Monitoring system and method for monitoring rubber shock insulation support

Abstract

The invention relates to the field of monitoring of rubber shock insulation supports, in particular to a monitoring system and method for monitoring a rubber shock insulation support. The system comprises an integrated sensor, a data acquisition module, a communication module, a data processing and analysis module, a user interface and an alarm and notification module. Monitoring data of the rubber shock insulation support are collected through the integrated sensor, the abnormal mode, the potential fault and the service life of the rubber shock insulation support are predicted through the data processing and analyzing module according to the data, and then related personnel are notified through the alarm and notification module. The monitoring system for monitoring the state of the rubber shock insulation support in real time and in all directions is provided, meanwhile, long-term operation of the integrated sensor is maintained by using vibration energy or solar energy conversion energy in the integrated sensor, and the practicability of the monitoring system is improved; the model optimization unit is introduced, so that continuous optimization can be performed according to real-time data and performance feedback, high efficiency and accuracy of the model are kept, the system adapts to data and environment changes, and long-term monitoring performance is ensured.

Description

Technical Field

[0001] The present invention relates to the field of rubber seismic isolation bearing monitoring, and in particular to a rubber seismic isolation bearing monitoring system and method. Background Art

[0002] Seismic isolation technology is an effective means of shock absorption. By installing seismic isolation bearings between buildings or bridges and foundations, the transmission of seismic forces is reduced, thereby protecting the structure from serious damage. Rubber seismic isolation bearings have good elasticity and durability. Therefore, rubber seismic isolation bearings are usually commonly used materials in seismic isolation technology. Therefore, monitoring of rubber seismic isolation bearings is very important.

[0003] Traditional monitoring methods for rubber isolation bearings usually rely on regular physical inspections and simple strain or displacement measurements. Regular inspections and monitoring take a long time and provide limited measurement information. They cannot achieve real-time monitoring of rubber isolation bearings and make it difficult to fully assess their health status. They also consume a lot of human resources and are restricted by environmental conditions during the inspection process.

[0004] Therefore, there is an urgent need to develop a monitoring system and method for monitoring rubber isolation bearings to overcome the shortcomings of the existing technology. Summary of the Invention

[0005] In order to solve the problem that traditional monitoring methods cannot monitor the status of seismic isolation bearings in real time and comprehensively, the present invention provides a monitoring system for rubber seismic isolation bearings, including an integrated sensor, a data acquisition module, a communication module, a data processing and analysis module, a user interface, and an alarm and notification module;

[0006] The integrated sensor is used to collect key indicators of the rubber isolation bearing, including displacement, pressure, temperature, and strain;

[0007] The data acquisition module integrates edge computing technology to perform preliminary processing on the sensor data and adjust the sensor monitoring frequency according to the currently monitored environment and the state of the rubber isolation bearing;

[0008] The communication module supports wired and wireless communication technologies to transmit the monitoring data preliminarily processed by the data acquisition module to ensure reliable data transmission;

[0009] The data processing and analysis module is used to identify abnormal patterns in the data and predict potential failures and lifespans of the rubber isolation bearings;

[0010] The user interface is used to intuitively display the real-time data and health status of the rubber isolation bearing;

[0011] The alarm and notification module is used to automatically trigger an alarm based on a preset threshold or pattern recognition result, and notify relevant personnel through multiple channels.

[0012] Furthermore, the integrated sensor converts energy through vibration energy or solar energy to ensure long-term sustainable operation; at the same time, the integrated sensor has a fault diagnosis function, which can detect the health status of its own energy collection and storage system. When it detects that the performance of the energy collection or storage module has declined, it automatically performs fault diagnosis and notifies maintenance personnel to perform maintenance.

[0013] Furthermore, the data acquisition module adjusts the sensor monitoring frequency according to the currently monitored environment and the state of the rubber isolation bearing, including: the data acquisition module adjusts the energy usage strategy of the integrated sensor according to the energy consumption of the integrated sensor and the collected energy dynamics, to ensure the continuous operation of key monitoring tasks under limited energy supply.

[0014] Furthermore, the data processing and analysis module identifies abnormal patterns in the data in the following specific steps:

[0015] S4.1 Using data fusion technology, fuse different types of data from the rubber isolation bearing to construct a comprehensive data model to fully understand the state of the rubber isolation bearing;

[0016] S4.2 performing feature extraction on the data to extract key features that can accurately reflect its status;

[0017] S4.3 uses an internally deployed machine learning algorithm for anomaly monitoring and deep learning technology to predict potential failures and lifespan of rubber isolation bearings;

[0018] S4.4 sends the real-time data, health status and early warning analysis of the rubber isolation bearing to the user interface and the alarm and notification module, the user interface displays the data, and the alarm and notification module issues specific alarms and notifications.

