Intelligent monitoring system for predicting service life of UPS battery

Abstract

The invention discloses an intelligent monitoring system for predicting the service life of a UPS battery. A multi-parameter data acquisition module is used for collecting parameters such as voltage, current, temperature, charge-discharge state, internal resistance, self-discharge rate, cycle index and battery health index of the UPS battery in real time; the advanced data processing module is used for preprocessing the collected data, including filtering, denoising, data standardization, abnormal value detection and data fusion; the deep feature extraction module is used for extracting key features related to the service life of the battery from the preprocessed data, wherein the key features comprise a time sequence feature, a frequency domain feature, a statistical feature and a deep learning feature. By collecting various parameters (such as voltage, current and temperature) of the UPS battery in real time, the system can comprehensively analyze the working state of the battery, so that a user can comprehensively predict possible faults and performance degradation, early warning is realized, and the loss caused by battery failure is effectively reduced.

Description

Technical Field

[0001] The invention belongs to the technical field of UPS battery life monitoring, and in particular is an intelligent monitoring system for predicting UPS battery life. Background Art

[0002] In modern power systems, the reliability and performance of uninterruptible power supply (UPS) batteries are crucial to ensuring the normal operation of equipment. As the demand for power continuity for electronic equipment and data centers continues to increase, the management and monitoring of UPS batteries becomes increasingly important.

[0003] In existing related technologies, traditional UPS battery monitoring usually relies on periodic inspections and cannot reflect the battery's working status in real time. This not only reduces the ability to warn of potential failures, but also often causes users to be passive when the battery fails.

[0004] In addition, traditional data collection methods are often interfered by environmental noise and outliers, resulting in poor data quality, which not only affects the reliability of subsequent analysis but also leads to erroneous fault warnings and judgments. Summary of the Invention

[0005] In view of the above situation, in order to overcome the defects of the prior art, the present invention provides an intelligent monitoring system for predicting the life of UPS batteries, so as to at least partially solve the above technical problems.

[0006] The technical solution adopted by the present invention is as follows:

[0007] The present invention proposes an intelligent monitoring system for predicting UPS battery life, comprising:

[0008] Multi-parameter data acquisition module, used to collect UPS battery parameters such as voltage, current, temperature, charge and discharge status, internal resistance, self-discharge rate, cycle number and battery health index in real time;

[0009] Advanced data processing module for pre-processing collected data, including filtering, denoising, data normalization, outlier detection, and data fusion;

[0010] A deep feature extraction module is used to extract key features related to battery life from preprocessed data, including time series features, frequency domain features, statistical features, and deep learning features;

[0011] An adaptive life prediction model for predicting the remaining life of UPS batteries based on extracted features. The model uses a hybrid deep learning algorithm, such as long short-term memory (LSTM), gated recurrent unit (GRU), and convolutional neural network (CNN), and incorporates battery aging mechanisms and environmental factors for model optimization.

[0012] Intelligent user interface module, used to display battery status, prediction results and maintenance suggestions to users, and provide operation interfaces, including real-time monitoring interface, historical data query interface, prediction result display interface and maintenance suggestion interface;

[0013] The data processing module further includes a data storage unit and a data security unit for storing pre-processed data and prediction results, and supporting batch export and import of data while ensuring data security and privacy;

[0014] The lifespan prediction model is based on a machine learning algorithm that can automatically adjust prediction parameters based on historical data, support online updates and offline training of the model, and has adaptive learning capabilities that can continuously update and optimize the prediction model based on new data.

[0015] The user interface module includes an intelligent alarm system that automatically triggers an alarm when the prediction result shows that the battery life is lower than the preset threshold, and supports customizing the alarm threshold, alarm mode and alarm notification content;

[0016] The system also includes a battery health assessment module for comprehensively assessing the health of the battery and providing recommendations for battery maintenance and replacement, while supporting customized assessment cycles and assessment items;

[0017] Adaptive learning module, used to continuously update and optimize the prediction model based on new data to improve prediction accuracy, while supporting custom learning rates and learning cycles;

[0018] The environmental factor analysis module is used to analyze the environmental factors of the UPS battery, such as temperature and humidity, and incorporate them into the life prediction model to improve the accuracy and robustness of the prediction.

