Skills

C/C++

90%

Quantitative analysis

70%

Photography

10%

Java/J2EE

30%

Experience

 
 
 
 
 
February 2017 – Present
San Jose, California

Senior Research Engineer

GEIRI North America

Dr. Yu has been employed at GEIRI North America since February of 2017. In his position, he is responsible for data analysis and algorithm developments. Particularly, he is working on bad data detection, data process and analysis, pattern recognition, pattern extraction, training of AI models, and algorithm design. He is also developing software to implement algorithms and visualizing the analysis results. Responsibilities include:

  • Data Processing
  • Model Training
  • R&D
 
 
 
 
 
January 2011 – December 2016
Ithaca, New York

Research Assistant

Cornell University

In this research position, Dr. Yu performed various duties, including teaching assistant (TA) and research assistant (RA). As a TA, he helped me to prepare lectures, give recitations, lead labs, and design exams. As a RA, he is responsible to formulate problems, collect and process data, build and validate models, develop algorithms and control, carry out numerical experiments, write paper, and present results. Responsibilities include:

  • Analysing
  • Modelling
  • Algorithm Design
  • Course Teaching

Selected Publications

The problem of using metric learning in forced oscillation source locating is considered.
In T&D, 2018

The problem of stochastic deadline scheduling is considered. A constrained Markov decision process model is introduced in which jobs arrive randomly at a service center with stochastic job sizes, rewards, and completion deadlines.
In IEEETAC, 2017

The problem of online detection of low-frequency oscillation in power system is considered. A distributed consensus algorithm is proposed based on extended Kalman filter.
In CSEE, 2017

The diffusion of electric vehicles (EVs) is studied in a two-sided market framework consisting of EVs on the one side and EV charging stations (EVCSs) on the other.
In TRD, 2016

The problem of modeling and stochastic optimization for home energy management is considered. Several different types of load classes are discussed, including heating, ventilation, and air conditioning unit, plug-in hybrid electric vehicle, and deferrable loads such as washer and dryer. A mixed integer multi-time scale stochastic optimization is formulated for the scheduling of loads of different characteristics.
In IEEETSG, 2012

Recent Publications

More Publications

A distribution estimation method of load model coefficients is proposed.

The problem of DER sizing for commercial building in a hybrid microgrid is considered.

The problem of recovering the lost PMU data using ADMM algorithm is considered.

The problem of using dynamic time warping in online coherence identification is considered.

The problem of synchronized reconnection of a multi-bus microgrid is considered and a distributed algorithm is proposed.

The problem of using extended Kalman filter to enhance Hilbert-Huang transform in low-frequency oscillation detection in power system …

The problem of recovering the lost PMU data using ADMM algorithm is considered.

The problem of using metric learning in forced oscillation source locating is considered.

The problem of transmission expansion planning is considered.

The problem of stochastic deadline scheduling is considered. A constrained Markov decision process model is introduced in which jobs …

Projects

Load Modeling and prediction

Load monitoring, modeling and forecasting.

IoT

Internet of Things and Edge Computing in power system stability.

WAMS

Wide Area Monitoring.

Oscillation detection and source locating in power systems

Explore the application of machine learning approaches in oscillation detection and source locating in power systems

Deadline Scheduling

Scheduling of jobs with deadlines, such as EV charging, internet streaming, and communications.

Home Energy Mangement System

Modeling and control algorithm in home energy management systems

iEMS

An Intellegent energy management system that controls the admittance and scheduling of EV charging.

Two-sided market of Electric Vehicles and EV charging facilities

Modeling the indirect network effect between EV and EV charging facilities.

Contact

  • (412)620-3071
  • yzae2623
  • 250 W Tasman Drive, STE 100, San Jose, California, 95136, USA
  • Weekday 10:00 to 18:00
    Otherwise email to book an appointment