A comprehensive reviewing of existing interharmonic analysis and estimation methodologies irrespective of application is carried out. The paper is enlisting the characteristics of an appropriate method to analyse and estimate interharmonics in PV systems, and linking these characteristics with the features of the reviewed methodologies. The distinctive characteristics of interharmonic emissions related to PV systems are therefore presented. The various methodologies are classified, summarized, and a checklist is prepared to emphasize the areas to be paid attention to while establishing an apt method for interharmonic analysis in PV systems. The priorities for selection of a method by a practicing engineer vary case by case. This paper will serve as a guideline for selection and further development of a suitable method for interharmonic analysis in a PV included power system.

A comparative analysis of ‘Desynchronized processing technique’ (non-parametric) and ‘Sliding window ESPRIT method’ (parametric) for interharmonic estimation in PV systems is carried out. Both methods were applied to field measurements and semi-measured signals to test their feasibility under different conditions. In Desynchronized processing technique, in the second stage of estimation, a Short-term Fourier transform was used to address the time-varying nature of interharmonics. In Sliding window ESPRIT method, the signal subspace dimension was fixed in all analysed cases. An adaptive algorithm was used to find the best rank of the Hankel matrix and apt order of the filter to ensure stability. A set of critical parameters affecting the performance of these methods in interharmonic estimation are identified. This paper emphasizes the significance of using appropriate methods for accurate interharmonic estimation and also demonstrates through the illustrated results that different inferences can be drawn for the same measurements analysed using different methods.

The main objective of the paper is to investigate theexistence of interharmonic emissions from an MPPT drivengrid-connected PV inverter, identify their severity andpersistence. The presence of interharmonics in the measuredcurrent from a PV installation is linked to direct and diffusedsolar irradiation as well as a high ramping rate of theirradiation causing variations in both active and reactive power.The paper sets forth a set of observations and inferences, whichis an appendage to the ongoing research on the power qualityaspects of solar power. Three different case studies areevaluated in detail using signal processing tools like STFT andFFT.

Widely existing circuit topologies and inverter control strategies for photovoltaic (PV) systems allow customer flexibility but also introduce different kinds of interharmonics into the grid. A complete understanding of interharmonics from PV systems, with reasons behind their origin, remains needed. In addition, the time-varying nature of interharmonics and the potential impacts on other equipment are yet to be understood. In this paper, laboratory and field measurements of seven different inverter types at multiple locations are presented. A comprehensive analysis is performed to understand the existence, persistence, and propagation of interharmonics in PV systems on the dc side as well as grid side for different power levels. The origins of the interharmonics are established with experimental evidence and through a comparative analysis. A rural low voltage six customer network, with two different impedance profiles caused by the installation of PV, is considered to show the potential impact on customer voltage. To address the time-varying nature of interharmonics, a sliding window ESPRIT method is preferred over fast fourier transform (FFT)-based methods.

A model to analyze the probability distribution of the peak value of the interharmonics aggregating at the point of common connection is developed. With multiple sources of interharmonics, the fluctuations in the peak value of voltage or current matters. It is shown by virtue of Monte Carlo Simulations that the distribution of aggregated interharmonic peak value follows either a Gumbel distribution with random number of interharmonics or a normal distribution with constant number of interharmonics. Approximate mathematical expressions are formulated from the estimated parameters of the probability distribution functions to predict the highest probability of 95 percentile peak value occurence. The results from the model are verified using real data from a PV installation and a wind park with multiple converters. The significance of probabistic studies to understand the extremities due to interharmonic aggregation and also the analysis in time domain for evaluation of instantaneous aggregated peak value is emphasized in this paper

The harmonic interaction mechanism in a wind park is examined in this paper. The paper investigates the feasibility of a solution to the yet challenging harmonic contribution estimation from multiple sources in a wind park with limited available information, via an extension of a simple modeling approach as well as from detailed analysis of field measurements. The paper has two distinct objectives in assessing harmonic interactions (a) one to extend the classical Norton equivalent model to a multi-measurement wind park system and to suggest potential areas for further model developments from field measurement analysis, and (b) second to draw inferences from field measurements and to develop a new independent concept of analysis from long-term field measurements in the wind park. From practical experience, a new concept of analysis with a ‘harmonic interaction break-even point’ is introduced. With the help of it, one could identify whether the primary emission (emission from considered source) or secondary emission (emission from a distant source) dominates in the analysis period. In this way, the highest responsibility between different interacting time-varying harmonic sources is evaluated. It was concluded that from long-term measurements one can define a magnitude of power production where a certain harmonic order is canceled or reaches its lowest magnitude. If one finds this cancellation point, one can define a level of secondary/primary emission or at least a feasible range. This knowledge is a step forward towards harmonic contribution analysis.

Photovoltaic (PV) systems, Electric vehicles (EV) and LED lamps have gained significant popularity in our current society. It is therefore common to find customer installations with all three operating together. PV and EV are known sources of voltage variations on the grid. The impact of these voltage variations on LED lamps situated in close proximity to PV or EV in a low voltage installation, in terms of overvoltage, undervoltage, and rapid voltage changes is systematically studied in a laboratory environment in this paper. Such variations can cause malfunctioning of the lamp based on its immunity and tolerance level or be disturbing to the end-user based on the intensity of variations and rate of recurrence of being subjected to such variations. In this work, the observed impacts on LED lamps are illustrated as 15 different cases. The scope of this work is to identify the possible impacts due to voltage variations induced by PV and EV systems on LED lamps and the potential problems that could happen long term due to recurrent subjection of such voltage variations.

This paper proposes a deep learning (DL) method for the identification of spectral patterns of timevarying waveform distortion in photovoltaic (PV) installations. The PQ big data with information on harmonic and/or interharmonics in PV installations is handled by a deep autoencoder followed by feature clustering. Measurements of voltage and current from four distinct PV installations are used to illustrate the method. This paper shows that the DL method can be used as a starting point for further data analysis. The main contributions of the paper include: (a) providing a novel DL method for finding patterns in spectra; (b) guiding the manual post-processing based on the patterns found by the DL method; and (c) obtaining information about the emission from four PV installations.