Fruit fly Optimization with Deep Learning Based Reactive Power Optimization Model for Distributed Systems

Due to the massive number of distributed generators (DGs) combined with the power distribution systems, the difficulty of distribution system (DS) process was raised that resulting in greater needs for reactive power optimization (RPO). The RPO of the distributed systems could minimize the power loss and improve the voltage quality and economical functioning of the distributed systems. In this aspect, this paper presents a novel fruit fly optimization with deep learning based RPO (FFO- DLRPO) model for distributed systems. The FFO-DLRPO technique mainly focuses on the design of convolutional neural network long short term memory (CNN-LSTM) technique to learn the non-linear complex relationship as well as RP control solution. The previous information from DG is utilized to train the CNN-LSTM for identifying the relativity amongst system feature and power controls. For boosting the efficacy of the CNN- LS TM technique, the fruitfly optimiz ation algorithm (FFOA) was utilized. The experimental result analysis of the FFO-DLRPO technique is carried out and outcomes are inspected in several measures. The simulation outcome reported the enhanced performance of the FFO-DLRPO approach on existing algorithms.