model based design and simulation of control strategies to maximize the pv hosting capacity in isolated diesel networks using solar short term forecasts for predictive control of diesel generation

• DOI: 10.4229/eupvsec20162016-6eo.1.4
• Publication Date: 2016-01-01
• Authors: K. von Maydell; Dorothee Peters; Carsten Agert; Rasmus Völker; T. Kilper; C. Carter; Martina Calais; Thomas Schmidt
• Keywords: PV Applications and Integration; 13. Climate action; 0211 other engineering and technologies; 02 engineering and technology; 7. Clean energy; PV Applications Without a Centralised Grid

32nd European Photovoltaic Solar Energy Conference and Exhibition; 2532-2535<br/>Increasing Photovoltaic (PV) system penetration in small, isolated networks faces a number of technical, regulatory and economic challenges but also opportunities. The replacement of costly and polluting diesel generation is highly desirable, however, a key technical issue is system stability, particularly on intermittent cloudy days. In this context, the step load response capability and spinning reserve provided by diesel generation are critical to cover for the sudden loss of PV power. This work develops and tests the application of short-term forecasts using ‘sky imagers’ in the control of generation capacity. The benefits for system stability shall be examined using an example isolated network modeled in PowerFactory that simulates frequency and voltage at the generator busbar. To reduce forecast errors, ensemble forecasts for all PV systems in the network are used. The results show a favourable low impact of PV power on voltage and frequency, if realistic ramp rates for dispersed PV systems are used. The envisaged fuel savings from reducing spinning reserve requirements are achieved but stay below 1 %. With the inclusion of short-term storage capacity, further benefits are expected. In addition, the high dependency of possible PV penetration on generator dispatch and loading is shown.<br/>