[0019] Furthermore, the comprehensive data model is a multi-level and multi-dimensional data representation, which facilitates further analysis and processing by integrating and converting various types of data from different sensors.

[0020] Furthermore, the monitoring of abnormal patterns uses a support vector machine algorithm. The support vector machine model is trained using historical data, the weights of different sensor data are continuously adjusted, and the performance of the support vector machine model is evaluated on an independent test data set. The trained support vector machine model is deployed to the data processing and analysis module to monitor and analyze data in real time. The prediction of potential failures and lifespan of rubber isolation bearings using deep learning technology is the same as the monitoring process of abnormal patterns. The calculation formula for the training of the support vector machine model is:

[0021]

[0022] Where F(x) is the decision function in the support vector machine model, which is used to monitor the abnormal mode of the rubber isolation bearing and predict the state of the bearing based on the data collected from the rubber isolation bearing sensor; sgn() is the sign function used to convert the continuous output of the decision function F(x) into discrete categories, that is, normal mode or abnormal mode; α i Lagrange multipliers, used to help identify which historical data points are most critical for predicting the performance of rubber isolation bearings; i is the class label used to mark the normal or abnormal status of each data point; is the RBF kernel function, which is used to determine the attenuation degree of similarity between data points. -γ determines the sensitivity of the model to different states. i is the monitoring data of the rubber isolation bearing, x is the data of the new sample point to be predicted; b is the bias term, which is used to help the support vector machine model distinguish between normal and abnormal states more effectively.

[0023] Furthermore, the data processing and analysis module is provided with a model optimization unit, which updates the model according to real-time data, automatically evaluates the model performance, and adjusts the learning strategy according to performance feedback to achieve self-optimization of the model. The specific implementation steps are:

[0024] S7.1 obtains new monitoring data after preprocessing and inputs it into an incremental learning algorithm, wherein the incremental learning algorithm causes the support vector machine model to gradually receive the new monitoring data and adjust the support vector machine model parameters according to the new monitoring data;

[0025] S7.2 monitor the performance indicators of the model in real time, including accuracy and recall, to evaluate the model's adaptability to new data, and dynamically adjust the model parameters or structure based on performance feedback;

[0026] S7.3 regularly evaluate the performance of the model and set thresholds for key performance indicators based on business requirements and previous performance data of the model. When the performance indicator falls below the threshold, the model optimization unit automatically triggers the following optimization process;

[0027] S7.4 evaluates the current learning efficiency of the model, i.e., the rate and magnitude of improvement of the model on the new monitoring data, and dynamically adjusts the learning rate using a predetermined strategy based on changes in model performance. The learning rate adjustment result is then fed back to S7.3 to ensure the effectiveness of the learning rate adjustment strategy and to make adjustments as needed.

[0028] S7.5 uses the incremental learning algorithm to make subtle adjustments to the model parameters based on the performance monitoring results of S7.3 to improve the accuracy and generalization ability of the model. When the parameter adjustment cannot achieve a significant improvement in the model performance, the model structure is adjusted.

[0029] Furthermore, the user interface displays real-time data, health status and warning information in the form of graphics and charts, and provides explanatory analysis of model predictions, thereby increasing the transparency of the system and user trust; at the same time, users can query and analyze historical monitoring data of rubber isolation bearings through the user interface.

[0030] Furthermore, the alarm and notification module monitors the data obtained from the data processing and analysis by setting thresholds. When the set thresholds are exceeded, an alarm is issued, and key information is ensured to be conveyed to relevant personnel in a timely and accurate manner through multiple channels.

[0031] Furthermore, a method for monitoring rubber isolation bearings specifically includes the following steps:

[0032] S1 deploys the integrated sensor on the rubber isolation support and starts the integrated sensor to collect key indicator data;

[0033] S2: The data acquisition module collects the key indicator data, performs preliminary processing on the data, analyzes the data and obtains analysis results, and adjusts the sensor data acquisition frequency according to the analysis results;

[0034] S3: The data transmission module transmits the data pre-processed by the data acquisition module to the data processing and analysis module;

[0035] S4, the data processing and analysis module identifies abnormal patterns in the data through an internally deployed machine learning model, and predicts potential failures and lifespans of the rubber isolation bearings;

[0036] S5: when the alarm and notification module identifies an abnormal pattern or a potential fault in the data, it activates an alarm mode and sends a notification to the staff;

[0037] S6 The user interface displays the analysis results of the data processing and analysis module.

