Faulty measurement substitution and control reconfiguration by using a multivariate flow control loop.

Area of Science:

• Control Engineering
• Industrial Automation
• Systems Engineering

Background:

• Instrumentation and control systems require robust monitoring for sensors, equipment, and actuators.
• Developing effective fault detection and control reconfiguration strategies is crucial for system reliability.

Purpose of the Study:

• To design and build a multivariate control loop test bed for research.
• To develop and demonstrate a fault-tolerant control strategy using empirical models.
• To investigate fault detection and control reconfiguration in a real-world system.

Main Methods:

• A two-tank multivariate loop was designed and constructed as a research test bed.
• A fault-tolerant control strategy was experimentally developed and demonstrated.
• Data-based empirical models were used for fault detection and measurement substitution.
• Control reconfiguration was tested in the presence of actuator failure.

Main Results:

• The study successfully demonstrated a fault-tolerant control strategy in a single-tank configuration.
• Empirical models proved effective in detecting faults and substituting faulty sensor data.
• The multivariate control loop system showed resilience under degraded operating conditions.

Conclusions:

• Data-based empirical models are feasible for real-time fault detection and system adaptation.
• Accurate empirical models are essential for a control system to survive degraded conditions.
• This research validates the robustness of multivariate control loops with fault-tolerant strategies.