CFChyannie Amarillio Fahdzyana

Chyannie Amarillio Fahdzyana

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Chyannie Amarillio Fahdzyana has a total of 5 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Decomposition-Based Integrated Optimal Electric Powertrain Design

    by Chyannie Amarillio Fahdzyana, Mauro Salazar, M.C.F. Donkers, Theo Hofman

    This paper presents a decomposed co-design optimization framework for an electric vehicle powertrain equipped with a Continuously Variable Transmission (CVT). We jointly design the sizes of the powertrain components and the energy management strategy, considering the design and the dynamics of the transmission. Specifically, we divide the resulting nonlinear optimization problem into separate hierarchical optimization subproblems leveraging the concept of Analytical Target Cascading (ATC). Furthermore, we compare the proposed decomposition-based strategy with a centralized approach, in which we solve the same problem, and show that the method presented in this paper can be a potential alternative to a centralized optimization formulation for solving the large-scale, complex integrated plant and control system design. Finally, we compare the resulting performance of using a CVT on the vehicle total-cost-of-ownership with that equipped with a Fixed-Gear Transmission (FGT), which is what is commonly implemented in current electric vehicles. Our results show that by jointly optimizing the design of the transmission along with the control strategy, the CVT yields a lower energy consumption and total-cost-of-ownership compared to that of the FGT.


  • A Convex Optimization Framework for Minimum Lap Time Design and Control of Electric Race Cars

    by Olaf Borsboom, Chyannie Amarillio Fahdzyana, Theo Hofman, Mauro Salazar

    Lap time is the most relevant performance indicator in electric racing. In order to minimize it, it is important to carefully design the powertrain of the race car in terms of sizing and technology, and meticulously administer the energy available on-board. To facilitate efficient design, this paper presents a convex optimization framework to rapidly compute the minimum-lap-time control strategies for a battery electric race car. Specifically, we first identify a convex model of the electric powertrain, including the battery, the electric motor, and two transmission technologies: a fixed-gear transmission (FGT) and a continuously variable transmission (CVT). Second, assuming an expert driver, we formulate the minimum-lap-time control problem in a convex fashion for a given driving path, and compute its globally optimal solution with second-order conic programming algorithms. Third, we showcase our framework on the Le Mans track by comparing the performance achievable with an FGT and a CVT, and validate it with nonlinear simulations. Our results show that for the given setup a CVT can balance its lower efficiency and higher weight with a higher-efficiency and more aggressive motor operation, significantly outperforming the lap time achievable with an FGT. Finally, we leverage the computational efficiency of our framework to carry out parameter studies on the components, revealing that optimizing the size of the battery and the motor for the specific scenario can considerably improve the achievable lap time, and that the best transmission strongly depends on the sizing decisions.


  • Open Access

    A Decomposed Co-design Strategy for Continuously Variable Transmission Design

    by Chyannie A. Fahdzyana, Mauro Salazar, Theo Hofman

    This paper presents a decomposed co-design optimization framework to jointly design the geometry and the controller of a Continuously Variable Transmission (CVT), accounting for its low-level dynamics. Specifically, we first devise a model of the CVT and the feedback controller, and use it to formulate the optimal co-design problem with the goal of minimizing the transmission's mass as well as the losses that occur in the system, including the lower-level actuation. Second, we divide the resulting nonlinear multi-objective optimization problem into separate hierarchical optimization subproblems and we leverage the concept of Analytical Target Cascading (ATC) to solve the separate optimization subproblems using an interior-point optimization algorithm. Finally, we showcase our framework on a representative drive cycle. Our results demonstrate that the presented co-design method can achieve a more compact CVT design without compromising the desired ratio trajectory and reducing the overall losses by up to 14%.


  • Integrated Plant and Control Design of a Continuously Variable Transmission

    by Chyannie Amarillio Fahdzyana, Mauro Salazar, Theo Hofman

    This paper presents an optimization framework to design the components and the controller of a Continuously Variable Transmission (CVT) in an integrated manner. Specifically, we aim at reducing the mass of the transmission and the leakage losses that occur in the system. To do so, we first formulate the joint plant and control design problem including the corresponding objectives and constraints. Thereafter, we propose a proportional integral structure for the design of the CVT ratio control. The combined plant and control design problem is formulated as a nonlinear multi-objective optimization problem, and is simultaneously solved using an interior point optimization method. We evaluate the obtained design on the Worldwide Harmonized Light Vehicles Test Cycle (WLTC) as well as on more aggressive driving scenarios, and demonstrate that the optimized CVT design is always capable of realizing the required driving performance. Additionally, we study the impact of the plant design parameters on the control performance by analyzing the coupling strength between the subproblems. Thereby, the pulley radius is found to have the strongest influence in the resulting leakage losses that occur at the variator level. Finally, leveraging the presented design framework, we show that up to 13% and 18% reduction in the CVT variator mass and on leakage losses, respectively, can be achieved without compromising the desired ratio trajectory over a representative dynamic driving cycle.