Wardah Inam

MIT EECS PhD


Bio

I am the co-founder and CEO of Overjet. At Overjet we are building technology to make dental care affordable and convenient for everyon. Prior to this, I was the Lead Product Manager at Q bio, focused on simulation and quantification of the human physiology. I did a postdoc at MIT CSAIL focused on Emerald – a machine learning based device which uses wireless signals for remote monitoring of vital signs and tracking of physical movement. I received my PhD from MIT focused on autonomous microgrids, Over the course of my PhD, I did two internships at GE and Apple. I co-founded a startup, uLink, to build technology that lowers the cost of electricity access. uLink's AI driven electronics and analytics platform enabled energy companies to provide lowest cost electricity by forming plug & play microgrids. uLink was recognized by National Geographic as one of tech breakthroughs that could help power the world.



Awards

2017 : MIT Hacking Medicine's Grand Hack Winner, won prize for the best use of Robot (Loomo)
2016 : MIT D-Lab Scaleup fellowship, deploying uLink for povery alleviation
2014 : MIT Clean Energy Prize Track Winner, won Infrastructure and Resource Track for startup uLink
2013 : MIT Graduate Women of Excellence Award, recognition for leadership and service to MIT community
2013 : US Department of Energy's Better Buildings Case Competition Winner, team received most innovative and best proposal awards for energy efficiency recommendations for commercial applications and city planning
2010 : Ghulam Ishaq Khan Gold Medal, highest aggregate academic standing in the entire university
2010 : Quaid-i-Azam Gold Medal, best overall, extra-curricular and academic, performance in the entire university
2010 : Electronics Engineering Gold Medal, best academic performance in the department
      First student to receive all three medals in the history of GIK Institute

Press

National Geogrpahic : 8 Tech Breakthroughs of 2015 that could help power the world
MIT Technology Review : Power to the people - Researchers create a bottom-up approach to electric grids
World Economic Forum : How to bring energy microgrids to rural villages
MIT Spectrum : Solar breakthrough powered by MIT
Engadget : SOFT Rockers combine solar panels and moving furniture to charge your gadgets

Research


Autonomous Microgrids

Modelling, stability and control algorithm for ad hoc grid

Ad hoc electrical networks are formed by connecting power sources and loads without pre-determining the network topology. There are two core aspects to ad hoc system design:
1) designing units such that the microgrid formed from the arbitrary interconnection is always stable and
2) developing control strategies to autonomously manage the microgrid in a decentralized manner and under large uncertainty.
To address these challenges we apply a nonlinear control techniques to obtain conditions under which an ad hoc dc microgrid will have a suitable and asymptotically stable equilibrium point. Further, we propose and test a new decentralized control algorithm that coordinates many sources to achieve a specified power dispatch from each.


Simulated network with sources are depicted as squares, and loads as circles.

[PDF] Stability, Control, and Power Flow in Ad Hoc DC Microgrids

[PDF] Stability and Control in Ad Hoc DC Microgrids

Architecure of peer-to-peer ad hoc grid and design trade-off's to reduce cost

This work presents a new concept of peer-to-peer electricity sharing which creates a marketplace for electricity. In this marketplace, the people who can afford power generating sources such as solar panels can sell electricity to people who are unable to afford generating sources or who might have access to electricity but require more electricity at certain times. These ad-hoc microgrids created by sharing of resources provide affordable electricity
We propose a method to trade-off efficiency and cost of converters to decrease overall system cost and demonstrate this proposed methodology with the development and comparison of different types of converters.

[PDF] Architecture and System Analysis of Microgrids with Peer-to-Peer Electricity Sharing to Create a Marketplace which Enables Energy Access

[PDF] Analysis and Design Methodology for System Cost Reduction in Distributed Power Systems

Schedulling algortihm for demand response

The electricity sharing network poses a problem in multi-agent power scheduling that is not solved by current approaches. The scheduling algorithm must be able to explicitly deal with uncertainty in generation, take into account users' competing demands for power and reliability, and be computationally feasible for a distributed network of extremely low-cost microcontrollers. Our advancement is creating a Market for Reliability (M.f.R.) algorithm similar to tatonnement but, unlike traditional tatonnement, we explicitly price reliability. This allows us to account for uncertainty and provide users with a probabilistic guarantee on achieving their desired activities.

[PDF] A Market for Reliability for Electricity Scheduling inDeveloping World Microgrids


Power Circuits, Systems and Control

VFX Control technique to improve efficieny of laptop power supply

Variable frequency multiplier (VFX) technique enables the design of converters for wide input and/or output voltage ranges while preserving high efficiency. The technique is applied to an LLC converterto demonstrate its effectiveness for converters having a wide input voltage variation such as universal input power supplies. This technique compresses the effective operating range required of a resonant converter by switching the inverter and/or the rectifier operation between processing energy at a fundamental frequency and one or more harmonic frequencies. This technique demonstrataed on a laptop power supply to increase its efficieny.

[PDF] Variable Frequency Multiplier Technique for High Efficiency Conversion Over a Wide Operating Range

High efficiency converter for solar applications

This paper presents a new topology for a high-efficiency dc/dc resonant power converter that utilizes a resistance compression network (RCN) to provide simultaneous zero-voltage switching and near-zero-current switching across a wide range of input voltage, output voltage, and power levels. The RCN maintains desired current waveforms over a wide range of voltage operating conditions. The use of ON/OFF control in conjunction with narrowband frequency control enables high efficiency to be maintained across a wide range of power levels.

[PDF] High Efficiency Resonant dc/dc Converter Utilizing a Resistance Compression Network


Robotics and Computer Vision

Particle filter based self-localization using visual landmarks

An approach to vision-based self-localization using the combination of particle filter and preprocessed image database. The robot uses particle filter with odometery data and landmark pose for position tracking. Furthermore, it uses the image database to globally localize itself when it has been kidnapped or when no landmark is visible.

[PDF] Particle filter based self-localization using visual landmarks and image database

Projects

uLink

Led a team of researchers to develop and deploy ad hoc microgrids to provide lowest cost electricity access

Images of Pilot '16


Emerald

Machine learning based device which uses wireless signals for remote monitoring of vital signs and tracking of physical movement


Soft Rocker

Increasing the adoption of Solar using good design. Solar powered outdoor furniture that's comfortable, beautiful and can charge your phone


SAVIOUR

Led the team to build Pakistan's first Rescue Robot. This robot was semi-autonomous and relied heavily on computer vision and low-cost sensors to detect survivors and map its terrain.


Miscellaneous

Investing

Angel investor at Sarmaycar - a seed and angel fund focused on startups targeting the Pakistani market