insideBIGDATA Guide to Energy – Part 2

Sponsored Post

13 ways big data can improve efficiency and drive down costs in the energy industry

While these disruptors represent significant challenges, companies are finding ways toovercome those challenges by investing in big data analytics.

It’s worth noting that big data isn’t a simple or easy solution to any of the problemsenergy companies face. In order to do analytics well, you need the right mix of talent,hardware, and software. You need clearly defined problems, targeted objectives, and executive support to shepherd the project.

Putting all those elements into place is very difficult. According to some estimates,between 60% and 85% of all big data projects fail.

But the potential upside is so significant that most energy companies are investingsignificantly in big data technology.

What are those potential benefits? Here are 13 ways that big data analytics is helpingenergy companies manage the current level of disruption.

1. Better weather prediction

Today, energy companies have access to much more weather data than has ever beenavailable before. Many have their own weather sensors installed on key equipment, andthey also subscribe to both paid and free sources of weather data.

Many have also invested in advanced servers or built supercomputers that can enablethem to apply artificial intelligence (AI) or machine learning (ML) techniques to that data. This allows them to create highly accurate, localized weather forecasts. Armed with thisinformation, they can make adjustments that allow them to better prepare for andrespond to extreme weather events. While better weather prediction can’t stop or slowclimate change, it can make it easier for companies to deal with natural disasters causedby climate change.

2. Faster research

Big data analytics can also speed the process of conducting research. Whether companiesare searching for hidden oil reserves, inventing new types of photovoltaic panels, testingbatteries for energy storage, choosing locations for wind turbines, or doing other scientific work, big data analytics can help the process go faster.

Today’s servers are capable of volumes of calculations that would have been nearlyimpossible just a few years ago. That means companies can analyze more data muchmore quickly than ever before. That can help them make breakthroughs that might slowor mitigate global warming.

3. Preventive maintenance

Every energy company relies on equipment of some kind to help them produce, transmit,and/or deliver energy to consumers. Today, a growing number are installing IoT sensors that can detect minute changes in the way the equipment is operating. By performingadvanced analytics on this data, they can predict in advance when a particular piece ofequipment or part is going to need repair.

This information allows firms to schedule maintenance at a time when it is least disruptive to their operations. For example, if a utility knows that a transformer is likely to fail in thenext two weeks, they can take it offline for repairs overnight when demand is low, makingit easier for the rest of the grid to compensate.

4. Pipeline integrity

Companies that transmit oil and gas through pipelines can use a very similar process to detect and prevent future leaks. Spilling a large volume of oil and gas can be disastrousfor both the environment and a company’s reputation.

By installing sensors at key locations on the pipeline, companies can detect small changesin pressure, temperature, flow, density, or other factors that might indicate problems. Insome cases, they can also use computer vision techniques or ultrasound to detect cracksor dents in the pipe that could eventually result in a leak.

And if the worst happens and a leak occurs, these sensors provide the information to thecompanies right away, as well as allowing them to use analytics to determine the best response.

5. Security analytics

Most energy companies deal with near-constant cyberattacks—sometimes from stateactors as part of ongoing cyberwar and sometimes from ordinary cybercriminals hopingto make a buck or sow chaos.

Security professionals often feel like they are falling behind. As quickly as they can comeup with defenses, bad actors are coming up with brand new kinds of attacks thatcompanies have to figure out how to detect and prevent.

One of the most successful strategies for dealing with these evolving threats has been to rely on big data analytics. Many of today’s best cybersecurity tools use machine learningmodels to define a baseline “normal” level of activity on corporate networks and thenimmediately spot anything out of the ordinary. These tools aren’t foolproof, but they canmake energy companies safer.

6. Seismic surveys

For several decades, oil and gas companies have been relying on seismic surveys to helpthem locate deposits within the earth. After setting off small explosions, they use aseismic array to measure the waves as they flow through the earth’s crust, allowing themto create a visualization of what lies beneath the surface.

Today, geologists must look much deeper to find the oil and gas they are looking for. That requires larger arrays that generate much more data—generally terabytes or petabytes. To handle that much data, companies need hardware with scalable storage, fast processors,and advanced graphics processing units (GPUs) that will allow them to conduct analyticson their survey data to find the resources they are looking for.

7. Geophysics simulations

Scientists combine seismic survey data with other data to help them build geophysicalmodels. These models are incredibly valuable because they allow oil and gas companiesto predict with a high degree of accuracy where they will find underground reserves, aswell as the likely volume and quality of those reserves.

Today’s models are far more complex than those created in the past, relying on muchlarger volumes of data and frequently incorporating advance ML techniques. Again, thisrequires powerful servers similar to what is required for processing seismic surveys.

8. Talent management

Today, attracting and retaining high-quality workers is a make-or-break proposition formany energy companies. Because competition is so fierce, many companies are investingin talent management software to help them accomplish these goals. The best of these systems rely on big data analytics to identify the best candidates. Some companies arealso turning to predictive systems that attempt to identify staff members who are likely to leave the company so that managers can take action to try to get them to stay. But inorder to make these predictions accurately, the systems need a high volume of data.