[0038] Beneficial effects:

[0039] The present invention adds an energy self-sufficiency function to the integrated sensor of the rubber isolation bearing, so that the integrated sensor can continue to operate without an external power supply, especially in remote or difficult-to-reach areas, greatly improving the reliability and autonomy of the application, ensuring continuous monitoring of the rubber isolation bearing, a key structure, and reducing maintenance costs and complexity.

[0040] The data processing and analysis module in the present invention combines data fusion technology and machine learning algorithms, enabling the application to accurately identify abnormal patterns of rubber seismic isolation bearings and predict their potential failures, greatly advancing the failure warning time and providing the possibility for taking preventive maintenance measures, thereby avoiding or reducing potential losses; at the same time, by introducing a model optimization unit, the system can be continuously optimized according to real-time data and performance feedback, maintaining the efficiency and accuracy of the model, adapting to changes in data and environment, and ensuring long-term monitoring performance.

[0041] This application uses a user interface to systematically and intuitively display data and interpretative analysis, which not only enables non-professional users to easily understand complex monitoring data and prediction results, but also increases the transparency of the system and user trust; at the same time, the alarm and notification module notifies relevant personnel of key information in a timely and multi-channel manner, making the response in emergency situations faster and more effective. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 This is a system structure diagram of the present invention.

[0043] Figure 2 Flow chart of the method of the present invention.

[0044] Figure 3 This is a specific flow chart of the data processing and analysis module of the present invention for identifying abnormal patterns in data.

[0045] Figure 4 This is a flow chart for implementing the model optimization unit of the present invention. DETAILED DESCRIPTION

[0046] The present invention will be further described below with reference to the accompanying drawings and examples.

[0047] like Figure 1 As shown, a monitoring system for rubber isolation bearings specifically includes an integrated sensor, a data acquisition module, a communication module, a data processing and analysis module, a user interface, an alarm and notification module

[0048] Integrated sensors are used to collect key indicators of rubber isolation bearings, including displacement, pressure, temperature, strain, etc.

[0049] The integrated sensor converts energy through vibration or solar energy to ensure long-term sustainable operation. At the same time, the integrated sensor has a fault diagnosis function and can monitor its own energy collection efficiency and storage battery status in real time. When it detects performance degradation or failure of the energy collection or storage module, it automatically performs fault diagnosis and automatically sends fault reports and maintenance requests to the monitoring center or the maintenance personnel's mobile device through the communication module to ensure timely response and maintenance, thereby reducing system downtime.

[0050] The data acquisition module integrates edge computing technology to perform preliminary processing on sensor data and adjust the sensor monitoring frequency based on the current monitored environment and the status of the rubber isolation bearings;

[0051] The data acquisition module has built-in edge computing capabilities, which enable real-time processing of raw data collected by integrated sensors. This includes data cleaning, preliminary analysis, and compression to reduce the amount of data transmitted and improve response speed. The data acquisition module adjusts the sensor monitoring frequency based on the currently monitored environment and the status of the rubber isolation bearings. This includes: The data acquisition module adjusts the energy usage strategy of the integrated sensors based on their energy consumption and collected energy dynamics. When energy is sufficient, the frequency and accuracy of data collection are increased. When energy is limited, critical monitoring tasks are prioritized, such as reducing the frequency of non-critical data collection to ensure the continuous operation of critical monitoring tasks when energy supply is limited.

[0052] Communication module supports wired and wireless communication technologies to ensure reliable data transmission.

[0053] The communication module supports wired communication technology or wireless communication technology to transmit the data pre-processed by the data acquisition module, and adopts TLS / SSL encryption algorithm to protect the security of data transmission, prevent data from being intercepted or tampered, ensure the integrity and confidentiality of monitoring data, and can also dynamically adjust data transmission parameters such as retry interval and data compression level according to current network conditions such as signal strength and congestion conditions to optimize transmission efficiency and success rate.

[0054] The data processing and analysis module is used to identify abnormal patterns in the data and predict the potential failure of the rubber isolation bearing and the life of the rubber isolation bearing.

