Tuning a three-phase separator level controller via particle swarm optimization algorithm

Three-Phase Separators are used to separate well crudes into three portions; water, oil, and gas. A suitable control system should be in place to ensure the optimum function of the Three-Phase Separator. The current PID tuning technique does not provide an optimum system response of the separator. Overshoot response, offset, steady-state error and system instability are some of the problems faced. Besides, the current method used is purely based on trial and error which is time consuming. There is room for improvement of the current PID tuning technique. An artificial intelligence (AI) PID tuning technique called Particle Swarm Optimization (PSO) is introduced to improve the system response of the Three-Phase Separator. The PSO algorithm mimics the behaviour of bird flocking and fish schooling striving for its global best position. In our case, the global best position is replaced with the optimized PID tuning parameters for the separator. The PSO algorithm has been used in several other applications such as the Brushless DC motor and in the Control Ball Beam system. It has proven to be an effective tuning technique. Tuning of the Three-Phase Separator via PSO could prove to be an effective solution for Oil Gas industries.