TY - JOUR AU - Setiawan, Dedy Kurnia PY - 2015 TI - Pemulihan Tegangan Kedip Menggunakan Dynamic Voltage Restorer (DVR) dengan Kendali Constructive Levenberg Marquardt Neural Network JF - Jurnal Arus Elektro Indonesia; Vol 1 No 1 (2015) KW - N2 - Abstract In this paper proposed an algorithm to control dynamic voltage restorer (DVR). The proposed controller is using an artificial neural network with Levenberg Marquardt method Constructive Neural Network (CLMNN). The purpose of this control is to obtain a DVR with a quick and accurate response while improving the quality of the voltage due to the voltage flicker in low-voltage systems. CLMNN use that is based on the configuration of hidden layer neural network used can be changed to obtain the optimal value. CLMNN used in the detection of the change in voltage conditions, either in the form of fluctuations in the voltage amplitude and phase changes in load. Once detected, CLMNN generate a PWM signal as a control. To determine the performance CLMNN controller, then the simulation is used as a comparison of conventional controllers. Based on the simulation results it is known that overshoot voltage of sensitive load decreased by 20% when using CLMNN. In addition, the DVR is more stable with a faster settling time, ie less than 0.01 seconds, whereas when using a PI controller reaches 0.02 seconds. Keywords— Dynamic Voltage Restorer (DVR), voltage flicker, Levenberg Marquardt Constructive Neural Network   UR - https://jurnal.unej.ac.id/index.php/E-JAEI/article/view/1454