Loughborough University
Browse
Thesis-2017-Koutsourakis.pdf (11.56 MB)

Compressed sensing current mapping spatial characterization of photovoltaic devices

Download (11.56 MB)
thesis
posted on 2017-07-11, 15:50 authored by George Koutsourakis
In this work a new measurement technique for current mapping of photovoltaic (PV) devices is developed, utilising the compressed sensing (CS) sampling theory. Conventional current mapping measurements of PV devices are realised using the light beam induced current (LBIC) measurement method. For its realization, a light beam scans a PV device and the induced current is measured for every point, generating the final current map of the device. Disadvantages of the LBIC method are the low measurement speed, the complicated and usually expensive measurement layouts and the impractical application of the method on PV modules. With the development of CS current mapping in this work, the above issues can be mitigated. Instead of applying a raster scan, a series of illumination patterns are projected onto the PV sample, acquiring fewer measurements than the pixels of the final current map. The final reconstruction of the current map is achieved by means of an optimisation algorithm. Spatially resolved electrical simulations of CS current mapping demonstrate that theoretically the proposed method is feasible. In addition, it is shown that current maps can be acquired with even 40% of the measurements a standard LBIC system would require, saving a significant amount of measurement time. The performance of CS current mapping is the same, regardless of the features a sample may contain and measurements can be applied to any type of photovoltaic device. The ability of the method to provide current maps of PV modules is demonstrated. The performance of several reconstruction algorithms is also investigated. An optical measurement setup for CS current mapping of small area PV devices was built at the National Physical Laboratory (NPL), based on a digital micromirror device (DMD). Accurate current maps can be produced with only 40% of the measurements a conventional point by point scan would need, confirming simulation results. The measurement setup is compact, straightforward to realise and uses a small number of optical elements. It can measure a small area of 1cm by 1cm, making it ideal for current mapping of small research samples. A significant signal amplification is achieved, since the patterns illuminate half of the sample. This diminishes the use of lock-in techniques, reducing the cost for current mapping of PV devices. Current maps of an optical resolution up to 27μm are acquired, without the use of any demagnification elements of the projected pattern that the DMD generates. v A scale up of this new current mapping method is demonstrated using Digital Light Processing (DLP) technology, which is based on DMD chips. A commercial DLP projector is utilised for building a proof of concept CS current mapping measurement system at the Centre of Renewable Energy Systems Technology (CREST). Current maps of individual PV cells in encapsulated modules can be acquired, something that is extremely difficult to achieve with conventional LBIC systems. Direct current mapping of a PV module with by-pass diodes is successfully applied for the first time. Specific shading strategies are developed for this purpose in order to isolate the cell under test. Due to the application of compressive sampling, current maps are acquired even if the signal-to-noise-ratio levels are so low that a point by point scan is not possible. Through the above implementations of CS current mapping of this work, the proposed technique is studied and evaluated. The results demonstrate that this novel method can offer a realistic alternative approach for current mapping of PV cells and modules that can be cost effective and straightforward to implement. In addition, this work introduces the application of the CS theory and DLP technology to PV metrology in general.

Funding

EPSRC.

History

School

  • Mechanical, Electrical and Manufacturing Engineering

Publisher

© George Koutsourakis

Publisher statement

This work is made available according to the conditions of the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0) licence. Full details of this licence are available at: https://creativecommons.org/licenses/by-nc-nd/4.0/

Publication date

2017

Notes

A Doctoral Thesis. Submitted in partial fulfilment of the requirements for the award of Doctor of Philosophy of Loughborough University.

Language

  • en

Usage metrics

    Mechanical, Electrical and Manufacturing Engineering Theses

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC