Optimization of buttressed earth-retaining walls using hybrid harmony search algorithms

• Published in: Engineering structures Vol. 134; pp. 205 - 216
• Main Authors: Molina-Moreno, Francisca, García-Segura, Tatiana, Martí, José V., Yepes, Víctor
• Format: Journal Article
• Language: English
• Published: Kidlington Elsevier Ltd 01.03.2017; Elsevier BV
• Subjects: Algorithms; Buttresses; Concrete structures; Cost optimization; Design parameters; Friction; Harmony search; Heuristics; Optimization; Optimization algorithms; Project design; Reinforced concrete; Reinforcement; Retaining walls; Search algorithms; Soil bearing capacity; Soils; Structural design; Studies; Superplasticity; Wall friction; Structural design; Harmony search; Concrete structures; Cost optimization; Heuristics
• ISSN: 0141-0296; 1873-7323
• DOI: 10.1016/j.engstruct.2016.12.042

•A hybrid harmony search is a reliable procedure for optimum buttressed earth-retaining walls.•Geometrical dimensions can be reduced compared to usual rules of thumb.•The type of fill is the main factor affecting the cost of optimum buttressed walls.•Cost-optimum walls are sensitive to the maximum bearing pressure for each type. This paper represents an economic optimization of buttressed earth-retaining walls. We explore the optimum solutions using a harmony search with an intensification stage through threshold accepting. The calibration of the resulting algorithm has been obtained as a result of several test runs for different parameters. A design parametric study was computed to walls in series from 4 to 16m total height. The results showed different ratios of reinforcement per volume of concrete for three types of ground fill. Our main findings confirmed that the most sensitive variable for optimum walls is the wall-friction angle. The preference for wall-fill friction angles different to 0° in project design is confirmed. The type of fill is stated as the main key factor affecting the cost of optimum walls. The design parametric study shows that the soil foundation bearing capacity substantially affects costs, mainly in coarse granular fills (F1). In that sense, cost-optimum walls are less sensitive to the bearing capacity in mixed soils (F2) and fine soils of low plasticity (F3). Our results also showed that safety against sliding is a more influential factor for optimum buttressed walls than the overturning constraint. Finally, as for the results derived from the optimization procedure, a more suitable rule of thumb to dimension the footing thickness of the footing is proposed.