Ahmed Chiheb Ammari

Deriving a near-optimal power management policy using model-free reinforcement learning and Bayesian classification

To cope with the variations and uncertainties that emanate from hardware and application characteristics, dynamic power management (DPM) frameworks must be able to learn about the system inputs and environment and adjust the power management policy on the fly. In this paper we present an online adaptive DPM technique based on model-free reinforcement learning (RL), which is commonly used to control stochastic dynamical systems. In particular, we employ temporal difference learning for semi-Markov decision process (SMDP) for the model-free RL. In addition a novel workload predictor based on an online Bayes classifier is presented to provide effective estimates of the workload states for the RL algorithm. In this DPM framework, power and latency tradeoffs can be precisely controlled based on a user-defined parameter. Experiments show that amount of average power saving (without any increase in the latency) is up to 16.7% compared to a reference expert-based approach. Alternatively, the per-request latency reduction without any power consumption increase is up to 28.6% compared to the expert-based approach.

Generating Highly Accurate Predictions for Missing QoS Data via Aggregating Nonnegative Latent Factor Models

Automatic Web-service selection is an important research topic in the domain of service computing. During this process, reliable predictions for quality of service (QoS) based on historical service invocations are vital to users. This work aims at making highly accurate predictions for missing QoS data via building an ensemble of nonnegative latent factor (NLF) models. Its motivations are: 1) the fulfillment of nonnegativity constraints can better represent the positive value nature of QoS data, thereby boosting the prediction accuracy and 2) since QoS prediction is a learning task, it is promising to further improve the prediction accuracy with a carefully designed ensemble model. To achieve this, we first implement an NLF model for QoS prediction. This model is then diversified through feature sampling and randomness injection to form a diversified NLF model, based on which an ensemble is built. Comparison results between the proposed ensemble and several widely employed and state-of-the-art QoS predictors on two large, real data sets demonstrate that the former can outperform the latter well in terms of prediction accuracy.

An effective and distributed particle swarm optimization algorithm for flexible job-shop scheduling problem

Flexible job-shop scheduling problem (FJSP) is very important in many research fields such as production management and combinatorial optimization. The FJSP problems cover two difficulties namely machine assignment problem and operation sequencing problem. In this paper, we apply particle swarm optimization (PSO) algorithm to solve this FJSP problem aiming to minimize the maximum completion time criterion. Various benchmark data taken from literature, varying from Partial FJSP and Total FJSP, are tested. Experimental results proved that the developed PSO is enough effective and efficient to solve the FJSP. Our other objective in this paper, is to study the distribution of the PSO-solving method for future implementation on embedded systems that can make decisions in real time according to the state of resources and any unplanned or unforeseen events. For this aim, two multi-agent based approaches are proposed and compared using different benchmark instances.

Centralized Charging Strategy and Scheduling Algorithm for Electric Vehicles Under a Battery Swapping Scenario

Centralized charging of electric vehicles (EVs) based on battery swapping is a promising strategy for their large-scale utilization in power systems. The most outstanding feature of this strategy is that EV batteries can be replaced within a short time and can be charged during off-peak periods or on low electric price and scheduled in any battery swap station. This paper proposes a novel centralized charging strategy of EVs under the battery swapping scenario by considering optimal charging priority and charging location (station or bus node in a power system) based on spot electric price. In this strategy, a population-based heuristic approach is designed to minimize total charging cost, as well as to reduce power loss and voltage deviation of power networks. We introduce a dynamic crossover and adaptive mutation strategy into a hybrid algorithm of particle swarm optimization and genetic algorithm. The resulting algorithm and several others are executed on an IEEE 30-bus test system, and the results suggest that the proposed one is effective and promising for optimal EV centralized charging.

High level H.264/AVC video encoder parallelization for multiprocessor implementation

H.264/AVC (advanced video codec) is a new video coding standard developed by a joint effort of the ITU-TVCEG and ISO/IEC MPEG. This standard provides higher coding efficiency relative to former standards at the expense of higher computational requirements. Implementing the H.264 video encoder for an embedded system-on-chip (SoC) is a big challenge. For an efficient implementation, we motivate the use of multiprocessor platforms for the execution of a parallel model of the encoder. In this paper, we propose a high-level independent target-architecture parallelization methodology for the development of an optimized parallel model of a H.264/AVC encoder (i.e. a processes network model balanced in communication and computation workload).

