|dc.description.abstract||The Permanent Downhole Gauge (PDG) can monitor the reservoir in real time over a
long period of time. This produces a huge amount of real time data which can
potentially provide more information about wells and reservoirs. However, processing
large numbers of data and extracting useful information from these data brings new
challenges for industry and engineers.
A new workflow for processing the PDG data is proposed in this study. The new
approach processes PDG data from the view of gauge, well and reservoir. The gauge
information is first filtered with data preprocessing and outlier removal. Then, the
well event is identified using an improved wavelet approach. The further processing
step of data denoise and data reduction is carried out before analyzing the reservoir
The accurate production history is very essential for data analysis. However, the
accurate production rate is hard to be acquired. Therefore, a new approach is created
to recover flow rate history from the accumulated production and PDG pressure data.
This new approach is based on the theory that the relation between production rate and
the amplitude of detail coefficient are in direct proportion after wavelet transform.
With accurate pressure and rate data, traditional well testing is applied to analyze the
PDG pressure data to get dynamic reservoir parameters. The numerical well testing
approach is also carried out to analyze more complex reservoir model with a new
toolbox. However, these two approaches all suffer from the nonlinear problem of PDG
pressure. So, a dynamic forward modelling approach is proposed to analyze PDG
pressure data. The new approach uses the deconvolution method to diagnose the linear
region in the nonlinear system. The nonlinear system can be divided into different
linear systems which can be analyzed with the numerical well testing approach.
Finally, a toolbox which includes a PDG data processing module and PDG data analysis
module is designed with Matlab.||en_US