چكيده به لاتين
Electricity markets have undergone many changes since the past. The emergence of smart grids has also triggered another wave of changes in electricity markets. This study addresses the design of an electricity market for Multi-Microgrids. This electricity market includes a real time and a Day-ahead market. In order to model real time market, a multi-objective optimization method is used in this report, in which the system operator tries to solve the problem, and simultaneously, optimize several goals. The objectives of this market include maximizing network profits, grid profit, and system reliability. In order to maximize the reliability of the system, the network operator tries to maintain the maximum amount of storage on the batteries. Of course, it should be noted that other target functions act as a limiting factor, and all three objectives should be addressed to find the answer. For modeling the day-ahead market, several grids are connected to each other, and the power exchange between them is achieved by maximizing a social welfare function. For this purpose, each of the gridlines, according to its production and consumption, derives a piece-wise curve that can be used as the basis for the proposed price of the grid in the market. After collecting the price curves of each grid, the operator run the market and determines the market price. In this report, power transmission constraints between gridlines are considered as limiting factors. Also, at the uncertainty of electric charge, solar radiation and wind speeds are included in a scenario-based model that contributes to a more precise model. The results show that by implementing the real time market, the model has been able to determine the market price and find solutions to the system that can satisfy all of the goals. Using the day-ahead market, it was also possible for the system to execute its day-ahead market by using the limitation of the transmission constraints between gridlines, taking into account the system's uncertainty, and resulting a more accurate model of the problem.