[0055] The specific steps of the data processing and analysis module to identify abnormal patterns in the data are as follows: Figure 3 As shown:

[0056] S4.1 uses data fusion technology to fuse different types of data from rubber isolation bearings and construct a comprehensive data model to fully understand the status of rubber isolation bearings.

[0057] The comprehensive data model is a multi-level, multi-dimensional data representation. By integrating and converting various types of data from different sensors, the multi-dimensional data from different sensors are merged into a comprehensive feature vector to comprehensively reflect the status of the rubber isolation bearing, facilitating further analysis and processing to more accurately identify abnormal patterns or make status predictions.

[0058] S4.2 performs feature extraction on the data to extract key features that can accurately reflect the status of the rubber isolation bearing.

[0059] Specifically, in order to accurately reflect the health status and performance of rubber isolation bearings, data science and machine learning techniques are used to extract key features from the comprehensive data model, including statistical features in time series (mean, variance, kurtosis, skewness, etc.), frequency domain features (such as frequency distribution obtained by Fourier transform), and model-based features.

[0060] S4.3 uses an internally deployed machine learning algorithm for anomaly detection and deep learning technology to predict the potential failure and life of rubber isolation bearings.

[0061] The support vector machine algorithm is used to monitor abnormal modes of rubber isolation bearings. The support vector machine model is trained using historical data, and the weights of different sensor data are continuously adjusted. The performance of the support vector machine model is evaluated on an independent test data set. The trained support vector machine model is deployed to the data processing and analysis module to monitor and analyze data in real time. The use of deep learning technology to predict the potential failure and life of rubber isolation bearings is the same as the monitoring process of abnormal modes. The calculation formula for the training of the support vector machine model is:

[0062]

[0063] Among them, F(x) is the decision function in the support vector machine model, which is used to monitor the abnormal mode of the rubber isolation bearing. The state of the bearing is predicted based on the data collected from the rubber isolation bearing sensor, including displacement, pressure, temperature and strain data. Based on the model learned from the training data, the new monitoring data is classified into normal or abnormal state; sgn() is a sign function, which is used to convert the continuous output of the decision function F(x) into discrete categories, that is, normal mode or abnormal mode. In this way, according to the sensor data, each monitoring point in this application can be clearly marked as being in normal operation or abnormal, thereby triggering a corresponding alarm or maintenance process; α i is a Lagrange multiplier that helps identify which historical data points are most critical for predicting rubber isolation bearings, thereby optimizing the model's decision boundary to more accurately classify new observations; iis a class label used to mark the normal or abnormal status of each data point, i.e., each sensor data collected from the rubber isolation bearing; is the RBF kernel function, which is used to determine the attenuation degree of similarity between data points. In this application, the RBF kernel enables the support vector machine model to process nonlinearly separable data. By adjusting the value of γ, the sensitivity and accuracy of the support vector machine model for detecting abnormal states of rubber isolation bearings are optimized; -γ determines the sensitivity of the support vector machine model to different states (normal or abnormal); x i is the monitoring data of the rubber isolation bearing, which represents the sample points in the training set data. These sample points include the monitoring data of the rubber isolation bearing; x is the data of the new sample points to be predicted, which also include the monitoring data of the rubber isolation bearing; b is the bias term, which is used to help the support vector machine model distinguish between normal and abnormal states more effectively.

[0064] In order to improve the model's generalization ability in the face of unseen data, enable the model to continuously learn from new data to better understand the complex relationships and changing trends behind the data, so as to maintain a high prediction accuracy when facing new or unknown data, a model optimization unit is set in the data processing and analysis module. The model optimization unit can update the model according to real-time data, automatically evaluate model performance, and adjust the learning strategy according to performance feedback to achieve self-optimization of the model.

[0065] The specific implementation steps are as follows Figure 4 As shown:

[0066] S7.1 obtains new monitoring data after preprocessing and inputs the new monitoring data into the incremental learning algorithm. The incremental learning algorithm first determines whether the new monitoring data has a significant impact on the decision boundary of the model. If so, the incremental learning algorithm uses the new monitoring data as support vectors and updates the parameters of the model, such as adjusting the weights of the support vectors.

[0067] S7.2 Monitor model performance indicators in real time, including key performance indicators such as accuracy and recall, to assess the model's adaptability to new data. These indicators should be displayed through the user interface for easy monitoring. Based on performance feedback, if a decline in model performance is detected, dynamically adjust model parameters or structure.

