Systematic design optimization of grabs considering bulk cargo variability
| Autor: | Arjan J. van den Bergh, M. Javad Mohajeri, Dingena L. Schott, Jovana Jovanova |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
Computer science
business.industry General Chemical Engineering Process (computing) DEM-MBD co-simulation 02 engineering and technology 010402 general chemistry 021001 nanoscience & nanotechnology 01 natural sciences Multi-objective optimization Sustainable design Cohesive iron ore 0104 chemical sciences Latin hypercube sampling Mechanics of Materials Grabs Genetic algorithm Range (statistics) Sensitivity (control systems) 0210 nano-technology Process engineering business Engineering design process Bulk cargo |
| Zdroj: | Advanced Powder Technology, 32(5) |
| ISSN: | 0921-8831 |
| DOI: | 10.1016/j.apt.2021.03.027 |
| Popis: | Ship unloader grabs are usually designed using the manufacturer’s in-house knowledge based on a traditional physical prototyping approach. The grab performance depends greatly on the properties of the bulk material being handled. By considering the bulk cargo variability in the design process, the grab performance can be improved significantly. A multi-objective simulation-based optimization framework is therefore established to include bulk cargo variability in the design process of grabs. The primary objective is to reach a maximized and consistent performance in handling a variety of iron ore cargoes. First, a range of bulk materials is created by varying levels of cohesive forces and plasticity in the elasto-plastic adhesive DEM contact model. The sensitivity analysis of the grabbing process to the bulk variability allowed three classes of iron ore materials to be selected that have significant influence on the product performance. Second, 25 different grab designs are generated using a random sampling method, Latin Hypercube Design, to be assessed as to their handling of the three classes of iron ore materials. Of this range of grab designs, optimal solutions are found using surrogate modelling-based optimization and the NSGA-II genetic algorithm. The optimization outcome is verified by comparing predictions of the optimization algorithm and results of DEM-MBD co-simulation. The established optimization framework offers a straightforward and reliable tool for designing grabs and other similar equipment. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect