data analytics and machine learning approaches for utility ... - 中 … 2018/final_yannansun... ·...
TRANSCRIPT
![Page 1: Data Analytics and Machine Learning Approaches for Utility ... - 中 … 2018/Final_YannanSun... · 2018-10-19 · Energy-Use Forecasting- Methods • Several “Off-the-shelf”](https://reader034.vdocuments.site/reader034/viewer/2022050602/5fa99a1707838b43af0af262/html5/thumbnails/1.jpg)
Data Analytics and Machine Learning Approaches for Utility Companies
Modern Engineering and Technology Seminar 2018
Yannan Sun Data Scientist [email protected]
![Page 2: Data Analytics and Machine Learning Approaches for Utility ... - 中 … 2018/Final_YannanSun... · 2018-10-19 · Energy-Use Forecasting- Methods • Several “Off-the-shelf”](https://reader034.vdocuments.site/reader034/viewer/2022050602/5fa99a1707838b43af0af262/html5/thumbnails/2.jpg)
Outline
• Introduction
• Statistical forecasting (Building energy use)
• Physics-based models (Parameter estimation for distribution systems)
• Statistical classification/anomaly detection (Building event detection)
• Oncor use cases using machine learning
![Page 3: Data Analytics and Machine Learning Approaches for Utility ... - 中 … 2018/Final_YannanSun... · 2018-10-19 · Energy-Use Forecasting- Methods • Several “Off-the-shelf”](https://reader034.vdocuments.site/reader034/viewer/2022050602/5fa99a1707838b43af0af262/html5/thumbnails/3.jpg)
About Me
• BS in Math at University of Sci. and Tech. of China
• MS in Statistics, PhD in Math at WSU
• Scientist/Senior Scientist at Pacific NW Natl. Lab
• Data Scientist at Oncor Electric Delivery
![Page 4: Data Analytics and Machine Learning Approaches for Utility ... - 中 … 2018/Final_YannanSun... · 2018-10-19 · Energy-Use Forecasting- Methods • Several “Off-the-shelf”](https://reader034.vdocuments.site/reader034/viewer/2022050602/5fa99a1707838b43af0af262/html5/thumbnails/4.jpg)
Oncor – By the Numbers
3.3M AMS Meters
121k T&D Circuit Line Miles
500k Switch Points
250k SCADA Points
1.5 B SCADA X-actions / month
60 TB New Data Records/month
250 TB Current EDW
Data storage
• Open energy delivery platform
• Enabling competitive retail,
generation, and open market in-home services
• Agnostic to generation
technologies or locations
• Focus on reliability and information services
• Need advanced data analytical methods
![Page 5: Data Analytics and Machine Learning Approaches for Utility ... - 中 … 2018/Final_YannanSun... · 2018-10-19 · Energy-Use Forecasting- Methods • Several “Off-the-shelf”](https://reader034.vdocuments.site/reader034/viewer/2022050602/5fa99a1707838b43af0af262/html5/thumbnails/5.jpg)
Oncor’s Analytics Platform
![Page 6: Data Analytics and Machine Learning Approaches for Utility ... - 中 … 2018/Final_YannanSun... · 2018-10-19 · Energy-Use Forecasting- Methods • Several “Off-the-shelf”](https://reader034.vdocuments.site/reader034/viewer/2022050602/5fa99a1707838b43af0af262/html5/thumbnails/6.jpg)
Data Size and Other Problems
![Page 7: Data Analytics and Machine Learning Approaches for Utility ... - 中 … 2018/Final_YannanSun... · 2018-10-19 · Energy-Use Forecasting- Methods • Several “Off-the-shelf”](https://reader034.vdocuments.site/reader034/viewer/2022050602/5fa99a1707838b43af0af262/html5/thumbnails/7.jpg)
Building Energy-Use Forecasting
using Regression Trees
![Page 8: Data Analytics and Machine Learning Approaches for Utility ... - 中 … 2018/Final_YannanSun... · 2018-10-19 · Energy-Use Forecasting- Methods • Several “Off-the-shelf”](https://reader034.vdocuments.site/reader034/viewer/2022050602/5fa99a1707838b43af0af262/html5/thumbnails/8.jpg)
Building Energy-Use Forecasting
• Volume – For one building the data is small, but if we want to mine the entire history of many buildings, the data size grows rapidly.
