Frame weight and anti-fatigue co-optimization of a mining dump truck based on Kriging approximation model. (August 2016)

• Record Type: Journal Article
• Authors: Mi, Chengji; Gu, Zhengqi; Zhang, Yong; Liu, Shuichang; Zhang, Sha; Nie, Duzhong
• Is Part Of: Engineering failure analysis. Volume 66(2016:Aug.)
• Page Start: 99
• Page End: 109
• Publication Date: 2016-08
• Subjects: Mining dump truck -- Kriging approximation model -- Latin hypercube sampling -- Experiment design -- Genetic algorithm -- Lightweight; System failures (Engineering) -- Periodicals; Fracture mechanics -- Periodicals; Reliability (Engineering) -- Periodicals; Pannes -- Périodiques; Rupture, Mécanique de la -- Périodiques; Fiabilité -- Périodiques; Fracture mechanics; Reliability (Engineering); System failures (Engineering); Periodicals; Electronic journals; 620.112
• DOI: 10.1016/j.engfailanal.2016.03.021
• Languages: English
• ISSNs: 1350-6307

Abstract

Abstract: In order to reduce frame redundant weight caused by the traditional experience design and delay the crack initiation occurrences at local positions during early service period, a structural weight and lifetime co-optimization method for frame of a mining dump truck based on Kriging approximation model was proposed. Considered as the most sensitive factor to the frame fatigue life and weight, the thickness of several steel plates was chosen as design variables in terms of un-allowed change of structure shape, while achieving weight reduction and improving fatigue life were together regarded as the optimization targets. An experiment design with 20 sample points obtained by the Latin hypercube sampling method was conducted to figure out the sensitivity of design variables, whose response values were acquired through repeated simulations. Those data were supposed to set up the approximation model constructed based on the Kriging interpolation technique and its fitting precision was certified by comparison of the finite element computational results and the approximation model calculated ones. The non-dominated sorting genetic algorithm II (NSGA-II) was utilized to optimize the thickness of the steel plates based on the approximation model. The tolerance between the results of the simulation and the approximation model was less than 1% when using the optimal design variables and the weight of the optimized frame was lessened by 22.3% while the minimum fatigue life and maximum static stress were only decreased by 3.8% and 4.6%, compared with the initial frame. These optimized results were acceptable for frame lightweight almost without expense of fatigue life and static strength. Highlights: Design of experiment for 4 variables using the Latin hypercube method Multi-objectives optimization based on the Kriging approximation model. Weight and anti-fatigue co-optimization without expense of fatigue life and static strength