State of Charge (SoC) estimation is an indispensable feature in Battery Management Systems (BMS) and is an important function to be implemented for Electric Vehicles (EVs), Renewable Energy Storage (RES) and Portable electronics. For the nonlinearity of lithium-ion Batteries, the current model-based ones generally suffer from uncertain parameters and lose accuracy under various operating conditions. This paper discusses four traditional machine learning algorithms: Extreme Gradient Boosting (XGBoost), Random Forest (RF), Extra Trees (ET), Support Vector Regression (SVR) and explores them in the context of a common evaluation methodology with regard to SoC estimation. In order to make a fair comparison between the machine learning algorithms, a set of common methodology, which involves preprocessing, feature engineering, splitting into train and test set, and hyperparameters tuning are evaluated in all these machine learning techniques. Various performance parameters have been taken into account for assessing the performance of the actual prediction and to decide the extent to which the selected models are appropriate to be utilized in actual BMS applications.
Sudharshana, Subramanya Bhat, Bhushith M K· 2026 International Conferenc...· 0 citations
Industrial machinery operating in manufacturing and process environments is frequently subjected to adverse operating conditions such as excessive temperature rise, abnormal current consumption, and mechanical vibrations, which may lead to performance degradation, unexpected failures, production losses, and safety hazards. To address these challenges, this paper presents the design and implementation of an Internet of Things (IoT)-enabled real-time machine health monitoring and protection system based on the ESP32 microcontroller platform. The proposed system integrates a DHT11 sensor for temperature and humidity monitoring, an ACS712 Hall-effect sensor for current measurement, and an MPU9250 inertial measurement unit (IMU) for vibration analysis. Sensor data are continuously acquired, processed, and transmitted through Wi-Fi to a cloud-based Firebase Realtime Database, enabling remote access and centralized monitoring. A responsive web dashboard hosted on GitHub Pages provides real-time visualization of machine operating parameters, status indicators, and fault notifications. To enhance operational safety and equipment reliability, threshold-based fault detection algorithms are implemented to identify abnormal operating conditions. When predefined critical limits are exceeded, the ESP32 automatically initiates protective actions by disconnecting the machine through a relay module, activating a visual alarm, and updating the fault status on the cloud platform. The dashboard additionally supports bidirectional communication, allowing authorized operators to remotely restart the machine, while a local push-button interface enables manual system recovery. Furthermore, the developed platform incorporates a browser-based logging mechanism that records timestamped sensor measurements, machine status transitions, fault events, and downloadable CSV trend data for maintenance analysis and performance evaluation. Experimental validation demonstrates reliable real-time monitoring with a data refresh interval of approximately 3 s, accurate threshold-based fault detection, dependable cloud connectivity, and effective remote supervisory control. The proposed solution offers a low-cost, scalable, and practical framework for predictive maintenance and industrial equipment condition monitoring in smart manufacturing environments.
Sudharshana, Kratika V Ulman, Kishan K Kulal et al.· 2026 International Conferenc...· 0 citations