Recompense Load Demand and Improve Power Quality in Wind Connected System Using JNN Based FACTS Device
摘要
Integrating renewable energy sources to the electricity grid presents additional obstacles even though they have been considered to an alternative energy source. However, because of the state of the environment, power generated from renewable energy sources fluctuates constantly. Similar to this, adding wind power to an electric grid has an impact on the quality of the power because of the wind’s erratic behaviour and the relatively recent developments in wind generator technology. This includes reactive power, active power, voltage sag, flicker, voltage swell, harmonics, and electrical behaviour during switching operations. In order to address these problems, a novel learning network controller was introduced to control the compensator operation. The suggested approach aims to mitigate power quality issues by connecting FACTS devices, such as SVC, STATCOM, and UPQC at a common coupling point. To enhance the quality of power, the grid-connected wind energy resulting in the system can be managed through learning network approach based FACTS device. Wind system and its corresponding loads are designed as per a specific value. In between source and load, various FACTS devices are linked to improve power quality. Creating various faults like sag, interruption and swell to generate a dataset which contains the voltage value of the coupling point. Based on this data, a Jordan neural networks (JNNs) model is designed to control the FACTS action. The voltage values of the coupling point were analyzed at every second in order to provide a pulse signal for the FACTS device, which retains the power flow in the system constant. To evaluate the performance of the proposed model at various fault conditions including swell, sag, harmonics, and interruptions. The compensators supply low harmonic content, which was noted as STATCOM having 2.10