Nombre: MARCOS WAGNER JESUS SERVARE JUNIOR
Fecha de publicación: 27/10/2022
Supervisor:
Nombre |
Papel |
|---|---|
| HELDER ROBERTO DE OLIVEIRA ROCHA | Advisor * |
Junta de examinadores:
| Nombre |
Papel |
|---|---|
| GISELE DE LORENA DINIZ CHAVES | External Examiner * |
| GLAYDSTON MATTOS RIBEIRO | External Examiner * |
| HELDER ROBERTO DE OLIVEIRA ROCHA | Advisor * |
| JOSE LEANDRO FÉLIX SALLES | Co advisor * |
| MARCIA HELENA MOREIRA PAIVA | Internal Examiner * |
Páginas
Sumario: In this Thesis, a mixed integer linear programming (MILP) model and two mixed integer nonlinear programming (MINLP) models are proposed to plan the material flow of a stockyard in a port terminal, aiming to minimize the power consumption. The proposed models are solved by a commercial solver, using the Heuristic Based on Linear Relaxation (LBRH) and the heuristic based on Rolling Horizon. The first model is solved by the LRBH algorithm to optimize the power consumption of the small stockyard-port system, operating in the long term (more than 60 days) and with multiple products. The other two are solved by the Rolling Horizon algorithm to optimize the power consumption
of the large stockyard-port system in the short-term horizon (24 hours), with multiple products, different powersuppliers and the possibility of using a battery bank. In addition, operational uncertainties in large stockyards are considered and that decisions are made in real time (hourly). These heuristics will allow to the planner to find solutions that reduce power costs in iron ore stockyards in port terminals for large instances, that is, WHERE it is not possible to obtain an optimal solution by solving the MILP and MINLP models. Using
numerical simulations, comparisons are made between the optimal solutions, obtained by the MILP and MINLP models, and approximate solutions, obtained by the LBRH and Horizon Rolling heuristics for some small and medium-sized instances. In these instances, LRBH provides viable solutions with an average distance from the objective function of 3.99% in relation to the optimal solution (GAP) and the approach using the Horizon Rolling based heuristic provides viable solutions with a GAP of 3.21%. In large instances, the solutions obtained with the proposed heuristics have an accessible computational
time. In these instances, the proposed model with batteries solved with the HR algorithm, provides an energy cost reduction of up to 17.8%.
Keywords: Iron Ore Stockyard Energy planning, Linear Relaxation-Based Heuristic, Rolling horizon, Scheduling, Mathematical programming.
