Design of Oxidative Pyrolysis Control Algorithm Based on Fuzzy Safety Area and Center Definition

Abstract

This article examines an approach to controlling complex dynamic industrial technology using fuzzy logic, which is one of the artificial intelligence “AI” trends helping improve quality of decision making and thereafter the quality of industrial technology management systems. Industrial technologies function under the risk of various accidental violations when process state estimates are complicated and therefore raise a number of control problems, and the solution is aimed at compensating the random prime causes in abnormal situations. Therefore, the solution of dynamic process safety management problems in semi-structured and poorly formalized environments is a relevant and urgent task. An acetylene production control system by oxidative pyrolysis of natural gas is considered with the process peculiarities being taken into account. A sequence of interconnected heaters for initial components and an oxidative pyrolysis reactor is used as control object. The control system has to provide the required pyrolysis gas composition to the reactor outlet in condition that ensures safety of the process. A two-level structure is suggested for the control system process including a local parameter adjustment at the lower level and a local circuit task estimation with a fuzzy controller adjustment at the upper level. Depending on the parameter values determining the process safety, some control actions, i.e., natural gas discharge rate, ratio of natural gas, oxygen and water flow for pyrolysis gas “hardening,” are estimated. A control system with a fuzzy controller shows the best performance.