The cutting edge of shale technology?
…the most productivity-enhancing set of technical papers of any year in the database. Recent areas of innovation include completion design, fracturing fluids, EORand machine learning. We also break down the…
…the most productivity-enhancing set of technical papers of any year in the database. Recent areas of innovation include completion design, fracturing fluids, EORand machine learning. We also break down the…
…across transistors, DRAM, GPUs and deep learning. GPU efficiency will inevitably increase, but compute increases faster. AI most likely uses 300-2,500 TWH in 2030, with a base case of 1,000…
…the 1980s, 1990s and early 2000s. Total development capex actually declined from 1970s levels over this time frame, in real terms, due to learning curve effects. Activity levels have also…
…the CEOs of Fortune 500 companies. I am never going to impress them with a question about machine learning. No. I like to ask dumb questions. And I make money…
…nuanced in savanna landscapes and must be balanced with other environment goals, especially biodiversity (pages 8-11). This is especially true for fire suppression (pages 12-14). Learning curves are crucial (pages…
Learning curves and cost deflation are widely assumed in new energies but overlooked for nature-based CO2 removals. This 15-page note finds the CO2 uptake of well-run reforestation projects could double…
…completion designs (page 8), optimizing completion fluids (page 8), Shale-EOR (page 9) and a step-change in machine learning algorithms (page 10-11). The leading companies are highlighted on page 12, ranked…
…more extensive than expected. For example, Stanford’s “Deep Solar” project, has used machine learning to identify over 1.5M solar installations from 1bn satellite images. 5% of houses in California are…
…can reduce refinery CO2 intensities by 5kg/bbl. Pages 18-21 highlight breakthrough, digital technologies to improve the development of new catalysts, including super-computing and machine learning techniques. Pages 23-24 screen 35…
…learning software is used to identify sharks with 90% accuracy, compared with 30% for human operators. Training a drone to identify sharks versus dolphins is computationally similar to identifying vulnerable…