefficient determined according to experience, which is used to adjust the influence degree of different rainfall intervals on the threshold adjustment, is the frequency of displacement exceeding the threshold during rainfall, is the frequency of displacement exceeding the threshold in normal weather;

[0076] The threshold adjustment coefficient of the wind speed information is:

[0077] , where is the coefficient determined according to the accuracy requirements and stability characteristics of the monitoring facilities, which is used to adjust the influence degree of the wind speed, is the shaking amplitude of the monitoring facilities, is the threshold of the allowable shaking amplitude;

[0078] The threshold adjustment coefficient of the mining depth information is:

[0079] , where is the initial set reference value of the depth threshold, which is a depth value determined in advance according to the geological conditions and engineering experience of the mine is the increase in mining depth;

[0080] The weights of the threshold adjustment coefficients of geological comprehensive information, rainfall information, wind speed information, and mining depth information are calculated respectively by the analytic hierarchy process, and are respectively: ; The calculation method is as follows:

[0081] The problem of determining the weights of the threshold adjustment coefficients of the mine displacement monitoring and early warning system is divided into an objective layer, a criterion layer, and a scheme layer. The main task of the objective layer is to reasonably determine the weights of the threshold adjustment coefficients. The criterion layer includes the threshold adjustment coefficients corresponding to geological comprehensive information, rainfall information, wind speed information, and mining depth information. The scheme layer is a specific weight distribution scheme;

[0082] A judgment matrix is constructed by comparing the relative importance of the factors in the criterion layer; for example, for the comparison of the weights of the threshold adjustment coefficients corresponding to the mining depth and geological conditions, mine engineering experts, geological experts, etc. are invited to make judgments based on experience and professional knowledge. If the experts believe that the importance of the mining depth to the displacement threshold adjustment is twice that of the geological conditions, then the value at the intersection of the row corresponding to the mining depth and the column corresponding to the geological conditions in the judgment matrix is 2, and vice versa, the value at the intersection of the row corresponding to the geological conditions and the column corresponding to the mining depth is 1 / 2.

[0083] Perform eigenvalue decomposition on the judgment matrix, find the largest eigenvalue and its corresponding eigenvector, and normalize the eigenvector to obtain the weight vector of the threshold adjustment coefficients of each factor. For example, assuming that the calculated eigenvector is [0.4, 0.1, 0.1, 0.4], then the weight of the threshold adjustment coefficient of geological comprehensive information is 0.4, the threshold adjustment coefficient of rainfall information is 0.1, the threshold adjustment coefficient of wind speed information is 0.1, and the threshold adjustment coefficient of mining depth information is 0.4, and so on.

[0084] Calculate the comprehensive threshold adjustment coefficient: ;

[0085] Calculate the dynamically adjusted dynamic displacement threshold: ; where is the initial displacement warning threshold.

[0086] Compare the displacement prediction value with the dynamic displacement threshold, and trigger an early warning when the prediction value is equal to or exceeds the dynamic displacement threshold;

[0087] The edge computing module sends the displacement data, prediction values of the displacement trend prediction model, and early warning information monitored in real time by the three-dimensional displacement measurement unit to the cloud server through the data transmission unit.

[0088] A three-dimensional displacement sensor, as Figure 2 shown, includes:

[0089] A three-dimensional displacement measurement unit for collecting displacement time series data of the monitoring point;

[0090] An edge computing module connected to the three-dimensional displacement measurement unit, the edge computing module includes:

[0091] A data preprocessing unit that receives the displacement time series data from the three-dimensional displacement measurement unit and filters it to remove noise interference;

[0092] A trend analysis and prediction unit for predicting the displacement value of the monitoring point and outputting a displacement prediction value;

[0093] A dynamic displacement threshold adjustment unit for adjusting the dynamic displacement early warning threshold according to the geological comprehensive information, rainfall information, wind speed information, and mining depth information of the monitoring point sent by the cloud server to the edge computing module.

[0094] An intelligent decision-making and control unit compares the displacement prediction value with the dynamic displacement early warning threshold, and triggers an alarm signal or controls the action of related devices when the displacement prediction value is equal to or greater than the dynamic displacement early warning threshold;

[0095] A data transmission unit connected to the edge computing module for transmitting the data processed by the edge computing module to the cloud server. The data transmission unit uses an optimized transmission protocol to reduce protocol overhead, and combines a reliable transmission mechanism and a real-time transmission protocol to ensure stable and fast data transmission.

[0096] The three-dimensional displacement measurement unit is composed of a laser interferometric, capacitive or other principle that can realize three-dimensional displacement measurement.

[0097] The data preprocessing unit in the edge computing module uses Kalman filtering, mean filtering or other effective filtering algorithms to remove noise, and uses dictionary coding, wavelet transform or other suitable data compression algorithms. The implementation of the Kalman filtering algorithm is as follows: First, define the state equation and measurement equation of the system, and determine the relevant parameters in the Kalman filtering algorithm according to the sampling period of the sensor and the characteristics of the measurement noise and system noise, such as the state transition matrix, measurement matrix, process noise covariance matrix and measurement noise covariance matrix, etc. In the code, through a loop structure, the prediction and update steps are continuously carried out to perform real-time filtering processing on the collected original displacement data. The implementation of the data compression algorithm based on dictionary coding first constructs a dictionary data structure. During the data processing process, the continuously occurring data sequences are matched with the entries in the dictionary. If the match is successful, the original data sequence is replaced by the index in the dictionary for transmission, thus realizing data compression. When decompressing, the original data sequence is restored according to the dictionary index.

[0098] The present invention enhances the data processing ability of the device side by introducing an edge computing module at the sensor device side, and reduces the computing burden of the cloud server. At the same time, the security warning threshold of the present application can be dynamically adjusted according to the changes in the environment, improving the safety of mining activities.

