Advanced Calibration of Multi-System Hot-Wire Anemometers Using PID Temperature Control and ANN Techniques

MEMS wind sensors manufactured from polysilicon semiconductors require precise operating conditions, primarily temperature control, to ensure accurate performance in hot-wire anemometers. These sensors operate mainly in two modes: Constant Temperature Anemometer (CTA) and Constant Current Anemometer (CCA). While CTA offers high precision and fast response, it is often constrained by the complexity, cost, and power consumption of its circuitry. To address these challenges, this paper presents a novel temperature control system based on a Peltier module regulated by a Proportional Integral Derivative (PID) controller. To enhance accuracy, a hexagonal hot-wire anemometer chip structure was selected and mounted on a Peltier module to maintain a constant, stable operating temperature during calibration and measurements. This procedure proved to be effective, reducing the system complexity and power consumption while enhancing response time and accuracy, as demonstrated through experimental validation. The system was calibrated in a wind tunnel. Subsequently, the Artificial Neural Networks (ANNs) technique was adopted to estimate wind parameters (speed and direction) while compensating for the effect of ambient temperature variation. The results confirm the accuracy and reliability of the proposed system.