| Grant number: | 18/04943-0 |
| Support Opportunities: | Scholarships abroad - Research Internship - Doctorate (Direct) |
| Start date: | July 01, 2018 |
| End date: | November 14, 2018 |
| Field of knowledge: | Engineering - Chemical Engineering - Industrial Operations and Equipment for Chemical Engineering |
| Principal Investigator: | Jorge Andrey Wilhelms Gut |
| Grantee: | Robert Eduard Franzoi Junior |
| Supervisor: | Christos T. Maravelias |
| Host Institution: | Escola Politécnica (EP). Universidade de São Paulo (USP). São Paulo , SP, Brazil |
| Institution abroad: | University of Wisconsin-Madison (UW-Madison), United States |
| Associated to the scholarship: | 17/03310-1 - Integrated scheduling optimization in the crude-oil refinery industry: from crude-oil unloading to fuel deliveries, BP.DD |
Abstract So far, the literature on crude-oil scheduling optimization covered the problem from the crude-oil unloading and storage up to the distillation straight-run streams considering an offline environment. To go further, this research extends the scope of the problem from the raw material deliveries up to product liftings through the refinery process-shop in an online environment integrated within the scheduling cycle. Routine process feedback data and laboratory measurements from the field are considered in both open- and closed-loop strategies for better process predictions. In the open-loop strategy, the inputs of the process are updated in the online scheduling cycle such as catalyst activities, residence time of reactors and fill-draw-delays in tanks, the latter for better mixing or segregations of streams. On the other hand, in a closed-loop strategy rather than only updating the inputs, measured process outputs can calibrate gains and biases as ymeasured = gain × ymodel + bias, whereby the data updating in y considers both process yields and variables such as throughputs, flows, holdups and properties, whose effects propagate throughout the process network. After a simultaneous data reconciliation and estimation computation, the parameter feedback will be applied in a complete crude-oil refinery blend scheduling problem considering real tank topology, cascaded distillation towers, process-shops and blend-shops to effectively optimize the complex process system. The feedback strategy is solved for the refinery production scheduling within an iterative mixed-integer linear and nonlinear programming (MILP-NLP) decomposition by updating NLP results of process-shop's yields and properties, and blend-shops' recipes in the next MILP solution until convergence is achieved. (AU) | |
| News published in Agência FAPESP Newsletter about the scholarship: | |
| More itemsLess items | |
| TITULO | |
| Articles published in other media outlets ( ): | |
| More itemsLess items | |
| VEICULO: TITULO (DATA) | |
| VEICULO: TITULO (DATA) | |