A NEW SELF-TUNING FUZZY PI CONTROLLER FOR INTEGRATING PROCESSES WITH DEAD-TIME
Published In: INTERNATIONAL CONFERENCE ON ADVANCES IN COMPUTING, CONTROL AND COMMUNICATION
Author(s): RAJANI K. MUDI , RITU RANI DE
Abstract: We propose a robust non-fuzzy self-tuning scheme for fuzzy PI controllers (FPIC) for a class of systems. The output scaling factor (SF) of the proposed non-fuzzy self-tuning FPIC, termed as NF-STFPIC, is modified on-line based on the normalized change of error of the controlled variable. Performance of the proposed controller is evaluated for a pure integrating plus dead-time (IPD) and first-order integrating plus dead-time (FOIPD) processes with a wide variation in dead-time under both set-point change and load disturbance. Detailed performance comparison with conventional PI (both fuzzy and non-fuzzy) controllers as well as a self-tuning fuzzy PI controller (STFPIC) reported in the leading literature is provided with respect to a number of performance indices. Unlike STFPIC, which uses 49 expert's defined self-tuning fuzzy rules our NF-STFPIC uses a single deterministic rule. Experimental results justify the effectiveness of the proposed NF-STFPIC.
- Publication Date: 17-Jun-2012
- DOI: 10.15224/978-981-07-2579-2-426
- Views: 0
- Downloads: 0
UPGRADED TANGO TREE TO SOLVE THE DICTIONARY PROBLEM AND ITS APPLICATIONS
Published In: INTERNATIONAL CONFERENCE ON ADVANCES IN COMPUTING, CONTROL AND COMMUNICATION
Author(s): SUNEETA AGARWAL , V S ANIRUDHA KAKI
Abstract: In 1989 Wilber [2] conjecture a lower bound O (log n) for a query using any balanced existing binary search tree with static data. Later in 2004 Domain et al.., [1] Came up with new lower bound O (loglogn) called interleave lower bound and he also claimed that these two lower bound O (log n) and O (loglogn) acts as a good tight intervals for any binary search tree. Tango tree was recently introduced by Demaine et al., [1], having O (loglogn) - competitive ratio. Tango tree [1] supports only lookup (search) operations, where most of the online algorithms (like dictionary problem, a cache problem, adaptive data compression, etc.,) need additions and removals of collections (key, value) too, which are not supported by tango trees [1]. In this paper, we propose a new upgraded version of tango tree which supports addition and removal of collections (key, value) dynamically without knowing the sequence before hand in O (loglogn) time. We show run-time analysis with experimental results. We a
- Publication Date: 17-Jun-2012
- DOI: 10.15224/978-981-07-2579-2-445
- Views: 0
- Downloads: 0