Monitoring of photovoltaic systems: good practices and systematic analysis


Trends in PV System Performance

Researchers have observed a continuous improvement of performance ratios (PR) over the last 20 years. Detailed and more reliable monitoring campaigns have played a significant role in improving this trend.

The International Energy Agency’s Photovoltaic Power Systems Programme (IEA PVPS) includes a taskforce dedicated to improving the performance, reliability and therefore yield of PV systems (Task 13). This international task-force focuses on performance assessment techniques and specifically on monitoring as one of the remaining levers for improving performance of PV systems further. 3E is a key contributor of the IEA PVPS Task 13 expert group. We presented new IEA PVPS results at EUPVSEC in October 2013, in the paper “Monitoring of Photovoltaic Systems: Good Practices and Systematic Analysis”. The paper is part of the EUPVSEC proceedings from WIP.

The paper outlines good practices in PV monitoring in terms of set-up, equipment and processes before providing new techniques for data analysis. Read the full paper here

Good Practice in PV Monitoring: What to Watch for


In reviewing the state of the art in monitoring techniques, our paper highlights the need to track indicators across the whole energy conversion chain, including both capture losses and system losses.The most prominent factors causing reduced PRs are:

  1. Module temperature
  2. Dirt, dust and shading, which can account for 70% of all losses in some locations
  3. Mismatch and wiring losses
  4. DC to AC conversion losses

Being able to rapidly detect and diagnose these factors and issues specifically is therefore of key importance in monitoring for optimal performance.


Getting the Right Input

Being aware of the limitations and uncertainties related to measurement equipment is a first step in getting more insight from the data collected.

For electricity yield measurements, energy meters or true-rms power meters should be used. The inverter-integrated measurements are usually not sufficiently precise but can deliver useful information for relative analyses over time.

To distinguish performance from the variability of the solar resource and retrieve accurate PRs from a monitoring system, reliable irradiance data is essential. The choice of a data source depends on accuracy levels but also on budgets and maintenance requirements (which can be quite high). The paper provides an overview of the different technologies available. The 3E research team has further investigated accuracy levels of different data sources and can provide guidance and support in setting up the right system. Video_Icon_3E-pyranometer Video_Icon_3E-01 Video_Icon_3E-12Choosing the right irradiance data source is about finding the right balance between accuracy and costs.

The way irradiance data is processed and integrated in PR calculations, for example through third-party monitoring software, will also be key to getting maximum value from monitoring systems. Ensuring that irradiance data is available at all times and treated transparently, is a strong feature for a monitoring software platform which should be assessed carefully. 3E’s own independent platform, SynaptiQ, integrates the lessons learned from our research team and provides multiple reliable fall-back options and clear categorizations of PRs indicating the underlying irradiance source.

Data Mining and Analysis

The main purpose of a monitoring system is to assess PV system performance to identify design flaws and malfunctions.

3E_monitoring_IEA-PVPSFigure: Deviations in hourly system yield versus reference yield, plotted per week for an installation In Belgium subject to shading. There are a variety of methodologies to handle and automate performance assessments, which all basically follow a simple process:

  1. Tracking the relevant indicators over periods of time
  2. Detecting deviation
  3. Assessing scale and profile of deviation
  4. Delivering (preliminary) diagnosis to take action

Our IEA PVPS Task 13 paper points out a few of the most well-known methodologies and introduces a new analysis based on regression-based linear modelling. This technique consists in describing key behaviour of a PV system with simple linear relationships to be mapped out over time and checked for deviations. Examples of various issues throughout the energy conversions chain show that with weekly reviews of these relationships, deviations can very clearly and quickly be identified.

To carry out such analyses and to manage the whole performance assessment process most effectively, a monitoring system must combine:

  1. Clear, reliable indicators of system behaviour and components
  2. Powerful visualization and analysis tools to check patterns,  see changes quickly, and automate processing
  3. Possibilities to drill-down and investigate issues from different angles

We are continuously enhancing our own SynaptiQ monitoring tool around these capacities. Contact us for more information about SynaptiQ’s new scatter plot and data mining tools.

20_ideas_grey Find out more about our general research activities here

20_SQ-grey  Find out more about our independent PV monitoring software here