I love to make stuff with a purpose! It started with making widgets, evolved toward making software, and for the past several years has been focused on making Virtual Peaker. I started Virtual Peaker because of the tremendous opportunity for utilities to use consumer internet connected products to help balance the grid and mitigate the variability of renewable energy. My mission is to make demand as controllable as a gas turbine.
I'm a dyed in the wool engineer, and my defining quality as an engineer is the ability to conceptualize and manipulate the big picture. This systems thinking lead me to obtain a PhD in Mechanical Engineering with a focus on control systems and AI. By thinking about large interconnected system in terms of control theory, control engineers are able to manage and manipulate the actions of complex dynamic systems. This framework is perfect for manipulating billions of discreet connected devices to control their energy consumption!
Timeline
work
Job
school
Education
folder
Project
Résumé
Jobs
verified_user
0% verified
Founder and CEO
Virtual Peaker
Apr 2015 - Current(11 years 6 months)
VIrtual Peaker is a software as a service company that provides holistic control and scheduling of residential demand for the electric utility industry. We specialize in real-time optimization in the presence of dynamic pricing tariffs, but we also provide more traditional demand response and demand scheduling services. In my role as CEO, I am responsible for technology and business development.
B
Hustle and Muscle
Billstron Industries
May 2014 - Current(12 years 5 months)
I do engineering contract work. I'm a mechanical engineer with over 10 years of software development experience. I write code and develop hardware products.
CEO
Calculingua
May 2012 - Dec 2014(2 years 8 months)
Calculingua is a web application for scientists and engineers that is modernizing knowledge creation. I'm the founder, and I currently wear all of the hats -- development, marketing, and sales. I'm hoping to change that very soon.
Advanced Systems Engineer
GE
Jun 2010 - May 2014(4 years)
Broadly, my responsibility is the research and development of advanced algorithms and software related to home energy management and the connected home. In my role, I work on technology that is from 6 months to 5 years from adoption. The following is a sample of successful projects I lead: I developed an HVAC energy learning algorithm. I designed an application store paradigm for the connected home. I specified and prototyped a distributed restful architecture for real-time communications over XMPP.
Algorithm Consultant
EcoFactor
Sep 2009 - Apr 2010(8 months)
I developed learning and identification software for residential energy customers. I wrote software in Java and implemented it in the cloud.
Graduate Student Researcher
UC Berkeley
Aug 2006 - May 2010(3 years 10 months)
My research focused on the application of Control Theory to grid level energy consumption systems. Primarily, I focused on controlling the power consumption of groups of homes using programmable communicating thermostats (PCTs). In 2007, I was part of the team that prototyped the first PCT. In that team, my role was in software development and design. Beginning in 2008, I examined how to control the aggregate power demand of neighborhoods of PCTs using feedback control techniques and game theoretic algorithms.
Control Engineering Consultant
Holochip Corporation
Aug 2005 - Sep 2005(2 months)
I designed and built a prototype fluid-filled lens. I was responsible for the control hardware, electronics development, and testing.
N
Development Engineer
NanoMuscle
Aug 2004 - Nov 2004(4 months)
I worked on product development for automotive actuators. I was tasked with solving a problem he had with switch contact that shortened the life of the automotive actuators. I designed and build the Billstron. It was an Instron like machine with the sole purpose of testing and sorting our automotive actuators. This project exercised all of my analytical, software, and mechanical design skills. I designed the test fixtures, load cells, and moving stage. I wrote the software using LabView. Through experimentation, I determined the correct test procedure.
The start-up failed in Nov 2004.
N
Field Applications Engineer
NanoMuscle
Jan 2004 - Aug 2004(8 months)
I was responsible for sales related application support of our rotary actuators. My role included product prototyping and design, as well as sales calls and attending to customers.
Development Engineer
LEXMARK INTERNATIONAL INC
May 2000 - Sep 2003(3 years 5 months)
I worked in hardware development and motion control design and analysis for ink-jet printers and scanners.
Education
verified_user
0% verified
Doctor of Philosophy (PhD, Mechanical Engineering, Controls, Energy, Machine Learning
UNIVERSITY OF CALIFORNIA BERKELEY
Jan 2005 - Jan 2010(5 years 1 month)
Control of Residential Load Management Networks Using Real Time Pricing
Bachelor of Science (BS, Mechanical Engineering
University of Kentucky
Jan 1995 - Jan 2000(5 years 1 month)
I studied Mechanical Engineering, and I graduated Summa Cum Laude.
Projects (professional or personal)
verified_user
0% verified
C
Calculingua
Jun 2012 - Current(14 years 4 months)
Calculingua is a web application for scientists and engineers to do mathematical computation and analysis. It is like Google Docs for math.
Awards
verified_user
0% verified
I
Summa Cum Laude
Issued by University of Kentucky
Publications
verified_user
0% verified
I
Real-time Dynamic House Thermal Model Identification for Predicting HVAC Energy Consumption
IEEE Green Technologies Conference Apr
This paper presents a real-time algorithm to predict the energy consumption of the heating, ventilation, and air conditioning (HVAC) system at home. The autoregressive model with exogenous inputs (ARX model) is used to identify the house thermal model. The ARX model, with the thermostat controller, is simulated to obtain the future state of the HVAC system with the knowledge of the weather forecast data obtained from a weather server. The utility bill for the HVAC system can be estimated if a real-time price model is provided, thereafter. The proposed method is validated by experimentation in a particular home using GE Nucleus energy management system for data aggregation and algorithm implementation. The experimental results show that the
T
Low-Frequency Pulse Width Modulation Design for HVAC Compressors
The ASME International Design Engineering Technical Conferences Aug
I
Residential Electricity Auctions with Uniform Pricing and Cost Constraints
IEEE North American Power Symposium Oct
This paper considers electricity auctions as auctions of a divisible good when the bidders have soft cost constraints. Soft cost constraints are such that the bidder proportionally reduces the desired quantity when the price is over a defined threshold. We prove a policy-consistent mechanism to allocate the goods using uniform pricing. Further, the mechanism is communication efficient and polynomial time computable. We finish by describing how the auction mechanism could work for residential real time electricity pricing using intelligent programmable communicating thermostats.
N
Residential Load Managing Products in Smart Grid
NEMA Electro Industry Jun
I
Smart Appliances Energy Study
Int Journal of Computers and their Applications Sep
The purpose of this paper is to discuss an in-depth demand side management (DSM) study being performed by General Electric (GE) in conjunction with the Tennessee Valley Authority (TVA), and the Glasgow Electric Plant Board (GEPB) of Glasgow, Kentucky. GE has contracted the University of Louisville to perform data analysis for this study. The study is separated into multiple phases each of which is approximately three (3) months in length. Different demand response (DR) strategies are tested during these phases and energy consumption is measured. The study is ongoing, but this paper will be focused on completed phases and their results. These results include the energy performance of ENERGY STAR smart appliances, as well as the effects of tw