Six-degree-of-freedom safety monitoring system

Abstract

The invention relates to the technical field of safety monitoring, and discloses a six-degree-of-freedom safety monitoring system, which comprises the following modules: a hardware acquisition module; the system comprises a high-precision displacement meter, an IMU (Inertial Measurement Unit) which is installed at the same point with the high-precision displacement meter, and an environment sensor for monitoring temperature and humidity, the edge calculation and data fusion module adopts an embedded data acquisition terminal and acquires and fuses the data acquired by the hardware acquisition module; and the cloud processing module uploads the monitoring data to a cloud server in real time through a 4G / 5G wireless network to form a real-time data stream. By combining a high-precision sensor and a real-time data acquisition technology, high-precision and real-time monitoring of structural health is realized. The system reduces data noise through Kalman filtering and Butterworth filtering, and can carry out temperature compensation according to the environment temperature to ensure the measurement precision. The multi-sensor data dynamic fusion algorithm automatically adjusts the sensor weight, and the data accuracy and the system stability are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of safety monitoring, and particularly relates to a six-degree-of-freedom safety monitoring system. BACKGROUND

[0002] With the rapid development of industrialization and modernization, the safety of engineering structures has become a focus of attention. Especially in the operation process of key infrastructure such as bridges, buildings, tunnels, etc., the monitoring and maintenance of structural safety are crucial. Traditional structural health monitoring methods mainly rely on manual inspection and simple sensor measurement, but these methods have problems such as long cycle, low efficiency, high cost, and difficulty in real-time feedback.

[0003] Currently, there are some applications in the field of structural health monitoring, but most of the monitoring systems mainly focus on single type of sensor collection, or lack real-time data analysis and prediction capabilities. Especially in complex engineering structures, a single monitoring method often cannot fully assess the health status, and cannot timely discover potential damage and risks, leading to engineering accidents.

[0004] The traditional health monitoring method also has the following problems:

[0005] Data processing lag: Many existing systems rely on offline analysis after data collection, which leads to data processing delay and cannot provide immediate feedback.

[0006] Lack of accurate prediction capability: Many monitoring systems cannot effectively combine historical data and real-time data to predict the future health status of the structure, especially cannot accurately assess the remaining life of the structure.

[0007] Insufficient maintenance and repair decisions: Existing systems often cannot provide accurate early warning and timely repair decisions, leading to inaccurate maintenance work.

[0008] In order to solve the above problems, modern monitoring systems begin to introduce multi-sensor data fusion technology, digital twin technology, real-time simulation technology and prediction model based on big data, in order to provide more accurate and real-time structural health status assessment and prediction. However, existing technologies still face problems of data diversity, complex calculation and model updating.

[0009] Based on this, the present application proposes a six-degree-of-freedom safety monitoring system, which aims to solve the problems in the current technology by combining multiple sensor data, edge computing and cloud processing technology, and real-time simulation and digital twin model updating. SUMMARY

[0010] To solve the technical problems proposed in the background art, the present application provides a six-degree-of-freedom safety monitoring system.

[0011] The application adopts the following technical solutions: a six-degree-of-freedom safety monitoring system comprising the following modules:

[0012] a hardware acquisition module; the hardware acquisition module comprises a high-precision displacement meter, an IMU inertial measurement unit installed at the same point as the high-precision displacement meter, and an environmental sensor for monitoring temperature and humidity; through the design and configuration of the above hardware acquisition module, the system can comprehensively monitor the six degrees of freedom (including three-axis translation and three-axis rotation) of the structure, and provide high-precision raw data for subsequent data analysis and processing.

[0013] an edge computing and data fusion module, which adopts an embedded data acquisition terminal to collect and fuse the data collected by the hardware acquisition module;

[0014] a cloud processing module, which uploads the monitoring data to a cloud server in real time through a 4G / 5G wireless network to form a real-time data stream;

[0015] a real-time simulation and digital twin self-evolution module, which judges the structural damage state in real time by constructing a real-time updated finite element or data-driven model, dynamically inverts material parameters (such as elastic modulus E) by using a genetic algorithm, and realizes real-time evolution of the digital twin;

[0016] a structure damage and life prediction module, which applies an LSTM network to perform time series analysis based on the updated digital twin model and historical data, predicts the fatigue damage trend of the structure, and accurately calculates the remaining life of the key components of the structure by using a Coffin-Manson model.

[0017] Further, the data acquisition and preprocessing in the edge computing and data fusion module comprises:

[0018] synchronous data acquisition; preliminary filtering and denoising of data; temperature compensation algorithm; dynamic fusion algorithm of multi-sensor data; data fusion and output; edge data compression and uploading.

