Industry best practice for performance testing of utility scale solar PV plants
Performance testing is an important technical requirement of a solar EPC contract. Not only does it provide assurance to project owners of solar PV plant design and construction quality but also performance testing results are typically tied to guarantees, liquidated damages and provisions set in financing agreements.
The performance ratio (PR) has been one of the most common metrics in determining acceptable performance of solar projects. IEC 61724-1 Photovoltaic system performance – Part 1: Monitoring provides some reasonable suggestions for PR calculation methodologies, however, calculation methods usually proposed by EPC contractors vary from project to project and do not always set a particularly high pass threshold. PR is simple to calculate and flexible in terms of test duration; however, this is only one side of the coin. When negotiating PR calculation requirements in EPC contract, what is typically neglected are the inherent uncertainties associated with variation of weather conditions such as temperature, irradiance and seasonal variability, as well as plant operating conditions, e.g. inverter clipping or non-coplanar irradiance sensors. Such uncertainties can skew the results and inaccurately reflect the performance of a given solar plant.
Considering the importance of performance testing in confirming solar PV design and construction quality and supporting EPC guarantees and financing agreements, DNV GL recommends a different test approach is used in place of PR. This is in part supported by the following international standards.
- ASTM E2848-11 Standard test method for reporting photovoltaic non-concentrator system performance: regression based short term acceptance test
- IEC TS 61724-2 Photovoltaic system performance – Part 2: Capacity evaluation method: non-regression based short term acceptance test
- IEC TS 61724-3 Photovoltaic system performance – Part 3: Energy evaluation method: long term performance test
These standards address different issues associated with short term and long term testing, which are often ignored in PR calculations. The fundamental differences between the short term and long term approaches are the data validation process and the way in which the performance metrics are derived.
ASTM E2848-11 has been around for quite some time and is adopted widely in the US PV market. The standard proposes regression-based approach with a rather rigorous data validation process, which helps reduce the uncertainties associated with the short-term nature of solar PV plant acceptance testing. Perhaps for obvious reasons, ASTM E2848-11 is not widely used in other markets, even though it would result in a lower uncertainty for short term acceptance testing when compared to using PR. IEC TS 61724-2, recently published in late 2016, attempts to close this gap. The IEC Technical Specification shares some similarity with the ASTM standard in terms of data validation, however leaves some of the methodology open to negotiation between the owner and EPC contractor.
IEC TS 61724-3 on the other hand, discusses requirements on long term performance test using and actual-versus-expected energy test. In this approach, a project model and measured weather data are used to provide expected energy and compare it to the actual measured energy. It eliminates the need to rely on historical weather data and the corresponding pre-determined PR threshold.
Recognising that the current standards, as is often the case, set minimum requirements and in some instances, may open to interpretation, DNV GL recommends that further refinements be introduced to the solar PV plant performance testing methodology. This can be done to various aspects of the tests, such as which input data to use, which transfer functions are more robust, or simply how to preserve the project model for use after a few years of operation. All these improvements are designed to close gaps which can increase uncertainty, and in turn ensure a more accurate way to assess the performance of a solar PV plant before takeover.
So, when the industry norm may not be best practice when it comes to PV performance testing, are project developers and owners equipped with the necessary tools for testing the performance of their solar PV projects? To support developers and owners DNV GL has included its recommended performance test methodology in its bespoke EPC Employer’s Requirements for utility scale solar projects in Australia. For more information, click here or contact DNV GL on +61 3 9600 1993 and email email@example.com.