[0019] In one embodiment of the present invention, the data acquisition module includes a plurality of high-precision sensors corresponding to different parameters of the UPS battery, and the sensors have a self-calibration function and can automatically calibrate measurement errors.

[0020] In one embodiment of the present invention, the data processing module further includes a data synchronization unit and a data compression unit, which are used to ensure the temporal consistency of data from different sensors and reduce the overhead of data storage and transmission.

[0021] In one embodiment of the present invention, the life prediction model adopts a variety of deep learning algorithms, such as convolutional neural networks (CNN), recurrent neural networks (RNN), generative adversarial networks (GAN) and reinforcement learning algorithms, and combines the battery aging mechanism to optimize the model to improve the accuracy and robustness of the prediction.

[0022] In one embodiment of the present invention, the user interface module includes remote access function and multi-user management function, allowing users to access system status through the Internet, and supporting multi-user simultaneous access and operation, and also has permission management and operation log recording functions.

[0023] In one embodiment of the present invention, the historical data analysis module uses data mining and machine learning technologies to analyze the historical performance data of the battery to optimize the prediction model, and supports customized analysis dimensions, analysis cycles and analysis result presentation methods.

[0024] In one embodiment of the present invention, the adaptive learning module uses online learning and incremental learning techniques to continuously update and optimize the prediction model based on new data, improve prediction accuracy, and support customized learning rate, learning cycle and learning strategy.

[0025] In one embodiment of the present invention, the environmental factor analysis module uses environmental monitoring sensors and machine learning algorithms to monitor the environmental factors of the UPS battery in real time and incorporate them into the life prediction model to improve the accuracy and robustness of the prediction.

[0026] The beneficial effects of the technical solution of the present invention are:

[0027] By collecting various parameters of UPS batteries (such as voltage, current, and temperature) in real time, the present invention enables the system to fully understand the working status of the batteries. This comprehensiveness enables users to predict possible failures and performance degradation, achieve early warning, and effectively reduce losses caused by battery failure. The application of high-precision sensors ensures the accuracy of the data, thereby providing a reliable basis for subsequent data analysis.

[0028] This invention preprocesses collected data through filtering and denoising techniques to improve data quality. The application of outlier detection and data fusion technologies enhances data validity and consistency, ensuring the stability of the lifespan prediction model. Data standardization transforms data from different sources into a unified format for further analysis and model training.

[0029] Through deep feature extraction, the module in this paper can extract important factors related to battery life from complex battery operating data, identify potential influencing factors, and enhance the model's predictive capabilities. In particular, by utilizing time series and frequency domain features, the system can capture the patterns of battery behavior over time, thereby more accurately reflecting the battery's health.

[0030] By utilizing hybrid deep learning algorithms (such as LSTM, GRU, and CNN), the model can accurately and dynamically predict the remaining life of UPS batteries. The algorithm combines the battery aging mechanism to further improve the accuracy and robustness of the model. The adaptive characteristics ensure that the model can automatically adjust with the introduction of new data, continuously optimize the prediction results, and ensure that users obtain the latest and most accurate information.

[0031] This invention utilizes a data storage unit and a data security unit to not only securely store data but also ensure privacy protection. Considering the security of data during transmission, the system takes necessary encryption measures to protect user data from being leaked. Support for batch export and import improves data processing efficiency.

[0032] Additional aspects and advantages of the present invention will be set forth in part in the description which follows and, in part, will be obvious from the description which follows, or may be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0034] Figure 1 This is a schematic diagram of an intelligent monitoring system for predicting UPS battery life proposed in an embodiment of the present invention. DETAILED DESCRIPTION

[0035] The following describes embodiments of the present invention in detail, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and are not to be construed as limiting the present invention.

[0036] An intelligent monitoring system for predicting UPS battery life according to an embodiment of the present invention will be described below with reference to the accompanying drawings.

[0037] like Figure 1 As shown, an embodiment of the present invention provides an intelligent monitoring system for predicting UPS battery life, including: a multi-parameter data acquisition module for collecting parameters such as UPS battery voltage, current, temperature, charge and discharge status, internal resistance, self-discharge rate, cycle number and battery health index in real time;

[0038] Advanced data processing module for pre-processing collected data, including filtering, denoising, data normalization, outlier detection, and data fusion;

[0039] A deep feature extraction module is used to extract key features related to battery life from preprocessed data, including time series features, frequency domain features, statistical features, and deep learning features;

[0040] An adaptive life prediction model is used to predict the remaining life of UPS batteries based on extracted features. The model uses a hybrid deep learning algorithm, such as long short-term memory (LSTM), gated recurrent unit (GRU), and convolutional neural network (CNN), and combines battery aging mechanisms and environmental factors for model optimization;

[0041] Intelligent user interface module, used to display battery status, prediction results and maintenance suggestions to users, and provide operation interfaces, including real-time monitoring interface, historical data query interface, prediction result display interface and maintenance suggestion interface;

[0042] The data processing module further includes a data storage unit and a data security unit, which are used to store pre-processed data and prediction results, and support batch export and import of data while ensuring data security and privacy;

[0043] The lifespan prediction model is based on a machine learning algorithm that can automatically adjust prediction parameters based on historical data. It supports online model updates and offline training, and has adaptive learning capabilities, enabling it to continuously update and optimize the prediction model based on new data.

[0044] The user interface module includes an intelligent alarm system that automatically triggers an alarm when the predicted results show that the battery life is lower than the preset threshold. It also supports customizing the alarm threshold, alarm method and alarm notification content.

[0045] The system also includes a battery health assessment module, which comprehensively evaluates the health of the battery and provides recommendations for battery maintenance and replacement. It also supports customized assessment cycles and assessment items.

[0046] Adaptive learning module, used to continuously update and optimize the prediction model based on new data to improve prediction accuracy, while supporting custom learning rates and learning cycles;

[0047] The environmental factor analysis module is used to analyze the environmental factors of the UPS battery, such as temperature and humidity, and incorporate them into the life prediction model to improve the accuracy and robustness of the prediction.

[0048] In specific applications of the embodiments of the present invention, the collected raw data often contains noise and outliers. The advanced data processing module ensures the accuracy and consistency of the data through filtering, denoising, and data standardization steps. At the same time, the outlier detection mechanism can promptly detect and process anomalies in the data to avoid adverse effects on the prediction results. The data fusion technology further integrates data from different sources to improve the integrity and reliability of the data. Based on the pre-processed data, the deep feature extraction module uses time series analysis, frequency domain analysis, statistical analysis, and deep learning methods to extract key features that are closely related to battery life. It not only reflects the current performance status of the battery, but also reveals the trends and laws of battery aging, providing strong support for subsequent predictions.

[0049] Based on the extracted features, the adaptive life prediction model uses hybrid deep learning algorithms such as LSTM, GRU, and CNN, combined with battery aging mechanisms and environmental factors for model optimization. The hybrid algorithm can fully utilize the advantages of different algorithms to improve the accuracy and robustness of the prediction. At the same time, the model has adaptive learning capabilities and can continuously update and optimize prediction parameters based on new data to ensure the timeliness of the prediction results. The intelligent user interface module provides users with a multi-functional operation interface including real-time monitoring interface, historical data query interface, prediction result display interface, and maintenance recommendation interface. Through these interfaces, users can intuitively understand the battery status, prediction results, and maintenance recommendations. At the same time, the intelligent alarm system can automatically trigger an alarm when the prediction results show that the battery life is below the preset threshold. It supports custom alarm thresholds, alarm methods, and alarm notification content to ensure that users can take timely countermeasures.

[0050] The data storage unit and data security unit together form the core of the data processing module. They not only store preprocessed data and prediction results, but also support batch data export and import, facilitating data backup and analysis. Furthermore, data security mechanisms ensure data privacy and security, preventing data leakage and misuse. The battery health assessment module comprehensively evaluates battery health and provides users with recommendations for battery maintenance and replacement. It supports customizable assessment cycles and items, which can be flexibly adjusted to suit individual user needs and actual circumstances, enabling users to more effectively manage battery resources and extend battery life.