[0019] Further, in the synchronous data acquisition

[0020] the displacement meter is arranged at a key node of the structure, has a range of ±200 mm and an accuracy of 0.01 mm, the angular velocity measurement range of the IMU inertial measurement unit is ±300° / s, the temperature accuracy is ±0.5℃, and the humidity accuracy is ±3%;

[0021] the displacement meter data is collected by an ADC, and the sampling frequency is ≥100Hz;

[0022] the IMU sensor is collected through an SPI interface, and the sampling frequency is ≥200Hz;

[0023] the environmental temperature data is collected through an I2C interface, and the sampling frequency is ≥1Hz.

[0024] Further, the data preliminary filtering denoising adopts Kalman filtering or Butterworth filtering to denoise the collected data in real time.

[0025] Further, the temperature compensation algorithm corrects the displacement data in real time according to the environmental temperature data.

[0026] ΔL corr = ΔL raw × [1-α×(T-25)]

[0027] Wherein, α = 0.0001 / ℃.

[0028] Further, the multi-sensor data dynamic fusion algorithm fuses the data, and the algorithm comprises:

[0029] Step 1: Calculate the vibration intensity index of the structure.

[0030] Based on the IMU sensor measured three-axis acceleration vector module length calculation;

[0031]

[0032] In the formula, a x ,a y ,a z is the three-axis acceleration component measured by IMU (unit: m / s 2 ), and 20.0 is the maximum safe acceleration value of the structure selected by experience;

[0033] Step 2: Initial weight dynamic self-adaptive adjustment;

[0034] According to the above vibration index, the initial weight of the displacement meter and the IMU is calculated; wherein:

[0035] The displacement meter weight w disp = 0.8×(1-vibration_index)};

[0036] The IMU weight w imu = 0.2+0.6×vibrate i ndex

[0037] Step 3: Real-time error monitoring and dynamic correction

[0038] Real-time monitoring of the difference between the displacement meter data and its historical average value (U disp ), when the real-time data deviation is greater than 0.1mm, it is considered that the displacement meter measurement accuracy is reduced, and the weight is dynamically adjusted.

[0039] If |Unisp Usisp|>0.1mm, then:

[0040] The updated displacement meter weight is w disp,new = w disp × 0.7

[0041] The updated IMU weight is w imu,nsw = w imu × 1.3

[0042] Step 4: Ambient temperature compensation

[0043] According to the current temperature T (unit: ℃), the displacement meter weight temperature compensation is α = 0.0001 / ℃

[0044] w disp,temp = w disp,new × [1-0.0001×(T-25)]

[0045] w disp,temp = w disp,new × [1-0.0001×(T-25)]

[0046] Step 5: Weight normalization

[0047] The final weight is normalized to ensure that the sum of the weights is 1

[0048]

[0049] Step 6: Multi-sensor data fusion output

[0050] The final three-axis translation displacement is

[0051]

[0052] The final rotation angle calculation is

[0053] θ xyz = ∫(w imu,final ·ω gyro +w disp,final ·θ disp )dt.

[0054] Further, in the data fusion and output

[0055] Displacement data fusion

[0056] Rotation angle data fusion θ xyz = ∫(w gyro ·ω gyro +w angle ·θ disp )dt.

[0057] Further, the edge end data compression and uploading uses MQTT or a lightweight communication protocol to upload the processed data to a cloud platform in real time through 4G / 5G.

[0058] Further, in the real-time simulation and digital twin self-evolution module, an initial structure finite element model is established, and initial structure parameters (material, boundary condition, load, etc.) are input; the twin model is refreshed once every 5-15 minutes according to the real-time data uploaded by the sensor, and the specific process is as follows:

[0059] Step 1, damage sensitivity factor (DSF) calculation; calculate the spatial curvature change of the structure according to the real-time data;

[0060]

[0061] Step 2, establish a parameter inversion objective function; define the objective function as the error square sum between the damage sensitivity factor DSF obtained by model simulation calculation simulated and the measured damage sensitivity factor DSF measured ;

[0062]

[0063] DSF simulated ,i(E) represents the damage sensitivity factor at the i-th position based on the finite element model with the current elastic modulus E;

[0064] Step 3, genetic algorithm inversion;

[0065] The genetic algorithm optimizes the objective function to seek the optimal elastic modulus Enew, and the specific steps are as follows:

[0066] Initialize the population: randomly generate an initial population within the possible range of elastic modulus;

[0067] Fitness evaluation: based on the objective function Loss(E), calculate the fitness (i.e. error size) of each individual;

[0068] Selection: according to the fitness size, select individuals with smaller errors to enter the next generation;

[0069] Crossover: randomly select two individuals and exchange their partial elastic modulus genes to generate new individuals;

[0070] Mutation: randomly select the elastic modulus gene of an individual to perform random small perturbation to generate a new genotype;

[0071] The genetic algorithm is iterated until any of the following convergence conditions is met:

[0072] The error is less than a preset threshold or the maximum number of iterations is reached;

[0073] The optimization result is the optimal elastic modulus;

[0074] E new = argmin E (Loss(E)

[0075] Step 4, real-time updating of the model; the elastic modulus E obtained by using the genetic algorithm is used for new updating the stiffness matrix K in the finite element model new = f(E new );

[0076] Step 5, model updating triggering strategy optimization.

