Condition Monitoring February Maintenance Preventive Maintenance Reliability

Pervasive Sensing Saves Money and Trouble

EP Editorial Staff | February 21, 2014


Deployed anywhere and everywhere in a plant, today’s intelligent wireless technologies supply critical equipment and process information that end-users could only dream of in the past.

By Jane Alexander, Deputy Editor

The more information you can have, the better? If you could see everything that’s going on in your facility and know the operating status of every piece of equipment, you could avoid countless problems. There would be far fewer surprises from unexpected failures and unplanned downtime. Plant asset performance would improve dramatically.

Alas, for end-users in the real world, several things have stood in the way of this “living the dream” scenario. First, how could a site capture so much data? Sensors would be needed everywhere: on rotating or reciprocating machinery (to monitor vibration); on pipes (to measure flow velocity); on vessels and other plant assets (to measure temperature); and in the surrounding environment (to detect leaks or fugitive emissions). Second, how many plants could afford this all-encompassing blanket of technology? While the price tag for such an extensive collection of required sensors might seem high, the added cost of wired systems to carry the data those devices generate back to where it’s needed would make the approach unfeasible for most. At least, it would in the past.

But things are changing. Pervasive sensing is becoming within reach at every site. The latest generation of intelligent wireless sensors includes products so economical and simple to install that they can be put to work all over an operation—in any area of the facility and on any part of its equipment systems or infrastructure. Once these smart devices are in place, their ability to supply a stream of crucial, real-time information can quickly pay for their widespread deployment. Given such benefits, how long will it be before the majority of sensors in a typical plant are wireless?

The future is now
The term “pervasive sensing” isn’t new, nor has it been floated solely in an industrial context. David J. Nagel (currently a Research Professor at George Washington University) used the term as the title of a technical paper in 2000.[1] In it, he made a case for, among other things, the potential pervasive sensing had in monitoring weather, energy usage and equipment health in electrical-distribution networks and information systems, and in improving process control, transportation, security and medicine.

Progress in pervasive sensing has been fueled by advancements in sensor design, coupled with legal requirements for additional monitoring. Consider the family car: Until the late 1980s in Europe and the 1990s in the United States, the only way to know the pressure in its tires was to manually check them with a gauge. If the tires were hot, the reading could be misleading. Going through this exercise on a weekly basis was so inconvenient that most people seldom, if ever, did it. In turn, fuel was wasted, tires wore prematurely and property and lives were lost when tires failed. Even those who checked their tire pressures regularly could not know about on-the-road tire-pressure problems until they occurred.

In time, tire-pressure-monitoring systems appeared, first in luxury cars. Following passage of the TREAD (Transportation Recall Enhancement, Accountability and Documentation) Act in 2000, they were required for all cars manufactured after 2007. Since then, these systems have gone from merely sending low-tire warnings to providing automated warnings with diagnostics to supplying complete tire-health visibility. As the amount of technology in them has increased, the amount of human effort required has decreased—and both fuel economy and safety have improved.

As explained by Peter Zornio, Chief Strategic Officer of Emerson Process Management, at the 2013 Emerson Global Users Exchange last fall, a similar evolution is taking place in industry, thanks to improvements in sensors and reductions in their cost. Measurements that used to be difficult and expensive to perform—and, consequently, were done only when necessary—have become easy and inexpensive. The benefits to an operation include lower operating costs, heightened process and site safety and enhanced reliability and energy efficiency.

Gathering what’s critical
According to Zornio, the data from pervasive sensing, once subjected to strategic interpretation, can generate actionable information that can significantly improve operating efficiency, emissions levels and safety, and help prevent unplanned shutdowns (see Sidebar, page 26). Much of this improvement is due to the spread of wireless sensors. They can be installed wherever needed, and without wired infrastructure that could add hundreds or thousands of dollars to the cost of each device.

Lots of these devices can be mounted without any process penetration, which saves time and money on installation (i.e., no drilling, tapping, welding or process shutdown required) and eases relocation of the units when circumstances dictate. Many require no commissioning and are maintenance-free—maintaining their accuracy without periodic calibration and offering lifetime reliability. (Note: Some wireless sensors do require process penetration. While they’re not easily relocated, wireless connectivity reduces their installation cost below equivalent wired versions.)

The first wireless sensors to become popular were vibration monitors for keeping tabs on rotating-equipment health. They can detect bearing wear, misalignment and chipped gear teeth. In pumps, they can spot cavitation, alignment issues and impeller or blade damage. By establishing baseline levels of vibration, facilities can keep track of equipment health and schedule maintenance before equipment failure causes a shutdown or worse.

Other variables that can be measured wirelessly include temperature and flow (using clamp-on ultrasonic sensors), plus sensors to detect flames, smoke, airborne chemicals and acoustic emissions from leaks and valve openings. Wireless devices in use today monitor valve position, liquid hydrocarbon spills, use of safety showers and opening of safety valves. Consider the following applications.