[0051] The adaptive learning module continuously updates and optimizes the prediction model based on new data, improving prediction accuracy. It supports customizable learning rates and learning cycles, allowing for flexible configuration based on actual user needs. Furthermore, the environmental factor analysis module analyzes environmental factors such as temperature and humidity within the UPS battery and incorporates these factors into the lifespan prediction model.

[0052] In one possible implementation, the data acquisition module includes multiple high-precision sensors corresponding to different parameters of the UPS battery. The sensors have a self-calibration function and can automatically calibrate measurement errors. The data processing module also includes a data synchronization unit and a data compression unit to ensure the temporal consistency of data from different sensors and reduce the overhead of data storage and transmission.

[0053] In specific applications, the system employs multiple high-precision sensors, each dedicated to monitoring key UPS battery parameters such as voltage, current, temperature, and internal resistance. Each parameter is accurately and stably measured. The self-calibration function relies on a built-in reference source and algorithm. The sensors periodically compare themselves to the reference source. If deviations are detected, the measurement parameters are automatically adjusted according to a pre-set algorithm, ensuring accurate results.

[0054] Since the system uses multiple sensors to simultaneously monitor different parameters of the UPS battery, the data synchronization unit uses a timestamp mechanism to add an accurate time tag to the data collected by each sensor. When the data processing module receives this data, it will sort and match the data according to the time tag to ensure that the data of each parameter is collected at the same time point. The system can then conduct real-time and comprehensive monitoring of the UPS battery status, providing an accurate data basis for subsequent life prediction.

[0055] The data compression unit utilizes a highly efficient data compression algorithm, significantly reducing data size without losing critical information. This allows the system to retain more useful information at lower storage and transmission costs. Furthermore, compressed data is easier to transmit and process, improving overall system efficiency. A high-precision sensor array collects key UPS battery parameter data. A self-calibration function ensures the accuracy of this data, while a data synchronization unit ensures its temporal consistency, enabling the system to provide real-time, comprehensive monitoring of the UPS battery status. Finally, the data compression unit efficiently compresses this data to reduce storage and transmission overhead.

[0056] In one possible implementation, the life prediction model adopts multiple deep learning algorithms, such as convolutional neural networks (CNN), recurrent neural networks (RNN), generative adversarial networks (GAN) and reinforcement learning algorithms, and combines the battery aging mechanism to optimize the model to improve the accuracy and robustness of the prediction; the user interface module includes remote access and multi-user management functions, allowing users to access the system status through the Internet, and supports simultaneous access and operation by multiple users, and also has permission management and operation log recording functions.

[0057] In specific applications of the present invention, the feature extraction capabilities of CNN are also applicable to feature extraction in battery life prediction. By performing convolution operations on historical battery data (such as voltage, current, and temperature), CNN can automatically learn and extract key features related to battery aging, providing strong support for subsequent predictions. RNN can capture the time dependency in data and accurately model battery aging trends. Through the recurrent layer, RNN can remember historical information, thereby making more accurate predictions of future battery status.

[0058] GANs excel at data generation and can be used to enhance model generalization. In battery life prediction, GANs can generate synthetic data similar to real-world battery data, thereby expanding the training sample set and improving the model's prediction accuracy. GANs can also be used for data augmentation, further enhancing model robustness. Reinforcement learning algorithms simulate battery usage environments, continuously experimenting and adjusting prediction strategies to maximize prediction accuracy. In battery life prediction, reinforcement learning algorithms can be used to optimize prediction models, ensuring they maintain high accuracy across a wide range of conditions.

[0059] In addition to deep learning algorithms, the prediction model also incorporates knowledge of the mechanisms of battery aging. By analyzing the chemical reactions and physical structural changes within the battery, the model can gain a deeper understanding of the nature of battery aging, thereby further optimizing the prediction results. The combination of deep learning algorithms and domain knowledge gives the prediction model both powerful data-driven capabilities and a solid theoretical foundation.

[0060] The user interface module provides remote access, allowing users to access system status anytime, anywhere via the internet. This improves system usability and convenience, enabling users to view real-time battery status and forecast results from anywhere, at any time. The system supports simultaneous multi-user access and operation, meeting the needs of team collaboration and resource sharing. Through multi-user management, different users can have different permissions and roles, allowing them to participate in battery life prediction and management.

