Interesting division between academia and industry about robust decision making in oilfield development, with an example of scenario discovery approach


, , , ,

Why there are so many methods developed for uncertainty quantification? Why many research works seem to be very far from actual industry practice? What is the reason for the big gap between industry and academia in thinking about practical problems? How to close the gap and make one’s research more meaningful?

In an earlier post Practical Elements of “Decision Making under Uncertainty”, I introduced the topic of Decision Making Under Uncertainty (DMUU)–the fundamental challenge in petroleum reservoir development planning. In this post, I want to discuss more details, partially based on our recently published paper Scenario Discovery Workflow for Robust Petroleum Reservoir Development under Uncertainty [1].

Continue reading


What’s new in 2017?

Last year, I finished an internship in Google it was about machine learning. Now, you can find more and more active studies on machine learning applications almost everywhere! Energy business is no exception! It is really a good time for me to keep reading, thinking and writing!

Expect some articles about uncertainty quantification in the near future! Also I am hoping  to have more interesting discussions about reservoir simulation, reduced-order modeling, and machine learning applications.

Check my ResearchGate page and Google Scholar Page for more updates on academic side! Also I will be attending the SPE Reservoir Simulation Conference in Texas!

Field Trip of California Power Sources


, , ,

Just come back from an interesting field trip (ENERGY 101A: Energizing California) to different types of power plants. It is very interesting to learn all sorts of knowledge that are related to our daily life. I created a map on Google map, see here.

Some interesting facts for the state of California

  1. California electricity generation in 2013 (exclude imports) is 200,077 GWhr, about 23GW power on average. Per capita usage: 750W or 0.75kW every person.
  2. California electricity generation in 2013 by source: 51% from natural gas, 11.8% from hydroelectric, 6.4% is from wind (huge increase recent years), 1.4% is from biomass. Other significant sources: geothermal, and nuclear.
  3. In US, California plays a role of “environment forerunner”. Many energy saving or environmental protection measures were first taken place in California. It is California actually drive the US standard up. Examples include the energy star system and higher environment standards. The federal sometimes “wait” for California standard to implement and then try to work on the other states in US.
2016-04-08 16.00.30

Giant wind turbine

Wind farm (Shiloh I Wind Power Plant by Iberdrola):

  1. Basics: about 100 GE 1.5MW wind turbines, capacity factor 28%, so on the average about ~30MW power generated. 10 full time employees.
  2. Electric price fluctuation: although Shiloh is under fixed price contract, the real-time electric price changes every 5 minutes. It can goes as high as $100/MWhr for peak hours & seasons and as low as maybe -$50/MWhr for low hours. (You need to pay money to output electricity to the grid!)
  3. Fully automated wind turbine rotor: the angle of the blade is adjusted to either parallel to the wind or at certain degree to turn the turbine on and off. The manager just pushed one button to shut down the turbine smoothly.
  4. The wind turbines placement: they are randomly scattered around the hills where there is lots of wind. As in late years, the wind has becoming more irregular.
  5. Wind farm is a huge distributed system: it is reliable because turbines won’t have problems all at the same time. However, some maintenance work, e.g. software updates can be really inconvenient…

Biomass power plant (Rio Bravo-Rocklin):

  1. Capacity 24MW power (quite continuous). It takes in 600-800 tonnes of biomass material per day. Tens of tonnes of ashes after burning can be used as things like fertilizer, construction material. There are 30 biomass plants in California.
  2. The main environmental benefit is that there is almost no pollution compared to burning those biomass materials in open air.
  3. Sources: construction wastes, agricultural things, wood pieces, saw dust, grains, little branches, pretty much any waste with high organic content.
  4. Biomass material will decompose naturally. Fresh wood pieces (yellowish) will become dark (dark or black) as time goes. Piled up biomass will look “smoky” as they decomposing!! “Wild fire” may be developed if they are overheat.
  5. Trucks will be lift up to 75 degree angle (!) to pour materials into the feeding machine. Looks like launching the truck rocket! See here.
  6. Unfortunately, the transportation of biomass from cities, forests to the biomass power plant is so high, so the electricity price is not favorable for them. So they have to shut down the plant.

Hydroelectric power plant (Folsom Powerhouse):

  1. Generates electricity from 1895-1952. Four generators. Total of 3MW power (equivalent to 2 wind turbines…). Now, it is a State Park/museum.
  2. Operates until the end of WWII. So it has the chance of survival (not being taken for military used).
  3. There is a telephone booth in the power plant. One need to literally jump into the booth and wear rubber gloves to make/take phone calls to avoid the electric shock!!
  4. First built Folsom prison, then the prisoners did the construction works.
2016-04-09 15.24.28

Typical gas field facility

Natural gas field by California Resources Corp.

