Jiang Guozhang

Case-based reasoning intelligent prediction model and its application in flue temperature of coke oven

Flue temperature is a key technical index in coke oven heating process, which is so difficult to measure directly online. The offline analysis has large-time delay and poor precision. An intelligent prediction model of flue temperature based on case-based reasoning was developed, which consists of four modules: data collection and pretreatment, prediction, online modification and effect estimate. The practical data of some coke oven were simulated. The industrial application results show that the prediction model can reflect the actual operation condition and meet the requirement of real-time control. Its effectiveness is proved evidently.

Temperature Predictive Fuzzy Control of Coke Oven

On the basis of analyzing production characteristics of coke oven, heating intelligent control model of coke oven was built. Then predictive fuzzy control algorithm was proposed combining predictive control with fuzzy control. In the algorithm, predictive control was responsible for control accuracy and fuzzy control was to provide control variables when predictive model was not convergent as well as uncorrected control variables when the model had converged. A good temperature control of coke oven with characteristics of large inertia and time delay was obtained by application of the algorithm. The practical running indicates that the algorithm has good control performance and great practical value.

Study of Contact Stress and Extrusion of Vane Seals via Numerical Method

Contact stress of vane seals is studied by using numerical and FEA method based on the linear elasticity theory. Results show that numerical model is appropriate to predict the distributing curve of contact stress when the maximum normal strain is lower than 10-15%. But, with sealed extrusion, finite element analysis (FEA) would be better due to the complexity of the material properties, local geometry and boundary conditions. Accordingly, nonlinear FEA analysis is adopted to study the effect of section shape and section structure on the extrusion process of vane seals. Results illustrate that extrusion occurs with elastomeric seals with rectangular section while a backup PTFE can minimize or eliminate the extrusion.