[0068] S7.3: Set a reasonable evaluation cycle, such as weekly or monthly. At the end of the cycle, conduct a comprehensive performance evaluation of the model. Based on business needs and the model's previous performance data, set thresholds for key performance indicators such as accuracy and recall. Use the thresholds as the benchmark for triggering the model optimization process. When the performance indicator falls below the threshold, the model optimization unit automatically triggers the following optimization processes, namely S7.4 and S7.5.

[0069] S7.4 evaluates the model's current learning efficiency, that is, the speed and magnitude of the model's improvement on the new monitoring data. Based on the model's performance changes, dynamically adjust the learning rate using a predetermined strategy. If the model's performance improves rapidly on the new data, reduce the learning rate appropriately to avoid overfitting. If the model's performance improves slowly on the new data, increase the learning rate to accelerate learning progress. The learning rate adjustment results are then fed back into the regular performance evaluation in S7.3 to ensure the effectiveness of the learning rate adjustment strategy and make necessary adjustments based on the model's actual performance.

[0070] S7.5 uses the performance monitoring results of S7.3 to enable the incremental learning algorithm to make subtle adjustments to the model parameters, including fine-tuning the parameters of the support vectors in the model or adjusting the parameters of the kernel function to improve the accuracy and generalization ability of the model; when parameter adjustment cannot achieve a significant improvement in model performance, adjust the model structure, including increasing or decreasing the number of support vectors in the model, modifying the kernel function in the model, or introducing new eigenvalues.

[0071] The real-time performance monitoring implemented in step S7.2 and the periodic performance evaluation in step S7.3 have different purposes and functions. The focus of step S7.2 is on continuously monitoring the performance of the model, aiming to promptly capture any fluctuations or declines in model performance so that immediate action can be taken when problems are discovered; while the focus of the periodic performance evaluation in step S7.3 is on analyzing the root causes of performance degradation and observing long-term trends. Periodic performance evaluation helps to identify potential problems or long-term trends that are easily overlooked in real-time performance monitoring. By setting a performance threshold in step S7.3, when the model performance falls below the threshold, the optimization process is automatically triggered to ensure that the model maintains an acceptable performance level in the long term.

[0072] By setting up a model optimization unit in the data processing and analysis module, the deployed model can process and learn from new monitoring data in real time without the need to retrain the entire model, thereby maintaining the accuracy and timeliness of the model's prediction capabilities. At the same time, the model automatically adjusts model parameters and structure based on performance feedback to improve future prediction performance through self-optimization, reducing the need for manual intervention and enabling the model to continuously improve its efficiency in a dynamic environment.

[0073] S4.4 sends the real-time data, health status and early warning analysis of the rubber isolation bearing to the user interface and the alarm and notification module. The user interface displays the data, and the alarm and notification module issues specific alarms and notifications.

[0074] User interface for visually displaying real-time data and health status of rubber isolation bearings.

[0075] The user interface displays real-time data, health status and warning information in the form of graphics and charts, and provides explanatory analysis of model predictions, increasing system transparency and user trust; based on the output of the data processing and analysis module, the user interface displays the health status score or grade of the rubber isolation bearing, as well as warnings of any abnormal patterns; users can query and analyze historical monitoring data of rubber isolation bearings through the user interface, including through time series analysis, trend lines and other functions to evaluate long-term performance changes; at the same time, users can obtain key monitoring indicators of rubber isolation bearings in real time through the dashboard, such as displacement, pressure, temperature and strain.

[0076] The alarm and notification module is used to automatically trigger alarms based on preset thresholds or pattern recognition results, and notify relevant personnel through multiple channels.

[0077] The alarm and notification module sets thresholds, monitors the data obtained through data processing and analysis, and issues an alarm when the set threshold is exceeded. It also ensures that key information can be conveyed to relevant personnel in a timely and accurate manner through multiple channels, including but not limited to SMS, email, and mobile APP notifications.

[0078] like Figure 2 As shown, a monitoring method for rubber isolation bearings includes:

[0079] S1 deploys integrated sensors on rubber isolation supports and starts collecting key indicator data;

[0080] The S2 data acquisition module collects key indicator data and performs preliminary data processing. It analyzes the data and obtains analysis results, and adjusts the sensor data acquisition frequency according to the analysis results.

