Dual-loop optimization of digital accelerometers based on advanced dual-degree-of-freedom Smith predictor

Area of Science:

• Control Systems Engineering
• Instrumentation and Measurement

Background:

• Traditional digital accelerometer control structures optimize only the disturbance loop, neglecting simultaneous performance enhancement of dual loops.
• A conflict exists in optimizing performance for both the setpoint (balancing) and disturbance (external input measurement) loops concurrently in conventional designs.

Purpose of the Study:

• To propose and validate an advanced dual-degree-of-freedom (ADOF) Smith predictive control for digital accelerometers.
• To structurally isolate the setpoint and disturbance loops for independent optimization.

Main Methods:

• Elaboration of closed-loop digital accelerometer principles and analysis of dual-loop performance contradictions.
• Proposal of ADOF Smith predictive control, comparing it with existing control structures.
• Design of setpoint and disturbance controllers using Dahlin algorithm and H ∞ optimal control for single-parameter tuning.
• Derivation of relationships between control parameters, margin indicators, and robust performance.

Main Results:

• The ADOF Smith predictive control structure demonstrates significantly superior performance over traditional closed-loop and filtered Smith predictive controls.
• Experimental validation confirms the effectiveness of the ADOF control structure and parameter tuning methods.
• The proposed ADOF control exhibits better nominal performance under low robustness parameters due to inherent delay compensation.

Conclusions:

• The ADOF Smith predictive control effectively addresses the limitations of traditional digital accelerometer control.
• Independent loop optimization via structural isolation leads to enhanced overall system performance.
• The proposed control strategy offers a robust and high-performing solution for digital accelerometer applications.