Matching demand with renewable resources using artificial intelligence techniques

The wide-spread integration of renewable energies in modern power systems is a vital pre-requisite to transform the global energy system towards sustainability. The very obstacle that prevents these sources from spreading is its intermittent nature which results in a fluctuating generated power profile and that considerably affects the ability of these supplies to satisfy the required demand independently. This mismatch issue between both demand and renewable sources profiles can be solved by aiding these sources with auxiliary fossil dependant ones to make for the shortage in the renewable power and reshaping the demand profiles of isolated systems in order to match the renewable sources profiles and hence increase the share of the renewable energy supplied to the demand. This paper introduces a developed Demand Side Management (DSM) technique that reshapes the demand profile through applying different DSM measures (load shifting and load shedding) to create greater flexibility in demand and better facilitate the integration of renewable energy technologies within the built environment. A heuristic optimization algorithm i.e. Genetic Algorithm Optimization (GA) is used to assure achieving the best match. Another AI technique i.e. Artificial Neural Networks (ANN) is used to implement a short-term load forecasting module integrated with the proposed DSM technique to estimate the load profile for the period of study (24 hour) in order to assure that the DSM actions for reshaping the load will achieve the required match.