Though the Support Vector Regression Machines (SVRM) is considered to be an effective method for time series prediction, its performance is greatly influenced by its parameters. In order to improve the rationality of parameter setting, the influences of the parameters (the number of support vectorsNSand the prediction lengthNE) and signal characteristics on the SVRM performance were discussed. The results proved that the existence of confliction between prediction accuracy and prediction efficiency, and SVRM may inappropriate to long-term prediction, andNSshould be greater than a threshold which depends on the signal characteristics for an accurate prediction result. The research results may provide a theoretical basis for the improvments of SVRM algorithm.
The Support Vector Regression Machines (SVRM) is considered to be an effective method for time series prediction, while it is not suitable for long-term prediction. In order to improve the SVRM application in eliminating the end effect of EMD, we apply the SVRM to achieve an accurate short-term extension in the end of the original signal before EMD, and then extend two extreme with the Self-adaptive Triangular Waveform Matching extension method (STWM) in the every sifting process of EMD. The results show that the new extension method can effectively contain end effect of EMD.
The xylanase from Leveking was used in the experiment of bleached Soda-AQ pulp of fast-growing poplar. The suitable conditions of the xylanase from Leveking were obtained in the pretreatment of the Soda-AQ pulp of fast-growing poplar that temperature was 58°C, pH was 8.5, time was 120min and the dosage of the xylanase was 8IU/g. The pulp brightness was increased by 3.7%ISO when compared XDPQ and DPQ bleaching pulp under identical conditions. The pulp brightness was increased by 4.1%ISO when compared XDED and DED bleaching pulp under identical conditions. The untreated and pretreated Soda-AQ pulp of fast-growing poplar by the xylanase of two-dimensional and three dimensional figure of fiber surface under AFM were analyzed to explain the xylanase effect in the pulp bleaching theoretically.
In order to increase the accuracy and real-time of grey prediction fuzzy direct torque control system, an improved grey prediction fuzzy direct torque control method was advanced. The equal-dimensional new information model was used to construct the new grey prediction model. The position angle of motor stator flux was divided into sectors, by which the simplified fuzzy algorithm was obtained. The simulation results show that the improved method can greatly overcome the shortcomings caused by the original method, such as the overlarge dimensions of original time sequence and the decrease of effective information amount due to the lasting of time. The calculation quantity in prediction process is reduced. The precision of prediction and the real-time of fuzzy control system are increased. The instability of stator flux is lowered and the speed respond of motor is meliorated.
Jet grading technology is an efficient process in different industries. In this research, tungsten powder with different particle size distribution was used as a raw material to produce tungsten products via isostatic pressing as well as sintering. The mechanism of jet grading and the morphology and particle size distribution of different precursors were analyzed. The results showed that jet grading technology had remarkable effect on tungsten powder classification. The appropriate grading treatment was helpful to the formation of tungsten products with high performance. After jet grading and the following process like pressing and sintering, the tungsten products with better properties were manufactured which was used fischer particle size of 3.0~3.5μm as the raw material. The obtained products’ density was 18.77g/cm3 and its hardness was 372.15HV0.3.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.