Time Series Prediction with Emphasis on Machine Learning Algorithms in the Context of Smart Grid

Speaker: Mostafa Majidpour
Affiliation: UCLA Ph.D. Candidate

Abstract:  In view of the success of machine learning based prediction algorithms (as opposed to statistical methods) in the recent years, in this study, we have employed a selection of these algorithms on some time series prediction problems in the context of smart grid. We have used real world data from the UCLA campus solar PV panels and parking lots. In the process of applying these algorithms on the Electric Vehicle (EV) charging load prediction problem, two new prediction algorithms have been proposed, namely Modified Pattern Sequence Forecasting (MPSF) and Time Weighted Dot Product Nearest Neighbor (TWDP-NN). One of the objectives when predicting the EV charging load is speed of prediction since it is intended to be used in a real time application (smart phone application for EV customers). Using our dataset, TWDP-NN decreased the processing time by a third.

As missing data is a significant concern in real world data, the effect of missing values on the prediction quality has been investigated. Six different imputation methods have been applied to compensate for missing values in EV charging data. Based on nonparametric statistical tests, suitable (or unsuitable) imputation methods for each prediction algorithm are recommended.

Forecasting of the Electric Vehicle (EV) charging load can be done based on two different datasets: data from the customer profile (charging record) and data from outlet measurements (station record). We found that charging records provide relatively faster prediction while putting customer privacy in jeopardy. On the other hand, station records provide relatively slower prediction while respecting the customer privacy. In general, both datasets generate comparable prediction error.

Forecasting solar power generation with application on real-time control of energy system has also been investigated. Since predictions are made on every minute for one minute ahead values, the designed system has to be rapidly responsive. This has been pursued by: first, we have solely relied on past values of solar power data (rather than external data), hence lowering the volume of input data; second, the investigated algorithms are capable of generating predictions in less than a second. The results show that kNN and SVR show lower error.

Biography: Mostafa Majidpour is a Ph.D. candidate in the Electrical Engineering Department and Smart Gird Energy Research Center at UCLA, developing Statistical and Machine Learning algorithms to analyze various data in the context of smart grid. He was a founding member and CEO of a start-up company developing software packages for regional electricity utilities from 2006 to 2007. In 2011 and 2012, he was a Research Intern and Independent Consultant for Fujitsu Laboratories of America on Electric Vehicle Charging Algorithms and Building Level Load Prediction.  He has extensive background in Computer Vision and Image Processing as well, with his M.S. thesis on Face Recognition and projects on Medical Imaging Analysis.

Mostafa is the recipient of, among others, the UCLA Graduate Division Fellowship Award in 2011, NeuroImaging Training Program Fellowship Award (NIH funded) in 2012, the Henry Samueli Outstanding Teaching Award for the best Teaching Assistant in 2014 and Outstanding Achievement in Doctoral Education Scholarship from APSIH in 2016. His research interest is concept extraction and pattern discovery in data, which depending on the context, is referred to as Machine Learning, Pattern Recognition, Data Mining, or System Identification.

For more information, contact Prof. Jason L. Speyer ()

Date(s) - May 20, 2016
2:00 pm - 4:00 pm

E-IV Maxwell Room #57-124
420 Westwood Plaza - 5th Flr. , Los Angeles CA 90095