As the central component of rotating machine,the performance reliability assessment and remaining useful lifetime prediction of bearing are of crucial importance in condition-based maintenance to reduce the maintenance cost and improve the reliability.A prognostic algorithm to assess the reliability and forecast the remaining useful lifetime(RUL) of bearings was proposed,consisting of three phases.Online vibration and temperature signals of bearings in normal state were measured during the manufacturing process and the most useful time-dependent features of vibration signals were extracted based on correlation analysis(feature selection step).Time series analysis based on neural network,as an identification model,was used to predict the features of bearing vibration signals at any horizons(feature prediction step).Furthermore,according to the features,degradation factor was defined.The proportional hazard model was generated to estimate the survival function and forecast the RUL of the bearing(RUL prediction step).The positive results show that the plausibility and effectiveness of the proposed approach can facilitate bearing reliability estimation and RUL prediction.
针对公交车行程时间预测存在数据稀疏、数据缺失及更新间隔长等问题,提出了一种基于相似路段划分并融合多线路信息的卡尔曼滤波算法。该算法对每条路段的属性特征和空间结构特征进行归一化处理,利用属性特征和空间结构的相似性及POI(Point of Interest)对交通影响的变化动态地划分相似路段;然后融合相似路段与目标路段上的多条公交线路的数据信息,用相似路段的数据丰富实验数据;最后结合卡尔曼滤波算法动态性高、实时性强等特点建立模型,从而实现短时预测,并对信息进行修正。选取沈阳市162线路和299线路作为实验线路,各划取一段相似路段进行基础数据采集并进行实验。通过相似路段上的信息来推断数据稀疏或缺失路段的信息,能够缩短数据更新间隔并提高算法预测的实时性及精准性,尤其在早高峰时段,提出的算法模型的绝对平均百分误差达到13.2%,能达到实时查询的性能需求。
The fishtail in head and tail of the slabs was studied during V-H hot rolling process. With the application of ANSYS/LS-DYNA, simulation analysis was used to research this process. The various factors which have a great influence on fishtail shapes were analysed, such as initial width, initial thickness, radius of the edger roll and horizontal roll, edging draught,horizontal reduction rate, and friction coefficient of the surface. Then the curves that can describe the shapes were obtained. After a certain time of self-learning, the optimized curves were given out. At last, through the fitting of the simulation test results, the math models for the area of fishtail defect changing with the presented factors were received. The experimental results show that the accuracy of the prediction for the fishtail shapes is more than 95%. With the application of the prediction for the fishtail shapes and the area of the fishtail defect, the loss rate of the slab is decreased by about 0.1%.
In order to overcome shortcomings of traditional BP neural network,such as low study efficiency, slow convergence speed,easily trapped into local optimal solution,we proposed an improved BP neural network model based on adaptive particle swarm optimization( PSO) algorithm. This algorithm adjusted the inertia weight coefficients and learning factors adaptively and therefore could be used to optimize the weights in the BP network. After establishing the improved PSO-BP( IPSO-BP) model,it was applied to solve fault diagnosis of rolling bearing. Wavelet denoising was selected to reduce the noise of the original vibration signals,and based on these vibration signals a wide set of features were used as the inputs in the neural network models. We demonstrate the effectiveness of the proposed approach by comparing with the traditional BP,PSO-BP and linear PSO-BP( LPSO-BP) algorithms. The experimental results show that IPSO-BP network outperforms other algorithms with faster convergence speed,lower errors,higher diagnostic accuracy and learning ability.
