초록 ▼

A method for forming a module (hydrodynamic or thermodynamic for example) intended for real-time simulation of the flow mode, at any point of a pipe, of a multiphase fluid stream comprising at least a liquid phase and at least a gas phase. The method comprises using a modelling system based on non-l

A method for forming a module (hydrodynamic or thermodynamic for example) intended for real-time simulation of the flow mode, at any point of a pipe, of a multiphase fluid stream comprising at least a liquid phase and at least a gas phase. The method comprises using a modelling system based on non-linear neural networks each having inputs for structure parameters and physical quantities, outputs where quantities necessary for estimation of the flow mode are available, and at least one intermediate layer. The neural networks are determined iteratively to adjust to the values of a learning base with predetermined tables connecting various values obtained for the output data to the corresponding values of the input data. A learning base suited to the imposed operating conditions is used and optimized neural networks best adjusted to the imposed operating conditions are generated.

대표청구항 ▼

1. A method for forming a module providing for real-time simulation of the flow mode, at any point of a pipe, of a multiphase fluid stream comprising at least a liquid phase and at least a gas phase, for imposed operating conditions concerning determined structural parameters relative to the pipe, a

1. A method for forming a module providing for real-time simulation of the flow mode, at any point of a pipe, of a multiphase fluid stream comprising at least a liquid phase and at least a gas phase, for imposed operating conditions concerning determined structural parameters relative to the pipe, and a set of determined physical quantities, with fixed variation ranges for the parameters and said physical quantities, by a modelling system based on non-linear neural networks each having inputs for structure parameters and the physical quantities, outputs where quantities necessary for estimation of the flow mode are available, and at least one intermediate layer, comprising determining the non-linear neural networks so as to adjust to values of a learning base with predetermined tables connecting values obtained for the output data from the output to corresponding values of input data to the inputs and wherein the learning base is suited to the imposed operating conditions and the determined non-linear neural networks are adjusting to the imposed operating conditions. 2. A method as claimed in claim 1, wherein the physical quantities are hydrodynamic quantities. 3. A method as claimed in claim 1, wherein the physical quantities are thermodynamic quantities. 4. A method as claimed in claim 2, wherein the module being integrated is a hydrodynamic and thermodynamic multiphase flow simulation model, the model being used to form the learning base to select the physical quantities best suited to the model, variation ranges fixed for the parameters and the physical quantities, and an optimized neural networks best suited to the learning base.