Jiang Qingyin

Modeling and simulation for the hydrocracking reactor

A simplified six-lump-model for one practical residue hydrocracking reactor has been developed according to the characteristics of hydrocracking reaction. The model, including six lumps (crude oil, diesel oil, aviation kerosene, heavy naphtha, light naphtha and gas), took both mass balance and kinetics equations into consideration. Only fifteen kinetic parameters of the model need to be estimated, which was half less than that of literature[1], and the parameters were estimated with the measured industrial data shown in literature[1] by Nelder-Mead minimization algorithm. The distribution of the hydrocracking product yields along the reactor position and the reaction time can be predicted by solving this simplified lump-model. The calculated result was compared with the measured industrial value and all the relative deviations were within 5% which was close to the literature[1].

Design of Interface Software for a Distributed System With Mixed Varied Intelligent Meters

Main problems of the common communication interface software (CIS) for an industry-control network based on RS485 bus are poor openness, flexibility and stability, so some functions of intelligent meters are limited. A new CIS was designed by the following ways: the RS485 serial communication process was composed of physical and application layers; the interface initializing process related to communication at physical layer was described by normalized communication module; the common parts of order message creation and responsive message transmission were described by a data structure and the special parts of them were described by text files respectively; configuration software was used to interpret these descriptions. The software has characteristics of broad generalization and operational facilitation, and can run on Windows completely. The results in application cases show that the software is effective and feasible

Synchronization control scheme for multi-process systems based on model predictive control

Multi-process system (MPS) is an important process system for modern industry. The parallel operating subsystems may have synchronization requirements. A generalized synchronization control scheme is thus developed in this paper based on the model predictive control framework by combining a generalized synchronization cost function and the predictive cost function. The resulted control algorithm indicates that the predictive control errors of each sub-process and the predictive synchronization errors between sub-processes are used together as feedback information in the control scheme to ensure the optimal control performances of each sub-processes as well as synchronization performance, which essentially leads to a multi-input and multi-output (MIMO) control for the MPS. With a proper selection of the synchronization error functions, ratio and distance synchronization controls are conducted with the numerical simulation on an MPS consists of three sub-processes. The results clearly prove the effectiveness, robustness and flexibility of the proposed synchronization control scheme.