Towards Enhancing the Performance of Grid-Tied VSWT via Adopting Sine Cosine Algorithm-Based Optimal Control Scheme

Abstract

The expanding trend of wind power technology motivates scholars to pursue more investigation on optimising energy extraction from the wind and integrating high-quality power into the utility grid. This paper is aimed at introducing a novel application of the sine cosine algorithm (SCA) which attempts to find the optimal gains of proportional-integral (PI) controllers used to control the power electronic converter (PEC) equipped with the Variable speed Wind turbine (VSWT) such that a maximum power extraction and performance enhancement can be realized. The PEC equipped with the VSWT combines a machine side converter (MSC) and a grid-side inverter (GSI). Both the MSC and GSI are controlled by the proposed SCA-based PI controllers through cascaded vector control schemes. The MSC is responsible for controlling the wind generator's rotational speed, active power, and reactive power. The GSI is used to regulate the dc-link voltage and to keep the terminal voltage at the desired frame set by the operator. To obtain the optimum PI gains, the SCA is applied to minimize the sum of the integral squared error (ISE) of twelve PI controllers error inputs in the control schemes simultaneously. Performances of the proposed SCA-PI control schemes are assessed under severe grid disturbance and random wind speed variation to mimic more realistic conditions. The effectiveness of the proposed SCA-PI is verified in the MATLAB/Simulink environment, and the results are compared to those obtained using a grey wolf optimizer and particle swarm algorithm-based optimal PI controller. The simulation findings confirm the SCA-PI can be regarded as an efficacious way to enhance the performance of the VSWT.INDEX TERMS Wind turbine control, power electronic converter, MPPT, PMSG, PI controller, sine cosine algorithm.

A state‐of‐the‐art comprehensive review of modern control techniques for grid‐connected wind turbines and photovoltaic arrays distributed generation systems

“…V sabc ) and stator current (i.e. I sabc ), respectively, ω s = ω r is the PMSG rotor speed or synchronous speed in rad/s, R s is the PMSG stator resistance and ψ fsd and ψ fsq are the dq-axis components of the stator flux linkage, respectively that can be described as [35]:…”

“…From ( 4) and ( 5), the dq-axis control outputs of the MSC CB (i.e. V isd and V isq ) can be expressed as [33,35]:…”

“…V sabc ) and stator current (i.e. I sabc ), respectively, ω s = ω r is the PMSG rotor speed or synchronous speed in rad/s, R s is the PMSG stator resistance and ψ fsd and ψ fsq are the dq ‐axis components of the stator flux linkage, respectively that can be described as [35]: $$\begin{equation} \def\eqcellsep{&}\begin{array}{l} {\psi _{fsd}} = {L_d}{I_{sd}} + {\psi _f}\\[8pt] {\psi _{fsq}} = {L_q}{I_{sq}} \end{array} .\end{equation}$$…”

“…The PMSG comprehensive mathematical model in the dq ‐axis reference frame can be expressed as [35]: $$\begin{equation} \def\eqcellsep{&}\begin{array}{l} {V_{sd}} = {R_s}{I_{sd}} + \left( {{{d{\psi _{fsd}}} \mathord{\left/ {\vphantom {{d{\psi _{fsd}}} {dt}}} \right. \kern-\nulldelimiterspace} {dt}}} \right) - {\omega _s}{\psi _{fsq}}\\[8pt] {V_{sq}} = {R_s}{I_{sq}} + \left( {{{d{\psi _{fsq}}} \mathord{\left/ {\vphantom {{d{\psi _{fsq}}} {dt}}} \right.$$…”

“…The SCs, QCs and CCs for the MSC/RSC CBs are classically executed using PI control technique with fixed control gains [11,12,32]. However, the performance of conventional PI control-based SC in the MSC CB is further improved using modern techniques, such as adaptive FLC (A-FLC) [33], grey wolf optimizer-based tuned PI (GWO-PI) control [34], sine cosine algorithm-based PI (SCA-PI) control [35] and A-FLC with passivity-based backstepping control (A-FLC-PBC) [36]. Similarly, to avoid the traditional PI control [37], the modern SCs/QCs for the RSC CB are applied using adaptive backstepping control [38], FLC-sliding mode control (FLC-SMC) [39], H ∞ control [40], FLC-SMC using particle swarm optimization (PSO)-gravitational search algorithm (GSA) [41], recursive least squares (RLS) control [42], non-linear adaptive control (NAC) [43] and model-assisted active disturbance rejection control (MA-ADRC) [44].…”