Optimal Off-line Solution Research Articles

Assessment of Energy Management in a Fuel Cell/Battery Hybrid Vehicle

The energy management strategy plays a major role in hybrid platforms powered by fuel cells (FCs) and batteries. This paper presents an assessment of energy management focused on fuel economy and battery degradation. Particularly, a proposed heuristic strategy and the widely known equivalent consumption minimization strategy are compared with the optimal solution obtained offline via dynamic programming. The case study is based on a real FC hybrid vehicle. Accordingly, the powertrain model of the vehicle used for the simulations is validated experimentally, and the profile of the power demand is measured from the real application. The results show that the proposed strategy offers the same performance as the equivalent consumption minimization strategy when the battery degradation is prioritized, and in comparison with the optimal off-line solution, it can be seen that there is still margin for improvement in terms of battery degradation.

Design and robustness analysis of autonomous vehicles in intersections

Abstract The paper proposes the design of a neural-network-based control strategy of autonomous vehicles in intersections. The motivation of the neural network approach is to reduce the numerically-intensive computation of the optimization problem in which the motions of autonomous vehicles are formed. In the method the neural network is trained through a preliminary optimal off-line solution. Moreover, a robustness analysis on the neural network based control strategy is proposed. The focus of the analysis is to consider the impact of the position and speed estimation errors on the motion of vehicles. The design and the analysis are illustrated through CarSim simulation examples.

Energy Optimal Wireless Data Transmission for Wearable Devices: A Compression Approach

Wearable devices are designed to have a small size and be lightweight. Consequently, the battery life is constrained and becomes a crucial limitation. In this paper, we use both data compression and wireless transmission speed control to minimize the energy consumption of wearable devices for data transmission, subject to a deadline constraint. We consider both an off-line setting where future channel gains are known ahead of time and a stochastic setting where channel gains change stochastically according to a Markov process. For the first case, we present an efficient $(1+\epsilon)$ approximation algorithm for an arbitrarily small $\epsilon$ , while in the latter case we give a stochastic algorithm to minimize the total expected energy use. We also conduct experimental studies on the proposed algorithms and show that the stochastic algorithm, despite not knowing future channel gains, closely approximates the performance of the nearly optimal off-line solution with less than 0.1% difference in energy consumption on an average. We also compared the stochastic algorithm with several other practical algorithms and showed that our algorithm achieves significant improvements in the overall energy use.

Real-Time Energy Storage Management for Renewable Integration in Microgrid: An Off-Line Optimization Approach

Microgrid is a key enabling solution to future smart grids by integrating distributed renewable generators and storage systems to efficiently serve the local demand. However, due to the random and intermittent characteristics of renewable energy, new challenges arise for the reliable operation of microgrids. To address this issue, we study in this paper the real-time energy management for a single microgrid system that constitutes a renewable generation system, an energy storage system, and an aggregated load. We model the renewable energy offset by the load over time, termed net energy profile, to be practically predictable, but with finite errors that can be arbitrarily distributed. We aim to minimize the total energy cost (modeled as sum of time-varying strictly convex functions) of the conventional energy drawn from the main grid over a finite horizon by jointly optimizing the energy charged/discharged to/from the storage system over time subject to practical load and storage constraints. To solve this problem in real time, we propose a new off-line optimization approach to devise the online algorithm. In this approach, we first assume that the net energy profile is perfectly predicted or known ahead of time, under which we derive the optimal off-line energy scheduling solution in closed-form. Next, inspired by the optimal off-line solution, we propose a sliding-window based online algorithm for real-time energy management under the practical setup of noisy predicted net energy profile with arbitrary errors. Finally, we conduct simulations based on the real wind generation data of the Ireland power system to evaluate the performance of our proposed algorithm, as compared with other heuristically designed algorithms, as well as the conventional dynamic programming based solution.

On-line ranking of split graphs

Graphs and Algorithms A vertex ranking of a graph G is an assignment of positive integers (colors) to the vertices of G such that each path connecting two vertices of the same color contains a vertex of a higher color. Our main goal is to find a vertex ranking using as few colors as possible. Considering on-line algorithms for vertex ranking of split graphs, we prove that the worst case ratio of the number of colors used by any on-line ranking algorithm and the number of colors used in an optimal off-line solution may be arbitrarily large. This negative result motivates us to investigate semi on-line algorithms, where a split graph is presented on-line but its clique number is given in advance. We prove that there does not exist a (2-ɛ)-competitive semi on-line algorithm of this type. Finally, a 2-competitive semi on-line algorithm is given.

The distributed wireless gathering problem

We address the problem of data gathering in a wireless network using multi-hop communication; our main goal is the analysis of simple algorithms suitable for implementation in realistic scenarios. We study the performance of distributed algorithms, which do not use any form of local coordination, and we focus on the objective of minimizing average flow times of data packets. We prove a lower bound of Ω ( n ) on the expected competitive ratio of any acknowledgment-based distributed algorithm minimizing the maximum flow time, where n is the number of nodes of the network. Next, we consider a distributed algorithm which sends packets over shortest paths, and we use resource augmentation to analyze its performance when the objective is to minimize the average flow time. If interferences are modeled as in Bar-Yehuda et al. [R. Bar-Yehuda, O. Goldreich, A. Itai, On the time complexity of broadcast in multi-hop radio networks: an exponential gap between determinism and randomization, Journal of Computer and Systems Sciences 45 (1) (1992) 104–126] we prove that the algorithm is ( 1 + ϵ ) -competitive, when the algorithm sends packets a factor O ( log ( δ / ϵ ) log Δ ) faster than the optimal off-line solution; here δ is the radius of the network and Δ the maximum degree. We finally extend this result to a more complex interference model.

COMPETITIVE COMPLEXITY OF MOBILE ROBOT ON-LINE MOTION PLANNING PROBLEMS

This paper classifies common mobile robot on-line motion planning problems according to their competitive complexity. The competitiveness of an on-line algorithm measures its worst case performance relative to the optimal off-line solution to the problem. Competitiveness usually means constant relative performance. This paper generalizes competitiveness to any functional relationship between on-line performance and optimal off-line solution. The constants in the functional relationship must be scalable and may depend only upon on-line information. Given an on-line task, its competitive complexity class is a pair of lower and upper bounds on the competitive performance of all on-line algorithms for the task, such that the two bounds satisfy the same functional relationship. The paper classifies the following on-line motion planning problems into competitive classes: area coverage, navigation to a target, and on-line search for an optimal path. In particular, it is shown that navigation to a target whose position is either apriori known or recognized upon arrival belongs to a quadratic competitive complexity class. The hardest on-line problem involves navigation in unknown variable traversibility environments. Under certain restriction on traversibility, this last problem belongs to an exponential competitive complexity class.

On-line tracing supervisory control of discrete-event dynamic systems based on outlooking

The conventional off-line approach to supervisor synthesis for discrete-event dynamic systems contains critical drawbacks in building overall automata models and calculating the complete control policy for these models. To resolve the complexity of this off-line approach, the supervisory control scheme based on limited lookahead control policies was proposed by Chung et al. and improved by Kumar et al. recently. However, these methods do not guarantee the same closed-loop behavior as an optimal off-line solution. Therefore, we propose an on-line tracing supervisory control scheme founded on component discrete-event models, which ensures the same closed-loop language as an optimal off-line scheme.