• Variety – numerical, categorical, ordinal, all kinds might be available from sensors.
• Velocity – hourly data with day ahead predictions, potentially not an issue.
• Veracity – weather reports, sensor malfunctions, lack of sensors
![Page 9: Data Analytics and Machine Learning Approaches for Utility ... - 中 … 2018/Final_YannanSun... · 2018-10-19 · Energy-Use Forecasting- Methods • Several “Off-the-shelf”](https://reader034.vdocuments.site/reader034/viewer/2022050602/5fa99a1707838b43af0af262/html5/thumbnails/9.jpg)
Building Energy-Use Forecasting
Inputs
Output
Holiday
![Page 10: Data Analytics and Machine Learning Approaches for Utility ... - 中 … 2018/Final_YannanSun... · 2018-10-19 · Energy-Use Forecasting- Methods • Several “Off-the-shelf”](https://reader034.vdocuments.site/reader034/viewer/2022050602/5fa99a1707838b43af0af262/html5/thumbnails/10.jpg)
Energy-Use Forecasting- Methods
• Several “Off-the-shelf” methods work well for prediction. – Regression tree, Gaussian Processes, Support Vector Machine (SVM),
(Deep) Neural Network
• Regression trees are fast for training and testing, so it is easy to use with a “rolling” window.
root
a>5 b>2
b<2
a<5
y = c11*a+c12*b
y = c21*a+c22*b
y = c31*a+c32*b
![Page 11: Data Analytics and Machine Learning Approaches for Utility ... - 中 … 2018/Final_YannanSun... · 2018-10-19 · Energy-Use Forecasting- Methods • Several “Off-the-shelf”](https://reader034.vdocuments.site/reader034/viewer/2022050602/5fa99a1707838b43af0af262/html5/thumbnails/11.jpg)
1-Day Ahead Forecasts – Regression Tree
• One month of initial training data • RMSE is about 12 • The model works well for all prediction windows, but the real dependency
is the weather forecast.
![Page 12: Data Analytics and Machine Learning Approaches for Utility ... - 中 … 2018/Final_YannanSun... · 2018-10-19 · Energy-Use Forecasting- Methods • Several “Off-the-shelf”](https://reader034.vdocuments.site/reader034/viewer/2022050602/5fa99a1707838b43af0af262/html5/thumbnails/12.jpg)
Normalized Error of Forecasts – 1 Day
Holidays
Training set too small
![Page 13: Data Analytics and Machine Learning Approaches for Utility ... - 中 … 2018/Final_YannanSun... · 2018-10-19 · Energy-Use Forecasting- Methods • Several “Off-the-shelf”](https://reader034.vdocuments.site/reader034/viewer/2022050602/5fa99a1707838b43af0af262/html5/thumbnails/13.jpg)
Parameter Estimation for Distribution
Systems
![Page 14: Data Analytics and Machine Learning Approaches for Utility ... - 中 … 2018/Final_YannanSun... · 2018-10-19 · Energy-Use Forecasting- Methods • Several “Off-the-shelf”](https://reader034.vdocuments.site/reader034/viewer/2022050602/5fa99a1707838b43af0af262/html5/thumbnails/14.jpg)
Parameter Estimation for Distribution Systems
• Volume – power injection, voltage, power flow, AMI measurements
• Variety – numerical data (multi-phase, multi-measurement)
• Velocity – SCADA data, 5-minute data generated by GridLAB-D (a power distribution system simulation tool)
• Veracity – bad data, missing data, parameter error, measurement noise
![Page 15: Data Analytics and Machine Learning Approaches for Utility ... - 中 … 2018/Final_YannanSun... · 2018-10-19 · Energy-Use Forecasting- Methods • Several “Off-the-shelf”](https://reader034.vdocuments.site/reader034/viewer/2022050602/5fa99a1707838b43af0af262/html5/thumbnails/15.jpg)
Parameter Estimation using the Kalman Filter
• Keeps track of the estimated state of the system and the variance or uncertainty of the estimate
• Many applications: dynamic positioning, satellite navigation systems, weather forecasting, power system state estimation, etc.