[0077] Further, in the life prediction module, the updated twin model parameters are transmitted to the life prediction engine, and the remaining life is calculated based on the Coffin-Manson model

[0078] T remain = C·(Δε) m

[0079] Wherein, C, m are calibrated according to historical tests, and Δε is obtained in real time through the twin model.

[0080] The six-degree-of-freedom safety monitoring system provided by the application further comprises a remote emergency response module; the remote emergency response comprises:

[0081] I-level response: displacement exceeds 50% of the design value, and on-site sound and light alarm;

[0082] II-level response: when the remaining life is less than 30 days, an automatic maintenance plan is pushed;

[0083] III-level response: when the rotation angle of the key node exceeds 0.5°, the relevant departments are notified to start the emergency disposal process.

[0084] The application further provides a safety monitoring method, which comprises the following steps:

[0085] Step 1, the hardware acquisition module (high-precision displacement meter, IMU inertial measurement unit and environmental sensor) is installed to achieve the effect of collecting the displacement, angular velocity and temperature and humidity data of the key nodes of the structure.

[0086] Step 2, through the data synchronous acquisition function, the data of the displacement meter, the IMU sensor and the environmental sensor are acquired in real time, so that the effect of ensuring data synchronization and accurate acquisition is achieved.

[0087] Step 3, the collected data is preliminarily filtered and denoised through Kalman filtering or Butterworth filtering, so that the effects of reducing data noise and improving data accuracy are achieved.

[0088] Step 4: Implement an algorithm for temperature compensation to correct displacement data in real time based on environmental temperature, achieving the effect of eliminating the impact of temperature changes on displacement data.

[0089] Step 5: Use a multi-sensor data dynamic fusion algorithm to calculate the vibration intensity index and dynamically adjust the weights of the displacement meter and IMU sensor, achieving the effects of improving data fusion accuracy and system reliability.

[0090] Step 6: Through real-time error monitoring and dynamic correction, monitor the difference between displacement meter data and historical average, and dynamically adjust sensor weights, achieving the effect of real-time correction of measurement accuracy.

[0091] Step 7: Adjust sensor weights based on current temperature through temperature compensation, achieving the effect of eliminating the impact of temperature changes on sensor measurement accuracy.

[0092] Step 8: Through weight normalization processing, ensure that the sum of sensor weights is 1, achieving the effect of balancing the fusion accuracy of different sensor data.

[0093] Step 9: Output the final three-axis translation displacement and rotation angle through data fusion, achieving the effect of accurately obtaining real-time displacement and attitude information of the structure.

[0094] Step 10: Through edge data compression and upload, use MQTT or lightweight communication protocols to upload data to the cloud platform in real time, achieving the effect of realizing remote real-time monitoring and data storage.

[0095] Step 11: Through real-time simulation and digital twin self-evolution module, update the finite element model of the structure using real-time data, achieving the effect of real-time judgment of structural damage state based on sensor data.

[0096] Step 12: Use genetic algorithm to invert material parameters (such as elastic modulus) and update the stiffness matrix in the finite element model, achieving the effect of optimizing the structure model to accurately assess damage.

[0097] Step 13: Calculate the remaining life using the Coffin-Manson model based on the updated digital twin model to predict the remaining life of the structure, achieving the effect of real-time prediction of the remaining life of the key components of the structure.

