Prediksi produksi padi menjadi penting dilakukan untuk menunjang pembangunan nasional sektor pertanian pada suatu negara atau wilayah. Artificial Neural Network (ANN) termasuk metode yang terbaik dalam melakukan prediksi. Masalah utamanya adalah bagaimana menentukan jumlah neuron dan hidden layer yang optimal sehingga akurasi prediksinya tinggi. Artikel ini bertujuan untuk merancang arsitektu ANN unutk melakukan prediksi terhadap produksi padi menggunakan ANN dengan algortima backpropagation. Tahapan penelitian yang dilakukan adalah mengumpulkan data produksi padi, melakukan pre-processing data, memproses prediksi, dan pengujian akurasi dan error serta implementasi. Dalam memproses prediksi dilakukan sesuai dengan rancangan model prediksi, yaitu parameter epoch, momentum, learning rate, hidden layer untuk menghasilkan keakuratan yang tinggi. Temuan yang diperolah berupa rancangan optimal untuk melakukan prediksi yaitu dengan menggunakan multilayer. Hasil pengujian sistem prediksi produksi padi yang terdiri dari 75 kali pengujian pada di 19 daerah di Sumatera Barat, diperoleh tingkat akurasi mencapai 88,14% atau dengan tingkat error yang relatif rendah yaitu 11,86%.
<p><em>T</em>he paper presents grid-tied PV monitoring system using wireless sensor networks. The temperature and humidity parameters were measured using DHT22. While ACS 712 5A current sensor and Arduino voltage sensor modules were used to measure photovoltaic output current and voltage respectively. Web application has been developed in the base station using PHP programming web server to access the sensor nodes through Zigbee wireless data communication. The user can access the HTML web interface of Photovoltaic monitoring system through local Ethernet or Wi-Fi connection. The residential 1.25 kWp grid connected photovoltaic system used to test the developed monitoring system. The data received exactly same as data sensed from remote area with average delay time 3 to 4 seconds. The result shows the photovoltaic power generation caracteristics under clear sky, cloud cover, and rainy weather conditions. The power 608.12 Wp has been generated by the solar panel from 7.00 am to 6.00 pm or 6.7 kWh per day during clear sky. During intermittent cloud covered the photovoltaic power graph have shown fluctuation power profile and energy conversion were decreased as well as in raining weather condition. The electrical power has been generated by photovoltaic for almost 12 hours per day in tropics area, but energy conversion is highly influenced by weather conditions, especially cloud cover, overcast and rainy.</p>
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.