[0081] S3 The data transmission module transmits the data pre-processed by the data acquisition module to the data processing and analysis module;

[0082] The S4 data processing and analysis module uses an internally deployed machine learning model to identify abnormal patterns in the data and predict potential failures and lifespan of rubber isolation bearings;

[0083] The S5 alarm and notification module activates the alarm mode and sends a notification to the staff when it identifies abnormal patterns or potential faults in the data;

[0084] The S6 user interface displays the analysis results of the data processing and analysis module.

[0085] Specifically, in step S2, the data is analyzed and the analysis results are obtained, and the sensor data acquisition frequency is adjusted according to the analysis results. Specifically, the data acquisition module adjusts the energy usage strategy of the integrated sensor according to the energy consumption of the integrated sensor and the collected energy dynamics. When energy is sufficient, the frequency and accuracy of data acquisition are increased. When energy is limited, the operation of key monitoring tasks is limited, such as reducing the acquisition frequency of non-critical data, to ensure that the continuous operation of key monitoring tasks is prioritized when energy supply is limited.

[0086] Specifically, in S2, the data acquisition module uses built-in edge computing capabilities to instantly process the raw data collected by the integrated sensors, including data cleaning, preliminary analysis and compression. Although the data acquisition module has performed preliminary processing on the data, the data may be affected by environmental interference during the collection and transmission process, resulting in abnormal values ​​or missing values ​​in the monitoring data. Therefore, in S4, the data processing and analysis module still needs to further preprocess the received data, such as filling missing values, handling abnormal values, etc., to ensure the quality and consistency of the data.

[0087] The above-described embodiments merely represent preferred embodiments of the present invention. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art would be able to make various modifications, improvements, and substitutions without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims.

Claims

1. A monitoring system for rubber seismic isolation bearings, characterized in that: It includes integrated sensors, data acquisition modules, communication modules, data processing and analysis modules, user interface, alarm and notification modules; The integrated sensor is used to collect key indicators of the rubber isolation bearing, including displacement, pressure, temperature, and strain; The data acquisition module integrates edge computing technology to perform preliminary processing on the sensor data and adjust the sensor monitoring frequency according to the currently monitored environment and the status of the rubber isolation bearing; the communication module supports wired and wireless communication technologies to transmit the monitoring data preliminarily processed by the data acquisition module to ensure reliable data transmission; The data processing and analysis module is used to identify abnormal patterns in the data and predict potential failures and lifespans of the rubber isolation bearings; The user interface is used to intuitively display the real-time data and health status of the rubber isolation bearing; The alarm and notification module is used to automatically trigger an alarm based on a preset threshold or pattern recognition result, and notify relevant personnel through multiple channels.

2. The rubber isolation bearing monitoring system according to claim 1 is characterized in that: The integrated sensor converts energy through vibration energy or solar energy to ensure long-term sustainable operation; at the same time, the integrated sensor has a fault diagnosis function and can detect the health status of its own energy collection and storage system. When it detects that the performance of the energy collection or storage module has deteriorated, it automatically performs fault diagnosis and notifies maintenance personnel to perform maintenance.

3. The rubber isolation bearing monitoring system according to claim 1, characterized in that: The data acquisition module adjusts the sensor monitoring frequency according to the currently monitored environment and the status of the rubber isolation bearing, including: the data acquisition module adjusts the energy usage strategy of the integrated sensor according to the energy consumption of the integrated sensor and the collected energy dynamics, ensuring the continuous operation of key monitoring tasks under limited energy supply.

4. The rubber isolation bearing monitoring system according to claim 1, characterized in that: The specific steps of the data processing and analysis module to identify abnormal patterns in data are: S4.1 Using data fusion technology, fuse different types of data from the rubber isolation bearing to construct a comprehensive data model to fully understand the state of the rubber isolation bearing; S4.2 performing feature extraction on the data to extract key features that can accurately reflect its status; S4.3 uses an internally deployed machine learning algorithm for anomaly monitoring and deep learning technology to predict potential failures and lifespan of rubber isolation bearings; S4.4 sends the real-time data, health status and early warning analysis of the rubber isolation bearing to the user interface and the alarm and notification module, the user interface displays the data, and the alarm and notification module issues specific alarms and notifications.

5. The rubber isolation bearing monitoring system according to claim 4 is characterized in that: The comprehensive data model is a multi-level, multi-dimensional data representation that facilitates further analysis and processing by integrating and converting multiple types of data from different sensors.