[0099] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A real-time transmission method for a three-dimensional displacement sensor, characterized in that Including the following steps: The edge computing module obtains the displacement time series data of the monitoring points through data transmission with the three-dimensional displacement measurement unit; The edge computing module conducts a stationarity test and differencing process on the displacement time series data to determine the order of the ARIMA model; Build a displacement trend prediction model based on the ARIMA model within the edge computing module to predict the future displacement of the monitoring points and obtain displacement prediction values; The edge computing module conducts data transmission with the cloud server through the data transmission unit. The cloud server sends information on factors that affect the displacement monitoring values to the edge computing modules arranged at each monitoring point at regular intervals, including the comprehensive geological information, rainfall information, wind speed information, and mining depth information of each monitoring point; Dynamically adjust the displacement warning threshold based on the new comprehensive geological information, rainfall information, wind speed information, and mining depth information to obtain the dynamically adjusted displacement warning threshold after comprehensive adjustment. The method is as follows: Calculate the threshold adjustment coefficients of the comprehensive geological information, rainfall information, wind speed information, and mining depth information respectively; The threshold adjustment coefficient of the comprehensive geological information is: , where represents the comprehensive geological influence factor around the i-th monitoring point, and its value range is 0-1, which is determined comprehensively according to the formation lithology, fault distribution, and joint fracture development degree. is the influence intensity coefficient of geological conditions; the threshold adjustment coefficient of rainfall information is: , where is a weight coefficient determined based on experience and is used to adjust the influence degree of different rainfall intervals on the threshold adjustment, is the frequency of displacement exceeding the threshold during rainfall, is the frequency of displacement exceeding the threshold under normal weather conditions; The threshold adjustment coefficient of the wind speed information is: , where is a coefficient determined according to the accuracy requirements and stability characteristics of the monitoring facility, and is used to adjust the degree of influence of the wind speed, is the sway amplitude of the monitoring facility, is the threshold of the allowable sway amplitude; the threshold adjustment coefficient of the mining depth information is: ,in, It is the reference value of the depth threshold initially set. It is a depth value predetermined based on the geological conditions of the mine and engineering experience. Increase the amount for mining depth; Calculate the weights of the threshold adjustment coefficients of the comprehensive geological information, rainfall information, wind speed information, and mining depth information respectively through the analytic hierarchy process; Multiply the threshold adjustment coefficient of each factor information that affects the displacement monitoring value by its corresponding weight and then sum them to calculate the comprehensive threshold adjustment coefficient; Multiply the comprehensive threshold adjustment coefficient by the initial displacement warning threshold to obtain the dynamically adjusted displacement warning threshold after comprehensive adjustment; Compare the displacement prediction value with the dynamic displacement threshold. When the prediction value is equal to or exceeds the dynamic displacement threshold, a warning is triggered; The edge computing module sends the displacement data, prediction values of the displacement trend prediction model, and warning information monitored in real time by the three-dimensional displacement measurement unit to the cloud server through the data transmission unit.

2. A real-time transmission method for a three-dimensional displacement sensor according to claim 1, characterized in that The method for conducting a stationarity test and differencing process on the displacement time series data to determine the order of the ARIMA model is as follows: Judge whether the displacement time series data has stationarity characteristics; If the displacement time series data is not stationary, conduct a differencing process to make the data stationary; Determine the appropriate order combination of the ARIMA model by observing the autocorrelation function and partial autocorrelation function.

3. A real-time transmission method of a three-dimensional displacement sensor according to claim 2, characterized in that, The formula of the displacement trend prediction model based on the ARIMA model is as follows: ； where p is the autoregressive order and q is the moving average order, is the predicted value of the i-th monitoring point at time, represents the number of future time steps, is the mean of the time series, are the autoregressive coefficients, are the moving average coefficients, is the index variable, is the actual displacement of the i-th monitoring point at time, is a white noise sequence.

4. A three-dimensional displacement sensor for implementing the real-time transmission method of a three-dimensional displacement sensor according to any one of claims 1-3, characterized in that, Including: A three-dimensional displacement measurement unit for collecting the displacement time series data of the monitoring points; An edge computing module connected to the three-dimensional displacement measurement unit. The edge computing module includes: A data preprocessing unit that receives the displacement time series data from the three-dimensional displacement measurement unit and conducts filtering processing on it to remove noise interference; A trend analysis and prediction unit for predicting the displacement value of the monitoring points and outputting displacement prediction values; A dynamic displacement threshold adjustment unit for adjusting the dynamic displacement warning threshold according to the comprehensive geological information, rainfall information, wind speed information, and mining depth information of the monitoring points sent by the cloud server to the edge computing module; The intelligent decision-making and control unit compares the displacement prediction value with the dynamic displacement warning threshold. When the displacement prediction value is equal to or greater than the dynamic displacement warning threshold, it triggers an alarm signal or controls the action of relevant devices. The data transmission unit is connected to the edge computing module and is used to transmit the data processed by the edge computing module to the cloud server. The data transmission unit adopts an optimized transmission protocol to reduce protocol overhead, and combines a reliable transmission mechanism and a real-time transmission protocol to ensure stable and fast data transmission.

5. The three-dimensional displacement sensor according to claim 4, wherein: The three-dimensional displacement measurement unit is composed of a laser interferometric, capacitive or other principle capable of realizing three-dimensional displacement measurement.

6. The three-dimensional displacement sensor according to claim 4, wherein: The data preprocessing unit in the edge computing module uses Kalman filtering, mean filtering or other effective filtering algorithms to remove noise, and uses dictionary coding-based, wavelet transform or other suitable data compression algorithms.

Images