Research On Optimal Design Of Load Frequency Active Disturbance Rejection Control Based On Intelligent Optimization Algorithms

Aiming at the frequency and power stability problems caused by load disturbances in the power system,active disturbance rejection control(ADRC)technology is carried out on the design of load frequency control(LFC)research to improve the control performance of the classic control strategy in this paper.In order to solve the intricate parameter optimization problem of active disturbance rejection control,intelligent optimization algorithms are introduced.Therefore,the active disturbance rejection control of load frequency based on intelligent optimization algorithm is proposed.The specific research content and main innovations are as follows:1.The state space equation of the interconnected area is expressed through the dynamics of the components of the power system based on which an active disturbance rejection control scheme is designed.Load disturbances and tie-line power deviations are regarded as total system disturbances,which are estimated in the extended state observer and compensated in the nonlinear state error feedback.Consequently,the original system is transformed into a standard tandem integral type.2.In order to reasonably optimize the parameters of the active disturbance rejection controller,this paper introduces the double chains quantum genetic algorithm(DCQGA)with better convergence speed and stability,and designs an active disturbance rejection control scheme based on DCQGA.The performance is compared with the control algorithms in the proportional-integral-derivative(PID)control,fuzzy PID control,etc.in the traditional reheated thermal two-area power system,the thermal two-area power system considering the governor dead band(GDB),the water-thermal-gas multi-source power system,and the interconnected three-area power system considering the generation rate constraint(GRC).Robustness test considering parameter perturbation and different disturbances are also carried out.The result proves the effectiveness of the proposed DCQGA-ADRC strategy.3.On the parameter optimization problem,this paper introduces a new type of memetic intelligence algorithm that integrates reinforcement learning into the judgment and decision-making of the particle swarm algorithm search process strategy,that is,the improved memetic particle swarm algorithm based on reinforcement learning(RLMPSO).A RLMPSO-based active disturbance rejection controller parameter optimization scheme is designed.The performance comparison of other control algorithms such as PID,FOPID and model predictive control(MPC)in the nominal-parameters and parameter-perturbation traditional non-reheat thermal two-area power system as well as non-linear power system considering governor dead zone(GDZ)and GRC highlights the advantages of the proposed RLMPSO-ADRC method and proves its superiority.4.An improved beetle swarm algorithm based on chaos and fractional differentiation ideas,namely chaos fractional order beetle swarm optimization(CFBSO)is presented.After elaborating the improvement ideas,the optimization problems of unimodal,multimodal and fixed-dimension multimodal benchmark functions are compared with the beetle swarm algorithm and the chaotic particle swarm algorithm,which proves the effectiveness of the improved strategy.Then the algorithm is used in the optimal design of the active disturbance rejection control.The proposed CFBSO based active disturbance rejection control scheme is adopted for participating in load frequency control in a solar-thermal power interconnected power system considering GDZ,GRC and transmission delay.The results show that compared with MPC,PID and other strategies,the proposed method has smaller overshoot and undershoot as well as shorter settling time,which means that it better meets the high performance requirements of LFC.