Hyperspectral Imaging in the Oil & Gas Industries: A Review
The oil and gas industry, aka petroleum industry, is a nearly $6000 Billion dollar global industry with a Compound Annual Growth Rate (CAGR) of 25%. In the United States alone they extract some 12 million barrels per day at the average price of $55 dollars a barrel. Indeed since the discovery of the Drake Well in Pennsylvania in 1859 this commodity has been so integral to the functioning and livelihood of Americans that it earned the name “black gold”. Even with the introduction of alternative energies, oil and gas has remained the predominant form of energy and has in fact only grown in size and scope employing roughly 10 million jobs in the United States.
The industry is composed of three general segments: The upstream exploratory and acquisition segment, the midstream transportation segment, and the downstream refinery segment. This blog will focus on the downstream refinery segment and its quality control.
Crude oil and gas has been found in 31 states, with Texas, North Dakota, and New Mexico currently leading in onshore extraction. Primary hydrocarbon reserves are normally found in Mesozoic, Devonian, and Permian formations. Extraction yields a combination of hydrocarbons in the form of alkanes, naphthenes, aromatics, and asphaltics. It also includes some level of oxygen, sulfur, and metals, most notably lead.
In order to determine the appropriate end-product, of which there are many, as well as to ensure its quality, a laundry list of analyses must be conducted. These include but are not limited to:
- Total Acid Number (TAN): corrosiveness due to acid content (Titration)
- Density (hydrometer)
- Viscosity (capillary test)
- Pour point (water bath)
- Copper corrosiveness due to sulfur compounds (copper strip test, x-ray)
- Refractive index (refractometer)
- Lead test (x-ray fluorescence spectroscopy)
- TBC polymerization inhibitor (UV spectroscopy)
These tests all have one thing in common – they are performed off-line. That is to say, they aren’t performed on crude oil or during various stages of refining within pipelines or holding tanks. They are not automatic. Samples are manually removed and taken “off-line” to a laboratory setting where various instruments (e.g. hydrometers and refractometers) are used. The problem with this is two-fold
1) It takes time and manpower to perform tests in this way
2) It limits the number of tests that can be performed
Anytime you’re moving samples away from the production line it’s going to amount to increased cost, decreased time, OR limited oversight. What you really want to shoot for is test methods that
1) Can be conducted at-line or in-line
2) Can be more frequent or continuous
3) Can be rapid
4) Can monitor multiple parameters
5) Requires minimal supervision
Hyperspectral imaging provides these characteristics. Optical-based sensors can be placed near production lines (at-line) because they don’t require the reagents or sample prep that would prevent them from being conducted in that environment. These sensors can even be placed in the piping itself to take readings (in-line). Even at-line can dramatically increase frequency and speed, and in-line provides opportunities for continuous monitoring. The lack of sample prep and sample movement minimizes supervision. Furthermore, sensors can be tailored to monitor multiple parameters simultaneously streamlining quality control.
Some of the analyses conducted in the quality control of crude oil refining, unfortunately, cannot be conducted in this way. Density, viscosity, pour point, and lead content are examples. However, a number of analyses currently conducted off-line such as TAN, sulfur content, refractive index, and TBC polymerization inhibitors can be monitored with optical-based hyperspectral imaging.
It’s not an all or nothing proposition. In an industry this large, even if a few of the quality control methods that are plaguing the refining process can be replaced with better methods the savings can be astronomical. For this reason alone, the oil and gas industry would do well for itself to explore the introduction of hyperspectral imaging in the quality control of crude oil and its byproducts.