Friday, November 24th, 2017

Estimating the maximum power of a High Concentrator Photovoltaic (HCPV) module using an Artificial Neural Network

Available online 26 March 2013
Publication year: 2013
Source:Energy

The peculiarities of High Concentrator Photovoltaic (HCPV) modules make it very difficult to estimate their output. There are some methods to estimate the maximum power of an HCPV module which provide good results, but they are not easy to apply for the following reasons: a) they do not offer a comprehensive explanation of the entire procedure, b) they need some intrinsic parameters of MJ (multi-junction) solar cells which are difficult to obtain, c) it is necessary to have either complex and expensive devices for indoor measurements, and/or an accurate and complete set of outdoor measurements. This paper is intended to propose a model based on Artificial Neural Networks (ANNs) to predict the maximum power of an HCPV module using easily measurable atmospheric parameters. To this end, a group of atmospheric parameters together with the maximum power of an HCPV module have been measured throughout a whole year at a research centre located in the south of Spain. The results showed that using atmospheric parameters, the proposed ANN is capable of estimating the maximum power of an HCPV module with a mean square root error of 3.29%. This model could be extended to other modules and other places.

Highlights

► Estimating the output of an HCPV module, due to its special features, is a very complex task. ► The aim is to develop an ANN to predict the maximum power ( P max ) of an HCPV module. ► For achieving this, an HCPV module have been measured for a year in the south of Spain. ► The ANN developed can be used to estimate P max of an HCPV module with an acceptable error. ► This model could be extended to other modules and other places following the structure proposed.


Speak Your Mind

Questions or comments? We'd love to hear from you!