A major cost in many plants is created by leaking or otherwise defective steam traps, which waste enormous amounts of energy every year in the United States. According to the Department of Energy’s Office of Energy Efficiency and Renewable Energy (EERE), “in steam systems that have not been maintained for three to five years, between 15% to 30% of the installed steam traps may have failed.”[2]

EERE estimates that an open steam trap with a 1/8-inch orifice on a 150 psig steam line will lose 75.8 pounds of steam per hour at an annual cost of $6640 (assuming steam costs $10 per thousand pounds). Multiply that by the number of leaking steam traps in a typical plant, and consider that a steam trap could leak for a year before it is spotted, and the cost becomes obvious. A wireless ultrasonic sensor can detect a leaking steam trap and report it to a monitoring system immediately; some of these detectors are
available with software like Armstrong’s SteamLogic that will provide real-time information on the condition of a monitored steam-trap population.

Ultrasonic sensors detect the turbulent flow associated with leaking valves of all kinds: not only steam traps and pressure-relief valves, but also in shutoff, check, isolation and bypass valves. Leaking relief valves create several problems. They waste process materials and, depending on what is leaking, can increase VOC releases and create a personnel hazard or a fire or explosion hazard. Even a compressed air leak wastes money, and it’s not unusual to find plants with leaks that amount to 20% to 30% of compressor output, according to EERE.[3]

Many valves are checked for condition based on a set schedule, for obvious reasons: Removing one to take it to the shop for examination involves process shutdown or disruption and considerable expense. Use of ultrasonic detectors can help organizations shift from preventive maintenance (or, worse, reactive maintenance) to predictive maintenance. As a valve begins to leak, it emits ultrasonic energy that can be detected immediately and, like vibration levels of rotating equipment, be compared with a baseline to allow proper maintenance scheduling.

There are a number of ways to detect leaks of flammable or toxic gas. Probably the simplest is the same as that used for steam or compressed air: a wireless ultrasonic or acoustic emission sensor. This type of device will react instantly to leaks at ranges of 10 meters—regardless of what is leaking—and is not affected by air currents near the sensor.

The fact that an ultrasonic sensor will react to a leak of any kind can also be a drawback because it cannot identify what, precisely, is leaking. For that reason, it can be appropriate to back up ultrasonic detectors with one or more wireless fixed-point gas detectors. Battery-operated units that can be easily placed wherever needed are available.

Software makes it work
Pervasive sensing generates copious amounts of data and requires appropriate software to make sense of it all. Software like SteamLogic, among others, is often available from the sensor manufacturer. These tools can usually be set up to communicate with OSIsoft’s PI system or other asset-management packages. MT&AP

Additional Reading
1. David J. Nagel, “Pervasive Sensing,” SPIE Proceedings Vol. 4126 (28 November 2000), pgs. 71-82.

2. Steam Tip Sheet #1 (January 2012).

3. Compressed Air Tip Sheet #3 (August 2004).

Improving Business-Critical Decisions Across an Enterprise

For Emerson Process Management, pervasive sensing makes perfect sense: The ability to get more and deeper data, in ways they couldn’t before, about all aspects of their enterprises, expands end-users’ visibility into operating safely, reliably and profitably. To that end, as announced at last fall’s Emerson Exchange, the company is extending its focus beyond traditional process control and safety systems to address applications like site safety, security, reliability and energy efficiency in oil and gas, refining, chemical, power, mining and other industries, where installing additional sensors has traditionally been physically difficult, expensive or technically challenging. 

“Our customers are like anyone else,” notes Peter Zornio, Emerson’s Chief Strategic Officer. “They want actionable information that can make their lives safer, more predictable, while reducing costs, risk and time. This goes beyond the control room and optimizing process performance.” Pervasive sensing, he says, provides the clarity and certainty of conditions end-users need for making effective, business-critical decisions.

Emerson estimates that over the next 10 years, the pervasive sensing market will more than double the existing $16 billion traditional measurement market. It’s already seeing customers move aggressively to leverage this approach and technologies. The experience of a large Eastern European oil-processing plant is telling: The site has been able to increase its total number of sensors by 60% using wireless devices, with 2000 devoted to personnel safety, 8000 to plant reliability and 2000 for energy cost reduction.

Pervasive Sensing in Brief

The process is built on a three-pillared foundation:

• Innovative sensors that are multivariable, non-intrusive and cover wide areas

• Easily commissioned components that are wireless, self-powered and configuration-free

• No-maintenance devices that are accurate, calibration-free and have lifetime reliability

The extensive new data that comes out of this foundation is delivered to a Strategic Interpretation level, which sorts through it using sensor-awareness functions, new algorithms, industry knowledge and human expertise and, ultimately, presents the findings to users at the Actionable Information level.






View Comments

Sign up for insights, trends, & developments in
  • Machinery Solutions
  • Maintenance & Reliability Solutions
  • Energy Efficiency
Return to top