Iphimedia research and interviews with clients, show that most of the utility companies are interested to increase their inspection rate of electric distribution and are reluctant due to high costs involved. Unmanned aerial systems for top of pole image collection, cloud data repositories to handle the tremendous volume of data that these systems collect, and machine learning to process this data technologies have become mature enough to be deployed to reengineer the electric distribution inspection processes. This technology can reduce the costs of inspections, increase the reliability of the grid, and optimize the maintenance of the equipment.
A common problem amongst many utility companies that machine learning could solve is the ability of algorithms to automatically classify images into an inventory of equipment that is on an electric distribution pole. In order to teach a computer algorithm to find an element in an image, at commercial grade accuracy, one must provide the algorithm with the training set. A set of images that will ‘clarify’ to the computer what should it look for, or what it may ignore. The need for a high-quality training set of images, in fact, thousands of images that were manually verified as relevant by humans. Those images are the instructions for the algorithm. It will try to recognize patterns that will describe the learning goal --- to provide an accurate inventory of the equipment on the electrical grid.
Companies that attempt to achieve usable accuracy must get a hold of this data. Most likely by collaborating with someone who has the access and resources to analyze that huge amount of data. Iphimedia has a strategic intent to capture and curate a usably accurate database of top of distribution pole images then developing a data analytics company-- Pecos Data Systems -- to analyze this data set.
Iphimedia's innovation in production unmanned aerial system flying of electric distribution poles has proven the concept of cost effective top of pole utility equipment image collection capabilities. The challenges involved in implementing this capability for an electric utility include:
•The manpower to analyze the data manually to improve machine learning algorithms.
•Various lighting conditions and camera settings that fluctuate in every image, opposing challenges to recognize objects even if the algorithm has seen them before.
•Lack of experienced personal for research in the field.
Predictive maintenance is defined as the process of using analytics to determine as precisely as possible when an utility’s part should be replaced . While this approach to equipment maintenance has been available for several years, electric utilities maintenance operations are only just now understanding it to help ensure the full lifespan of parts is utilized, minimize the number of equipment checks, and maximize the reliability of the electric distribution system. Iphimedia's innovation in lightning arrester predictive maintenance through measuring the exposure of lightning pulses each lightning arrester has had over time is an example of a predictive maintenance algorithm. The challenges involved for implementing this capability for an electric utility are primarily the lack of experienced personal to develop this predictive maintenance approach.
To explain the extent of the problem, take for example a broken insulator on high voltage power line. The number of damages and faults that could occur are numerous, and for each type, we need set of learning images, that would have to be analyzed manually and classified as learning goal for the algorithm. We are talking about data that is large in quantity but also with wide variance too. Some objects and issues are easier for machine learning to identify than others. For example, it is easy for them to identify heat spots when using thermal inspections. The more inspections are made, the more relevant data there is, the better algorithms can be trained.
When it comes to technological and business disruption, the electric utility industry has had it comparatively easy over the past couple of decades. Electric utilities regulated monopolies have provided few incentives for changes or innovations in approach. Many other industries--- music distribution, taxis, retailers as examples---- have experienced disruptive innovations by the same technologies discussed in this article. And no doubt, the biggest lesson of all may be that the ultimate winners in almost every industry end up being those already figuring out how to get ahead of disruption before it even starts.