[0098] Step 14: Through the remote emergency response module, when the displacement exceeds 50% of the design value, the remaining life is less than 30 days, or the key node rotation angle exceeds 0.5°, start the corresponding emergency response process, achieving the effect of timely starting structure maintenance and emergency disposal

[0099] Compared with the prior art, the beneficial effects of the present application are:

[0100] The application realizes high-precision and real-time monitoring of structural health by combining high-precision sensors and real-time data acquisition technology. The system reduces data noise through Kalman filtering and Butterworth filtering, and can perform temperature compensation according to the environmental temperature to ensure measurement accuracy. The multi-sensor data dynamic fusion algorithm automatically adjusts the sensor weight, improving the accuracy of the data and the stability of the system. In addition, the system uses real-time simulation and digital twin technology to dynamically evaluate the structural damage state by continuously updating the model, and optimizes the material parameters through genetic algorithm to accurately predict the service life of the structure. Combined with LSTM network for time series analysis, the system can predict the trend of structural fatigue damage in advance and provide residual life estimation. The remote emergency response module can start the emergency disposal process in time when an abnormality occurs at a key node to ensure the safety of the structure. Through cloud processing and remote monitoring, the system can realize real-time transmission and storage of data, providing scientific support for structural maintenance decision-making. These technical effects make the application have significant advantages in structural monitoring and management, improving safety, efficiency and the accuracy of decision-making. BRIEF DESCRIPTION OF DRAWINGS

[0101] Figure 1 The structural diagram of the six-degree-of-freedom safety monitoring system proposed by the application is shown in the figure.

[0102] Figure 2 The sensor arrangement diagram of the hardware acquisition module in the six-degree-of-freedom safety monitoring system proposed by the application is shown in the figure. DETAILED DESCRIPTION

[0103] In the following, the application will be further described in conjunction with the drawings and specific embodiments. It should be noted that the following described embodiments or technical features can be combined in any way to form new embodiments without conflict.

[0104] Embodiment 1:

[0105] Reference Figure 1 The six-degree-of-freedom safety monitoring system proposed by the application aims to monitor, evaluate and predict the life of the structure in real time. The system combines hardware acquisition, edge computing, data fusion, cloud processing, real-time simulation and digital twin technology to comprehensively detect the safety of the structure by comprehensively applying sensor data, real-time simulation models and life prediction algorithms, and timely respond to and predict the potential fatigue damage and residual life of the structure.

[0106] In general, the safety detection system mainly includes the following modules:

[0107] Hardware acquisition module; it includes high-precision displacement meter, IMU inertial measurement unit installed at the same point as the high-precision displacement meter, and environmental sensor for monitoring temperature and humidity;

[0108] An edge computing and data fusion module adopts an embedded data acquisition terminal to collect and fuse data collected by a hardware acquisition module.

[0109] A cloud processing module uploads monitoring data to a cloud server in real time through a 4G / 5G wireless network to form a real-time data stream, ensuring remote access and real-time monitoring of data. The specific functions include: 1. Data upload: using MQTT or other lightweight communication protocols, real-time upload of processed data to the cloud to ensure timely transmission and efficient storage of data; 2. Data storage and processing: the cloud server stores and further analyzes the real-time data stream to ensure long-term tracking and management of data.

[0110] A real-time simulation and digital twin self-evolution module judges the structural damage state in real time by constructing a real-time updated finite element or data-driven model, and dynamically inverts material parameters (such as elastic modulus E) using a genetic algorithm to realize real-time evolution of the digital twin.

[0111] A structural damage and life prediction module applies an LSTM network for time series analysis based on the updated digital twin model and historical data to predict the structural fatigue damage trend and accurately calculates the remaining life of key structural components using the Coffin-Manson model. This module evaluates the damage state of the structure in real time through a real-time updated finite element model or data-driven model.

[0112] In this scheme, the main task of the edge computing and data fusion module is to collect, preprocess and fuse the data of the hardware acquisition module to provide accurate real-time structural state information. Data collection and preprocessing in the edge computing and data fusion module includes synchronous data collection; preliminary filtering and denoising of data; temperature compensation algorithm; multi-sensor data dynamic fusion algorithm; data fusion and output; edge data compression and upload.

[0113] In detail, synchronous data collection synchronously collects displacement meter, IMU sensor and temperature and humidity data through ADC, SPI and I2C interfaces to ensure real-time and consistency of data collection.

[0114] In detail, data preprocessing uses Kalman filtering or Butterworth filtering to denoise the collected data, eliminating noise interference in the measurement process and ensuring the effectiveness and accuracy of the data.

[0115] In detail, the temperature compensation algorithm corrects displacement data by real-time collection of environmental temperature data, eliminating the influence of temperature changes on displacement measurement and improving measurement accuracy.

[0116] It is worth mentioning that multi-sensor data dynamic fusion includes:

[0117] Vibration intensity index calculation: Based on the three-axis acceleration vector module measured by the IMU sensor, the vibration intensity is calculated.

[0118] Dynamic self-adaptive adjustment of initial weights: According to the vibration intensity index, the initial weights of the displacement meter and the IMU are calculated and dynamically adjusted to adapt to real-time environmental changes.

[0119] Real-time error monitoring and dynamic correction: Real-time monitoring of the deviation between the displacement meter data and the historical mean value, when the deviation is greater than 0.1mm, automatically adjust the weights of the displacement meter and the IMU to maintain the measurement accuracy.

