Intelligent Alarm Handling in Process Plant using Genetic Evolution

Abstract

In this article an attempt has been made to combine Genetic Algorithms (GAs) and Genetic programming (GP) to solve the problem of alarm handling in a given process plant in which various alarms are to be handled judiciously, and accordingly, fast recovery actions have to be taken up. The proposed technique will help the operators when they face situations where many alarms, both unwanted and relevant ones, are activated. Through the use of GP, state in which the system is can be known, and if this is a non-acceptable state, values of control variables are changed through the use of GAs in order to take the system to an acceptable state. Benefit of each individual technique has thus been used in the design of a fully automatic system for handling alarms in a process plant with controlled redundancy. The control parameters are encoded in the chromosome. Simulated results are shown in tabular form and a graph exhibiting the performance of the proposed system during three cases, best of generation, average of generation and worse of generation, are provided.

Indexing terms:

• Hierarchical composition
• Parsetree
• Reproduction
• Crossover
• Mutation
• Standardized fitness
• Alarm system
• Selection
• Postfix Expression

Additional information

Notes on contributors

G P Basal

G P Basal holds the degree of ME in Applied Electronics and ME in Computer Engineering in 2001, he has been awarded PhD degree on the topic: Investigation in design and development of Real Time Knowledge Based System. He possess more than 35 years of experience in teaching undergraduate and post graduate classes. He has published 5 technical papers in the journals of international status and 10 papers in the proceedings of International Conferences and 24 papers in the proceedings of National Conferences/Seminars/Workshops. His fields of interest are; Evolutionary algorithms, Soft computing, Micro-processor applications in process control, Cyber protection using process control methods etc. He is currently appointed as Principal of Engineering College at AD Patel Institute of Technology, New VV Nagar, Anand, Gujarat. He is member of IEEE, ACM, LM of ISTE, Fellow of IETE, SM of ICCPI.

Bhupendra Verma

Bhupendara Verma passed his BE in Computer Technology from SATI Vidisha. He is working as a Lecturer in department of Computer Science and Engineering, SATI Vidisha. Presently he is doing MTech in Computer Science and Engineering. His area of interest is Genetic Algorithms and Neural Networks. He has published some research papers in national and international conferences. He is a life member of ISTE.

A K Tiwari

A K Tiwari passed his BE and ME from GEC Jabalpur and did his PhD from IIT Delhi. He is presently working as a professor in department of Electronics and Computer Science in Maulana Azad College of Technology, Bhopal. His area of interest is CAD of millimeter waves circuits and neural networks. He has published many research papers in national and international journals and guided several PhDs. He is fellow of Institute of Engineers, Institute of Electronics and Telecommunication engineers and member IEEE.

P k Chande

P K Chande graduated in Electrical (Communication) Engi-neering and subsequently earned his Master's and PhD degrees in Electronics Engineering, specializ-ing in Automation. He has 20 years of experience in teaching, research and administration. He has participated in conferences and deliver lectures, including visiting professor at Kumamoto University, Japan. Dr Chande has been honoured for his contributions in the technical fields, such as International Socholarship Award by SICE, Japan, in 1992; Radhkrishnan Award in 1992., Pandit Lajja Shankar Jha Award in 1994; Award of merit at ICAUTO in 1995 and Anna University—Indian Society for Technical Education (ISTE) national award for best academic in 1996 and Bharat Mata Award.