Technology takes great leaps forward while software and hardware provide enhanced value.
Light detection and ranging (LiDAR) — or more accurately, remote sensing and geospatial technologies — is the latest buzz in this smart world in general and in the electric utility industry specifically. For the past couple of years, utilities' interest in everything LiDAR has grown, thanks to practices like vegetation management and the North American Electrical Reliability Corp. (NERC) FAC-003, FAC-008 and FAC-009 standards.
The technology has been growing exponentially. LiDAR-based technologies have been evolving and maturing almost as fast as an idea for a new application is conceived, which is amazing, considering the complexity of the technology. Actually, it is several separate technologies working together simultaneously. The typical LiDAR system consists of a laser scanning system, precision gyroscopes, an inertial measuring unit and a GPS.
Growing Expectations
It seems like there is always the need to get better, more accurate data faster. Nothing seems to match LiDAR's abilities or potentials. There are three advantages LiDAR brings to the table, which are the driving force in the continuing development of the technology. First, LiDAR improves safety by permitting data collection without placing personnel in harm's way. Second, LiDAR is fast, and its speed reduces data gathering and design time from weeks to days, which reduces costs. Third, it is accurate. Its data is accurate to a few centimeters and is approaching millimeters in some applications.
Where once users were satisfied with the look-down perspective, today they expect to see the entire subject (top, bottom and sides), and that is another area LiDAR-based technologies have improved. The workhorse of LiDAR has been the airborne platform. Fixed-wing aircraft and helicopters have been gathering data, but that is only one of three configurations available for today's technology-enhanced digital world. There also are mobile and terrestrial LiDAR systems, which collect data from cars, trucks, trains, boats and tripods to get the rest of the picture.
LiDAR's Technical Triad
All LiDAR technological evolution has to address three key technical issues: data collection and analysis; error and accuracy; and the integration of LiDAR and photogrammetry technologies, LiDARgrammetry. When one aspect is altered, one of the other aspects is affected. If speed increases, it impacts accuracy and challenges the collection technology. Solving one problem frequently brings others where there was none. It is a frustrating process, but this is how to move to the next generation in any technology.
Customer demands and growing competition have hastened the evolution of LiDAR hardware even more. Consider the laser portion of the LiDAR system. Greater miniaturization has made lasers more portable, allowing adaptation to a greater variety of platforms. A wider selection of scanning patterns and wavelengths also are available today.
At the same time, developers have been increasing the pulse repetition frequency, too. Ten years ago, a repetition of 5,000 pulses per second was state of the art. The average system operates at 100 kHz today, but now 200 kHz is becoming very common. Then again, Riegl's Q680 device is rated 400 kHz, as is Optech's Pegasus HD 400; furthermore, Leica's ALS70 model is reported to be a 500-kHz device while flying at 1,000 m (3,281 km).
Jim Dow, CEO of Aerotec, reports, “Aerotec is using the TopEye Mk II elliptical laser-scanning system that integrates two 22-megapixel color cameras in the airborne data acquisition system. The innovative elliptical scan pattern provides a continuous forward, side-to-side and backward scan, which gives the system the ability to scan multiple sides of an object in one pass. The technology reduces flight time, but still detects even the smallest obstacles in the corridor.”
Another technology simplifying the process is the airborne meteorological sensor. Recently, GeoDigital announced it is using an air data-sensor probe on its aircraft, which is a technology originally developed for NASA. The probe measures wind speed and direction as well as ambient air temperature, humidity and solar load as the aircraft performs the LiDAR survey.
This data is recorded along each flight line in the vicinity of the wires on a span-by-span basis. GeoDigital takes the airborne data along with the standard line parameters and then, using Power Line Systems' PLS-CADD software, calculates the span sags and identifies the clearances.
Point Clouds Go High-Def
Improvements in sensor technology are another part of the evolution. Early sensors were large, bulky and limited in capability. They maxed out around 15,000 points per second for data gathering. Now, data is gathered above 200,000 points per second.
The rating term points per second differs from pulses per second. When the pulse is emitted, it hits objects and reflects from them. The improved sensors can catch four or five of these reflections from each pulse. Therefore, a LiDAR system with laser-emitting pulses at 100 kHz, combined with sensors capturing four or five returns per pulse, can be expected to collect 400,000 points per second or 500,000 points per second.
Because of this greater granularity, LiDAR providers can now create higher-quality 3-D models and other visualization products for their clients that look like imagery. It is not unusual for today's mobile LiDAR to provide readable street signs in its models.
Matthew Bethel, manager of systems engineering for Merrick & Co., says, “High data density is more than just how many times you fly back and forth over the transmission corridor. Sensors are the key; they have to stay in calibration, and the quality manufacturers are concentrating on that element. Merrick has performed testing with the equipment they use and has recorded a single strand of wire measuring less than 1 mm in diameter. That is an extreme, but it gives them a high degree of certainty in collecting varying types of field data.”
It also helps that improved direct georeferencing has been integrated into the latest systems. Data collection (points) can be located with absolute positioning more precisely, which becomes a real challenge as the scan and capture rates increase.