• State-vector augmentation with a Kalman filter
𝑥𝑘 = 𝑓(𝑥𝑘−1, 𝑢𝑘−1, 𝑤𝑘−1)
𝑧𝑘 = ℎ 𝑥𝑘 , 𝑣𝑘
where 𝑥 ≔ 𝑥𝑠 , 𝑥𝑝 with covariance 𝑃 ≔ 𝑃𝑠 00 𝑃𝑝
.
![Page 16: Data Analytics and Machine Learning Approaches for Utility ... - 中 … 2018/Final_YannanSun... · 2018-10-19 · Energy-Use Forecasting- Methods • Several “Off-the-shelf”](https://reader034.vdocuments.site/reader034/viewer/2022050602/5fa99a1707838b43af0af262/html5/thumbnails/16.jpg)
Parameter Estimation using the KF
Estimated parameter error with no measurement noise and single snapshot using residual analysis (rN) and the KF approach
Estimated parameter error with 1% measurement noise and single snapshot using residual analysis (rN) and the KF approach
![Page 17: Data Analytics and Machine Learning Approaches for Utility ... - 中 … 2018/Final_YannanSun... · 2018-10-19 · Energy-Use Forecasting- Methods • Several “Off-the-shelf”](https://reader034.vdocuments.site/reader034/viewer/2022050602/5fa99a1707838b43af0af262/html5/thumbnails/17.jpg)
Parameter Estimation using the KF
Estimated parameter error with 1% measurement noise and single time steps using residual analysis (rN) and the KF approach
Estimated parameter error with 1% measurement noise and 12 combined time steps using residual analysis (rN) and the KF approach
![Page 18: Data Analytics and Machine Learning Approaches for Utility ... - 中 … 2018/Final_YannanSun... · 2018-10-19 · Energy-Use Forecasting- Methods • Several “Off-the-shelf”](https://reader034.vdocuments.site/reader034/viewer/2022050602/5fa99a1707838b43af0af262/html5/thumbnails/18.jpg)
Parameter Estimation using the KF
𝜎(𝑧𝑘) 𝑡𝑐 = 1 𝑡𝑐 = 12 𝑡𝑐 = 24
RA KF RA KF RA KF
0.00% 0.0190 0.0178 0.0097 0.0094 0.0188 0.0188
0.50% 0.0174 0.0150 0.0091 0.0061 0.0180 0.0154
1.00% 0.0119 0.0180 0.0167 0.0190 0.0163 0.0192
1.50% 0.0026 0.0180 0.0048 0.0097 0.0138 0.0192
Mean of PE errors using RA and KF
𝜎(𝑧𝑘) 𝑡𝑐 = 1 𝑡𝑐 = 12 𝑡𝑐 = 24
RA KF RA KF RA KF
0.00% 0.0022 0.0043 0.0168 0.0174 0.0020 0.0041
0.50% 0.0626 0.0054 0.0259 0.0174 0.0132 0.0060
1.00% 0.1262 0.0042 0.0373 0.0039 0.0261 0.0035
1.50% 0.1922 0.0042 0.0595 0.0172 0.0393 0.0034
Standard deviation of PE errors using RA and KF
![Page 19: Data Analytics and Machine Learning Approaches for Utility ... - 中 … 2018/Final_YannanSun... · 2018-10-19 · Energy-Use Forecasting- Methods • Several “Off-the-shelf”](https://reader034.vdocuments.site/reader034/viewer/2022050602/5fa99a1707838b43af0af262/html5/thumbnails/19.jpg)
Building Event Detection using
Bayesian One-Class SVM
![Page 20: Data Analytics and Machine Learning Approaches for Utility ... - 中 … 2018/Final_YannanSun... · 2018-10-19 · Energy-Use Forecasting- Methods • Several “Off-the-shelf”](https://reader034.vdocuments.site/reader034/viewer/2022050602/5fa99a1707838b43af0af262/html5/thumbnails/20.jpg)
Building Event Detection
• Volume – for one building the data is small, but if we want to mine the entire history of many buildings, the data size grows rapidly.
• Variety – numerical, categorical, ordinal, all kinds might be available from sensors.
• Velocity – hourly data generated by EnergyPlus (a whole building energy simulation tool)
• Veracity – weather reports, sensor malfunctions, lack of sensors
![Page 21: Data Analytics and Machine Learning Approaches for Utility ... - 中 … 2018/Final_YannanSun... · 2018-10-19 · Energy-Use Forecasting- Methods • Several “Off-the-shelf”](https://reader034.vdocuments.site/reader034/viewer/2022050602/5fa99a1707838b43af0af262/html5/thumbnails/21.jpg)
Event/Anomaly Detection
• To predict binary outcomes (true/false) we could use supervised learning, this is called a binary classifier.
