Parameter Optimization Of Doubly-Fed Induction Generator Based On Proportional-Integral-Derivative Optimization Algorithm

Energy is an inexhaustible driving force for the development of human society.Ensuring green,stable and safe energy supply is the mission entrusted by the times’ development.In the world,fossil energy stock crisis and huge carbon emissions drive the supply-side energy reform.The permeability of renewable energy resources and clean energy is increasing,and wind energy utilization is one of the key research objectives.The wind energy received by the wind turbine is converted into electrical energy and incorporated into the grid to realize the utilization of wind energy.A doubly-fed induction generator(DFIG)has the advantages of low cost,high wind energy utilization efficiency,and bidirectional energy flow with the grid,it becomes a wind turbine with high priority.The control system of wind turbine is the key to ensure the stability of power generation.For the control system of DFIG,two smart optimization algorithms are proposed,i.e.,proportional-integral-derivative optimization algorithms(PIDOA)and multi-objective proportional-integral-derivative optimization algorithms(MOPIDOA).The idea of conventional proportional-integral-derivative controller inspires the proposed optimization algorithm.The proposed optimization algorithm consists of two types of controllers,i.e.,explorative controllers with variable parameters and exploitative controllers with fixed parameters.In the approach’s exploration process,multiple explorative controllers with variable parameters move toward the global optimal solution.In the approach’s exploitation process,multiple exploitative controllers with fixed parameters moving toward to local optimal solution.Eight benchmark functions are used to verify the performance of PIDOA,and 24 optimization algorithms are compared.The results show that PIDOA has excellent exploration and exploitation ability.To provide a better non-dominated solution set for multi-objective problems,MOPIDOA is introduced into multi-objective optimization problems.Compared with PIDOA,MOPIDOA adds the control objective concept to strengthen the interaction between the non-dominated solution set and the MOPIDOA.The non-dominated solution set is used as the control objective of MOPIDOA to facilitate the maintenance and updating of the non-dominated solution.MOPIDOA designs exploration controller and exploitation controller,and introduces the concept of disturbance in control system,i.e.,control individual disturbance and control target disturbance,to enhance the exploration ability of MOPIDOA.MOPIDOA and eight comparative optimization algorithms are tested in eight benchmark functions.The results show that MOPIDOA has high convergence performance and coverage performance.Finally,the DFIG system structure is analyzed,the maximum power point tracking control strategy is selected,and the control framework of DFIG is established.The proposed PIDOA and MOPIDOA are applied to optimize the PI controller’s parameters under the control framework,respectively.The results show that PIDOA can virtually explore and exploit the global optimal solution to obtain the minimum deviation of rotor angular velocity and reactive power under three working conditions.MOPIDOA can obtain a relatively smooth and dense Pareto front,which proves the convergence ability and coverage ability of MOPIDOA.