Electronic Theses and Dissertation

Dalam sistem otomasi, motor DC banyak digunakan sebagai aktuator karena kemampuannya mengatur kecepatan secara akurat, namun fluktuasi beban dapat menurunkan stabilitas sistem. Penelitian ini bertujuan untuk mengoptimasi parameter pengendali PID menggunakan algoritma genetika (GA) berbasis multi transfer function serta mengevaluasi performanya pada variasi beban nyata. Metode yang digunakan meliputi pemodelan sistem motor DC, optimasi parameter PID menggunakan GA dengan fungsi objektif Integral of Time-weighted Absolute Error (ITAE), serta pengujian simulasi dan eksperimen pada kondisi tanpa beban, beban ringan (100 gram), dan beban berat (300 gram). Hasil optimasi menghasilkan parameter PID optimal yaitu Kp = 1,9361, Ki = 1,4719, dan Kd = 0,0183. Pada pengujian tanpa beban, model tunggal menghasilkan overshoot 9,19% dan ITAE 14,9581, sedangkan model multi transfer function menghasilkan 4,89% dan 16,4952. Pada beban ringan, model tunggal menghasilkan 18,76% dan 25,6879, sedangkan model multi menghasilkan 14,01% dan 19,4926. Pada beban berat, model tunggal menghasilkan 18,90% dan 32,9437, sedangkan model multi menghasilkan 14,69% dan 27,9574. Pada perubahan beban langsung, model multi juga menunjukkan overshoot 6,45% dan ITAE 100,6118, lebih baik dibandingkan model tunggal sebesar 10,62% dan 113,2819. Dengan demikian, pendekatan multi transfer function memberikan respon sistem yang lebih stabil terhadap variasi beban.

In automation systems, DC motors are widely used as actuators due to their ability to regulate speed accurately; however, load variations can reduce system stability. This study aims to optimize PID controller parameters using a Genetic Algorithm (GA) based on a multi transfer function approach and to evaluate its performance under real load variations. The method includes DC motor modeling, PID parameter optimization using GA with the Integral of Time-weighted Absolute Error (ITAE) as the objective function, and both simulation and experimental testing under no- load, light-load (100 g), and heavy-load (300 g) conditions. The optimization results yield optimal PID parameters of Kp = 1.9361, Ki = 1.4719, and Kd = 0.0183. Under no-load conditions, the single model produces 9.19% overshoot and 14.9581 ITAE, while the multi model produces 4.89% and 16.4952. Under light load, the results are 18.76% and 25.6879 for the single model, and 14.01% and 19.4926 for the multi model. Under heavy load, the results are 18.90% and 32.9437 for the single model, and 14.69% and 27.9574 for the multi model. Under sudden load changes, the multi model also shows better performance with 6.45% overshoot and 100.6118 ITAE compared to 10.62% and 113.2819. These results indicate that the multi transfer function approach provides a more stable response to load variations.