This is an old revision of the document!
Forecasting Meeting Minutes for Week of October 25, 2016
Present Tuesday: Austin, Gordon, Brieanna, Jaimie
Present Wednesday:
Updates
- Sharif will meet with us on Wednesday
Meeting with Dr. Kuh
Expectations
- All of this is normalized (w/ Zenith Angle)
- X' = [X / 1 1 1 1 1 1]
- X'.T = [X.T | 1]
- X'*X'.T = [X*X.T | X1]
- [1.T*X.T | m ]
- (X'*X'.T)/m
- Mean: X1/m
- Correlation(R) : X*X.T/m
- Samples along diagonal of correlation matrix should be similar
- Like toeplitz matrix
- Right-hand bottom corner needs to be 1
- Number of columns (m) in array rows (n)
- Number of times you are doing prediction
- Estimate 9-5 (8 hours, every 5 mins, 96 solar irradiance times/day *365 days/year= 35040 solar irradiance data points/year)
- Diagonals same and symmetrical
- 1's on bottom are for the zenith normalized data
- Do not need for the statistically normalized data
- Types of Prediction:
- Average
- Persistence: Most recent value for prediction (tap=1)
- Python programming help from Sharif tomorrow
- MSE value: D-Y=E(365*96)
- Inner product of E/entries = 2D array
- (Sum |E|^2)^.5
- Root Mean Squared Working this week
- Training to use same data for RMS
- Go down with number of taps increasing
- Take square root of the error
- (90,000)^.5 = 300
- Weight for taps
- Positive values
- Decrease in value
- Train on 2011 and test on 2012
- Train on 2012 and test on 2011
- RMS graphs for the 2011 (Test and train) and 2012 (test and train)
- Higher on test than training
- Can become higher though
- More time then train based on seasons
- Winter 2011/2012, Spring 2011/2012, Summer 2011/2012, and Fall 2011/2012
- Interested in power solar panels will be making