| Summary: | 碩士 === 國立臺灣科技大學 === 電機工程系 === 102 === Non-intrusive load monitoring (NILM) system is an energy demand monitoring and load identification system that only uses voltage and current sensors that are installed at the power service entrance of an electric system. The system is better than traditional intrusive monitoring system because it is able to reduce the cost of sensors and installations. In this study, a real single-phase three-wire unbalanced 220V/110V distribution system model of a residential building is designed and implemented, and some non-intrusive techniques are executed in the Intel Atom Embedded System and a LabView program. To enhance the performance, this thesis proposes Particle Swarm Optimization (PSO) to optimize the parameters of Back-propagation Artificial Neural Network (BP-ANN) for training steady-state power signatures, ex. real and reactive power (P and Q) and current harmonics. The results indicate that the simulation result for using real and reactive power (P and Q) is 75% with actual accuracy of 82.35%, while simulation result for using P, Q, and current harmonics is 98% with actual accuracy of 100% for the single-phase two-wire 110V distribution system; the simulation result for using Ptotal and Qtotal signatures is 97% with 100% actual accuracy; while using Pright, Qright, Pleft, and Qleft has simulation result and actual accuracy of 100% for the single-phase three-wire 220V/110V unbalanced distribution system.
|
|---|