Lexicographic Multiobjective Integer Programming for Optimal and Structurally Minimal Petri Net Supervisors of Automated Manufacturing Systems

Based on Petri net (PN) models of automated manufacturing systems, this paper proposes a deadlock prevention method to obtain a maximally permissive (optimal) supervisor while minimizing its structure. The optimal supervisor can be achieved by forbidding all first-met bad markings (FBMs) and permitting all legal markings in a PN model. An FBM obtained via a single transitions firing at a legal marking is a deadlock or marking that inevitably evolves into a deadlock. A lexicographic multiobjective integer programming problem with multiple objectives to be achieved sequentially is formulated to design such an optimal and structurally minimal supervisor. As a nonlinear function, the quantity of its directed arcs is minimized. A conversion method is proposed to convert the nonlinear model into a linear one. With the premise that each place in the supervisor is associated with a nonnegative place invariant, the controlled net holds all legal markings of the net model, and the supervisor has the minimal structure. Finally, some examples are used to illustrate the application of the proposed approach.

Inductive Power Transfer System With Self-Calibrated Primary Resonant Frequency

Inductive power transfer (IPT) is a commonly employed technique for wirelessly supplying power to implantable medical devices. A major limit of this approach is the sensitivity of the inductive link to coupling factor variations between transmitting and receiving coils. We propose in this paper a new method for compensating these variations and improving the inductive link efficiency. The proposed technique is based on a mechatronic module that dynamically tunes the primary resonant capacitor value in order to maintain the resonance state of the IPT system. The module is able to maintain resonance state for apparent primary inductance range at least from 0.5 to 5 μH using a high capacitance resolution of 0.032 pF. Experimentations conducted on a 13.56MHz IPT system showed a 65% higher power transfer compared to a traditional IPT system.

Two stage particle swarm optimization to solve the flexible job shop predictive scheduling problem considering possible machine breakdowns

The flexible job shop scheduling problem under machine breakdowns is considered.A two stages particle swarm optimization is proposed to solve the problem.The proposed algorithm optimizes makespan, robustness and stability of the solution.A predictive schedule witch is more robust and stable is obtained. In real-world industrial environments, unplanned events and unforeseen incidents can happen at any time. Scheduling under uncertainty allows these unexpected disruptions to be taken into account. This work presents the study of the flexible job shop scheduling problems (FJSP) under machine breakdowns. The objective is to solve the problem such that the lowest makespan is obtained and also robust and stable schedules are guaranteed. A two-stage particle swarm optimization (2S-PSO) is proposed to solve the problem assuming that there is only one breakdown. Various benchmark data taken from the literature, varying from Partial FJSP to Total FJSP, are tested. Computational results prove that the developed algorithm is effective and efficient enough compared to literature approaches providing better robustness and stability. Statistical analyses are given to confirm this performance.

System-level performance evaluation of a H.264/AVC encoder targeting multiprocessors architectures

The system-level modeling and simulation framework Sesame/Artemis aims to efficiently explore the design space of heterogeneous embedded multimedia architectures. The Sesame environment provides several methods and tools to quickly and separately build the application process network model, the target architecture model, and the mapping model of the application onto the architecture. In addition, this tool is designed to allow the refining simulation models smoothly across different abstraction levels and to include support for refining only parts of an architecture model, creating a mixed-level simulation model. In this paper, the Sesame software framework is selected to implement at the black-box architecture model level a parallel H.264/AVC video encoding application targeting multiprocessors platforms.

A YAPI system level optimized parallel model of a H.264/AVC video encoder

H.264/AVC (Advanced Video Codec) is a new video coding standard developed by a joint effort of the ITU-TVCEG and ISO/IEC MPEG. This standard provides higher coding efficiency relative to former standards at the expense of higher computational requirements. Implementing the H.264 video encoder for an embedded System-on-Chip (SoC) is a big challenge. For an efficient implementation, we motivate the use of multiprocessor platforms for the execution of a parallel model of the encoder. In this paper, we propose a high-level independent target-architecture parallelization methodology for the development of an optimized parallel model of a H.264/AVC encoder. This methodology is used independently of the architectural issues of any target platform. It is based on the exploration of the task and data levels forms of parallelism simultaneously, and the use of the parallel Kahn Process Network (KPN) model of computation and the YAPI programming C++ runtime library. The encoding performances of the obtained parallel model have been evaluated by system-level simulations targeting multiple multiprocessors platforms.