A microgrid is a set of components that includes power generation sources, energy storage systems and consumers. Designed for flexibility and environmental protection, microgrids often include renewable energy sources such as wind turbines, geothermal generators, biomass / biogas generators and photovoltaic panels to meet electricity demand. As the world’s governments begin to impose legislation to cover an increasing percentage of energy needs using “green” sources, optimizing the operating costs of microgrids is becoming a necessity.
This thesis, entitled “Smart Microgrids. Management Applications Development”, aims to optimize the operating cost of an isolated microgrid and experimentally validate the solutions.
The main purpose of the thesis is to minimize the operating cost of an islanded microgrid by using Day-Ahead Scheduling methods (algorithms). Minimizing the operating cost is, in itself, an optimization problem with a single objective: to obtain a daily operating cost as low as possible while fully covering the energy needs and the using the microgrid elements within nominal parameters. The limitation of Day-Ahead Scheduling methods consists in predicting the weather conditions and the energy requirements during the day. These optimization methods will actually obtain the calculated results only if the two predictions are perfect. Thus, the secondary purpose of the thesis is to develop and validate a real-time control method for the microgrid that uses the operating program obtained after optimizing the cost and to adapt it in situations where at least one of the predictions does not correspond to reality.
Personal contributions:
- Mathematical formulation of the optimization problem, highlighting the differences between equations and reality.
- Defining the purpose of optimizing the operating cost of the microgrid and its limitations.
- Implementing the harmony search method for the defined optimization problem.
- Analysis of the solutions generated by the HS method and their correction to ensure their viability.
- Identifying the cases where the solutions generated by the HS method, although correct, are not ideal.
- Implementing the particle swarm optimization method for the defined optimization problem.
- Analyzing the solutions generated by the PSO method and their correction to ensure their viability.
- Identifying the cases in which the solutions generated by the PSO method, although correct, are not ideal.
- Identifying the cases in which the general optimization methods used have imposed difficulties.
- Proposing, developing and implementing a personalized optimization method that offers the advantages of general methods and avoids their disadvantages.
- Demonstrating that the proposed goals for the SD method were achieved by comparing the results obtained using it with those obtained using the general optimization methods presented earlier.
- Creating a simulation model for the proposed microgrid.
- Adapting the results of the optimization methods in order for them to be simulated.
- Simulating the results of the 3 optimization methods and demonstrating the efficiency of the personalized optimization method by analyzing the obtained results.
- Developing a testing methodology for the experimental validation of the SD optimization method.
- Adapting the optimization problem, the optimization method and the simulation model for the experimental microgrid.
- Simulating the experimental microgrid together with the new form of the optimization method.
- Experimental validation of the SD optimization method.
- Demonstrating the usefulness of the cost optimization method by comparing the results obtained experimentally with those obtained by testing the same scenario using an energy management algorithm based on State of Charge.
- Development and simulation of a real-time control method of the microgrid.