1. Introduction
The centrifugal fan blades are mainly used for cooling of traction motors, main transformer cabinets and auxiliary transformer cabinets of high-speed trains. As one of the main components of rail transportation equipment, the fan blades’ safety and reliability deserves more attention and study. The phenomenon of fatigue is one of the main reasons for material failure. Due to the complex operating environment of the high-speed-train system, the blades are subjected to periodic changes in alternating stress, vibration excitation and so on when working. These loads usually can be described by deterministic formulas or stochastic theory [1]. On the other hand, when the fan starts and stops, the alternating stress will increase suddenly. High cycle fatigue life consumption [2], caused by alternating stress during starts and stops, will be considered and examined. In short, the centrifugal fan blades will be damaged by fatigue working long hours, which will affect the normal operation of the high-speed-train group system. Therefore, the assessment of fatigue life is necessary for the design of fan blade and an important index for evaluating reliability. It has important theory meanings and practical applying value for centrifugal fan blades fatigue life forecasting.
At present, many investigations on the fatigue performance of fan blades have been conducted. In general, fatigue life prediction for fan blades is based on nominal stress [3-5]. Static and dynamic stress analyses and fracture mechanics criterion were accomplished to simulate fatigue crack growth for compressor blade [6, 7].The most commonly adopted approach to study fatigue performance of fan blades is finite element technique, which makes the analysis of coupling of fluid and solid and is the key technique of understanding fatigue characteristics and failure mechanism [8-10]. Finite element analysis and Neu/Sehitoglu (N/S) TMF model provided a life assessment tool for aero jet engine blades by Abu, AO [11]. Finite element analysis could be used to assess low cycle fatigue life of steam turbine blade [12], which could predict the catastrophic failure of turbine blade arisen from fatigue [13]. The influence of interaction of low cycle fatigue (LCF) and high cycle fatigue (HCF) on aerodynamic and structural properties of fan blades were also studied [14-16]. The fatigue caused by combined cycle loadings was noticed [17-20]. Miner’s rule was used to calculate the fatigue life of helicopter main rotor blade according to the classical life estimation method under random load [21]. In actual railway projects, the Miner’s rule [22] had been frequently used to assess combined cycle loadings, however, it ignored the effects of load interactions and the coupling damage [23].
Considering the impact of structural dynamic response, it is very important to verify relevant models for the development of fan blades [24-26]. Meanwhile, The research on the dynamic characteristics of fan blades mainly focuses on gas turbine blade [27], compressor blade [28], steam turbines [29], wind turbine blade [30, 31], wings [32], helicopter blade [33] and there are few studies on fan blades in the ventilation and cooling system of railway locomotives. It’s critical to design the fan blades of many railway vehicles, but fatigue design experiments of each type of fan structure are very costly and time consuming. Therefore, reliable fatigue life prediction is an important requirement for new product development.
This paper aimed to develop a model for the fatigue life prediction of fan blades in the ventilation cooling system of high-speed-train. Focusing on the causes of the dynamic stress of the fan blade, a typical centrifugal fan blade was taken as the research object. Chemical composition analysis, static tensile test and dynamic fatigue test were carried out on the blade material DC51D+Z, the basic composition and mechanical property parameters of the material were obtained, and theS-N curve of the material DC51D+Z was fitted. By using the finite element model of fan, considering the influence of friction and contact force of its structural parts, the stress distribution at the critical point of blade components was calculated. Considering the loading form, size effect, surface quality and stress concentration factor (SCF) of the fan blade comprehensively, this paper introduced an improved method based on the concept of nominal stress. Finally, in order to prevent the fan blade from damaging caused by fatigue under unexpected circumstances, this paper established a new model for predicting the fatigue life of fan blades in the ventilation cooling system of high-speed-train. It was combined with the Palmgren-Miner cumulative damage criterion and S-N curve to further studying details concerning the feasibility of the calculation model.