Rongxiu Lu

  • Citations Per Year
Learn More
A new multiple-model predictive control scheme is proposed for the control of the component content in the rare earth extraction process. Multiple local linear models with twoinput and two-output of the rare earth countercurrent extraction process are constructed, each of which is established under an operation condition. A component content predictive(More)
Aiming at the complicated characteristic of rare earth extraction process and combining the material balance model, a multiple models modeling and control method is proposed. Based on the data collected in an industrial field, an improved subtractive clustering algorithm is employed to obtain steady operation points for the process; the recuresive least(More)
The problems of small sample, non-linearity, high dimensions and local minimal value can be well solved by support vector machine in soft-sensor modeling. In consideration of the online measurement of the component content in rare earth counter-current extraction separation process, two algorithms of SVM and LS_SVM with RBF kennel was applied to the(More)
For constructing a span-lateral inhibition neural network (S-LINN) with optimal architecture and parameters for actual application, a self-organizing optimization approach is proposed in this paper to tune the architecture and parameters simultaneously. This self-organization pruning algorithm is to build a modified significant index function to evaluate(More)
In the process of rare earth cascade extraction, content of each component accumulates to different extent in the corresponding extraction tanks, and shows its characteristic color. When the production process conditions changed, the distribution of content of each component will be changed, and the result is the movement of characteristic color. It(More)
  • 1