[0120] Weight normalization: The maximum weight is normalized to ensure consistent weight proportion of different sensors in data fusion.

[0121] In detail, the multi-sensor data fusion output outputs three-axis translation displacement and rotation angle according to the maximum weight, providing real-time state information of the structure.

[0122] In this scheme, in detail, in the data synchronization collection:

[0123] High-precision displacement meter is placed at the key nodes of the structure, with a range of ±200mm and an accuracy of 0.01mm. The angular velocity measurement range of the IMU inertial measurement unit is ±300° / s; the temperature accuracy is ±0.5℃, and the humidity accuracy is ±3%. The displacement meter is used to measure the displacement of the key nodes of the structure in X, Y, Z three directions, which directly reflects the translational freedom of the structure.

[0124] ADC collects displacement meter data with a sampling frequency of ≥100Hz;

[0125] IMU sensor collects through SPI interface with a sampling frequency of ≥200Hz; it is installed with the displacement meter at the same point and is used to measure the angular velocity of the structure. The angular velocity measurement range of the IMU is ±300° / s, with high dynamic response capability, sampling frequency ≥200Hz, ensuring high-precision collection of angle change data. IMU contains accelerometer, gyroscope and magnetometer, which are used to measure the acceleration, angular velocity and direction of the structure. This is the core of realizing the rotational freedom monitoring. Among them: three-axis accelerometer: IMU measures the acceleration in X, Y, Z three directions. These data can be used to calculate the translation displacement and vibration intensity, but they play an indirect role in the realization of rotational freedom. Three-axis gyroscope: gyroscope is used to measure the angular velocity (i.e. rotation rate) around X, Y, Z axis, which is crucial for capturing the rotational freedom (angle change). By integrating these data, the rotation angle change of the object around each axis can be obtained, thus effectively capturing the rotational freedom of the structure.

[0126] Reference can be made to Figure 2, determine the monitoring point of the structure to be tested, fix the IMU sensor directly at the monitoring point, usually through high-strength adhesive or bolt fixation, and accurately align the displacement meter probe (such as a laser displacement meter or an LVDT measuring rod) with the monitoring point, while the support of the displacement meter is installed on an independent stable base (such as the ground or a support frame not affected by vibration, to ensure that it measures the absolute displacement of the monitoring point relative to the stationary reference system.

[0127] In order to accurately reflect the six degrees of freedom, the hardware acquisition module synchronously acquires the displacement meter and IMU data, and combines environmental data such as temperature and humidity for fusion. In this way, the hardware can provide complete dynamic behavior information about the structure in three-dimensional space, including translational displacement and rotational angle.

[0128] The environmental temperature data I2C interface is collected, and the sampling frequency is ≥1Hz. It is used to monitor the temperature and humidity changes of the structure environment in real time, and the temperature accuracy is ±0.5℃, the humidity accuracy is ±3%, and the data can be corrected according to different environmental conditions.

[0129] In the present scheme, the data preliminary filtering and denoising adopts Kalman filtering or Butterworth filtering to denoise the collected data in real time.

[0130] Further, the temperature compensation algorithm is used to correct the displacement data in real time according to the environmental temperature data.

[0131] ΔL corr = ΔL raw × [1-α×(T-25)]

[0132] Wherein, α = 0.0001 / ℃.

[0133] In the present scheme, the multi-sensor data dynamic fusion algorithm is used to fuse the data, and the algorithm includes:

[0134] Step 1: Calculate the vibration intensity index of the structure;

[0135] Based on the three-axis acceleration vector module length measured by the IMU sensor;

[0136]

[0137] In the formula, a x ,a y ,a z are the three-axis acceleration components measured by the IMU (unit: m / s 2 ), and 20.0 is the maximum safe acceleration value of the structure selected by experience;

[0138] Step 2: Initial weight dynamic self-adaptive adjustment;

[0139] According to the above vibration index, the initial weights of the displacement meter and the IMU are calculated; wherein:

[0140] Displacement meter weight w disp = 0.8 x (1-vibration_index)};

[0141] IMU weight w imu = 0.2+0.6 x vibration_index i

[0142] Step 3: Real-time error monitoring and dynamic correction

[0143] Real-time monitoring of the difference between the displacement meter data and its historical average (U disp ), when the real-time data deviation is greater than 0.1 mm, it is considered that the displacement meter measurement accuracy is reduced, and the weight is dynamically adjusted;

[0144] If |Unisp-Uisp|>0.1mm, then:

