Early train detection is vital for ensuring the safety of railway personnel, particularly in remote locations where fixed signaling infrastructure is unavailable. Unlike many existing solutions that rely on high-power, high-cost sensors and compute platforms, this work presents a lightweight, low-cost, and portable framework designed to run entirely on resource-constrained microcontrollers with just kilobytes of Random Access Memory (RAM). The proposed system uses vibration data from low-cost accelerometers and employs a simple yet effective Linear Regression (LR) model for almost real-time prediction of train arrival times. To ensure feasibility on low-end hardware, a parallel-processing framework is introduced, enabling continuous data collection, Machine Learning (ML) inference, and wireless communication with strict timing and energy constraints. The decision-making process, including data preprocessing and ML prediction, completes in under 10 ms, and alerts are transmitted via LoRa, enabling kilometer-range communication. Field tests on active railway lines confirm that the system detects approaching trains 15 s in advance with no false negatives and a small number of explainable false positives. Power characterization demonstrates that the system can operate for more than 6 days on a 10 Ah battery, with potential for months of operation using wake-on-vibration modes.
31 October 2024 in Eastern-European Journal of Enterprise Technologies
The object of the study is the conceptual prototype of a microprocessor resource-saving relay protection system. Currently, relay protection ensures electrical networks reliable and efficient work, however, the traditional architecture is proprietary, not allowing to fix and replace damaged parts without a company specialist. Therefore, the open-architecture relay protection is a very pressing issue, but the problem lies in meeting the relay protection requirements. The data transmission protocols nRF and ESP-NOW, Hall sensors evaluation for AC current measurement determination and sensor accuracy improvement was implemented. Experimental validation demonstrated that nRF and ESP-NOW protocols meet the delay and reliability requirements, however, the nRF protocol is more suitable due to its flexibility and obstacle penetration. The data demonstrated that the most effective conditions are without obstacles at 15 meters from the modem and with obstacles 5 meters from the modem. The experiment of Hall sensors characteristics determination demonstrated the accuracy of current measurement with the set values of the opening and closing currents. Nevertheless, it is not accurate (12.45 %) for the relay protection application. Therefore, the application of the changing values of the opening and closing currents is more effective and accuracy reaches 6.92 %. As a result, the service life of the Hall sensor was determined, and even after 10 million openings, the open state time remained unchanged. Therefore, the approximation function for current amplitude determination depending on open state time was found. On the other hand, Hall sensors may suffer from temperature drift and require further optimization to be fully reliable. The study limitation is the current range from 0 to 800 A
30 September 2024 in Bulletin of Toraighyrov University. Energetics series
Координация работы релейных защит является важным аспектом при возникновении серьезных неисправностей и должна обеспечивать быстрое выключение. Одним из решений является использование интернета вещей. Объектом данного исследования является архитектура системы релейной защиты, в которой используется только локальный интернет, а в открытом доступе-глобальный. Целью исследования является оценка эффективности использования интернета в релейной защите, а также актуальности её архитектуры. Эта архитектура передачи, сбора, анализа данных релейной защиты, конкретных протоколов своевременного подключения и безопасности данных является справедливой процессуальной единицей. Качество релейной защиты обеспечивает хорошо организованная структура для точного и эффективного сбора, передачи и анализа данных. Для исследовательских целей были проведены эксперименты, чтобы подтвердить возможность гипотезы. Эксперимент был проведен для нескольких устройств, чтобы определить наиболее эффективные из предложенных. Полученный результат представлен в виде таблицы и содержит значения как для беспрепятственных условий, так и для препятствий в виде стен. Выбран протокол данных ESP-now, который является открытым. Условия также предназначены для определения основных параметров радиоволн, в первую очередь скорости передачи в общее время работы релейной защиты, а также количества пакетов на тысячу с разной памятью. Было обнаружено, что наличие стены в качестве барьера приводит к потере мешков. При выполнении одного из условий исследования интернет вещей может использоваться в релейной защите. Ключевые слова: релейная защита, скорость срабатывания, магнитный датчик, магнитное поле, открытые стандарты, промышленный интернет вещей, эксперимент.
Internet of Things (IoT) devices can execute increasingly complex Machine Learning (ML) tasks at the edge instead of continuously transmitting raw data to the cloud. However, updating vision-based ML models remains challenging in large-scale IoT networks because updates may be large relative to the available wireless capacity, energy budget, and possible duty-cycle constraints. This work evaluates a cross-layer approach for transferring compact model knowledge over constrained IoT radios. First, several partial model update strategies are evaluated for edge image classification. The results show that very small updates, such as 1\% of the weights, provide limited accuracy improvement, while 5\% updates offer a stronger accuracy-communication trade-off. At this update level, weight binning achieves the best accuracy among the evaluated strategies, reaching $\sim$95\% accuracy. Second, three encoding strategies are compared, with bitmap encoding providing the smallest encoded model-update size across the evaluated cases. Finally, a peer-assisted dissemination protocol is introduced, where devices with newer model versions advertise, broadcast encoded updates, and selectively repair missing chunks for neighboring devices. The protocol is evaluated using a SimPy-based simulator. The results show that LoRa 2.4\,GHz SF7 with CAD provides the most consistent dissemination speed among the evaluated radio configurations, at the cost of higher energy consumption. IEEE 802.11ah exhibits on average 29\% lower dissemination speed combined with much lower energy consumption. Compared with literature baselines, the proposed protocol achieves faster dissemination while using lower or comparable average per-ED energy, reaching approximately two-times faster dissemination and about 40-50\% lower average energy usage in most evaluated configurations.