Field and Laboratory Results for a New Pattern Recognition Adaptive Controller

This paper presents laboratory and field test results for a new pattern recognition adaptive controller (PRAC) that adjusts the gain and integral of proportional-integral controllers while under closed loop control. The laboratory results demonstrate how PRAC tunes a static pressure control loop with aggressive and sluggish initial conditions. Field test results are presented for a static pressure control loop, supply air temperature control with a heating coil, supply air temperature control with dampers, and supply air temperature control with a cooling coil. PRAC is easy to use and provides near-optimal performance for a range of systems and noise levels. Also, PRAC is computationally efficient and does not have large memory requirements. Thus, PRAC can be used in todays digital control systems. Using PRAC to control HVAC processes will result in a number of economic and environmental rewards: time for commissioning new control systems will decrease, time for retuning control loops will be eliminated, actuator life will increase, energy use will decrease, and indoor air quality and the safety of building occupants will improve. PRAC has successfully tuned control systems for heating, ventilating, and air-conditioning equipment in office buildings, high schools, universities, national laboratories, department stores, hospitals, clinics, and large sports stadiums.