9. Supply chain management

While no amount of data can make computer chips or other equipment magically appearwhen none are available, big data can enable better visibility into the supply chain, andbig data analytics can improve forecasts about which supplies are likely to be necessary.Energy companies have long used supply chain management tools to keep tabs on theflow of equipment and goods. By combining these resources with big data from otherparts of the organization, firms can improve the quality of the insights they are gaining,speed up operations, and reduce risk.

10. Predictive consumption models

Using advanced predictive analytics and ML algorithms, data scientists can create moreaccurate models of consumer energy use under various scenarios. Using these tools toanalyze historical energy data can’t tell you when international conflict or extreme weather is going to occur, but it can tell you what is likely to happen when events like these takeplace. That can help firms plan ahead so that they can better meet demand and keep theworld supplied with the energy it needs to function. It can also help them reduce the riskthat they will miss out on potential revenue because they are unable to keep up with demand.

11. Predictive price modeling

Data scientists can also apply similar modeling techniques to pricing, allowing them toforecast with some certainty what is likely to happen to energy prices in differentsituations. This information can help oil and gas companies decide when, where, andwhether to drill. It can help refineries decide whether to increase capacity or close plants. It can help utilities more accurately set prices for the energy they deliver to businessesand consumers. And it can help energy companies of all kinds become more competitive.

12. Speed

The process of delivering energy to end users is long and complex. Big data analyticsdoesn’t make any one piece of this process dramatically faster. However, it can makenearly every step a little more efficient. Taken as a whole, these improvements can have acumulative effect of making companies able to execute on their plans significantly more quickly. That speed can be tremendously important as companies seek to keep up withthe competition and respond to the current disruption in the marketplace.

13. Agility

Speed is closely related to agility. By their nature, most energy companies are notnaturally agile. You can’t drill an oil well or build a new power plant in a day. And onceprojects like this are underway, changing your mind carries a huge amount of risk. But thespeed afforded by big data analytics can help organizations make good decisions morequickly. In an industry not known for quickly reacting and adapting to change, anyimprovements in this area can make a significant impact on the bottom line.

Looking ahead

Most analysts believe that this period of intense disruption in the energy industry is likely to continue at least through the end of this decade. And the effects of climate change willprobably only intensify for many decades to come.

Fortunately, organizations have growing amounts of big data from a wide variety ofsources to help them deal with this disruption. The firms that navigate this period of timemost successfully could very well be those that do the best job of converting their big data into actionable insights that can guide their decision making.

How to build an environmentally friendly data center

Energy companies face a dilemma: In order to deal with the challenges created by global warming, they need very powerful computing infrastructure that can perform the big data analytics that can give them the insights they need. But those powerful computers themselves can add to global warming, making it difficult for companies to reach their sustainability goals.

Fortunately, it is possible to create a very powerful data center that is also environmentally friendly.

For example, the University of Cambridge Research Computing Services built one of theworld’s greenest supercomputers, the Wilkes 3, with Dell PowerEdge XE8545 servers. Infact, the Wilkes 3 is currently fourth on the Green500 list of the world’s most energy efficient supercomputers.

The Wilkes 3 system includes 80 nodes with 26,880 cores in its AMD EPYC 7763processors. The EPYC chips are the world’s highest performing x86 server CPUs, which isideal for workloads like big data analytics. Augmenting those CPUs are 320 NVIDIA A100GPUs, which help the system achieve 4.5 to 5 petaFLOPS of computational power whiledriving down the overall energy consumption.

Another organization that used Dell PowerEdge servers featuring NVIDIA GPUs to createan environmentally friendly supercomputer was the Italian firm Eni. In this case, thecompany chose to use solar power for its data center, making the installation even moresustainable.

Dell Technologies is committed to advancing sustainability through its processes and products, including its advanced server hardware for big data analytics.

Over the next couple of weeks we’ll explore these topics:

Download the complete insideBIGDATA Guide to Energy technology guide courtesy of Dell Technologies and AMD.

Filed Under: Big Data, Big Data Hardware, Big Data Services, Big Data Software, Energy, Featured, Google News Feed, News / Analysis, Sponsored Post, Uncategorized, White Papers Tagged With: AMD, dell technologies, energy, energy exploration, Gas and Oil, Weekly Featured Newsletter Post

What’s so Important about Data Quality?

In this special guest feature, David Kolinek, Vice President of Product, Ataccama, asks why data quality is so important? Sometimes taking a step back and reviewing the basics can help clear things up. DQ is not a single activity but a series of actions that draw upon multiple resources and functions, all focused on making data usable in a purposeful way. It falls under data management, which has an overriding mission of delivering a view of datasets from various perspectives, enabling the type and quality level of data to be assessed.

[READ MORE...]

Latest Video

White Papers

Garbage in, Garbage Out – How We Got Here and Why We Must Get Out Now

This whitepaper from our friends over at Profisee, reflects on why the state of data in mostorganizations is as dismal as it is, and why there is such a challenge involved in demonstrating the value of trusted data available across mission-critical operations and analytics in an enterprise.

DownloadSee More White Papers »