Anomaly Detection for Multivariate Time Series

Multivariate time series analysis provides early detection of anomalies in Oracle Customer assets and products, allowing customers to avoid costly failures and outages.

Project Details

Anomaly Detection for Multivariate Time Series

Anomaly Detection for Multivariate Time Series

Multivariate time series analysis provides early detection of anomalies in Oracle Customer assets and products, allowing customers to avoid costly failures and outages.

Project Overview

Our multivariate time series analysis is based on a nonlinear, nonparametric, multivariate pattern recognition technique called the Multivariate State Estimation Technique (MSET). It was originally developed to discover anomalies in time series sensor data in nuclear power applications and has evolved to big data prognostic applications commonly seen in safety-critical industries such as aerospace, utilities, and computer systems. MSET builds a model that predicts individual sensor readings based on the correlations among sensor data. It uses a statistical technique called the Sequential Probability Ratio Test (SPRT) to determine when the differences between predicted and actual values is significant. This provides the earliest possible anomaly detection and can even differentiate between sensor failures and monitored asset anomalies, providing potentially huge savings in maintenance costs.

MSET2 is an ecosystem that has been built around MSET. It provides a suite of data preprocessing algorithms to check for uncorrelated signals that can be monitored via univariate techniques, flat or ramp signals, data spikes (e.g., voltage) that need to be handled separately, and many other data input issues. It even has an algorithm to find and remove anomalies in training data. MSET2 also has a synthetic signal generator to obviate data privacy issues and a test harness that allows comparison of multivariate time series analysis techniques.

Principal Investigator

Guang Wang

Principal Member of Technical Staff

I am a Principal Machine Learning Researcher in the Modeling, Simulation, and Optimization Group at Oracle Labs. My field of technical expertise is time series anomaly detection, in particular, Oracle’s anomaly detection technique, aka MSET2. As the main point-of-contact for MSET2 initiative and productization, I primarily work with internal groups in Oracle Labs, Database, National Security Group, Public Sector Consulting, Cloud Engineering, Cloud Infrastructure, and also collaborate with outside partnership and universities.

I have worked on many complex projects resulting in several technology transfers, the major projects include developing MSET2 ecosystem, performing three Customer PoCs and interacting with the account teams, benchmarking Oracle Roving Edge Device (RED) to support the RED launch event, and researching EMI fingerprint technology for identifying counterfeit electronic components. Currently, I am heavily involved in tech-transferring MSET2 ecosystem to OCI Anomaly Detection Service, and have been playing a key role from the inception of building the Service to the GA release.

My daily duties include developing new and improved approaches for algorithms to progress various research projects, coming up with solutions to problems emerged out of MSET2 productization, filing patents to protect Oracle’s intellectual property, and mentoring interns on research tasks and priorities.

Prior to joining Oracle Labs, I had done many interesting research projects as a graduate researcher and had extensive experiences in time series renewable energy forecasting, distributed energy resources planning, Electric Vehicle charging scheduling optimization, and photovoltaic power modeling. I received a Master’s degree in Aerospace at The Ohio State and earned my PhD in Energy Science at UC San Diego. Outside work, I am a PC geek, driving enthusiast, landscape photographer, adventurous traveler, and aviation lover.