System Performance

Aim

The aim of Strathclyde’s research was to improve understanding of PV array operation under non-uniform conditions, and in particular to develop improved maximum power point tracking (MPPT) for such situations. In addition, an outdoor measurement system was established with external access to real time data.

Research development

This research was concerned with accurate, cell by cell, modelling of arrays so as to capture the real operational I-V characteristics. These models were to be validated by data collected on real outdoor systems at Strathclyde University. The validated system models were then be used to develop new MPPT algorithms which will be implemented on a solar DC/DC converter for experimental performance verification. A DC/DC solar converter capable of implementing different MPP algorithms was to be built and tested for the case of part shading. A PV test system has been installed in Strathclyde University to assist the development of the research. The experimental arrangement comprises a PV system, instrumentation for monitoring the ambient data (radiation, wind and temperatures) and the PV performance, plus a Data Logger for controlling and recording all these data. On line weather station will be provided that includes the possibility to monitor real-time data and to download averaged data.

Main findings/impact

A novel MPPT algorithm was developed are tested and its performance has been compared to a standard Hill climbing approach. This algorithm makes use of an artificial neural net (ANN) trained with data coming from PV modules which implement only one bypass diode. Other simulations were run to test the performance of the algorithm with different module configurations. Condition of partial shading and fast irradiance on PV modules with 2 bypass diodes have been simulated to check the performance of the algorithm; the results have shown that the algorithm can track the true maximum even for these PV modules without need to be trained with different data. The algorithm was then optimized to detect when the PV system is affected by partial shading: in fact the algorithm tests the system to detect if the losses in power are due to partial shading or decreasing irradiance. In this way the controller executes the scan only when strictly necessary (ie partial shading).

Weather station. The weather station is now been equipped with a sun tracker, purchased for the project.

The program written in Labview has been upgraded including codes for reading averaged measurements (5 minutes and 1 hour averages) and for sending the data to a database. Thus a database in mySQL has been created to host the data for a website. The website can be seen at the following address http://weatherstation.heliohost.org/home.html and includes the following pages:

– Home: a brief description of STAPP has been included with a link to STAPP home page

– Experiment: this page describes the experiment in details

– Most recent measurements: this pages shows 8 graphs. On the left side measurements of global and diffuse radiation, wind speed and wind direction, ambient temperature and relative humidity are plotted with a time scale of seconds (raw measurements). On the right side the same graphs are plotted but the measurements shown are in a time scale of 5 minutes average. The page automatically updates every 30 seconds. This page shows only the last 1000 measurements (about 8 hours for the plots in seconds scale and 3 days for the averaged time scale).

– Advance plotting facilities are available

 Contact: David Infield