Dynamic Dispatch Of Wind-thermal Units Based On Multi-objective Differential Evolution

Against the fact that dynamic dispatch is high dimension, nonlinear, multi-constrained under the wind power integration, this paper carries on the further research on the multi-objective differential evolution algorithm, and addresses corresponding improvement strategies. In the field of population initialization, differential operation, parameter control and constraint implementation, this paper proposes some methods involved in orthogonal population initialization, hybrid differential strategy, self-adaptive parameter control and constrain-domination. It has been proved by TNK test function that the algorithm proposed has better data mining ability, and provides the non-inferior solution a better distribution characteristic. The static cleaning dispatch of power system shows that the algorithm has the ability to solve complex problems。For the interfaces dispatch model of power system, applying ramp rate constraint in a certain period of time to the schedule restraints, this paper constructs a dynamic multi-objective Environment/Economic Dispatch model in order to get the characteristics of the thermal power unit output and the timing coupling characteristic of system load under the guidance of dynamic dispatch model. The test results of10units test system show that the algorithm with constraint multi-objective differential evolution has an advantage.Based on the dynamic Environment/Economic Dispatch model as well as the grid system voltage safety requirement under the wind power integration, the dynamic power dispatch model with multi-objective Security constraints is put forward with optimization goal of economic, clean and safe. The model involves the output restrictions of the unit, the limitation of that grid voltage and the constraints of the transmission line power. As a result, the optimization of the system cost of power units, pollutant emissions and static voltage safety margin are achieved by optimizing the output of thermal power units.In the IEEE30-bus system, constrained multi-objective differential evolution algorithm is adopted to solve the established model. The results shows that the applicability of the proposed power dispatch policy, and finally explores optimization results impacted by different gird interconnection points.