Suwoko Suwoko
Mahasiswa Jurusan Teknik Mesin, Fakultas Teknik, Universitas Diponegoro

Published : 2 Documents
Articles

Found 2 Documents
Search

Error Analysis of ON-OFF and ANN Controllers Based on Evapotranspiration

TELKOMNIKA Indonesian Journal of Electrical Engineering Vol 12, No 9: September 2014
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Original Source | Check in Google Scholar | Full PDF (860.336 KB)

Abstract

This paper presents error analysis of ON-OFF controller as well as ANN controller which is applied to an automated irrigation system. Controlling irrigation system was conducted  based on revised Penman-Monteith evapotranspiration and compared to a reference of soil moisture. Input parameters of the evapotranspiration included temperature, heat radiation, atmosphere pressure and wind speed. The output data of evapotranspiration was then used as input variable of ON-OFF controller as well as ANN controller. The performance of such controllers were evaluated and compared based on error of both controllers. The simulation results show that the error of the ON-OFF controller can be adjusted by setting the sampling time of the dead zone discretization. Error of ON-OFF controller with sampling time of 0.05 second is equal to error of ANN controller, that is 14.3% of reference signal amplitude.

RANCANG BANGUN SISTEM PENGENDALI IRIGASI BERBASIS ANALISIS EVAPOTRANSPIRASI DENGAN KONTROLER PI

JURNAL TEKNIK MESIN Vol 3, No 1 (2015): VOLUME 3, NOMOR 1, JANUARI 2015
Publisher : Jurusan Teknik Mesin, Fakultas Teknik, Universitas Diponegoro

Show Abstract | Original Source | Check in Google Scholar

Abstract

Food is a basic requirement for life so that the availability is absolutely necessary of all time, while the source of the water decreases and becomes scarce during the dry season. One way to meet the water needs of plants properly is to calculate the crop evapotranspiration with Penman-Monteith equation. While the tools to realize the results of these calculations is to create an irrigation controllers system. Irrigation control system that is made has inputs from weather environment obtained with the sensor and has an output valve to drain the water. The type of valve used is linear and also installed on/off valve for comparison that performance. The main parameters are used as input air temperature, wind speed, relative humidity, and solar radiation. All parameters acquired digitally except solar radiation, solar radiation are acquired from correlation shaping using artificial neural networks algorithm with input temperature and light intensity. Artificial neural networks are used feedforward structure with neurons numbered 15. The controller used for each valve is on/off valve is controlled by the controller on/off and linear valves are controlled by a PI controller with constant Kp = 1.15 and Ki = 0.95.The test results are established neural network showed good results, where the error is 1.13%. The simulation results of the control system on/off produces an error of 0.68% and PI control system generates an error of 0.02%, while the application of control system on/off produces an error of 2.20%, and the PI control system generates an error of 1.74%