[0145] The updated displacement meter weight is w disp,new = w disp x 0.7

[0146] The updated IMU weight is w imu,nsw = w imu x 1.3

[0147] Step 4: Environmental temperature compensation;

[0148] According to the current temperature T (unit: ℃), the displacement meter weight temperature compensation α=0.0001 / ℃ is carried out

[0149] w disp,temp = w disp,new x [1-0.0001 x (T-25)]

[0150] w disp,temp = w disp,new x [1-0.0001 x (T-25)]

[0151] Step 5: Weight normalization;

[0152] The final weight is normalized to ensure that the sum of the weights is 1

[0153]

[0154] Step 6: Multi-sensor data fusion output;

[0155] The final three-axis translational displacement is

[0156] ​

[0157] The final rotation angle is calculated as

[0158] θ xyz =∫(w imu,final ·ω gyro +w disp,final ·θ disp )dt

[0159] In the present solution, the data fusion and output in

[0160] Displacement data fusion

[0161] Rotation angle data fusion θ xyz =∫(w gyro ·ω gyro +w angle ·θ disp )dt.

[0162] In detail, the edge-end data compression and uploading uses MQTT or a lightweight communication protocol to upload the processed data to a cloud platform in real time through 4G / 5G.

[0163] It should be noted that in the real-time simulation and digital twin self-evolution module, a finite element model of the initial structure is initially established, and initial structural parameters (material, boundary condition, load, etc.) are input; according to the real-time data uploaded by the sensor, the twin model is refreshed every 5 to 15 minutes, and the specific process is as follows:

[0164] Step 1, damage sensitive factor (DSF) calculation; calculate the spatial curvature change of the structure according to the real-time data;

[0165]

[0166] Step 2, establish a parameter inversion objective function; define the objective function as the sum of the squares of the errors between the damage sensitive factor DSF simulated obtained by model simulation calculation and the measured damage sensitive factor DSF measured ;

[0167]

[0168] Where DSF simulated ,i(E) represents the damage sensitive factor at the i-th position based on the finite element model simulated with the current elastic modulus E;

[0169] Step 3, genetic algorithm inversion;

[0170] The genetic algorithm optimizes the objective function to seek the optimal elastic modulus Enew, and the specific steps are as follows:

[0171] Initialize the population: Randomly generate an initial population within the possible range of elastic modulus;

[0172] Fitness assessment: Based on the objective function Loss(E), calculate the fitness (i.e., error magnitude) of each individual;

[0173] Selection: Based on fitness, individuals with smaller errors are selected to enter the next generation;

[0174] Crossover: Randomly select two individuals, exchange some of their elastic modulus genes, and generate new individuals;

[0175] Mutation: Randomly select an individual's elastic modulus gene and subject it to random, small perturbations to generate a new genotype;

[0176] The genetic algorithm terminates when it iterates until any of the following convergence conditions are met;

[0177] The error is less than the preset threshold or the maximum number of iterations is reached;

[0178] The optimization result is the optimal elastic modulus;

[0179] E new =argmin E (Loss(E)

[0180] Step 4: Real-time model update; using the elastic modulus E obtained by the genetic algorithm. new Update the stiffness matrix K in the finite element model new =f(E new );

[0181] Step 5: Optimize the model update trigger strategy.

[0182] In this scheme, the lifetime prediction module passes the updated twin model parameters to the lifetime prediction engine, and calculates the remaining lifetime based on the Coffin-Manson model.

[0183] T remain =C·(Δε) m

[0184] Among them, C and m are calibrated based on historical experiments, and Δε is obtained in real time through a twin model.

[0185] In this solution, the safety monitoring system also includes a remote emergency response module, which includes...

[0186] Level I response: Displacement exceeds the design value by 50%, triggering an on-site audible and visual alarm;

[0187] Level II Response: When the remaining service life is less than 30 days, an emergency maintenance plan will be automatically pushed out.

[0188] Level III response: When the key node corner exceeds 0.5°, notify the relevant departments to start the emergency disposal process.

[0189] Example 2:

[0190] This scheme is based on the safety detection system of Example 1, and proposes a safety detection method, which includes the following steps:

[0191] Step 1: Collect displacement, angular velocity and temperature and humidity data of the key nodes of the structure by installing hardware acquisition modules (high-precision displacement meters, IMU inertial measurement units and environmental sensors).

[0192] Step 2: Real-time acquisition of displacement meter, IMU sensor and environmental sensor data through data synchronization acquisition function, to ensure the synchronization and accuracy of different sensor data.

[0193] Step 3: Preliminary filtering and denoising of collected data through Kalman filtering or Butterworth filtering, to eliminate noise in sensor signals and ensure data quality.

