The secure and reliable operation of an electric power grid is critical to national security. Power grid components such as the state estimator used to monitor the operating state of a power system are subject to cyber-attacks. Previous works show that an intruder can compromise the state estimation by injecting the pre-designed false data into meters without being detected if the detailed knowledge of a transmission grid is known.
Electric power is almost entirely transmitted through polymer insulated cables or wires in every home, factory, plant, or apparatus. If the temperature of a cable increases, it would be an indication that some accidents or malfunctions such as inflow of excess electric current occur in the cable. The generated heat, indeed, degrades the polymer insulations in cable, thus, making it unsuitable and unsafe for extra service. Therefore, it would be markedly valuable if the thermally-degraded portion in the cable can be located accurately without destroying the cable.
Deep learning (DL) algorithms have achieved phenomenal success in different AI applications in recent times. Training DL algorithms require huge computational resources. Therefore, cloud or high-performance computing at the edge are obvious choices for this task. However, during inference cloud computing is not a suitable choice because of latency issues. There are billions of devices and sensors connected to the Internet, and data generated from these cannot be transferred and processed in geographically distant cloud data centers without incurring delays.
In this project the intern will work with time series data containing different parameters from a flight simulator. The intern will take these data and assess different learning as well as change point detection algorithms that can identify and segment pilot reactions to malfunctions and assign these reactions a proficiency metric. One of the possible approaches would be to assess the use of change point detection algorithms using a data driven approach. This will allow the partner organization to understand important segments of the flight data where large changes have taken place.
Increasing visibility into asset inventory and gaining situational awareness in industrial control system (ICS) environment are of critical importance for electric utilities to effectively manage cybersecurity risks. In this project, we aim to investigate the solution design for electric utilities to automate asset management in an ICS, addressing the unique challenges such as device heterogeneity and legacy technology.
This project aims to monitor the communications among the devices connected in a building automation system. It is important to detect any changes in the normal communication pattern of the devices. Such changes often signify a change of operational behavior, caused either by a failure or an impending breakdown. In the worst-case scenario, anomalies deviated from normal communication pattern may indicate the system is under a cyberattack, which would have serious consequences to the people inside and the building itself.
Computer hardware is enjoying a widespread renaissance with the emergence of the compute-intensive and challenging machine learning workloads. ABR develops and maintains NENGO, a biologically-plausible model for neural networks and is keen to develop hardware support for efficient realizations of these networks. FPGA (Field-Programmable Gate Arrays), are an attractive target for this if we can overcome the communication bottlenecks that limit the effectiveness of these designs.
The major contributing factor to waterborne outbreaks in Canada in small drinking water systems is the operators’ lack of technical expertise. Training of small system operators do not cover hands-on training specific to the treatment technologies used in their plants. A simplified smartphone app with real-time monitoring can assist the operators with the decision making process. Aqua Intelligent Technology Inc. is providing this smart solution for small water treatment systems. The vital step in this technology is receiving real-time data from the sensors in such facilities.
simulation package available) to simulate a zooming system composed of tunable liquid crystal lenses
that is capable of aberration correction. The first component of the project consists of simulating a liquid
crystal lens in COMSOL. The liquid crystals rearrange themselves based on a space varying electric
field created by the placement of multiple electrodes and special materials. The crystal rearrangements
produce a rapid change in local refractive index.
The team of interns will together build a complete framework for collecting quality data, building flood models, and visualizing the results in an understandable and comprehensive manner. The benefit to the partner organization will be the development of the services to communities at risk of flooding in British Columbia. The deliverable product developed by the intern team will be a flood forecasting application hosted with cloud services.