en
0.25
0.5
0.75
1.25
1.5
1.75
2
Novel Geo-spatial Interpolation Analytics for General Meteorological Measurements
Published on Oct 07, 20141953 Views
This paper addresses geospatial interpolation for meteorological measurements in which we estimate the values of climatic metrics at unsampled sites with existing observations. Providing climatologica
Related categories
Chapter list
Novel Geo-spatial Interpolation Analytics for General Meteorological Measurements00:00
IBM Smarter Energy Research Institute: Advance the Future of Utility 00:46
Electrical (Power) Engineering101 02:36
Electric Utility is a Capital Intensive Industry 05:00
Much more Distribution Transformers than Power Transformers 05:52
Optimal Asset Maintenance and Capital investment is of Importance 06:25
But How to Obtain Ambient Temperature of Transformers? 08:22
Challenge: there are only Limited Number of Weather Stations 09:13
Paucity of Weather Station Network is a Common Issue 10:14
Physical-based Model is typically Computationally Intensive 11:03
Learning-based Model: Trade-off between Accuracy and Speed 11:43
Major Contributions of this Work 12:04
Estimate DCT Coefficients to Minimize Reconstruction Error with Student-t prior13:40
Student-t is a better Prior than Laplace for Meteorological Metrics 14:18
BCST as a General Meteorological Interpolation Service 14:39
Experimental Setup 15:04
Temperature Interpolation Results on KNMI (metric: percentage) 15:20
Runtime Comparison (in ms) 15:46
Example: Barometric Pressure Map in NYC 15:57
Conclusions & Future Work 16:05