Search

This paper presents a reinforcement learning framework for solving battery scheduling problems in order to extend the lifetime of batteries used in electrical vehicles (EVs), cellular phones, and embedded systems. Battery pack lifetime has often been the limiting factor in many of today&rsquo, s smart systems, from mobile devices and wireless sensor networks to EVs. Smart charge-discharge scheduling of battery packs is essential to obtain super linear gain of overall system lifetime, due to the recovery effect and nonlinearity in the battery characteristics. Additionally, smart scheduling has also been shown to be beneficial for optimizing the system&rsquo, s thermal profile and minimizing chances of irreversible battery damage. The recent rapidly-growing community and development infrastructure have added deep reinforcement learning (DRL) to the available tools for designing battery management systems. Through leveraging the representation powers of deep neural networks and the flexibility and versatility of reinforcement learning, DRL offers a powerful solution to both roofline analysis and real-world deployment on complicated use cases. This work presents a DRL-based battery scheduling framework to solve battery scheduling problems, with high flexibility to fit various battery models and application scenarios. Through the discussion of this framework, comparisons have also been made between conventional heuristics-based methods and DRL. The experiments demonstrate that DRL-based scheduling framework achieves battery lifetime comparable to the best weighted-k round-robin (kRR) heuristic scheduling algorithm. In the meantime, the framework offers much greater flexibility in accommodating a wide range of battery models and use cases, including thermal control and imbalanced battery.

Energy consumption is a major issue in Wireless Sensor Networks (WSNs), as nodes are powered by chemical batteries with an upper bounded lifetime. Estimating the lifetime of batteries is a difficult task, as it depends on several factors, such as operating temperatures and discharge rates. Analytical battery models can be used for estimating both the battery lifetime and the voltage behavior over time. Still, available models usually do not consider the impact of operating temperatures on the battery behavior. The target of this work is to extend the widely-used Kinetic Battery Model (KiBaM) to include the effect of temperature on the battery behavior. The proposed Temperature-Dependent KiBaM (T-KiBaM) is able to handle operating temperatures, providing better estimates for the battery lifetime and voltage behavior. The performed experimental validation shows that T-KiBaM achieves an average accuracy error smaller than 0.33%, when estimating the lifetime of Ni-MH batteries for different temperature conditions. In addition, T-KiBaM significantly improves the original KiBaM voltage model. The proposed model can be easily adapted to handle other battery technologies, enabling the consideration of different WSN deployments.

System lifetime is a major design constraint for battery-powered mobile embedded systems. The increasing gap between the energy demand of portable devices and their battery capacities is further limiting durability of mobile devices. Thus, the guarantees over Quality of Service (QoS) of battery-constrained devices under strict battery capacities are of primary interest for mobile embedded systems’ manufacturers and stakeholders. This paper presents a novel approach for deriving QoS of applications modelled as synchronous dataflow (SDF) graphs. We map these applications on heterogeneous multiprocessor platforms that are partitioned into Voltage and Frequency Islands, together with multiple kinetic battery models (KiBaMs). By modelling the whole system as hybrid automata, and applying model-checking, we evaluate, (1) system lifetime; and (2) minimum required initial battery capacities to achieve the desired application performance. We demonstrate that our approach shows a significant improvement in terms of scalability, as compared to a priced timed automata based KiBaM model. This approach also allows early detection of design errors via model checking.