[0194] Step 4: Real-time correction of displacement data according to environmental temperature through temperature compensation algorithm, to avoid errors caused by temperature changes in displacement measurement.

[0195] Step 5: Calculation of vibration intensity index through multi-sensor data dynamic fusion algorithm, and dynamic adjustment of displacement meter and IMU sensor weights according to vibration intensity, to improve data fusion accuracy and system reliability.

[0196] Step 6: Real-time error monitoring and dynamic correction to monitor the difference between displacement meter data and historical average, and dynamically adjust sensor weights if the data deviation is large, to ensure the accuracy of data measurement.

[0197] Step 7: Adjustment of sensor weights according to current temperature through temperature compensation, to further eliminate the impact of temperature fluctuations on sensor measurement accuracy.

[0198] Step 8: Weight normalization processing to ensure that the sum of sensor weights is 1, to ensure the balance of each sensor in the data fusion process.

[0199] Step 9: Data fusion output of the final three-axis translation displacement and rotation angle, to provide accurate real-time displacement and attitude data for subsequent structure state analysis.

[0200] Step 10: Edge data compression and upload using MQTT or lightweight communication protocol to upload processed data to the cloud platform in real time, to realize remote monitoring and storage of data.

[0201] Step 11, update the finite element model of the structure with real-time sensor data through real-time simulation and digital twin self-evolution module, providing dynamic support for structure health state judgment.

[0202] Step 12, update the stiffness matrix in the finite element model by inverting material parameters (such as elastic modulus) through genetic algorithm, ensuring that the model accurately reflects the actual state of the structure.

[0203] Step 13, based on the updated digital twin model, calculate the remaining life by using the Coffin-Manson model to predict the structure's life and assess the fatigue damage trend and remaining service life of the structure.

[0204] Step 14, through the remote emergency response module, when the displacement exceeds 50% of the design value, the remaining life is less than 30 days, or the key node rotation exceeds 0.5°, the system automatically starts the corresponding emergency response process to ensure timely maintenance or repair measures.

[0205] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system, device and module described above can refer to the corresponding process in the foregoing method embodiments, which will not be repeated here.

[0206] In several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the modules is only a logical function division, and actual implementation can have another division manner, for example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed ones can be indirect coupling or communication connection through some interfaces, devices or modules, which can be electrical, mechanical or other forms.

[0207] The modules described as separate components can or can not be physically separated, and the components displayed as modules can or can not be physical modules, which can be located in one place or distributed on multiple network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the embodiment scheme.

[0208] In addition, the functional modules in each embodiment of the present application can be integrated in one processing module, or each module can exist physically, or two or more modules can be integrated in one module.

[0209] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the parts that contribute to the prior art or parts of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0210] Finally: the above only for the preferred embodiments of the present application, and not for limiting the present application, any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A six degrees of freedom safety monitoring system, characterized by, Comprise: Hardware acquisition module; It includes high-precision displacement meter, IMU inertial measurement unit installed at the same point with the high-precision displacement meter, and environmental sensor for monitoring temperature and humidity; Edge computing and data fusion module, which adopts embedded data acquisition terminal to collect and fuse the data collected by the hardware acquisition module; Cloud processing module, which uploads the monitoring data to the cloud server in real time through 4G / 5G wireless network to form real-time data stream; Real-time simulation and digital twin self-evolution module, which judges the structural damage state in real time by constructing a real-time updated finite element or data-driven model, and realizes the real-time evolution of digital twin body by dynamically inverting material parameters (such as elastic modulus E) using genetic algorithm; Structural damage and life prediction module, which applies LSTM network for time series analysis based on the updated digital twin model and historical data to predict the structural fatigue damage trend, and accurately calculates the residual life of the key components of the structure using Coffin-Manson model.

2. A six degree of freedom safety monitoring system as claimed in claim 1, wherein, Data acquisition and preprocessing in the edge computing and data fusion module includes: Data synchronous acquisition; data preliminary filtering and denoising; temperature compensation algorithm; multi-sensor data dynamic fusion algorithm; data fusion and output; edge data compression and upload.

3. A six degree of freedom safety monitoring system as claimed in claim 1, wherein, In the data synchronous acquisition The displacement meter is arranged at the key nodes of the structure, with a range of ±200mm and an accuracy of 0.01mm; the angular velocity measurement range of the IMU inertial measurement unit is ±300° / s; the temperature accuracy is ±0.5℃, and the humidity accuracy is ±3%; The displacement meter data is collected by ADC with a sampling frequency of ≥100Hz; The IMU sensor is collected through SPI interface with a sampling frequency of ≥200Hz; The environmental temperature data is collected through I2C interface with a sampling frequency of ≥1Hz.

