A novel proposal for harmonic compensation in the grid-connected photovoltaic system under electric anomalies

• Published in: Electric power systems research Vol. 223; p. 109512
• Main Authors: Patel, Nirav, Elnady, A., Bansal, Ramesh C., Hamid, Abdul-Kadir, Adam, Ali Ahmed
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.10.2023
• Subjects: Abnormal grid; Adaptive RLLAD filter; Harmonic mitigation; Photovoltaic systems; Power quality; Voltage source converter; Photovoltaic systems; Voltage source converter; Abnormal grid; Harmonic mitigation; Power quality; Adaptive RLLAD filter
• ISSN: 0378-7796; 1873-2046
• DOI: 10.1016/j.epsr.2023.109512

•To meticulously detect the fundamental active and reactive weight components of the load current from harmonically contaminated signals, this paper proposes a novel adaptive RLLAD filter. From a practical standpoint, the devised filtering scheme guarantees comparable convergence performance and ensures zero oscillations in FWCs. To this end, the proposed RLLAD filter ensures accurate extraction of the sinusoidal reference current in comparison with that identified by existing state-of-the-arts filtering methods. This way it complies with the 5% stringent total harmonic distortion (THD) limit specified in the IEEE revised standard 519- 2014 and 1547.2–2018.•The envisioned PV inverter together with the proposed adaptive RLLAD filter is configured to provide dynamic reactive power support at the point of connection (PoC). This way, it ensures unity power factor (UPF) operation on the grid-side and thereby satisfies the grid interconnection codes described in the IEEE revised std-1547.•Based on author's survey, it is the first time a detailed and step-by-step procedure is documented to derive feedforward compensator equations and working principle to counter the impact of nonlinearity introduced by the PV array. Consequently, it ensures minimum voltage overshoot/undershoot and recovery interval in DC-link voltage under preceding event of insolation transition.•To this end, the performance of the proposed RLLAD filter is experimentally investigated using the OPAL-RT control suite under: (1) abrupt change in irradiance; (2) parametric uncertainties in nonlinear loading; (3) distorted grid voltage; (4) utility grid voltage change; and (5) imbalanced grid voltage with negative sequence imposed.•The suggested control architecture is devoid of classic phase-locked loop and employs ESOGI scheme to detect the fundamental positive sequence component of the electric grid voltage. Practically, it exhibits a unique feature of rejecting the DC-offset and inter-harmonics introduced by the power-electronic devices. This way, it enables efficient synchronization of PV inverter and current control unit with the electric grid even during electrical anomalies. The accurate estimation and online calibration of the fundamental weight components (FWC) from harmonically contaminated load current is crucial to evaluate the performance of grid-integrated photovoltaic (PV) units especially when exposed to grid anomalies. Driven by this motivation, an adaptive regularized least logarithmic absolute difference (RLLAD) filter is proposed for FWC estimation during nonlinear loading. The proposed filter pertinently captures the FWC of the harmonically distorted load currents. The resulting FWC guarantees balanced and sinusoidal grid currents even under disturbances in the grid voltage and load current thereby achieving different power quality (PQ) improvement targets i.e., grid current balancing and harmonic suppression, active and reactive power management, neutral current compensation, and power factor correction. The proposed RLLAD filter ensures comparable convergence performance and minimum steady-state oscillations in the grid reference currents. Proceeding further, to counter the nonlinearity introduced by the PV array, a feed-forward compensator is judiciously equipped. From a practical standpoint, it ensures superior DC-link voltage stabilization especially under abrupt transition in solar insolation level. The performance of the RLLAD filter is comprehensively assessed through numerical simulations in MATLAB/Simulink software. Furthermore, the practical viability of devised RLLAD filter is confirmed under rigorous experimental scenarios through OPAL-RT control suite. The experimental results demonstrate practicability of the proposed adaptive RLLAD filter and thus, considered to be a promising solution as compared to existing benchmarks and state-of-the-art approaches for online FWC estimation and PQ improvement. Most importantly, in terms of nonlinear load tracking capability, the proposed RLLAD filter is shown to outperform the classical LLAD filter with a response speed of less than 400 μs.