At a recent Consumers Electronics show, Nokia subsidiary Navteq introduced a mobile mapping system that has 64 rotating lasers capturing 1.5 million data points per second, at the posted speed limit. In addition, panoramic cameras capture additional detail to improve street-level imagery. Microsoft announced it will use this system for its Bing Maps service.
Increase in Memory
With the higher resolution and greater data collection, another problem has surfaced — memory. Scanning a linear mile of transmission can generate 30 million points of data (depending on sampling rates) or more, depending on the pulse repetition frequency and sensor capability. At that rate, scanning 100 miles (161 km) of transmission corridor could generate 3 × 109 raw data points.
Processing those data points into 3-D models will increase the file size. That means it would not be terabytes (1012) of storage anymore; it would be petabytes (1015) or more. Since the folks in the utility industry are a conservative lot, all of this data would be backed up several times. Technology has created the problem and solved it with another technology called cloud computing. Massive storage is available at reasonable costs and backup is part of the package.
Artificial Intelligence
Massive amounts of data have caused another problem: How does all of this information get processed? Any 5 year old can go to Google Earth and pick out objects, name them and see groupings. The best super computer cannot. It takes a lot of coding to do even the simplest object recognition, but this is coming.
Software developers are working on software to identify topographical features like hills, buildings or transmission lines. This capability is called automatic extraction, and the software can separate predefined features and display them in the LiDAR image, which is a big plus since manual costs are always higher than automated process. Machine learning took a quantum leap with IBM's Watson beating Jeopardy contestants earlier this year, showing computers can be trained to learn as they go along.
In the meantime, some LiDAR providers have solved this problem by going offshore with the data processing. They have contracted with companies in India, China and Romania to crunch the raw data into 3-D models and other deliverables in an effort to reduce costs. In the short term, data is being processed, but what about the sensitivities of detailed imaging of North American's power grid in the hands of foreign entities? That has to be a touchy subject for the likes of NERC and the Department of Homeland Security.
LiDAR's Age of the Machine
Speaking of security, another sensitive subject is the use of unmanned aerial vehicles (UAV) with LiDAR systems. Private companies are not allowed to use UAVs in the United States, but they are being used in other parts of the world.
The 7th International Symposium on Mobile Mapping Technology, held this year in Kraków, Poland, had several presentations on the subject of LiDAR and UAVs. UAVs are ideal for this application. By flying low and slow, they can really take advantage of the higher lasers' pulse repetition frequency. UAVs also are energy efficient and have a very small carbon footprint compared to full-sized aircraft.
Another promising unmanned technology is the Electric Power Research Institute's (EPRI's) transmission line inspection robot presently under development.
Andrew Phillips, director of transmission and substation research at EPRI, reports, “EPRI has been testing it at its laboratory in Lenox, Massachusetts. It attaches to the transmission line's shield conductor and has been outfitted with a variety of sensors including LiDAR. It is schedule to be deployed on a 138-kV transmission line in 2014.”
It is only a matter of time until utilities and regulators realize surveying for the recent NERC alert has to be a continuing event, and it is reasonable to expect the voltage levels to drop to lower levels. The transmission system is always changing due to storm events, new construction and various events that encroach on the right-of-way. Reliability is too important and the penalties too high for it not to be a continuing task. Technologies like UAVs and EPRI's robot will become necessary tools to control costs.
In the Cockpit
LiDAR system manufacturers and providers are developing applications that allow the LiDAR operator real-time verification of the data quality through a graphical user interface. The application lets the operator monitor the system while collection takes place. The operator can see actual measurements and even change the scanning parameters if necessary.
Earlier this year, Utility Risk Management Corp. (URMC) and EPRI began testing a new thermal imaging technology developed by URMC, called Thermal Direct. The Thermal Direct system is mounted directly on the aircraft, where it measures the actual temperature of the transmission line conductor. Controlled testing at EPRI's High-Voltage Laboratory continued through June 2011 when Thermal Direct measurements were compared to EPRI line monitors.
Adam Rousselle, president and CEO of URMC, reports, “Thermal Direct was able to measure the conductor temperature with a root-mean-squared error (RMSE) below 5°C (41°F) when conductor temperature varied in a range typical to operating transmission lines. Precision in the measurement of conductor temperature has always been considered the Holy Grail for utilities. I am thrilled with our recent breakthrough and look forward to sharing it with transmission operators across the country.”
LiDAR's Future is Bright
LiDAR has evolved from a basic mapping doodad to a sophisticated field data-collection system. Once an interesting gizmo, it has migrated to being a technological necessity. Airborne LiDAR systems have been joined by mobile mapping schemes and static terrestrial 3-D scanning. Nevertheless, probably the most remarkable feature about LiDAR is the fact it is no longer an experts-only tool.
Manufacturers, software developers and LiDAR providers have been focused on making user-friendly products without dumbing down the end product. Clients are demanding access to and the ability to manipulate digital elevation models (DEMs) for themselves.
Even Google Earth has been providing users streaming tools for quick visualization of LiDAR data and generating raster DEMs. Today 3-D building models are commonplace on smartphone digital maps and navigation systems found in cars, and they are beginning to make their way into the utility world, too. LiDAR technology is here to stay.
Companies mentioned:
Aerotec www.aerotecusa.com
EPRI www.epri.com
GeoDigital www.geodigital.net
