Micromem Technologies Inc. (MMTIF)

[sjbexpress]

Chevron Launching Predictive Maintenance to Oil Fields, Refineries
Company aims to have thousands of pieces of equipment sensor-enabled by 2024
By Sara Castellanos
Sep 5, 2018 5:04 pm ET
0 COMMENTS
Heat exchangers at Chevrons El Segundo, Calif. refinery
Heat exchangers at Chevrons El Segundo, Calif. refinery PHOTO: CHEVRON CORP.
Chevron Corp. has launched an effort to predict maintenance problems in its oil fields and refineries, a capability that many companies have been working for years to cultivate and is just now gathering momentum.

Advances in the functionality and economics of sensors, data analytics and cloud computing are behind the rise of so-called predictive analytics, which Chevron executives say could lead to savings of millions of dollars annually.

Working with Internet of Things services from Microsoft Corp., Chevron aims to enable thousands of pieces of equipment with sensors by 2024 to predict exactly when equipment will need to be serviced.

“In the past, we had to figure out how equipment was performing,” said Chief Information Officer Bill Braun in an email. “In the future, the equipment will tell us how it’s performing. This represents a big shift.”

Microsoft’s $5 billion, four-year investment in the sector was announced in April. There will be an estimated 25.1 billion devices connected to the internet by 2021, up from 6.3 billion in 2016, according to recent research from Gartner Inc.

Total spending on IT, including data center systems, enterprise software and connected devices is expected to reach $4 trillion in 2021, up from $3.4 trillion in 2015, according to Gartner.

Chevron CIO Bill Braun says the Internet of Things has “very broad-reaching potential for an industry that is equipment-intensive.”
Chevron CIO Bill Braun says the Internet of Things has “very broad-reaching potential for an industry that is equipment-intensive.” PHOTO: CHEVRON CORP.
Chevron in 2017 signed a seven-year deal to make Microsoft’s Azure its primary cloud provider. The partnership will give Chevron’s engineers access to data in one cloud repository instead of in different silos within the organization, said Mr. Braun. That means it’ll be easier to gain insights from a wider set of data including information from sensors connected to equipment in the field.

“That, we believe, is a competitive differentiator,” Mr. Braun said. Internet of Things has “very broad-reaching potential for an industry that is equipment-intensive like ours,” he added.

In recent years, advancements have been made in the quality and affordability of sensors, as well as the cloud-based platforms required to gather and analyze data being streamed from devices in the field. It’s also become easier and quicker to outfit machines with wireless sensors, whereas in the past, sensors typically required weeks worth of wiring and installation.

Chevron expects to outfit oil machinery with sensors for predictive maintenance by 2019 in a wide-scale pilot program, with full adoption for many of the machines expected by 2024.

Over the past six months, it has worked on a small experiment connecting additional sensors to a few heat exchangers, which are widely used in processing oil and gas. The massive machines, which range in size from large trucks to small buildings, are similar to radiators that cool down car engines, and an unplanned outage could take days to resolve, translating to significant financial losses, said Deon Rae, head of Chevron’s Industrial Internet of Things Center of Excellence.

In the past, two sensors on the heat exchanger would collect information such as temperature of the cooling fluid and the oil flowing through the heat exchanger. But with limited sensor data, Chevron could only know about the current and past state of the machine.

In the experiment, four wireless sensors were put in strategic places along the machine, which captured a wider dataset, including information about temperatures and oil flow. With the additional data and Microsoft’s cloud-based predictive analytics applications, Chevron’s data scientists can predict when the heat exchanger will become dirty and need cleaning, Mr. Rae said.

There are 5,000 heat exchangers across the company’s oil and gas operations worldwide, Mr. Rae said. Over the next few years, sensors and data could be used to consistently monitor the health of the machines in real-time, helping data analysts better predict when it’s necessary to service each of them. That would reduce equipment downtime and unnecessary repairs, he said.

“Savings could be in the millions if you can monitor and predict the health across all of our exchangers,” Mr. Rae said.

By 2024, Chevron aims to have sensors connected to much of the high-value equipment that could significantly disrupt oil and gas operations and create lost profit opportunities if they ever broke down, he said. Such equipment includes compressors that are used to reduce the volume of gas by increasing pressure, and pumps used to move liquids.

“The value creation is going to be from the combination of assets that we can monitor,” Mr. Rae said.

Share this:
https://on.wsj.com/2NQKQe9
BILL BRAUN CHEVRON INTERNET OF THINGS PREDICTIVE ANALYTICS PREDICTIVE MAINTENANCE SENSORS

PREVIOUS
State Agencies Clinging to Mainframe Computers

NEXT
The Morning Download: Chevron Bringing Predictive Maintenance to Oil Fields, Refineries



Does anyone see a connection here with MMTIF Maybe?