Model-predictive optimization techniques for smart traffic and smart grids

Summary form only given. In this talk, we will discuss how the well-known model-predictive optimization (MPO) techniques can be applied to emergent applications such as smart traffic and smart grids. Basically, we need accurate prediction models for predicting the future trends/requirements of traffic conditions and energy usage/generation. Based on the historical data and predicted data, a novel model-predictive optimization method is proposed. The optimization algorithms include genetic algorithm for smart traffic and particle swarm optimization for smart grids. Simulators have been developed to prove the benefits of leveraging model-predictive optimization. Results show that MPO incurs a reduction in traffic congestion by 29%, in waiting time by 27%, and in overall cost for energy usage/generation by 19%.