Performance-Optimized Model-Based Design for Hydraulic Mobile Machinery
Forstedt, Nella (2026)
2026
Automaatiotekniikan DI-ohjelma - Master's Programme in Automation Engineering
Tekniikan ja luonnontieteiden tiedekunta - Faculty of Engineering and Natural Sciences
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Hyväksymispäivämäärä
2026-08-27
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:tuni-202608279315
https://urn.fi/URN:NBN:fi:tuni-202608279315
Tiivistelmä
This master’s thesis investigates computationally efficient control models in mobile machinery. Control models for complex components in mobile machines are typically computationally demanding and require considerable resources from control hardware. The thesis examines how control models can be implemented using computationally efficient modeling methods to reduce their computational load. Computationally efficient model-based design is evaluated in the broader context of product development, while considering the constraints and challenges related to hydraulic mobile machinery in control model development.
To evaluate the performance of the control model, the thesis introduces performance metrics and Simulink tools for identifying computationally demanding model components. To construct an efficient control model, efficient modeling methods presented in research in the field are divided into three categories, corresponding to the optimization of the model environment, structure, and fidelity. Since the methods are applied to real-time control and product development of a hydraulic mobile machine, the modeling methods are evaluated from this perspective, and the most suitable methods are selected for implementation. The functional behavior of the model in the target environment is kept unchanged, and data-based methods and methods with limited interpretability are excluded from the scope of the thesis.
To assess the effectiveness of the modeling methods, the performance of the optimized model was measured in a simulation-based SIL environment and in a more realistic HIL environment using hardware of the target environment. Performance was evaluated based on execution time, CPU utilization, and stack memory usage, with a focus on maximum values. Based on the measurement results, the optimized model improved performance according to all selected metrics. The obtained results differ between the measurement environments, but they consistently indicate performance improvements. Maximum execution time and CPU utilization both decreased by 34% in the SIL environment and by 11% and 1% in the HIL environment. Stack memory usage decreased by 6% in the SIL environment and by 3% in the HIL environment. The differences between the SIL and HIL results may be caused by the effect of other machine functionalities on the measurements and computational load. The results indicate that efficient and optimized modeling can have a positive effect on model performance, in an embedded environment as well. Future development topics for the thesis include the broader implementation of different fidelities as part of model-based design and the use of data-based tools to support different methods.
To evaluate the performance of the control model, the thesis introduces performance metrics and Simulink tools for identifying computationally demanding model components. To construct an efficient control model, efficient modeling methods presented in research in the field are divided into three categories, corresponding to the optimization of the model environment, structure, and fidelity. Since the methods are applied to real-time control and product development of a hydraulic mobile machine, the modeling methods are evaluated from this perspective, and the most suitable methods are selected for implementation. The functional behavior of the model in the target environment is kept unchanged, and data-based methods and methods with limited interpretability are excluded from the scope of the thesis.
To assess the effectiveness of the modeling methods, the performance of the optimized model was measured in a simulation-based SIL environment and in a more realistic HIL environment using hardware of the target environment. Performance was evaluated based on execution time, CPU utilization, and stack memory usage, with a focus on maximum values. Based on the measurement results, the optimized model improved performance according to all selected metrics. The obtained results differ between the measurement environments, but they consistently indicate performance improvements. Maximum execution time and CPU utilization both decreased by 34% in the SIL environment and by 11% and 1% in the HIL environment. Stack memory usage decreased by 6% in the SIL environment and by 3% in the HIL environment. The differences between the SIL and HIL results may be caused by the effect of other machine functionalities on the measurements and computational load. The results indicate that efficient and optimized modeling can have a positive effect on model performance, in an embedded environment as well. Future development topics for the thesis include the broader implementation of different fidelities as part of model-based design and the use of data-based tools to support different methods.