Patents
Literature
Patsnap Copilot is an intelligent assistant for R&D personnel, combined with Patent DNA, to facilitate innovative research.
Patsnap Copilot

1421 results about "Polymerization reactor" patented technology

Performance of artificial neural network models in the presence of instrumental noise and measurement errors

A method is described for improving the prediction accuracy and generalization performance of artificial neural network models in presence of input-output example data containing instrumental noise and/or measurement errors, the presence of noise and/or errors in the input-output example data used for training the network models create difficulties in learning accurately the nonlinear relationships existing between the inputs and the outputs, to effectively learn the noisy relationships, the methodology envisages creation of a large-sized noise-superimposed sample input-output dataset using computer simulations, here, a specific amount of Gaussian noise is added to each input/output variable in the example set and the enlarged sample data set created thereby is used as the training set for constructing the artificial neural network model, the amount of noise to be added is specific to an input/output variable and its optimal value is determined using a stochastic search and optimization technique, namely, genetic algorithms, the network trained on the noise-superimposed enlarged training set shows significant improvements in its prediction accuracy and generalization performance, the invented methodology is illustrated by its successful application to the example data comprising instrumental errors and/or measurement noise from an industrial polymerization reactor and a continuous stirred tank reactor (CSTR).
Owner:COUNCIL OF SCI & IND RES

Process for fluid phase in-line blending of polymers

A process for fluid phase in-line blending of polymers. The process includes providing two or more reactor trains configured in parallel and a separator for product blending and product-feed separation; contacting in at least one of the parallel reactor trains olefin monomers having three or more carbon atoms, catalyst systems, optional comonomers, optional scavengers, and optional inert diluents or inert solvents, at a temperature above the solid-fluid phase transition temperature of the polymerization system and a pressure no lower than 10 MPa below the cloud point pressure of the polymerization system and less than 1500 MPa; forming a reactor effluent including a homogeneous fluid phase polymer-monomer mixture in each parallel reactor train; combining the reactor effluent from each parallel reactor; passing the combined reactor effluent through the separator; maintaining the temperature and pressure within the separator above the solid-fluid phase transition point but below the cloud point pressure and temperature to form a fluid-fluid two-phase system including a polymer-rich blend phase and a monomer-rich phase; and separating the monomer-rich phase from the polymer-rich blend phase. The separated monomer-rich phase is recycled to the polymerization reactor bank. The polymer-rich blend phase is conveyed to a downstream finishing stage for further monomer stripping, drying and/or pelletizing to form a polymer product blend.
Owner:EXXONMOBIL CHEM PAT INC

Performance of artificial neural network models in the presence of instrumental noise and measurement errors

A method is described for improving the prediction accuracy and generalization performance of artificial neural network models in presence of input-output example data containing instrumental noise and / or measurement errors, the presence of noise and / or errors in the input-output example data used for training the network models create difficulties in learning accurately the nonlinear relationships existing between the inputs and the outputs, to effectively learn the noisy relationships, the methodology envisages creation of a large-sized noise-superimposed sample input-output dataset using computer simulations, here, a specific amount of Gaussian noise is added to each input / output variable in the example set and the enlarged sample data set created thereby is used as the training set for constructing the artificial neural network model, the amount of noise to be added is specific to an input / output variable and its optimal value is determined using a stochastic search and optimization technique, namely, genetic algorithms, the network trained on the noise-superimposed enlarged training set shows significant improvements in its prediction accuracy and generalization performance, the invented methodology is illustrated by its successful application to the example data comprising instrumental errors and / or measurement noise from an industrial polymerization reactor and a continuous stirred tank reactor (CSTR).
Owner:COUNCIL OF SCI & IND RES
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products