[0061] To ensure system security and traceability, the user interface module also provides permission management and operation logging capabilities. With permission management, the system can control user access to specific resources and operational permissions; with operation logging, the system can record all user operations for tracking and auditing in the event of problems.

[0062] When a user accesses the system via the internet, the user interface module verifies their identity and permissions and, based on their request, sends a data request to the prediction model. Upon receiving the data, the prediction model processes and analyzes it using deep learning algorithms and battery aging mechanisms, outputting the predicted battery lifespan and other relevant information. This information is then displayed to the user through the user interface module, and user actions are recorded in an operation log.

[0063] In one possible implementation, the historical data analysis module uses data mining and machine learning techniques to analyze the historical performance data of the battery to optimize the prediction model and supports customized analysis dimensions, analysis cycles, and analysis result presentation methods.

[0064] The adaptive learning module uses online learning and incremental learning technologies to continuously update and optimize the prediction model based on new data, improve prediction accuracy, and support customized learning rates, learning cycles, and learning strategies. The environmental factor analysis module uses environmental monitoring sensors and machine learning algorithms to monitor the environmental factors of the UPS battery in real time and incorporate them into the life prediction model to improve the accuracy and robustness of the prediction.

[0065] In specific applications of the embodiments of the present invention, the historical data analysis module uses data mining technology to deeply explore the hidden laws and patterns in the historical performance data of the battery, such as voltage fluctuations, current changes, and temperature distribution. The module uses machine learning algorithms, such as decision trees, random forests, and support vector machines, to train and learn historical data and build an initial prediction model. The model can capture the trend of battery performance changes over time and provide a basis for subsequent life prediction. In order to meet the needs of different users and business scenarios, the historical data analysis module supports custom analysis dimensions, such as analysis by time, batch, and model dimensions; at the same time, users can set the analysis cycle as needed, whether it is daily, weekly, monthly or annual, and can respond flexibly. In addition, the module also provides a variety of analysis result display methods, such as charts, reports, and dashboards to ensure that users can intuitively and clearly understand the historical performance data and prediction results of the battery.

[0066] The adaptive learning module uses online learning and incremental learning technologies to achieve continuous iteration and optimization of the prediction model. Online learning allows the model to be updated instantly when processing new data, without the need to retrain the entire dataset, thereby improving learning efficiency. Incremental learning builds on this foundation by further utilizing new data to gradually adjust model parameters, adapting them to changes in data distribution, thereby maintaining the accuracy and robustness of predictions. To enhance the adaptability and flexibility of the model, the adaptive learning module supports user-defined learning rates, learning cycles, and learning strategies. The learning rate determines the step size for model parameter updates. Users can adjust the learning rate based on data changes and model performance to achieve better convergence. The learning cycle determines the frequency of model updates, which users can set based on business needs and data generation speed. The learning strategy covers specific methods for model updates, such as gradient descent, stochastic gradient descent, and the Adam optimization algorithm. Users can select and optimize based on actual conditions.

[0067] The environmental factor analysis module collects real-time data on temperature, humidity, and vibration environmental factors through environmental monitoring sensors deployed around the UPS batteries. The module uses machine learning algorithms to analyze and process these environmental factor data, extract key features related to battery life, and incorporate them into the life prediction model. The prediction model not only considers the performance changes of the battery itself, but also comprehensively considers the impact of environmental factors on battery life, thereby improving the accuracy and robustness of the prediction.

[0068] In the intelligent monitoring system, the historical data analysis module, the adaptive learning module and the environmental factor analysis module are not isolated, but are closely connected and work together. The historical data analysis module provides the initial basic data and training samples for the prediction model; the adaptive learning module continuously iterates and optimizes the model parameters based on the new data; the environmental factor analysis module further improves the accuracy and robustness of the prediction by monitoring environmental factors in real time and incorporating them into the prediction model.

[0069] Specifically, when new data is generated, the adaptive learning module first updates the model using online and incremental learning techniques. Simultaneously, the environmental factor analysis module collects real-time environmental factor data from environmental monitoring sensors and uses machine learning algorithms to extract key features. This new data and features are then fed into the historical data analysis module for further analysis and processing to optimize the prediction model. Ultimately, the optimized prediction model can more accurately predict UPS battery life, providing users with reliable decision support.

