System Identification and Controller Design for Hydraulic Actuator

Abstract: System Identification of Hydraulic Actuators is critical for analyzing their performance and designing a suitable Control System. Hydraulic actuators are extensively used in many applications, ranging from flight simulators, robotics, orthopaedic surgery, material testing, construction and many other industrial types of machinery. In the aviation industry, hydraulic actuators are currently being used in full flight simulators used for controlling the position and orientation of the motion platform. Every actuator has its own characteristics, therefore, the choice of excitation signals for System Identification must take into account the dynamics of the actuator under consideration. This work proposes the selection of excitation signals based on bandwidth of the hydraulic actuator. Validation of the proposed selection is done by performing system identification, obtaining a mathematical model and comparing it with a nonlinear hydraulic actuator model designed in Simscape. After validation, a nonlinear PID control has been tuned on the identified model and tested on the nonlinear model. Extensive simulations have been run and results show accurate mathematical modelling, as well as precise control has been achieved through the proposed methodology.

• The paper presents a novel approach to model hydraulic actuator dynamics using optimized excitation signals and system identification techniques.
• It validates model accuracy through Simscape simulations, comparing chirp and multisine responses with low RMSE values.
• An NPID controller with Ziegler-Nichols tuning is applied for precise position tracking, highlighting advances in control design.

System Identification and Controller Design for Hydraulic Actuator

Abstract

The study presents an analysis of system identification and controller design for electro-hydraulic servo actuator systems (EHSAS), which are vital in diverse industries such as aviation. It discusses the complexity involved in modeling these systems due to their dynamic nature and proposes a methodology for selecting excitation signals based on the system's bandwidth. The research evaluates the modeling accuracy through system identification techniques and validates with nonlinear models simulated in Simscape.

Introduction

Hydraulic actuators serve critical roles in industries due to their reliability and efficiency, particularly in flight simulators and industrial machinery. Precise control of these actuators is a significant focus area in dynamic modeling and control research. System identification is required to model the EHSAS accurately, which enables efficient control despite challenges like nonlinearity, friction, and noise.

The paper emphasizes using black-box modeling, with excitation signals tuned to the hydraulic actuator's dynamics for enhanced model fidelity. Post identification, a nonlinear PID (NPID) controller is applied to achieve precise motion control, with the research backing its claim through simulations that indicate high accuracy in mathematical modeling and control execution.

Figure 1: Example of a flight simulator with individual hydraulic actuator.

Methodology

The methodology targets dynamic modeling of EHSAS by leveraging software tools such as Simulink for nonlinear system simulation. The approach involves:

• Dynamics and Nonlinear Modeling: Modeling the hydraulic system, considering comprehensive dynamics, and simulating it with actual parameters using the Simscape toolbox.
• System Identification: Utilizing excitations like chirp and multisine signals for system dynamics mapping. The excitation parameters are rationalized based on bandwidth and resonance frequency, unlike previous methods relying on trial and error.

  Figure 2: Basic flow of EHSAS.

Experimental Validation and Controller Design

Validation of model accuracy is performed by comparing its output against the nonlinear model's. The paper highlights the RMSE between measured outputs from chirp and multisine signals, demonstrating high fidelity in captured dynamics.

Controller Design: An NPID controller is employed on the identified model to fine-tune the position tracking and performance parameters. Using Ziegler-Nichols tuning enhances the controller's response to track reference signals effectively.

Figure 3: Step Response Comparison for Chirp Identified Model.

Implications and Future Directions

This research establishes a precedent for selecting excitation signals concerning actuator dynamics, fundamentally increasing system identification accuracy. It presents a framework for subsequent works to explore varying models of actuators and control designs.

Experimentation with different control algorithms, their hybridization with AI methods, and real-world application validations form the basis for future advancements. Such continuous development could potentially iterate on this method's efficacy and adaptability across diverse actuator models.

Conclusion

The research offers a comprehensive framework for system identification and controller design in hydraulic actuators. By correlating excitation signals with system dynamics, it achieves unparalleled accuracy in modeling, which is crucial for implementing effective control systems in industrial and aviation applications. Future efforts will focus on extending these methodologies to real-world applications to further validate and refine the findings.