<![CDATA[AbstrakPemberian pakan ikan dapat ditentukan melalui pengamatan perilaku, kualitas air, dan ukuran ikan. Salah satu metode otomatisasi yang digunakan adalah ANFIS (Adaptive Neuro Fuzzy Inference System), dengan masukan berupa kualitas air (kekeruhan dan NH3) serta aktivitas gerak ikan menggunakan inframerah (IR). Sistem ini mendukung keputusan pemberian pakan secara optimal. Validasi alat dilakukan menggunakan model regresi linier dan evaluasi kinerja berdasarkan Mean Relative Error (MRE). Hasil menunjukkan akurasi perangkat mencapai 95,77%, lebih tinggi 6,55% dibandingkan perangkat tanpa kualitas air (89,22%). Model ini terbukti andal dan dapat diterapkan pada sistem akuaponik berbasis industri untuk meningkatkan efisiensi pemberian pakan ikan. Kata kunci: Akuaponik; biologi otomatis; rekayasa sistem; robotika; instrumentasiAbstractFish feeding can be determined by observing behavior, water quality, and fish size. One automation method used is ANFIS (Adaptive Neuro Fuzzy Inference System), which uses inputs such as water quality (turbidity and NH3) and fish movement activity detected by infrared (IR). This system supports optimal feeding decision-making. The tool validation was conducted using a linear regression model, and its performance was evaluated based on the Mean Relative Error (MRE). The results showed that the device achieved an accuracy of 95.77%, 6.55% higher than devices without water quality input (89.22%). This model has proven reliable and can be applied to industry-based aquaponic systems to enhance the efficiency of fish feeding.Keywords: Aquaponics; automated biology; systems engineering; robotics; instrumentation]]>
Copyrights © 2024
