Harmonic Distortion Prediction Model of a Grid-Tie Photovoltaic Inverter Using an Artificial Neural Network

Abstract

Expanding the number of photovoltaic (PV) systems integrated into a grid raises many concerns regarding protection, system safety, and power quality. In order to monitor the effects of the current harmonics generated by PV systems, this paper presents long-term current harmonic distortion prediction models. The proposed models use a multilayer perceptron neural network, a type of artificial neural network (ANN), with input parameters that are easy to measure in order to predict current harmonics. The models were trained with one-year worth of measurements of power quality at the point of common coupling of the PV system with the distribution network and the meteorological parameters measured at the test site. A total of six different models were developed, tested, and validated regarding a number of hidden layers and input parameters. The results show that the model with three input parameters and two hidden layers generates the best prediction performance.

A Comprehensive Study of PV Array Connection Strategy with effect of Temperature to Improve System Efficiency

“…Numerous interdependent relationships are procreated for PV Array temperature problems that are used in literature which include essential environment-related variables and Parameters which are must for the conduction of experiments. [15][16][17] The power generated from the PV Generator (PVG) is dependent on solar shielding and favourable temperature, usually, a PV array is placed in severe climatic conditions. The temperature at which the PV conversion process operates plays a pivotal role.…”

“…Here is electrical efficiency at reference temperature for Array for and solar radiations used here is 1000 2 .here is taken as temperature coefficient and ɣ is taken solar luminosity coefficient have material properties of values 0.004 K-1 and 0.12 for crystalline silica Arrays [15] The second one is normally taken as zero Evans [16] and equation 2 is reduced to…”

Novel time‐domain average model for harmonic current prediction in photovoltaic and wind power units

“…4,5 Estimation models need to be developed to reproduce the electrical behavior of real-world installations to investigate the harmonic effects caused by RESs. 6 Žnidarec et al 7 reported that real-time and long-term techniques can be used for predicting harmonic currents. The former is often employed in active filters and static synchronous compensators, whereas the latter is used by transmission system operators to assess and mitigate harmonic problems during system operation.…”

“…The former is often employed in active filters and static synchronous compensators, whereas the latter is used by transmission system operators to assess and mitigate harmonic problems during system operation. Various models used for long-term harmonic estimation have been developed; for example, frequency-domain impedance models (IMs), [8][9][10][11] artificial intelligence models (AIMs), 7,12,13 and time-domain detailed or simplified models (DMs and SMs, respectively). 6,[14][15][16] Norton and Thevenin equivalent circuits are popular examples of frequency-domain IMs.…”

“…The first technique was based on instantaneous power calculation ( p‐q theory); the second, on the ADALINE neural network. Žnidarec et al 7 proposed a long‐term model for predicting harmonic currents using a multilayer perceptron neural network. They developed, trained, and validated distinct models using meteorological data and power quality measurements recorded over a year by considering different numbers of hidden layers and input parameters.…”