• The difficulty with this approach for faults is that it is very time-consuming to annotate the training data.
• Incorrect labels are also a problem. What is a fault? Are there unseen/unknown faults?
• To get around the labeling problem we adjust the problem as anomaly detection. This is an unsupervised classification problem.
![Page 22: Data Analytics and Machine Learning Approaches for Utility ... - 中 … 2018/Final_YannanSun... · 2018-10-19 · Energy-Use Forecasting- Methods • Several “Off-the-shelf”](https://reader034.vdocuments.site/reader034/viewer/2022050602/5fa99a1707838b43af0af262/html5/thumbnails/22.jpg)
Bayesian One-Class SVM
• The one-class SVM is used for novelty/outlier detection
• Construct the population density of a dataset and decide whether new points are – Normal: in high population – Abnormal: in low population
• The model is trained using past normal days. Predictions are made for the next week. – Unsupervised – Dynamic (time evolving)
Note that the frontier can be any level set, e.g. the light blue hourglass. Image from http://scikit-learn.org
![Page 23: Data Analytics and Machine Learning Approaches for Utility ... - 中 … 2018/Final_YannanSun... · 2018-10-19 · Energy-Use Forecasting- Methods • Several “Off-the-shelf”](https://reader034.vdocuments.site/reader034/viewer/2022050602/5fa99a1707838b43af0af262/html5/thumbnails/23.jpg)
Building Event Detection – Similarity Metrics
Radial basis function (RBF) Similarity
Euclidean Similarity
RBF wrong parameters
RBF wrong parameters
![Page 24: Data Analytics and Machine Learning Approaches for Utility ... - 中 … 2018/Final_YannanSun... · 2018-10-19 · Energy-Use Forecasting- Methods • Several “Off-the-shelf”](https://reader034.vdocuments.site/reader034/viewer/2022050602/5fa99a1707838b43af0af262/html5/thumbnails/24.jpg)
Building Event Detection– Data
![Page 25: Data Analytics and Machine Learning Approaches for Utility ... - 中 … 2018/Final_YannanSun... · 2018-10-19 · Energy-Use Forecasting- Methods • Several “Off-the-shelf”](https://reader034.vdocuments.site/reader034/viewer/2022050602/5fa99a1707838b43af0af262/html5/thumbnails/25.jpg)
Classification Results
![Page 26: Data Analytics and Machine Learning Approaches for Utility ... - 中 … 2018/Final_YannanSun... · 2018-10-19 · Energy-Use Forecasting- Methods • Several “Off-the-shelf”](https://reader034.vdocuments.site/reader034/viewer/2022050602/5fa99a1707838b43af0af262/html5/thumbnails/26.jpg)
Oncor Use Cases
![Page 27: Data Analytics and Machine Learning Approaches for Utility ... - 中 … 2018/Final_YannanSun... · 2018-10-19 · Energy-Use Forecasting- Methods • Several “Off-the-shelf”](https://reader034.vdocuments.site/reader034/viewer/2022050602/5fa99a1707838b43af0af262/html5/thumbnails/27.jpg)
Meter to Transformer Connectivity Error Detection
![Page 28: Data Analytics and Machine Learning Approaches for Utility ... - 中 … 2018/Final_YannanSun... · 2018-10-19 · Energy-Use Forecasting- Methods • Several “Off-the-shelf”](https://reader034.vdocuments.site/reader034/viewer/2022050602/5fa99a1707838b43af0af262/html5/thumbnails/28.jpg)
Transformer Load Forecast & Over Load Prediction
![Page 29: Data Analytics and Machine Learning Approaches for Utility ... - 中 … 2018/Final_YannanSun... · 2018-10-19 · Energy-Use Forecasting- Methods • Several “Off-the-shelf”](https://reader034.vdocuments.site/reader034/viewer/2022050602/5fa99a1707838b43af0af262/html5/thumbnails/29.jpg)
Data Analytics and Machine Learning Approaches for Utility Companies
Modern Engineering and Technology Seminar 2018
Yannan Sun Data Scientist [email protected]