Reservoir parameter prediction method and apparatus based on geological characteristic constraint, and storage medium

A reservoir parameter prediction apparatus based on geological characteristic constraint, comprising:

• a data acquisition unit which acquires seismic attributes of different types in the seismic waves from the target stratum, including seismic data and well logging data;
• an attribute screening unit which analyzes relevance between the seismic attributes of different types of the target stratum and reservoir parameters by seismic data and well logging data of the target stratum, and to select dominant seismic attributes from the seismic attributes of different types according to a magnitude of the relevance;
• a waveform classification unit which classifies, on a basis of the dominant seismic attributes, seismic waveforms of the target stratum by a preset waveform classification network model and according to waveform features, so as to obtain a waveform classification result, with waveforms of different types correspondingly representing different geological characteristics;
• a model construction unit which constructs different deep neural network models corresponding to the different geological characteristics with the seismic attributes as an input, the reservoir parameters as an output, and the waveform classification result as constraint;
• a model training unit which trains the deep neural network models at different logging well positions based on the seismic data, the seismic attributes and the well logging data of the target stratum, so as to optimize model parameters of each of the deep neural network models by fusing different trained deep neural network models into a set of spatial variation neural network prediction models according to the following formula:
• V
• p
• (
• i
• ,
• j
• ,
• k
• )
• =
• {
• w
• 1
• ,
• i
• ,
• j
• ·
• f
• 1
• (
• x
• 1
• ,
• x
• 2
• ,
• …
• ⁢
• x
• N
• )
• x
• 2
• ,
• i
• ,
• j
• ·
• f
• 2
• (
• x
• 1
• ,
• x
• 2
• ,
• …
• ⁢
• x
• N
• )
• ⋮
• w
• k
• ,
• i
• ,
• j
• ·
• f
• k
• (
• x
• 1
• ,
• x
• 2
• ,
• …
• ⁢
• x
• N
• )
• ,
• wherein i and j collectively represent a location in a plane, Vp represents a reservoir parameter, fk(x1, x2, . . . xN) represents a deep neural network model under a k-th type of geological characteristics, wk,i,j represents a spatial variation coefficient of the deep neural network model under the k-th type of geological characteristics, and x1, x2, . . . xN represents 1 to N types of seismic attributes;
• a parameter prediction unit which predicts the reservoir parameters of the target stratum by the set of spatial variation neural network prediction models;
• and
• a computing system which applies the predicted reservoir parameters to well placement planning.