Machine learning for production optimization

07 November 2017 by Vegard Flovik
Fully autonomous production facilities will be here in a not-too-distant future. But even today, machine learning can make a great difference to production optimization.

In a previous blog post, Michael Link gave an introduction on how to get started with data and analytics projects. Here, I will take a closer look at a concrete example of how to utilize machine learning and analytics to solve a complex problem encountered in a real life setting.

What is production optimization?

Product optimization is a common problem in many industries. In our context, optimization is any act, process, or methodology that makes something – such as a design, system, or decision – as good, functional, or effective as possible. Decision processes for minimal cost, best quality, performance, and energy consumption are examples of such optimization.

Follow Kongsberg Digital on Linkedin

To further concretize this, I will focus on a case we have been working on with a global oil and gas company. Currently, the industry focuses primarily on digitalization and analytics. This focus is fueled by the vast amounts of data that are accumulated from up to thousands of sensors every day, even on a single production facility. Until recently, the utilization of these data was limited due to limitations in competence and the lack of necessary technology and data pipelines for collecting data from sensors and systems for further analysis.

Within the context of the oil and gas industry, production optimization is essentially “production control”: You minimize, maximize, or target the production of oil, gas, and perhaps water. Your goal might be to maximize the production of oil while minimizing the water production. Or it might be to run oil production and gas-oil-ratio (GOR) to specified set-points to maintain the desired reservoir conditions.


How complicated is product optimization?

The production of oil and gas is a complex process, and lots of decisions must be taken in order to meet short, medium, and long-term goals, ranging from planning and asset management to small corrective actions. Short-term decisions have to be taken within a few hours and are often characterized as daily production optimization. They typically seek to maximize the oil and gas rates by optimizing the various parameters controlling the production process.

In most cases today, the daily production optimization is performed by the operators controlling the production facility offshore. This optimization is a highly complex task where a large number of controllable parameters all affect the production in some way or other. Somewhere in the order of 100 different control parameters must be adjusted to find the best combination of all the variables. Consider the very simplified optimization problem illustrated in the figure below.

In this case, only two controllable parameters affect your production rate: “variable 1” and “variable 2”. The optimization problem is to find the optimal combination of these parameters in order to maximize the production rate. Solving this two-dimensional optimization problem is not that complicated, but imagine this problem being scaled up to 100 dimensions instead. Now, that is another story. This, essentially, is what the operators are trying to do when they are optimizing the production. Today, how well this is performed to a large extent depends on the previous experience of the operators, and how well they understand the process they are controlling.

Machine learning algorithms can accumulate unlimited datasets

This is where a machine learning based approach becomes really interesting. The optimization performed by the operators is largely based on their own experience, which accumulates over time as they become more familiar with controlling the process facility. This ability to learn from previous experience is exactly what is so intriguing in machine learning. By analyzing vast amounts of historical data from the platforms sensors, the algorithms can learn to understand complex relations between the various parameters and their effect on the production.

Follow us on Twitter

The fact that the algorithms learn from experience, in principle resembles the way operators learn to control the process. However, unlike a human operator, the machine learning algorithms have no problems analyzing the full historical datasets for hundreds of sensors over a period of several years. They can accumulate unlimited experience compared to a human brain.

How an optimization algorithm works

Having a machine learning algorithm capable of predicting the production rate based on the control parameters you adjust, is an incredibly valuable tool. Referring back to our simplified illustration in the figure above, the machine learning-based prediction model provides us the “production-rate landscape” with its peaks and valleys representing high and low production. The multi-dimensional optimization algorithm then moves around in this landscape looking for the highest peak representing the highest possible production rate.
By moving through this “production rate landscape”, the algorithm can give recommendations on how to best reach this peak, i.e. which control variables to adjust and how much to adjust them. Such a machine learning-based production optimization thus consists of three main components:

1. Prediction algorithm:

Your first, important step is to ensure you have a machine-learning algorithm that is able to successfully predict the correct production rates given the settings of all operator-controllable variables.

2. Multi-dimensional optimization:

You can use the prediction algorithm as the foundation of an optimization algorithm that explores which control variables to adjust in order to maximize production.

3. Actionable output:

As output from the optimization algorithm, you get recommendations on which control variables to adjust and the potential improvement in production rate from these adjustments.

A machine learning-based optimization algorithm can run on real-time data streaming from the production facility, providing recommendations to the operators when it identifies a potential for improved production. A typical actionable output from the algorithm is indicated in the figure above: recommendations to adjust some controller set-points and valve openings. It also estimates the potential increase in production rate, which in this case was approximately 2 %.

This machine learning-based optimization algorithm can serve as a support tool for the operators controlling the process, helping them make more informed decisions in order to maximize production.

Fully autonomous operation of production facilities is still some way into the future. Until then, machine learning-based support tools can provide a substantial impact on how production optimization is performed

Machine learning
About the writer
Vegard Flovik
Vegard Flovik is a principal engineer in the AI Center of Excellence at Kongsberg Digital where he solves real-world problems for various industry sectors using machine learning and advanced analytics approaches. He has a Ph.D. in physics from the Norwegian University of Science and Technology (NTNU) and is a Master of Science in condensed matter and materials physics. Vegard has previously worked as a researcher for Statoil and several universities in Norway and abroad.