TASK Quarterly

Empirical Model of a High-Temperature Proton Exchange Membrane Fuel Cell for Diagnostics Based on the Test Without Active Gases Procedure

Wojciech Rosiński, Andrzej Wilk, Szymon Potrykus — Tue, 10 Feb 2026 00:00:00 +0100

This paper presents a static electrical model developed to analyze results from the Test Without Active Gases (TWAG) procedure, which characterizes fuel cell behavior in the absence of electrochemically active gases. The model topology is inspired by electric double-layer supercapacitor circuits and was derived from first principles using Lagrangian formalism. It was validated using five experimental TWAG discharge curves recorded at temperatures between 40°C and 120°C. Despite its simplicity and low computational cost, the model achieved satisfactory accuracy. The extracted parameters indicate potential for further refinement, such as introducing temperature-dependent components. The approach provides insight into the intrinsic electrochemical properties of high-temperature proton exchange membrane fuel cells in states without active gases and may serve as a foundation for broader diagnostic and modeling applications. Future developments may include extending the RC circuit, incorporating nonlinear elements, or applying the model to other fuel cell technologies. Testing on deliberately degraded cells could also help correlate model parameters with cell health.

Distributed Machine Learning Optimization for Large-Scale Cloud Resource Scheduling: A Hybrid Deep Learning and Evolutionary Algorithm Approach

Mohamed Salem — Tue, 10 Feb 2026 00:00:00 +0100

Cloud computing resource management represents one of the most critical challenges in modern distributed systems, where efficient allocation of computational resources directly impacts system performance, energy consumption, and operational costs. This paper presents a novel hybrid approach combining deep reinforcement learning (DRL) with genetic algorithms (GA) for optimizing cloud resource scheduling in large-scale distributed environments. The proposed framework, termed Distributed Adaptive Learning Resource Scheduler (DALRS), integrates convolutional neural networks (CNN) for workload prediction with deep Q-networks (DQN) for dynamic resource allocation decisions. We evaluate our approach on a comprehensive simulation platform modeling realistic cloud infrastructure with heterogeneous resource configurations. Experimental results demonstrate that DALRS achieves 34.7% improvement in resource utilization, 28.3% reduction in task completion time, and 31.5% decrease in energy consumption compared to state-of-the-art baselines. Furthermore, the hybrid genetic algorithm component provides Pareto-optimal solutions balancing multiple objectives including cost, latency, and throughput. The paper also addresses scalability challenges through a distributed implementation using Apache Spark, enabling efficient processing of workloads exceeding 10,000 concurrent tasks. Our results validate the effectiveness of combining machine learning with evolutionary optimization for managing complex resource allocation problems in cloud infrastructure.

Comparative analysis of methods for calculating HRV values on heart rate monitoring devices

Bartosz Kołakowski; Piotr Noga, Paweł Mańczak, Mateusz Nowak — Tue, 10 Feb 2026 00:00:00 +0100

The aim of the study is to conduct a comparative analysis of HRV calculation methods and how they affect its measures (SDNN, SDNN Index, SDANN, RSA Index, Mean RR, RMSSD, pNN50) and their quality. Modern medical devices offer many possibilities, including heart rate and ECG measurement. An important issue in assessing heart function is HRV and its measures, which allow us to determine how the heart responds to different conditions, but the question of how to measure HRV and how its calculation works in different conditions remains open. The paper will show how HRV changes depending on frequency during rest, activity (walking), and sleep. HRV will be calculated using both an analytical method (using the Pan-Tompkins and U-Net algorithms) and a convolutional neural network.