Normalized decoupling control for high-dimensional MIMO processes for application in room temperature control HVAC systems

• Published in: Control engineering practice Vol. 18; no. 6; pp. 652 - 664
• Main Authors: Shen, Yuling, Cai, Wen-Jian, Li, Shaoyuan
• Format: Journal Article
• Language: English
• Published: Kidlington Elsevier Ltd 01.06.2010; Elsevier
• Subjects: Air conditioning. Ventilation; Applied sciences; Approximation; Arrays; Computer science; control theory; systems; Control system synthesis; Control systems; Control systems design; Control theory. Systems; Decoupling; Energy; Energy. Thermal use of fuels; Equivalent transfer function matrix; Exact sciences and technology; Heating, air conditioning and ventilation; High-dimensional MIMO processes; HVAC process; Industrial metrology. Testing; Inverse; Mechanical engineering. Machine design; Normalized decoupling; PI/PID tuning; Relative normalized gain array; Strategy; Transfer functions; Relative normalized gain array; High-dimensional MIMO processes; Equivalent transfer function matrix; HVAC process; PI/PID tuning; Normalized decoupling; Integral proportional control; Control room; Control synthesis; System with n degrees of freedom; Dimensional control; Delay time; Transfer function; Relative gain array; Air conditioning; MIMO system; Transfer matrix; Matrix function; Closed feedback; Dimensioning; Decoupling; Temperature control; Causality; Differential integral proportional control; Room temperature
• ISSN: 0967-0661; 1873-6939
• DOI: 10.1016/j.conengprac.2010.03.006

A novel decoupling control system design for high-dimensional multi-input, multi-output (MIMO) processes is presented. Based on the Relative Normalized Gain Array (RNGA), an Equivalent Transfer Function (ETF) for each element in the transfer function matrix was derived for the closed-loop control system and was used to approximate the inverse of the process transfer function matrix. The decoupler could be easily determined with each element in the First-Order-plus-Time-Delay (FOPTD) form and resulted in a stable, proper and causal decoupled matrix. A PI/PID controller could then be designed to meet the performance objectives. The main advantage of this method was its simplicity; it did not require extensive calculation effort. To demonstrate its effectiveness, the decoupling control strategy was applied to control the temperatures of four neighboring rooms in an HVAC process.