UNCERTAINTY OF FIELD I-V-CURVE MEASUREMENTS IN LARGE SCALE PV-SYSTEMS

Daniela Dirnberger, Johannes Bartke, Andreas Steinhuser, Klaus Kiefer, Frank Neuberger Fraunhofer Institute for Solar Energy Systems ISE, Heidenhofstrase 2, 79110 Freiburg, Germany Phone +49 761 4588 5758, Fax +49 761 4588 9758, Email: daniela.dirnberger@ise.fraunhofer.de ABSTRACT: Field I-V curve measurements are an important means of quality assurance. They provide a possibility to verify the actual module power on a representative sample of modules in the field. The result is “worth money”, so it is essential that the uncertainty is kept at a minimum. This requires good knowledge of PV measurement principles, even if measurement equipment is readily available. Operators have to be aware of the influence of environmental conditions during measurements on the result. We performed a detailed uncertainty analysis in order to improve our procedure and measurement equipment. With our procedure, which follows the principles of IEC 60904-1 and applies procedure 1 in IEC 60891, we achieve uncertainties between 2.2% and 5%. This requires traceable primary calibrated measurement equipment and a thorough determination of temperature coefficients and other needed parameters. The major contributions to the combined uncertainty of corrected power are discussed in this paper. As it is important that the performance of thin film technologies can be assessed, we included not only crystalline silicon technologies, but also CdTe and amorphous silicon technologies in our analysis. Keywords: Field I-V curve measurements, Measurement uncertainty, Translation to STC, PV System 1 INTRODUCTION Performing I-V curve measurements of PV arrays has always been a useful means of testing PV installations [1, 2]. Field I-V curve measurements allow testing a large sample of modules right where they are installed, and they reveal not only weak modules but also faulty connections, which differentiates them from laboratory measurements. Periodically performed field I-V curve measurements may also reveal degradation. The situation has changed compared to 15 years ago, when field measurements were mainly performed by well-trained scientists. Today, as considerable sums are invested in large scale PV systems, performing field I-V curves is one of several common quality assurance measures. Their most important outcome is by far the verification of actual module power. The demand of investors for determination of actual power in the field is growing. In order to fulfill this demand, methods to measure field I-V curves have to be non-time-consuming, easy to perform and still give reliable results with small uncertainties. The market offers various kinds of field I-V curve measurement systems which promise to fulfill these requirements. However, reliable field I-V curve measurements require that the operator has good knowledge of the main influences on the measurements, and does not only rely on the I-V-curve tracer and the implemented correction procedure. It is important that the operator is aware of the specific uncertainties of his system and the correction procedure applied. Is this paper, we review the most important points for performing field I-V curve measurements with the purpose of determining the actual module power. We discuss the sources of uncertainty when working according to IEC60891 [3] and present the results of our uncertainty analysis. 2 array, the I-V curve is measurDETERMINATION OF ACTUAL MODULE POWER IN THE FIELD As illustrated in Figure 1 the power determination consists of three steps: the measurement itself, the correction to STC, and the correction of external losses such as soiling, electrical losses or mismatch losses. Even if the latter is a crucial aspect – the assumptions about the external losses directly affect the decision, whether the modules meet label specifications or not – a detailed discussion is beyond the scope of this paper.