CN-121994193-A - Automatic change building settlement monitoring target device

CN121994193ACN 121994193 ACN121994193 ACN 121994193ACN-121994193-A

Abstract

The invention relates to the technical field of engineering measurement and structural health monitoring, in particular to an automatic building settlement monitoring target device which comprises a multidimensional environment sensing module, an abnormal characteristic early warning and cleaning module and a statistical outlier algorithm, wherein the multidimensional environment sensing module is used for synchronously acquiring original settlement measurement data, structural vibration acceleration data and multipoint medium temperature data of building monitoring points and outputting an original multidimensional sensing signal set, and the abnormal characteristic early warning and cleaning module is used for carrying out data integrity verification on signals by utilizing a preset numerical interval threshold and an electrical parameter detection algorithm based on the original multidimensional sensing signal set and removing outlier data exceeding the statistical threshold by utilizing the statistical outlier algorithm. According to the invention, the thermal error compensation of the physical layer and the self-adaptive filtering of the statistical layer are tightly combined through the edge fusion resolving module, the structural vibration data is utilized to dynamically adjust the observed covariance parameter of the Kalman filtering, and the misleading of high-frequency vibration interference on the measurement weight is automatically reduced while the long-period temperature drift error of the sensor is eliminated.

Inventors

MENG ZHAOHU
LI JINGBO
PENG BO
CHEN HUA
LIU HAIWEN

Assignees

山东正元建设工程有限责任公司

Dates

Publication Date: 20260508
Application Date: 20260128

Claims (7)

1. An automated building settlement monitoring target device, comprising: The multi-dimensional environment sensing module is used for synchronously collecting original settlement measurement data, structural vibration acceleration data and multi-point medium temperature data of building monitoring points and outputting an original multi-dimensional sensing signal set; The abnormal characteristic early warning and cleaning module is used for carrying out data integrity verification on signals by utilizing a preset numerical interval threshold value and an electrical parameter detection algorithm based on the original multidimensional sensing signal set, removing outlier data exceeding the statistical threshold value by utilizing a statistical outlier algorithm and outputting verified effective characteristic vectors; The edge fusion calculation module is used for carrying out weighted fusion and thermal error compensation on the original settlement measurement data by using the structural vibration acceleration data in the feature vector as an observation covariance parameter of Kalman filtering based on the verified effective feature vector and outputting a corrected structural displacement value; The self-adaptive strategy management and control module calculates the change rate of the corrected structural displacement value based on the corrected structural displacement value, compares the change rate with a preset state threshold value, determines a system state grade identifier according to the change rate, and generates a dynamic sampling scheduling instruction for dynamically switching a control device between a low-frequency inspection mode and a high-frequency burst tracking mode; The space-time trend prediction coding module calculates a displacement trend slope by utilizing a linear regression model based on the dynamic sampling scheduling instruction, the current corrected structural displacement value and the historical data, codes and packages the displacement value, the displacement trend slope and the system state grade identifier, and outputs a monitoring message containing state codes; And the remote interaction transmission module is used for transmitting the monitoring message containing the state code to the cloud server through a wireless network and outputting a dormancy trigger signal for controlling each module to enter a dormancy state according to a confirmation receipt returned by the server.
2. The automated building settlement monitoring target device according to claim 1, wherein in the multi-dimensional environment sensing module, the outputting of the original multi-dimensional sensing signal set specifically comprises synchronously and parallelly sampling physical quantity reflecting vertical displacement of the building, structural vibration acceleration component and medium thermodynamic distribution data in response to a collection instruction of the system; Converting the physical signals obtained by sampling into digital quantities, and respectively extracting original sedimentation measurement data, structural vibration acceleration data and multipoint medium temperature data; Associating uniform microsecond-level time stamps with the original sedimentation measurement data, the structural vibration acceleration data and the multipoint medium temperature data in the same sampling period; Based on a predefined data protocol, combining each data map carrying the microsecond time stamps into a multi-channel data vector, and generating the original multi-dimensional sensing signal set.
3. The automated building settlement monitoring target device according to claim 1, wherein in the abnormal feature pre-warning and cleaning module, the removing outlier data exceeding the statistical threshold using the statistical outlier algorithm specifically comprises: constructing a fixed-length first-in first-out sliding window sequence based on original sedimentation measurement data of a historical time sequence; Carrying out statistical distribution calculation on the numerical values in the sliding window sequence, and extracting an arithmetic mean value reflecting the local trend of the data and a standard deviation reflecting the discrete degree in real time; taking the arithmetic mean value as a reference, and combining the weighted standard deviation to construct a dynamic updated effective numerical confidence interval; comparing the currently acquired original sedimentation measurement data with the effective value confidence interval, if the data falls outside the effective value confidence interval, marking the data as a statistical outlier, and performing smooth substitution by using a neighborhood interpolation algorithm to finish the elimination and correction of the outlier data.
4. The automated building settlement monitoring target device according to claim 1, wherein in the edge fusion calculation module, the performing weighted fusion and thermal error compensation on the raw settlement measurement data, and outputting the corrected structural displacement value specifically comprises: Calculating thermal drift correction quantity generated by medium temperature change based on the multipoint medium temperature data and a preset temperature-displacement correlation model, and subtracting the thermal drift correction quantity from the original sedimentation measurement data to generate a temperature compensated intermediate variable; Calculating the modulus of the structural vibration acceleration data, and establishing a positive correlation mapping relation between the modulus and the observed covariance parameter through a nonlinear mapping function to determine the observed covariance parameter in a Kalman filtering algorithm; Constructing a state prediction equation based on the system state estimation value at the previous moment, and calculating a priori state estimation value and a priori error covariance matrix at the current moment; And calculating a Kalman gain matrix by using the observed covariance parameter and the priori error covariance matrix, carrying out weighted correction on the residual error of the intermediate variable after temperature compensation and the priori state estimated value by using the Kalman gain matrix, and outputting a posterior state estimated value as the corrected structural displacement value.
5. The automated building settlement monitoring target device according to claim 1, wherein in the adaptive strategy control module, the generating the dynamic sampling scheduling instruction for dynamically switching the control device between the low frequency patrol mode and the high frequency burst tracking mode specifically comprises: Calculating the displacement increment in unit time based on the corrected structural displacement value output at the current moment and the historical displacement value at the last moment, and generating an instantaneous sedimentation change rate; comparing the instantaneous sedimentation change rate with a preset safety trigger threshold, if the instantaneous sedimentation change rate is greater than or equal to the safety trigger threshold, calling a first preset time parameter, and setting the system state grade mark as a high-frequency burst tracking mode; if the instantaneous sedimentation change rate is smaller than the safety trigger threshold, a second preset time parameter is called, and the system state grade mark is set into a low-frequency inspection mode; And calculating the starting time of the next acquisition period based on the determined mode state and the corresponding time parameter, and generating a dynamic sampling scheduling instruction containing the starting time information.
6. The automated building settlement monitoring target device according to claim 1, wherein in the spatiotemporal trend predictive coding module, the calculating the displacement trend slope based on the current corrected structural displacement value and the historical data using the linear regression model specifically comprises: Extracting a history correction displacement value of the latest preset number of periods from a data buffer area, and combining the history correction displacement value with the corrected structure displacement value at the current moment to construct a short-time sequence regression sample set arranged in time sequence; establishing a unitary linear regression model with relative sampling time as an independent variable and a structural displacement amplitude as a dependent variable; Fitting operation is carried out on the short-time sequence regression sample set by adopting a least square method, and a regression straight line equation which minimizes the residual square sum of the sample points to a fitting straight line is solved; and extracting a first order coefficient in the regression line equation to be used as the displacement trend slope.
7. The automated building settlement monitoring target device according to claim 1, wherein in the remote interactive transmission module, outputting a sleep trigger signal for controlling each module to enter a sleep state according to a confirmation receipt returned by the server specifically comprises: After the monitoring message is sent, a downlink data monitoring window with preset duration is opened, and the data packet received in the window is analyzed to identify a handshake confirmation protocol frame; carrying out data integrity check on the handshake confirmation protocol frame, and if the data integrity check is passed, judging that the cloud synchronization task in the current period is completed; analyzing the next acquisition starting time determined in the dynamic sampling scheduling instruction, calculating a time difference value from the current system time to the next acquisition starting time, and generating a dormancy countdown parameter; And constructing a system-level interrupt instruction containing the dormancy countdown parameter, and outputting the system-level interrupt instruction as the dormancy trigger signal to trigger the system to suspend the current running process and enter a low-power consumption standby mode.

