Sistem Kontrol Kendaraan Electric Hybrid Roda Dua Terintegrasi Berbasis Kalman Filter

Abstrak

Abstract
Control system presents a significant challenge in maintaining of electric hybrid uses, particularly during electric and gasoline
switching. Specifically, the problem of this lies in the noise on potentiometer sensor, which imitates the throttle angle.
Therefore, it is necessary to design a control system that reduces potentiometer sensor noise for optimization. This control
system design is accomplished by implementing Kalman Filter algorithm in control system. Kalman Filter algorithm is directly
deployed and made through the Arduino IDE (C++ programming language basis). To evaluate the control system design, a
comparison between control systems with and without the Kalman Filter algorithm is carried out on Suzuki Spin 125 with
added electric modifications, such as a 60 Volt/1000 Watt brushless direct current (BLDC) motor, a 60 Voltage/14,7 Ah
Lithium-ion battery, and an Arduino Mega2650 and STM32 board. The comparison results show the control system with
Kalman Filter gives smaller sensor noise signal data compared to the system without Kalman Filter. The control system noise
difference analysis values indicates optimization by 97.3%. Furthermore, the evaluation process includes optimizing control
system performance between the throttle opening angle variable and rpm. The results show Mean Absolute Error (MAE) of
80.5% for electric system and 88.9% for gasoline-engine system. Overall, optimization values exceed the 80%, indicating
algorithm is worthy of implementation. In conclusion, Kalman Filter algorithm can be utilized as an control system
optimization regulating electric hybrid motorcycle. This approach holds great potential of energy management and reducing
sensor noise, thereby contributing to the efficient operation of electric hybrid vehicle.
Keywords: Hybrid Electric Motorcycle, System Control, Kalman Filter, MAE.