Comparison of linear programming and simulated annealing methods for optimal load control and energy scheduling in smart home

Many tariffs are used for electricity pricing. Electricity bills can be minimized by choosing the optimal working hours of appliances and for appropriate tariffs in smart homes. For this reason, Linear Programming (LP) and Simulated Annealing (SA) optimization methods are used. One of the aims of the thesis is to compare the results of these methods. The multi-time tariffs which have different energy unit prices at different times of the day are used in the thesis. Energy Management System (EMS) shifts time slots of some appliances to cheap energy unit prices. There may be overload in system because of shifting of working hours of appliances. Optimal load control is applied to prevent high peak demand as the second aim of the thesis. Mathematical model of the problem is created and GAMS application is used in the solution. The cost table consisting of the energy unit prices of each time slot is used as input. Also, SA technique which is one of the metaheuristic solution methods is used and the solution has been reached using C# program. Optimum electricity cost, working hours of the appliances and peak to average ratio are achieved by two different solutions and the results are compared. The results of the LP are better than SA.