Description

Automatic change building settlement monitoring target device Technical Field The invention relates to the technical field of engineering measurement and structural health monitoring, in particular to an automatic building settlement monitoring target device. Background In large-scale infrastructure construction and operation and maintenance management, the long-term real-time monitoring of settlement deformation of buildings and geological structures is a key means for guaranteeing structural safety, wherein the static leveling instrument is widely applied to vertical displacement monitoring scenes of subway tunnels, bridge pipe galleries and dangerous old houses because of moderate measuring ranges, high precision and easiness in networking, and the relative vertical settlement of each measuring point is inverted mainly by measuring the liquid level change in a communicating pipe. The anti-interference capability of the existing static level monitoring equipment on a complex field environment is relatively limited in practical application, on one hand, the internal filling liquid of a sensor and mechanical parts are influenced by the fluctuation of the ambient temperature to generate expansion and contraction, so that measurement data are superimposed with unrealistic long-period thermal drift errors, on the other hand, the liquid level high-frequency fluctuation is easily caused by the surrounding vehicles passing or mechanical vibration generated by construction operation, the actual structure micro deformation and the ambient coupling noise are often difficult to accurately distinguish by the existing conventional linear filtering algorithm under the dynamic environment, the fluctuation or artifact of the output displacement data is easily generated, and the signal to noise ratio and the reliability of the monitoring result are influenced to a certain extent. Disclosure of Invention In order to make up for the defects, the invention provides an automatic building settlement monitoring target device, which aims to solve the problem that the existing static level monitoring equipment is easily interfered by thermal drift caused by environmental temperature fluctuation and liquid level fluctuation caused by mechanical vibration in a complex site. The invention provides a technical scheme that an automatic building settlement monitoring target device comprises: The multi-dimensional environment sensing module is used for synchronously collecting original settlement measurement data, structural vibration acceleration data and multi-point medium temperature data of building monitoring points and outputting an original multi-dimensional sensing signal set; The abnormal characteristic early warning and cleaning module is used for carrying out data integrity verification on signals by utilizing a preset numerical interval threshold value and an electrical parameter detection algorithm based on the original multidimensional sensing signal set, removing outlier data exceeding the statistical threshold value by utilizing a statistical outlier algorithm and outputting verified effective characteristic vectors; The edge fusion calculation module is used for carrying out weighted fusion and thermal error compensation on the original settlement measurement data by using the structural vibration acceleration data in the feature vector as an observation covariance parameter of Kalman filtering based on the verified effective feature vector and outputting a corrected structural displacement value; The self-adaptive strategy management and control module calculates the change rate of the corrected structural displacement value based on the corrected structural displacement value, compares the change rate with a preset state threshold value, determines a system state grade identifier according to the change rate, and generates a dynamic sampling scheduling instruction for dynamically switching a control device between a low-frequency inspection mode and a high-frequency burst tracking mode; The space-time trend prediction coding module calculates a displacement trend slope by utilizing a linear regression model based on the dynamic sampling scheduling instruction, the current corrected structural displacement value and the historical data, codes and packages the displacement value, the displacement trend slope and the system state grade identifier, and outputs a monitoring message containing state codes; And the remote interaction transmission module is used for transmitting the monitoring message containing the state code to the cloud server through a wireless network and outputting a dormancy trigger signal for controlling each module to enter a dormancy state according to a confirmation receipt returned by the server. Preferably, in the multidimensional environment sensing module, the output of the original multidimensional sensing signal set specifically comprises the steps of responding to an acquisition instruction o