Abstract
The penetration of distributed power sources has been increasing with the continuous promotion of clean renewable energy sources. This paper seeks to improve the utilization rate of clean energy and reduce the cost of microgrid operation by first establishing a double-layer wind power prediction error model based on a comprehensive consideration of the time-of-use price and the operating characteristics of different types of clean energy sources, such as wind power, photovoltaic power, thermal power, and transmission tie lines. A combined cooling, heating, and power microgrid collaborative optimization model that considers wind power forecast uncertainty is established with the goal of minimizing economic cost, environmental cost, and degree of power-generation unit output asynchrony of the microgrid. The established multi-objective optimization model is solved using an improved intelligent optimization algorithm that combines the non-dominated sorting genetic algorithm (NSGA) with co-evolution theory and the beetle antennae search algorithm. This algorithm employs a variety of groups in the NSGA to help with correcting the approximations of group members through competition and cooperation. Therefore, the proposed algorithm can combine the excellent convergence of the NSGA and the powerful searching ability of co-evolutionary algorithms. Finally, a practical microgrid system in Northwest China is simulated as a case study, and the performance of the proposed algorithm is compared with that of the conventional NSGA. The simulation results demonstrate the superiority of the global search performance and the rapid convergence performance of the proposed hybrid algorithm.
Highlights
Clean energy sources, such as wind power (WP) and photovoltaic (PV) power, have been promoted worldwide due to their renewability and environmental friendliness
The conventional economic dispatching of a microgrid mainly relies on a reasonable management of the output of each controllable unit in the grid, and it addresses the uncertainties associated with the random nature of renewable energy sources like wind and PV power as much as possible under the premise of meeting the load demand [5], while minimizing the economic losses caused by abandoned WP and PV power and the operation and maintenance cost of controllable units [6], [7]
In analytic methods based on point prediction, a corresponding empirical model of prediction error is established according to the prediction scale, such as a normal distribution model (NDM) or a Poisson distribution model (PDM), and its probability density function or distribution accumulation function is determined
Summary
Clean energy sources, such as wind power (WP) and photovoltaic (PV) power, have been promoted worldwide due to their renewability and environmental friendliness. The costs of interactions with external electric power grids Fgrid are determined by the deviation Pun(t) between the output and load of the microgrid and the purchasing and selling price of electricity at each moment, which are given as follows: Psh(t)Bp(t) −. According to the operating data of thermal power units, the pollutant emissions of single gas turbine units and diesel units can be described by quadratic functions of their active power generation, which are given respectively as follows. B. CONSTRAINTS OF MULTI-OBJECTIVE OPTIMIZATION The operating constraints of the CCHP microgrids system include constraints associated with cooling, heating, and energy balance, the constraints of equipment output, and interaction constraints with external grids, which are given as follows. The various populations learn from each other in the process of obtaining the Nash equilibrium, and continuously reconstruct their performance to make them tend toward a globally optimal solution
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
