Final report of the research project "MeQuSo, Development of methods for quality assurance of large-scale solar thermal plants under real operating conditions". The project MeQuSo was supported by the Austrian Climate and Energy Fund and carried out as part of the Energy Research Program (FFG 848 766). Project start 2015-08-01, project end 2019-06-30. Abstract Large solar thermal plants (> 500 m² collector area, > 350 kW thermal power) are an important technology to provide renewable heat. The market has experienced considerable growth in the last decade, with solar district heating (SDH) applications in Denmark as the driving force. Austria has successfully established a market for large solar thermal plants and is home to one of the leading flat plate collector producers in the world. Building new large solar thermal plants requires high initial capital investment, which is paid back over the lifetime of the plant (typically 20 to 25 years) with the revenue from the produced solar heat. To ensure consistently high solar energy yields over the lifetime of the plant, both quality assurance at the start of the operation phase (commissioning) and on-going surveillance of the collector array performance are essential. If the solar yield is below expectations, possible reasons (weather conditions, malfunctioning components, soiling and dust, plant design, control strategy, inaccurate solar yield simulation) need to be distinguished to identify the responsible party and assess costs and benefits of optimization measures. The main outcome of the project MeQuSo is a proof of concept of a new in-situ collector array test method called D-CAT (Dynamic Collector Array Test). The method is able to characterize a collector array with a set of parameters that describe the collectors' behavior, based on real-world operating conditions (measurements directly at the installation). D-CAT provides a parameter-based characterization of collector arrays in the field, much like what the quasi-dynamic/Solar Keymark test does for single collectors in the laboratory. Performance relevant factors which concern the collectors as a component (e.g. soiling, broken insulation, faulty foil tension etc.) are reflected in the test parameters. For instance, collectors with broken insulation or faulty foil tension will have higher heat loss coefficients. The D-CAT method disentangles collector factors from other factors that do not accrue to the collector as a component (e.g. weather conditions, operating conditions, collector array geometry etc.). The D-CAT method can be used at plant commissioning (comparison with the data sheet), for on-going surveillance (changes in collector characteristics), improved model-based/model-predictive control and to build systematic knowledge on the collector performance under real-world conditions. D-CAT can be used in combination with thermal power and energy yield guarantees to offer clarity between collector manufacturers and plant operators. D-CAT is applicable to collector arrays with flat plate collectors using measurement data from the normal, fully dynamic plant operation without the need to run special test sequences, and without interfering with system control in any way. The test can be run fully automatically and can be repeated easily at periodic intervals, or before and after important plant events. The “FHW” large solar thermal plant is located in Graz (Austria) and feeds solar thermal energy into the Graz district heating system. In the MeQuSo project, the FHW plant has been equipped with high-precision measurement instrumentation and monitored for three years. Six separate collector arrays consisting of high-efficiency flat plate collectors of five European manufacturers were measured and analyzed. In MeQuSo, two dynamic collector array models have been developed; these models adapt to the specific needs of collector arrays (vs. single collectors), explicitly modeling the fluid transport along the main flow direction. The models have been validated against experimental data from the FHW plant. The model parameters are estimated using a grey-box approach. In the estimation process, a statistically optimal data selection method is used to obtain those data intervals which contain maximum information for the estimation of the model parameters (lowest-variance and best-separable parameters). The estimation process is based on a global optimization algorithm. The D-CAT method also has a radiation model which allows calculation of the average beam and diffuse irradiance on the collector array, using only total tilted irradiance as input. This makes the method applicable to typical measurement setups for large solar thermal plants, when only the total tilted irradiance is available. Steps for future work include applying the D-CAT method to a larger number of LST plants and gain experience with D-CAT based KPI calculations as well as long-term performance evaluations. Developing D-CAT as an open-source software tool would help to make D-CAT available to other scientific institutes for collaborative development on a European level and disseminate D-CAT on the solar thermal market. How to read this report: The report starts with an introduction on quality assessment of large solar thermal plants in chapter 1 and then gives an overview of existing test procedures and performance check methods to define the research need in chapter 2. Chapter 3 is dedicated to the measurement of the FHW plant. Chapters 4 to 8 contain a detailed description of the D-CAT method, and chapter 9 shows the application and validation of the method. Chapters 10 and 11 conclude with a summary of the main outcomes, a discussion of lessons learned and an outlook towards further research and applications.