Flow Control System Design Using Sliding Mode Control (SMC) With Neural Network In backloading At Terminal BBM P.T. Pertamina Perak Surabaya
Year: 2013
Proceedings Title : Proc. Indon. Petrol. Assoc., 37th Ann. Conv., 2013
Backloading is the distribution of oil products at Terminal BBM P.T. Pertamina Perak Surabaya. The procedure requires accuracy of flow rate, with a standardized value of 79 m3/hr. The system requires a robust control mechanism (able to work well when confronted with unexpected disturbance) to maintain stable flow, e.g. the Sliding Mode Control (SMC). SMC has characteristics that can stabilize an uncertain condition. For instance, when the expected flow value does not match with the realized output flow then the control system will detect it as a certain amount of error which is used as a reference so that the controller can determine the magnitude of the flow. Further validation of the output flow rate is confirmed using Neural Network with back propagation type. From the simulation without the SMC controller, the system Flow Output exceeds the set point at rate 86.29 m3/hr. While the result of the flow control using SMC controller, the Fout value approaches the set point at 78.935 to 78.999 m3/hr achieved in less than 0.01 hours. Then for validation testing with Neural Network Training reaches performance or Mean Squared Error (MSE) of 0.00000983 for 479 iterations, to the target value is 79 m3/hr.
Log In as an IPA Member to Download
Publication for Free.
or
Purchase from AAPG Datapages.