4. A six degree of freedom safety monitoring system as claimed in claim 1, wherein, The collected data is denoised in real time by Kalman filter or Butterworth filter.

5. A six degree of freedom safety monitoring system as claimed in claim 1, wherein, The displacement data is corrected in real time according to the environmental temperature data by the temperature compensation algorithm; ΔL corr = ΔL raw × [1 - α × (T - 25)] Wherein, α = 0.0001 / ℃.

6. A six degree of freedom safety monitoring system as claimed in claim 1, wherein, The multi-sensor data dynamic fusion algorithm fuses the data, and the algorithm includes: Step 1: Calculate the vibration intensity index of the structure; Calculate the three-axis acceleration vector module based on the IMU sensor measurement; where a x ,a y ,a z are the three-axis acceleration components measured by the IMU, and 20.0 is an empirically selected maximum safe acceleration value for the structure. Step 2: Dynamic self-adaptive adjustment of initial weight Calculate the initial weight of the displacement meter and the IMU according to the above vibration index; wherein: displacement meter weight w disp = 0.8 x (1 - vibration_index) IMU weight w imu = 0.2 + 0.6 x vibrate i ndex Step 3: Real-time error monitoring and dynamic correction Real-time monitoring displacement meter data and its historical mean (U disp ) difference, when the real-time data deviation is greater than 0.1mm, it is considered that the displacement meter measurement accuracy is reduced, and the weight is dynamically adjusted; If |Unisp Usisp|>0.1mm, then: The updated displacement meter weight is w disp,new = w disp x 0.7 The updated IMU weight is w imu,nsw = w imu x 1.3 Step 4: Environmental temperature compensation; According to the current temperature T, the displacement meter weight temperature compensation α = 0.0001 / ℃ is carried out w disp,temp = w disp,new x [1 - 0.0001 x (T - 25)] w disp,temp = w disp,new x [1 - 0.0001 x (T - 25)] Step 5: Weight normalization; The final weight is normalized to ensure that the sum of the weights is 1 Step 6: Multi-sensor data fusion output; The final three-axis translation displacement is The final rotation angle is calculated as θ xyz = ∫(w imu,final · ω gyro + w disp,final · θ disp ) dt.

7. A six degree of freedom safety monitoring system as claimed in claim 1, wherein, In the data fusion and output Displacement data fusion Rotational angle data fusion θ xyz = ∫(w gyro · ω gyro + w angle · θ disp )dt.

8. A six degree of freedom safety monitoring system as claimed in claim 1, wherein, The edge data compression and upload uses MQTT or lightweight communication protocol to upload the processed data to the cloud platform in real time through 4G / 5G.

9. A six degree of freedom safety monitoring system as claimed in claim 1, wherein, The real-time simulation and digital twin self-evolution module, initially establishes a structure finite element model, and inputs initial structure parameters; according to real-time uploading data of the sensor, the twin model is refreshed once every 5-15 minutes, and the specific process is as follows; Step 1, damage sensitive factor DSF calculation; according to real-time data, the spatial curvature change of the structure is calculated; Step 2, Establish parameter inversion objective function; define objective function as error square sum between model simulation calculation obtained damage sensitive factor DSF and measured damage sensitive factor DSF simulated and measured damage sensitive factor DSF measured . where DSF simulated，i (E) represents the damage sensitivity factor at the i-th location simulated based on the finite element model with the current elastic modulus E; Step 3, genetic algorithm inversion; The genetic algorithm optimizes the objective function, seeking the optimal elastic modulus E new The specific steps are as follows: Initialize population: randomly generate initial population within the possible range of elastic modulus; Fitness evaluation: based on the objective function Loss(E), the fitness (i.e. error size) of each individual is calculated; Selection: according to the fitness size, select the individual with smaller error to enter the next generation; Crossover: randomly select two individuals, exchange their part of elastic modulus genes to generate new individuals; Mutation: randomly select the elastic modulus gene of the individual to generate new genotype by random small perturbation; The genetic algorithm is iterated until any one of the following convergence conditions is met; Error is less than the preset threshold or reaches the set maximum iteration number; The optimization result is the optimal elastic modulus; E new = arg min E (Loss(E) Step 4, model real-time updating; using the elastic modulus E obtained by genetic algorithm new Updating the stiffness matrix K in the finite element model new = f(E new ); Step 5, model update trigger strategy optimization.

10. The six degree of freedom safety monitoring system of claim 1, wherein, In the life prediction module, the updated twin model parameters are transmitted to the life prediction engine, and the remaining life is calculated based on the Coffin-Manson model T remain = C • (Δε) m Wherein, C, m are calibrated according to historical test, and Δε is obtained in real time through the twin model.

Images