Design and tuning of a Decentralized Fuzzy Logic Controller for a MIMO type Pressurized Heavy Water Reactor

作者：Wallam Fahad; Abbas Adeel; Abbasi Abdul Rehman^*

来源：Annals of Nuclear Energy, 2015, 85: 1104-1114.

DOI：10.1016/j.anucene.2015.07.028

• DFLC
• PHWR
• MIMO
• NMA
• Tuning
• Optimization

A Pressurized Heavy Water Reactor (PHWR) is a highly complex and unstable system. Designing a safe, reliable and robust controller with good performance for such a large and complex system is an important control engineering problem. In this work, a Decentralized Fuzzy Logic Controller (DFLC) with 140 input and 70 output membership functions, is designed for a 70th order Multi-Input Multi-Output (MIMO) type PHWR. In order to obtain high performance of the controller, it needs to be tuned optimally, however, it is very challenging task to optimally tune the DFLC with such a large membership functions. Moreover, PHWR is a coupled system which imposes additional limitation in tuning the controller since the output of one PHWR's zone affects the outputs of other zones. In this work, an application of Nelder-Mead Algorithm (NMA) is presented for auto tuning the DFLC. The NMA performance depends upon objective function and initial points given to the NMA at the start of the tuning process. A novel method for selecting the optimal objective function and initial points for the NMA is also proposed since their selection is another complicated process. Although several objective functions have been proposed by the researchers for use with NMA, this work focuses five common indices (IAE, ISE, ITAE, ITSE and ISTE) as objective functions, which are simple and system independent. Finally, the optimally tuned high-performance DFLC is applied to the PHWR and evaluated by simulating different scenarios. The simulation results show that the controller is efficient, fast and robust and ensures the safety and reliability of the PHWR.

• 出版日期2015-11