  1. Basics: 150 gas wells ranging from 3000-12000 ft deep. Gas produced all sent to power plant, generates about 25MW electricity.
  2. Natural gas is the least carbon intensive fossil fuel.
  3. By contract, the six land owners would have direct access to the gas produced from the underground (not through PG&E). Causing extra work for the petroleum company.
  4. Land leases are for 99 years.
  5. There is a “scrapper” separates solids and liquids from the gas, then three-stage compression facilities compress gas into 800psi, and transported to power plant through the pipeline.
  6. Because of the low oil price and gas price, the production is actually low. No new wells being drilled. Of course no drilling engineers, completion engineers, neither reservoir engineers. They are mainly in Bakersfield. Only operation supervisor was introducing these to us.
  7. Newly replaced “smart”meters made by ABB can be remotely controlled on an iPad! The spirit of IoT!!

Some interesting fact about local hydrology (thanks Troy Barnhart):

  1. The Delta area (Sacramento–San Joaquin River Delta) is below sea-level. And water needs to be pumped out to make it dry and suitable for agriculture.
  2. Water is being pumped from the river out to South California. That became a huge debate between the water shortage in South California and the devastating ecology of the Delta area.
  3. The whole pumping system (from Delta to South California) uses about 800MW electricity!!! Human are reversing many many hydroelectric power plants.

It is a really fun trip to have. Also, it is so important for people who are so used to modern technology to understand what are the cost of the benefits, to understand where electricity comes from, what are limitations, etc. In that way, people can cherish our mother nature more.


CS229 Final Report: Exploring Price Volatility Using Topic Modeling


, ,

Now, we have summarized the work on oil prices study using topic modeling.  Some new things added:

  • Calculated the correlation between the topic trend and price with a temporal moving window. Users are now easier to identify the factors related to price volatility.
  • Topic modeling of a randomly sampled corpus with 338,828 New York Times articles. Topics are compared to oil prices.
  • Identified some interesting correlations, such as a positive correlation between entertainment-related topics and oil prices, and a negative correlation between the crime-related topics and oil prices. (Is that because media tend to focus on more important topics than crime?)
  • Discovered the correlation between the unemployment and oil prices by examining articles mentioning “unemployment”, which was suggested by LDA results.

For the full poster and report, refer to: the following links.

Exploring Commodity and Stock Volatility using Topic Modeling on Historical News Articles — Application to Crude Oil Prices [poster] [report]

Practical Elements of “Decision Making under Uncertainty”


, , ,

Why decision analysis? How to properly handle situations that are of uncertain nature? What are the tools used by business people to analyze the uncertainty, so that one can better make decision? I am trying to brought up several practical elements in regards to these questions.

Why decision analysis?

Making decision can be a very simple intuition process. However, it becomes much more complicated when the consequence of decision is really big, or, when the decision need to withstand critical eyes of many people.

Let’s assume that you, a company owner, just learned about a promising oil reservoir lying under the seafloor of Gulf of Mexico. However, the process of petroleum development involves hundreds of millions to billions of upfront investment before you get the first drop of oil (that is your return of investment). You are not so sure whether the oil recovered in the future can give you enough return of investment and make the company keep running. What should you do with this opportunity?

That is the real challenge that all oil companies need to face to everyday. Nevertheless, there are many other similar occasions for other policy makers and decision makers. In general, big risks, big influence and exposure to people’s critics lead to the emergence of a practice so-called decision analysis.

Continue reading

Lasso regression, the elegant method

As my study goes into somewhere with more statistical flavor, I would like to share a powerful tool that I learned by chance.


To study a complicated system, the data-driven approach focuses on the data. From analyzing the data, scientists try to make observations, discover patterns and regularities, draw conclusions, and so on. Then, scientists want to use these findings to make prediction. This approach is different from the physics-based methods, where people try to deduct/simulate the system behavior from some clearly defined principles.

Continue reading

Reservoir Simulation – Basics



As my research is about reservoir simulation, it would be good to go over the fundamentals of the standard reservoir simulation. What is it? Why people need it? What are its key physical principles that reservoir simulation is based on? What are the uses of it? It is important to understand those basics before using it. And it is even more important to present it in a fun and simple way!


Why do we need simulation? Essentially, we want to better take control of something. A model is a replica of the reality with respect to things that you are working on. You simulate the actual practice by playing around with the model. Just like to simulate the actual horse-riding using a horse simulator, or to practice outdoor rock climbing using the artificial climbing wall in the gym! All types of movements needed in outdoor climbing can be learned at the gym. Continue reading

S3GRAF — Reservoir Simulation Post Processing Tool



Stumbled upon the website of Sciencesoft, a software company focusing on reservoir simulation post-processing  (and pre-processing) tool. The main product of the software suite is called S3GRAF, featuring visualization and analysis of reservoir simulation results. It is designed to support almost all mainstream simulation software: Eclipse, CMG, VIP, 3DSL, UTCHEM, etc.

The beauty of this kind of software is that it address the practical need of reservoir engineers, such as visualization of the reservoir property, overlapping different types of visualization, manipulation of reservoir model according to one’s need, and even syntax check for the input files.

S3GRAF-3D is worth of special attention. The 3D visualization functionality is great.

S3GRAF 3D visualization effect -- polyline and ploygon

S3GRAF 3D visualization effect — polyline and ploygon

Continue reading