The neural network with learning and very strong non-linear processing ability, BP neural network used as the neural network model of parameter recognition for transient well test in this paper, the internal connection weights of neural network on representation of the model parameters. Static system by neural network identification method available well test interpretation parameters, BP neural network as the transient well test parameter recognition of the neural network model, overcoming the affecting of human subjectivity with conventional well testing or modern well test analysis methods and nonlinear optimization problems obtained the diversity effects of consequences. Finally, on-site examples given, have proved to be very significant.
[1]
Vladimir N. Vapnik,et al.
The Nature of Statistical Learning Theory, Second Edition
,
2000,
Statistics for Engineering and Information Science.
[2]
W. Pitts,et al.
A Logical Calculus of the Ideas Immanent in Nervous Activity (1943)
,
2021,
Ideas That Created the Future.
[3]
Vladimir Vapnik,et al.
Statistical learning theory
,
1998
.
[4]
Dale Schinstock,et al.
System Modeling and Identification of a Precision Fast Tool Servo System
,
1999
.