Modeling and Simulation of Insolation for Solar PV and Thermal Based Power Plants in India

Abstract

Present research has made recommendations on optimal CSP and PV plant configurations (with respect to solar multiplier, storage, and hybridization), with respect to different solar radiation data sources for India. Present research first lay out arguments for choosing an existing solar resource data set for India based on a number of sources, then applying that data to models that simulate PV and CSP plant configurations at specific locations. A brief introduction, historical background and the modelling techniques for solar plant design are described. There are several data sources for weather and solar radiation in India. In India, solar resource data is drawn from various sources. These include recently installed state-of-the-art 51 solar meteorological stations throughout the country, Solar Radiation Handbook particularly Solar Radiant Energy over India prepared jointly by IMD and MNRE, the Indian Meteorological Department, National Aeronautics and Space Administration (NASA’s) Surface Meteorology and Solar Energy data set, METEONORM’s global climatological database, and satellite derived geospatial solar data products from the United States National Renewable Energy Laboratory (NREL), MNRE and IMD joint venture and several satellite based data service providers. One of the major sources of uncertainty in satellite derived solar data as well as groundmeasurements is the extensive amount of dust and haze that occurs over India. This can have an impact on the future viability of CSP and PV technologies in India, and impact on plant operations and outputs. Other major challenge in the development of solar power in India is the absence of ground solar radiation data. Impact of Solar PV and thermal power plants in India needs detailed modeling and simulation techniques for design purposes. This thesis embodies the subject matter resulting out of this study. NASA data and RETScreen software are used to quantify the impact of solar radiation on the technical configuration of different solar PV and thermal plants. Simulation scenarios are run for various sites in India for technical and financial viability of solar power generation with photo-voltaic (PV) technology. RETScreen model is run for various simulation scenarios on the feasibility of sites in India to build a 5 MW PVgrid connected power plant from techno-economical and environmental points of view 6 are discussed. A model is run for 31 major sites with varied insolation in India to measure the viability of Solar PV plants at these sites. Financial incentives announced in national solar mission of India have also been used as an input to the model. Viability indicators like internal rate of return (IRR), net present value (NPV), cost of electricity (CoE), and benefit cost (B-C) ratio are identified on the basis of the model. A comparison of results is done and the best sites in India are reported. SAM (SAM, 2011) developed by Sandia laboratories and National Renewable Energy Laboratory (NREL) is used for the weather data, cost estimates, and local specific assumptions. NREL solar radiation data and SAM model have been used for simulations to develop a cost minimization model to investigate the techno-commercial viability of a 100 MW parabolic trough solar thermal plant (PTST) for power generation at different sites in India. Cost minimization model prescreens the solar potential in India on the basis of parameters like slope, direct normal irradiation (DNI), protected areas etc. In order to identify the least costly feasible option for the installation of the PTST, a parametric cost–benefit analysis was carried out by varying parameters, such as capacity factor, capital investment, operating hours, etc. Cost minimization model focuses on the assessment of CSP potential in India based on high resolution NREL satellite DNI data uses SAM to optimize and measure the performance of CSP plants. Preliminary screening of sites is done based on the climate and geographical parameters and 25 sites are shortlisted. Financial incentives announced by government of India have been considered as an input to the model. The proposed model predicts various viability indicators like IRR, NPV, CoE, B-C ratio for 100 MW solar CSP at 25 sites in India. A comparison of the results is done and Jodhpur, Rajkot and Indore are found to be the best sites for setting up of 100 MW solar CSP plant in India. On the basis of the present study, it is recommended to erect the PTST power plant in the western region of India. The next section investigates the economic and financial feasibility of a 100MW parabolic trough concentrating solar power plant with energy storage at Jodhpur, Rajasthan, India. Base case scenario is defined as a 100 MW plant without thermal energy storage. The resulting Power Purchase Agreement (PPA) price comes out to be 34.17 US cents/kWh, barely within acceptable range of specified feed-in tariff indicated by the Indian Government in Jawaharlal Nehru National Solar Mission (JNNSM). This 7 implies that although the system configuration in the base case may be financially feasible, it yields lower returns. Scenario analysis with varying solar multiples and thermal energy storage has been done in the present study. In the last section, it is concluded that the lowest PPA price of about 31.4 US cents/kWh can be achieved in a system configuration having a solar multiple of two and equipped with three hours of thermal energy storage system. Sensitivity analysis is done to measure the impact of uncertainty of solar radiation on project economics and performance. The uncertainty in measurement and prediction of solar radiation has a direct impact on the levelized cost of electricity (LCOE) and capacity factor of the 100 MW parabolic trough plant. For the optimal design, six different plant configurations have been compared for an initial analysis. The different plant designs include 2 solaronly plants (differing in solar multiples) without storage or hybridization, two solar-only plants with storage capacity of four and eight hours respectively, and one solar plant integrated with dry cooling technology. Trough plant steam cycle performance is modeled in cost minimization model and SAM using Jodhpur, Rajasthan as the reference site. Hourly net electricity, fossil energy consumption, and ambient temperature are estimated. The total cost for the installation of solar hardware is based on Central Electricity Regulatory Commission (CERC) cost assumptions. The results obtained from the present study provide information to establish technical criteria for the design of CSP plants, which optimizes the solar electricity produced and its generation cost. Based on the present study, it is recommended that CSP projects could include 50-100 MW capacity project with a minimum of four hours of storage, one 100 MW hybrid project with a maximum of 30 % gas fraction and one 20-25 MW project employing 100 % dry cooling technique and minimum water consumption. Accurate DNI database is the one of the dominant problem in the deisgn and simulation of CSP plants in India. Accuracy to within 5% is currently out of reach of satellite-derived data without additional ground readings. Due to inter-annual variability, 10-20 years of DNI data is needed for project-specific data to reach such accuracy. Although satellite-datasets cover many regions, and can help to identify good sites, most of them are biased, meaning they have significant systematic errors. Thus, although satellite DNI maps can be used for site-selection, these need to be verified by qualified 8 measurements at the project development stage. Since at least one year of measurements are recommended for due diligence, this can slow down development unless a suitable meteorological station is installed as soon as possible. One solution is to use multiple datasets to find a quality-weighted best estimate. This means combining well maintained, calibrated and screened ground-based measurements and qualified time-series satellite data to analyse long-term variability at proposed sites. Also important is the benchmarking of satellite-derived DNI products with sound measurements.