<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=Windows-1252">
<style type="text/css" style="display:none;"> P {margin-top:0;margin-bottom:0;} </style>
</head>
<body dir="ltr">
<div id="Signature">
<div>
<div class="WordSection1">
<p class="MsoNormal" align="center" style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;text-align:center; line-height:150%">
<b><span style="font-size:22.0pt; line-height:150%; font-family:"Times New Roman",serif">Department of Petroleum Engineering</span></b></p>
<p class="MsoNormal" align="center" style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;text-align:center; line-height:150%">
<b><span style="font-size:16.0pt; line-height:150%; font-family:"Times New Roman",serif">Master Thesis Defense</span></b></p>
<p class="MsoNormal" align="center" style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;text-align:center; line-height:150%">
<b><span style="font-size:14.0pt; line-height:150%; font-family:"Times New Roman",serif">Optimization of Injection Well Placement for Water flooding in Heterogeneous Reservoirs using Artificial Neural Networks Coupled with Reservoir Simulation</span></b></p>
<p class="MsoHeader" align="center" style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: Calibri, sans-serif;text-align:center; line-height:150%">
<b><span style="font-size:14.0pt; line-height:150%"> </span></b></p>
<p class="MsoHeader" align="center" style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: Calibri, sans-serif;text-align:center; line-height:150%">
<b><span style="font-size:14.0pt; line-height:150%; font-family:"Times New Roman",serif">Xinwei Xiong</span></b></p>
<p class="MsoHeader" align="center" style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: Calibri, sans-serif;text-align:center; line-height:150%">
<b><span style="font-size:11.0pt; line-height:150%"> </span></b></p>
<p class="MsoHeader" align="center" style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: Calibri, sans-serif;text-align:center; line-height:150%">
<b><span style="font-family:"Times New Roman",serif">Defense Date:</span></b><span style="font-family:"Times New Roman",serif"> Monday, April 22<sup>th</sup> 2019
</span></p>
<p class="MsoHeader" align="center" style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: Calibri, sans-serif;text-align:center; line-height:150%">
<b><span style="font-family:"Times New Roman",serif">Time: </span></b><span style="font-family:"Times New Roman",serif">12:00 pm – 2:00 pm</span></p>
<p class="MsoHeader" align="center" style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: Calibri, sans-serif;text-align:center; line-height:150%">
<b><span style="font-family:"Times New Roman",serif">Location Room:</span></b><span style="font-family:"Times New Roman",serif"> ERP9 Room 125</span></p>
<p class="MsoHeader" align="center" style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: Calibri, sans-serif;text-align:center; line-height:150%">
<b><span style="font-family:"Times New Roman",serif">Committee Members:</span></b></p>
<p class="MsoHeader" align="center" style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: Calibri, sans-serif;text-align:center; line-height:150%">
<span style="font-family:"Times New Roman",serif">Dr.</span><span style="font-family:"Times New Roman",serif">
</span><span style="font-family:"Times New Roman",serif">Kyung Jae Lee</span></p>
<p class="MsoHeader" align="center" style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: Calibri, sans-serif;text-align:center; line-height:150%">
<span style="font-family:"Times New Roman",serif">Dr. Mohamed Y Soliman</span></p>
<p class="MsoHeader" align="center" style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: Calibri, sans-serif;text-align:center; line-height:150%">
<span style="font-family:"Times New Roman",serif">Dr. Ahmad Sakhaee-Pour</span></p>
<p class="MsoHeader" align="center" style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: Calibri, sans-serif;text-align:center; line-height:150%">
<span style="font-family:"Times New Roman",serif"> </span></p>
<pre style="margin: 0in 0in 0.0001pt; font-size: 10pt; font-family: Courier;text-align:justify; text-indent:.25in; line-height:150%"><span style="font-size:12.0pt; line-height:150%; font-family:"Times New Roman",serif">Secondary recovery methods such as water-flooding and gas-flooding are often applied to depleted reservoirs for enhancing oil and gas production. Reservoir simulations are often applied to predict the hydrocarbon production of secondary recovery methods in heterogeneous fields. Given that a large number of discretized elements are required in simulations, it is usually not technically-and-economically feasible to run full-physics simulation for every possible case. Thus, a machine learning method is introduced to efficiently predict hydrocarbon production.</span></pre>
<pre style="margin: 0in 0in 0.0001pt; font-size: 10pt; font-family: Courier;text-align:justify; text-indent:.25in; line-height:150%"><span style="font-size:12.0pt; line-height:150%; font-family:"Times New Roman",serif"> Artificial Neural Network (ANN) is a machine learning method building black box proxy models. Hydrocarbon productions are predicted as a function of heterogeneity and injection well placement with relatively small number of training dataset, which are obtained with full-physics simulation models. We consider two different cases in this study—water flooding in homogenous reservoirs and heterogeneous reservoirs. In each case, waterflooding performance is quantitatively predicted by ANN models.</span></pre>
<pre style="margin: 0in 0in 0.0001pt; font-size: 10pt; font-family: Courier;text-align:justify; text-indent:.25in; line-height:150%"><span style="font-size:12.0pt; line-height:150%; font-family:"Times New Roman",serif">Considering the significant heterogeneity of reservoir models, we include the diverse predictor variables to train ANN models. Predictor variables include time-series data of production rates of oil and water, cumulative productions of oil and water, ratios between oil and water productions, injectivity, and pressure transient data at injector. Response variables include cumulative oil and water productions at the end of the waterflooding process.</span></pre>
<pre style="margin: 0in 0in 0.0001pt; font-size: 10pt; font-family: Courier;text-align:justify; text-indent:.25in; line-height:150%"><span style="font-size:12.0pt; line-height:150%; font-family:"Times New Roman",serif">Data-driven models based on ANN allow us to efficiently predict the performance of secondary recovery processes and optimize the injection well placement. Inclusion of most controlling input data for ANN modeling reduces the amount of data and observation time required to build the data-driven models. This suggested ANN modeling approach coupled with reservoir simulations will enable efficient decision making with reduced computational and monitoring cost.</span><span style="font-family:"Calibri",sans-serif"></span></pre>
<p class="MsoNormal" style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;line-height:150%">
</p>
<p class="MsoNormal" style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;">
</p>
</div>
</div>
</div>
</body>
</html>