Physics-informed Machine Learning for Real-time Unconventional Res- ervoir Management Maruti K. Mudunuru, Daniel O’Malley, Shriram Srinivasan, Jeffrey D. Hyman, Matthew R. Sweeney, Luke Frash, Bill Carey, Michael R. Gross, Nathan J. Welch, Satish Karra, Velimir V. Vesselinov, Qinjun Kang, Hongwu Xu, Rajesh J. Pawar, Tim Carr, Liwei Li, George D. Guthrie, Hari S. Viswanathan Earth and Environmental Sciences Division, Los Alamos National Laboratory, Los Alamos, NM-87544. Department of Geology & Geography, West Virginia University, Morgantown, WV 26506. {maruti,omalled,shrirams,jhyman,msweeney2796,lfrash,bcarey,michael_gross,nwelch,satkarra,vvv,qkang,hxu,ra- jesh,geo,viswana}@lanl.gov, {tim.carr,liwei.li}@mail.wvu.edu Abstract We present a physics-informed machine learning (PIML) workflow for real-time unconventional reservoir manage- ment. Reduced-order physics and high-fidelity physics model simulations, lab-scale and sparse field-scale data, and ma- chine learning (ML) models are developed and combined for real-time forecasting through this PIML workflow. These forecasts include total cumulative production (e.g., gas, wa- ter), production rate, stage-specific production, and spatial evolution of quantities of interest (e.g., residual gas, reservoir pressure, temperature, stress fields). The proposed PIML workflow consists of three key ingredients: (1) site behavior libraries based on fast and accurate physics, (2) ML-based inverse models to refine key site parameters, and (3) a fast forward model that combines physical models and ML to forecast production and reservoir conditions. First, synthetic production data from multi-fidelity physics models are inte- grated to develop the site behavior library. Second, ML-based inverse models are developed to infer site conditions and en- able the forecasting of production behavior. Our preliminary results show that the ML-models developed based on PIML workflow have good quantitative predictions (>90% based on R 2 -score). In terms of computational cost, the proposed ML- models are (10 4 ) to (10 7 ) times faster than running a high-fidelity physics model simulation for evaluating the quantities of interest (e.g., gas production). This low compu- tational cost makes the proposed ML-models attractive for real-time history matching and forecasting at shale-gas sites (e.g., MSEEL – Marcellus Shale Energy and Environmental Laboratory) as they are significantly faster yet provide accu- rate predictions. 1. Introduction Energy extraction from conventional resources involves producing crude oil, natural gas, and its condensates from Copyright © 2020, for this paper by its authors. Use permitted under Crea- tive Commons License Attribution 4.0 International (CCBY 4.0). rock formations that have high porosity and permeability (Bahadori, 2017). These rock formations are found below an impermeable rock. However, energy extraction from uncon- ventional hydrocarbon resources (Ahmmed and Meehan, 2016) involves using advanced drilling and stimulation techniques (e.g., long horizontal laterals and multi-stage hy- draulic fracturing) to extract crude oil and natural gas that are trapped in the pores of relatively impermeable sediment- ary rocks (e.g., shale, tight sandstones). Typically, unconventional reservoirs have porosity in the range of 0.04-0.08 and matrix permeability on the order of nanodarcies (10 -16 -10 -20 m 2 ) (Rezaee, 2015; Belyadi et al., 2019). Instead of the porous flow that dominates conven- tional reservoirs, fracture flow dominates unconventional reservoirs, with natural fractures dissecting the matrix and intersecting with the hydraulic fractures. As result, energy extraction is more difficult than conventional reservoirs. Model-based optimization of unconventional reservoirs is also challenging because due to the long horizontal laterals there is insufficient site data to inform high-fidelity physics models (Mohaghegn, 2017; Belyadi et al., 2019). Despite these challenges and due to the abundance of unconven- tional resources, with reserves projected to last for many decades, energy extraction from these resources have gained prominence in recent years (Briefing, 2013 and Weijermars, 2014). Current extraction efficiency from unconventional reservoir is very low (~5-10%) for tight oil and ~20% for shale gas (Sandrea, 2007; Muggeridge et al., 2014) com- pared to conventional reservoirs (~20-40%) (Zitha et al., 2008). This is because the impact of resource development