A hybrid algorithm of Ant Colony Optimization (ACO) and Iterated Local Search (ILS) for estimating electricity domestic consumption: Case of Turkey

• A hybrid algorithm of ACO and ILS was presented to estimate electricity consumption.
• Its performance was tested using some indicators (population, import, export and GDP).
• Turkey’s electricity domestic consumption for 2014–2030 years was estimated.
• The results are expected to gain a new viewpoint to researchers and policy makers.

This paper presents the forecasting of Turkey’s electricity domestic consumption using hybrid algorithm, which bases on Ant Colony Optimization (ACO) and Iterated Local Search (ILS) algorithms. The approximation forecasting of an important energy source like electric for countries such as Turkey, which greatly imported energy sources, is so vital. In this paper, electricity domestic consumption model is first proposed by hybrid approach of ACO and ILS. Both techniques are quite simple and powerful stochastic local search method. ACO uses pheromone update mechanism to escape the local optimum. ILS applies iteratively local search to an initial solution until finding a local optimum; then it perturbs the solution and restarts a local search. This paper aims combining the advantages of these two algorithms to perform successful estimations. We use some economic indicators such as population, gross domestic product (GDP), import and export when Hybrid Electricity Domestic Consumption Estimation (HEDCE) model is developed. HEDCE equations proposed here are linear and quadratic. Results show that Quadratic_HEDCE find better solution because it can reflect better the variations of used indicators. The HEDCE models forecast Turkey’s electricity domestic consumption until 2030 under dissimilar scenarios.