AI/ML for Smart Energy

Preventing Wildfires in Energy Transmission by Automatic Power Line Defects Detection Using Machine Learning and AI

The aging infrastructure, equipment failure, and high wind and storms all can cause fire, because when a power line breaks and hits the ground, it can arc and ignite dry vegetation. Wildfires may not only cause tremendous damages to property like destruction of homes, businesses, and infrastructure of roads and power lines, but can also result in loss of human lives. To save the tremendous costs as well as human lives, this paper first reviews the current state-of-the-art in preventing wildfires in energy transition, and then presents various innovative and advanced AI, deep learning, statistical learning methods for energy transition optimization and smart energy by addressing in particular wildfires prevention. Specifically, we will present some experimental results using our advanced image and video analysis technologies to inspect the electrical wires automatically for potential risks that may cause wildfires, especially in the woods and remote areas. To save the cost for image and video data collection, we will utilize the free image and video feeds from the Google Map’s street views as shown in Fig. 1. If some power line defects are automatically detected or some tree limbs may fall on the wires and break them, the maintenance crew should be informed to replace the wire with defects and trim the branches that endanger the power lines.

Fig. 1 Street view of electrical wires from Google Map

Machine Learning and AI for Optimizing and Safeguarding Energy Transmission in Storms by Automatic Inspection of Electrical Wires

Storms often come with high winds that can blow down trees and cause power line damages like downed lines, downed poles, which lead to power outage to communities. Such damages not only bring inconvenience to these communities, but may also endanger human lives because the falling live wires pose the potential danger of electrocuting people. Actually, storms cost often includes significant direct damage costs and rising insurance premiums. For example, Hurricane Sandy caused over $60 billion in damages in the New Jersey tri-state area, with New Jersey alone facing nearly $30 billion in losses. To save costs and protect human lives, this paper presents various innovative and advanced AI, deep learning, statistical learning methods for optimizing and safeguarding energy transmission in storms by means of automatic inspection of electrical wires in order to mitigate potential damages before they occur. Specifically, we will apply our advanced AI and deep learning approaches to automatically inspect and protect electrical wires from being damaged by the falling tree limbs during major storms. Such AI/ML inspections will take place well before the storms hit to avoid the costly damages as shown in Fig. 2 that happen time and again!

Fig. 2 Storm Damages in New Jersey.

Demos and Results

Some demos and preliminary results using our innovative AI, deep learning, computer vision, and pattern recognition methods are publicly accessible at (https://web.njit.edu/~cliu/AISH.html) for AI Doctor & Digital Assistant for providing the best care for anyone, anywhere, and anytime; at (https://web.njit.edu/~cliu/NJDOT/DEMOS.html) for automated traffic incidents detection and traffic congestion detection for improving traffic safety; at (https://frvp.njit.edu) for advanced facial detection and recognition for enhancing security and public safety. We will leverage these innovative methods for addressing energy transition optimization and smart energy tasks.