An integrated optimization model and analysis for the production planning and bidding in the day-ahead energy market for a system of cascading hydroelectric power plants

As in the whole world, the demand for electricity in our country is increasing day by day, as a result of this, interest and investments in renewable energy have also increased. Hydroelectric energy, one of Turkey's important energy resources, has a structure that can be started and stopped more easily when compared to other energy resources. Since it is flexible and controllable, it provides the opportunity for manufacturers to produce at the right time. Therefore, a company with a Hydroelectric Power Plant can increase its revenue by considering the state of the system and deciding on the best time to generate electricity. In this study, it is aimed to model and solve the problem of scheduling production and making bid decisions in the day ahead market for a system consisting of sequential HEPPs with more than one power plant. The mathematical model is an integrated model that includes two main parts, the production model and the bidding model. Unlike other studies in the literature, this integrated model takes the planning horizon into days in production planning and decides on hourly production quantities, while at the same time it creates hourly quantity and price offers in the day-ahead energy market. In the production model part, the working capacities of the power plant and the balance of water amounts in the reservoirs are taken into account. The market clearing prices used in the bidding model are determined on an hourly and daily basis in the day-ahead market where physical electricity is traded. It is important to give the correct bid amount for market participants, because bids above the market price will not be accepted. Since market clearing prices (MCP) are determined after bids are placed, the model needs to be resolved under MCP uncertainty. Therefore, a forecasting model has been developed to make MCP estimations for use in the model. Based on the margin of error of this forecasting model, it is aimed to obtain a scenario-based solution for MCPs. Regression analysis was performed on an hourly basis for MCP values that changed during the day, and different scenarios were produced according to the expected value and standard deviation found. The performance of the scenario-based integrated model is compared with the integrated model that only works with expected MCP values and the solutions of the problem that the production schedule is decided first and then bid optimization is made. In addition, a numerical study has been carried out showing the effect of the increase in the estimation error on the performance of the integrated optimization model. In addition, sensitivity analysis was performed to see how the change of selected parameters affects the objective function value of the integrated model.