6. The rubber isolation bearing monitoring system according to claim 5, characterized in that: The monitoring of abnormal patterns uses a support vector machine algorithm. The support vector machine model is trained using historical data, the weights of different sensor data are continuously adjusted, and the performance of the support vector machine model is evaluated on an independent test data set. The trained support vector machine model is deployed to the data processing and analysis module to monitor and analyze data in real time. The prediction of potential failures and lifespan of rubber isolation bearings using deep learning technology is the same as the monitoring process of abnormal patterns. The calculation formula for training the support vector machine model is: Where F(x) is the decision function in the support vector machine model, which is used to monitor the abnormal mode of the rubber isolation bearing and predict the state of the bearing based on the data collected from the rubber isolation bearing sensor; sgn() is the sign function used to convert the continuous output of the decision function F(x) into discrete categories, that is, normal mode or abnormal mode; α i Lagrange multipliers, used to help identify which historical data points are most critical for predicting the performance of rubber isolation bearings; i is the class label used to mark the normal or abnormal status of each data point; is the RBF kernel function, which is used to determine the attenuation degree of similarity between data points. -γ determines the sensitivity of the model to different states. i is the monitoring data of the rubber isolation bearing, x is the data of the new sample point to be predicted; b is the bias term, which is used to help the support vector machine model distinguish between normal and abnormal states more effectively.

7. The rubber isolation bearing monitoring system according to claim 6, characterized in that: The data processing and analysis module is provided with a model optimization unit, which updates the model according to real-time data, automatically evaluates the model performance, and adjusts the learning strategy according to performance feedback to achieve self-optimization of the model. The specific implementation steps are as follows: S7.1 obtains new monitoring data after preprocessing and inputs it into an incremental learning algorithm, wherein the incremental learning algorithm causes the support vector machine model to gradually receive the new monitoring data and adjust the support vector machine model parameters according to the new monitoring data; S7.2 monitor the performance indicators of the model in real time, including accuracy and recall, to evaluate the model's adaptability to new data, and dynamically adjust the model parameters or structure based on performance feedback; S7.3 regularly evaluate the performance of the model and set thresholds for key performance indicators based on business requirements and previous performance data of the model. When the performance indicator falls below the threshold, the model optimization unit automatically triggers the following optimization process; S7.4 evaluates the current learning efficiency of the model, i.e., the rate and magnitude of improvement of the model on the new monitoring data, and dynamically adjusts the learning rate using a predetermined strategy based on changes in model performance. The learning rate adjustment result is then fed back to S7.3 to ensure the effectiveness of the learning rate adjustment strategy and to make adjustments as needed. S7.5 uses the incremental learning algorithm to make subtle adjustments to the model parameters based on the performance monitoring results of S7.3 to improve the accuracy and generalization ability of the model. When the parameter adjustment cannot achieve a significant improvement in the model performance, the model structure is adjusted.

8. The rubber isolation bearing monitoring system according to claim 1, characterized in that: The user interface displays real-time data, health status and warning information in the form of graphics and charts, and provides explanatory analysis of model predictions, thereby increasing the transparency of the system and user trust; at the same time, users can query and analyze historical monitoring data of rubber isolation bearings through the user interface.

9. The rubber isolation bearing monitoring system according to claim 1, characterized in that: The alarm and notification module monitors the data obtained from the data processing and analysis by setting thresholds. When the set thresholds are exceeded, an alarm is issued and key information is ensured to be conveyed to relevant personnel in a timely and accurate manner through multiple channels.

10. A method for monitoring a rubber isolation bearing monitoring system according to any one of claims 1 to 9, characterized in that: The specific steps include: S1 deploys the integrated sensor on the rubber isolation support and starts the integrated sensor to collect key indicator data; S2: The data acquisition module collects the key indicator data, performs preliminary processing on the data, analyzes the data and obtains analysis results, and adjusts the sensor data acquisition frequency according to the analysis results; S3: The data transmission module transmits the data pre-processed by the data acquisition module to the data processing and analysis module; S4, the data processing and analysis module identifies abnormal patterns in the data through an internally deployed machine learning model, and predicts potential failures and lifespans of the rubber isolation bearings; S5: when the alarm and notification module identifies an abnormal pattern or a potential fault in the data, it activates an alarm mode and sends a notification to the staff; S6 The user interface displays the analysis results of the data processing and analysis module.

Images