Time Domain Particle Swarm Optimization of PI Controllers for Bidirectional VSC HVDC Light System

Abstract

This paper proposes a novel technique to tune the PI controllers of a bidirectional HVDC light system by embedding particle swarm optimization directly in the Simulink model in the design procedure. The HVDC light system comprises of a rectifier station, a DC link, and an inverter station. Each converter station requires four PI controllers to be tuned in the decoupled d-q vector control scheme, and with the bidirectional HVDC system, the required PI controllers are doubled. Tuning these many controllers using conventional methods is a challenging task, especially if the parameters of the converter stations are different. A novel approach to tune the PI controllers for a bidirectional HVDC system using the time-domain performance indices is presented in this paper. The time-domain performance indices are optimized using the particle swarm optimization (PSO) algorithm. The results of the proposed tuning method show that the proposed method not only gives superior results but also is less cumbersome to tune compared to conventional methods like modulus optimum (MO).

Quantitative criteria of considering AC infeed in DC fault assessment of modular multilevel converters

“…To achieve better dynamic performance and adequate speed response of the inner control loop, modulus optimum is conventionally used for tuning the current control parameters due to the fast response and simplicity [30][31][32][33]. The PI parameters then can be written as [ Ta is defined as 1/(2fsw) [30,31], where fsw is the switching frequency of the converter.…”

Electric Power Network Interconnection: A Review on Current Status, Future Prospects and Research Direction

“…The configuration of a bi-directional HVDC light system is depicted in Figure 12. Each converter station contains two poles, one with positive polarity and one with negative polarity, each with their neutral grounded [85].…”

Intelligent Multiobjective NSBGA-II Control of Power Converters in DC Microgrids

“…In the state-of-the-art SOO [26][27][28]30] and MOO [29] based control techniques, evolutionary algorithms are considered as an offline or online tuning tool. In other words, in [26][27][28]30], a basic controller, which can be PI or PID is used to generate a continuous signal valued in the interval 𝑢 ∈ [0,1] and then applied to the PWM block to generate the switching signal; and evolutionary approaches are utilized to tune the gains and characteristics of the basic controller.…”

“…The main distinguishing feature of the developed approach over the state-of-the-art results is that it uses the MOO to generate the pulsing signal of converters without the need for the PWM. Thereby, the suggested topology is different from state-of-the-art methods [26][27][28][29][30]. The proposed approach utilizes the non-dominating sorting binary genetic algorithm (NSBGA-II).…”