52 results about "Petroleum processor" patented technology

The invention is a composition that is suitable for reducing sulfur species from products produced by petroleum refining processes, especially gasoline products produced by fluidized catalytic cracking (FCC) processes. The composition comprises zeolite, yttrium, and at least one element selected from the group consisting of zinc, magnesium and manganese, wherein the yttrium and element are present as cations. The yttrium and zinc are preferably present as cations that have been exchanged onto the zeolite. The zeolite is preferably a zeolite Y.

The invention is a two-stage heavy oil catalytic cracking conversion method and device, belonging to petroleum refining process and device technical field, separating heavy oil reaction course into top and bottom reacting regions, making raw oil and regenerated catalyst firstly contact and react with each other in the bottom reacting region and successively separating the carbon deposit deactivated catalyst from reacting oil; making the oil gas go up into the second reacting region to mix with the catalyst used in gasoline reaction and continue reaction; a heavy oil reactor is connected in parallel with gasoline reacting part; the gasoline and catalyst complete reaction in the gasoline reactor and then make gas-solid separation in gasoline settler, where the oil gas is discharged through pipeline and the extracted catalyst is fed into pre-rising stages below the first and second heavy oil reacting regions to continue taking part in heavy reaction; a catalyst return pipe is arranged between the pre-rising stages of the gasoline settler and gasoline reactor to make a part of catalyst used in gasoline reaction return to the bottom of the gasoline reactor, so as to take part in gasoline reaction. And it can obviously raise efficiency of heavy oil reactor, improve product quality and need not consume energy because of catalyst change.

The invention discloses an adaptive control method for the liquid level of a fractionation tower in the petroleum refining process. In this method, the reverse error neural network and extreme value optimization method are used as adaptive tools, and the predictive PI control strategy is designed in combination with the input and output of the actual operation process. The present invention first uses the data acquisition device to collect the past input and output of the system, and then designs an extreme value optimization algorithm as a solver to adjust the weight of the neural network online, thereby efficiently and accurately training the neural network; after satisfying the set error constraints, The trained neural network weight is used as the identification parameter of the system, and a system model close to the actual situation is established. Finally, the predictive PI control is designed to control the liquid level of the fractionation tower in the process of petroleum refining. By adopting the method of the invention, the interference of model mismatch to the system can be effectively relieved, the control performance is improved, and the application of advanced control in industrial process control is also promoted.