[0070] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0071] The present invention and its embodiments are described above. This description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if a person skilled in the art is inspired by this and, without departing from the purpose of the present invention, designs structures and embodiments similar to this technical solution without inventiveness, they shall fall within the scope of protection of the present invention.

Claims

1. An intelligent monitoring system for predicting UPS battery life, characterized in that: include: Multi-parameter data acquisition module, used to collect UPS battery parameters such as voltage, current, temperature, charge and discharge status, internal resistance, self-discharge rate, cycle number and battery health index in real time; Advanced data processing module for pre-processing collected data, including filtering, denoising, data normalization, outlier detection, and data fusion; A deep feature extraction module is used to extract key features related to battery life from preprocessed data, including time series features, frequency domain features, statistical features, and deep learning features; An adaptive life prediction model for predicting the remaining life of UPS batteries based on extracted features. The model uses a hybrid deep learning algorithm, such as long short-term memory (LSTM), gated recurrent unit (GRU), and convolutional neural network (CNN), and incorporates battery aging mechanisms and environmental factors for model optimization. Intelligent user interface module, used to display battery status, prediction results and maintenance suggestions to users, and provide operation interfaces, including real-time monitoring interface, historical data query interface, prediction result display interface and maintenance suggestion interface; The data processing module further includes a data storage unit and a data security unit for storing pre-processed data and prediction results, and supporting batch export and import of data while ensuring data security and privacy; The lifespan prediction model is based on a machine learning algorithm that can automatically adjust prediction parameters based on historical data, support online updates and offline training of the model, and has adaptive learning capabilities that can continuously update and optimize the prediction model based on new data. The user interface module includes an intelligent alarm system that automatically triggers an alarm when the prediction result shows that the battery life is lower than the preset threshold, and supports customizing the alarm threshold, alarm mode and alarm notification content; The system also includes a battery health assessment module for comprehensively assessing the health of the battery and providing recommendations for battery maintenance and replacement, while supporting customized assessment cycles and assessment items; Adaptive learning module, used to continuously update and optimize the prediction model based on new data to improve prediction accuracy, while supporting custom learning rates and learning cycles; The environmental factor analysis module is used to analyze the environmental factors of the UPS battery, such as temperature and humidity, and incorporate them into the life prediction model.

2. The intelligent monitoring system for predicting UPS battery life according to claim 1, characterized in that: The data acquisition module includes a plurality of high-precision sensors corresponding to different parameters of the UPS battery. The sensors have a self-calibration function and can automatically calibrate measurement errors.

3. The intelligent monitoring system for predicting UPS battery life according to claim 1, characterized in that: The data processing module further includes a data synchronization unit and a data compression unit, which are used to ensure the temporal consistency of data from different sensors and reduce the overhead of data storage and transmission.

4. The intelligent monitoring system for predicting UPS battery life according to claim 1, characterized in that: The life prediction model adopts multiple deep learning algorithms, such as convolutional neural networks (CNN), recurrent neural networks (RNN), generative adversarial networks (GAN) and reinforcement learning algorithms, and combines the battery aging mechanism to optimize the model.

5. The intelligent monitoring system for predicting UPS battery life according to claim 1, characterized in that: The user interface module includes remote access function and multi-user management function, allowing users to access system status through the Internet, and supports multi-user simultaneous access and operation, and also has authority management and operation log recording functions.

6. The intelligent monitoring system for predicting UPS battery life according to claim 1, characterized in that: The historical data analysis module uses data mining and machine learning technologies to analyze the historical performance data of the battery to optimize the prediction model, and supports customized analysis dimensions, analysis cycles and analysis result presentation methods.

7. The intelligent monitoring system for predicting UPS battery life according to claim 1, characterized in that: The adaptive learning module adopts online learning and incremental learning technology to continuously update and optimize the prediction model according to new data, improve prediction accuracy, and support customized learning rate, learning cycle and learning strategy.

8. The intelligent monitoring system for predicting UPS battery life according to claim 1, characterized in that: The environmental factor analysis module uses environmental monitoring sensors and machine learning algorithms to monitor the environmental factors of the UPS battery in real time and incorporate them into the life prediction model.

Images