Forecasting Strategies for Predicting Peak Electric Load Days
Challenges in forecasting peak electricity demand
Transcript of Challenges in forecasting peak electricity demand
![Page 1: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/1.jpg)
Challenges in forecastingpeak electricity demand
Rob J Hyndman
Challenges in forecasting peak electricity demand 1
Part 2
![Page 2: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/2.jpg)
Challenges
1 How to evaluate forecast distributions
2 How to select the best forecasting method
3 How to account for off-grid generation
4 How to use smart metre data and networkdata when forecasting
5 How to improve forecasts
Challenges in forecasting peak electricity demand 2
![Page 3: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/3.jpg)
Challenges
1 How to evaluate forecast distributions
2 How to select the best forecasting method
3 How to account for off-grid generation
4 How to use smart metre data and networkdata when forecasting
5 How to improve forecasts
Challenges in forecasting peak electricity demand How to evaluate forecast distributions 3
![Page 4: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/4.jpg)
Forecast accuracy measures
MAE: Mean absolute errorMSE: Mean squared errorMAPE: Mean absolute percentage error
å Good when forecasting a typical future value(e.g., the mean or median).
Challenges in forecasting peak electricity demand How to evaluate forecast distributions 4
![Page 5: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/5.jpg)
Forecast accuracy measures
MAE: Mean absolute errorMSE: Mean squared errorMAPE: Mean absolute percentage error
å Good when forecasting a typical future value(e.g., the mean or median).
Challenges in forecasting peak electricity demand How to evaluate forecast distributions 4
![Page 6: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/6.jpg)
Forecast accuracy measures
MAE: Mean absolute errorMSE: Mean squared errorMAPE: Mean absolute percentage error
å Good when forecasting a typical future value(e.g., the mean or median).
Challenges in forecasting peak electricity demand How to evaluate forecast distributions 4
![Page 7: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/7.jpg)
Forecast accuracy measures
MAE: Mean absolute errorMSE: Mean squared errorMAPE: Mean absolute percentage error
å Good when forecasting a typical future value(e.g., the mean or median).
Challenges in forecasting peak electricity demand How to evaluate forecast distributions 4
![Page 8: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/8.jpg)
Forecast accuracy measures
MAE: Mean absolute errorMSE: Mean squared errorMAPE: Mean absolute percentage error
å Good when forecasting a typical future value(e.g., the mean or median).
å Useless for evaluating forecast percentiles(probability of exceedance values) and forecastdistributions.
Challenges in forecasting peak electricity demand How to evaluate forecast distributions 4
![Page 9: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/9.jpg)
Evaluating forecast distributions
Challenges in forecasting peak electricity demand How to evaluate forecast distributions 5
PoE (annual interpretation)
Year
PoE
Dem
and
2.0
2.5
3.0
3.5
4.0
98/99 00/01 02/03 04/05 06/07 08/09 10/11
10 %50 %90 %
●
●
●
●
●
●
●●
● ●
●
●
●
●
![Page 10: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/10.jpg)
Evaluating forecast distributions
Challenges in forecasting peak electricity demand How to evaluate forecast distributions 5
PoE (annual interpretation)
Year
PoE
Dem
and
2.0
2.5
3.0
3.5
4.0
98/99 00/01 02/03 04/05 06/07 08/09 10/11
10 %50 %90 %
●
●
●
●
●
●
●●
● ●
●
●
●
●
10 out of 13 above 90% PoE
5 out of 13 above 50% PoE
0 out of 13 above 10% PoE
![Page 11: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/11.jpg)
Evaluating forecast distributions
Qt(p) = PoE of yt, to be exceeded with per-centage p.
G(p) = percentage of times yt greater thanQt(p) in the historical data.
Challenges in forecasting peak electricity demand How to evaluate forecast distributions 6
![Page 12: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/12.jpg)
Evaluating forecast distributions
Qt(p) = PoE of yt, to be exceeded with per-centage p.
G(p) = percentage of times yt greater thanQt(p) in the historical data.
Challenges in forecasting peak electricity demand How to evaluate forecast distributions 6
![Page 13: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/13.jpg)
Evaluating forecast distributions
Qt(p) = PoE of yt, to be exceeded with per-centage p.
G(p) = percentage of times yt greater thanQt(p) in the historical data.
Challenges in forecasting peak electricity demand How to evaluate forecast distributions 6
If Qt(p) is accurate, then G(p) ≈ p
![Page 14: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/14.jpg)
Evaluating forecast distributions
Challenges in forecasting peak electricity demand How to evaluate forecast distributions 7
0 20 40 60 80 100
020
4060
8010
0
p = Forecast percentage exceedance
G(p
) =
Act
ual p
erce
ntag
e ex
ceed
ance
![Page 15: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/15.jpg)
Evaluating forecast distributions
Challenges in forecasting peak electricity demand How to evaluate forecast distributions 8
020
4060
8010
0
p = Forecast percentage exceedance
G(p
) =
Act
ual p
erce
ntag
e ex
ceed
ance
0 10010% 50% 90%
5/13
10/13
![Page 16: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/16.jpg)
Evaluating forecast distributions
Challenges in forecasting peak electricity demand How to evaluate forecast distributions 9
0 20 40 60 80 100
020
4060
8010
0
p = Forecast percentage exceedance
G(p
) =
Act
ual p
erce
ntag
e ex
ceed
ance
KS
KS = maxp |G(p)− p|
KS = Kolmogorov-Smirnov statistic= largest difference between G(p) and p.
![Page 17: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/17.jpg)
Evaluating forecast distributions
Challenges in forecasting peak electricity demand How to evaluate forecast distributions 10
0 20 40 60 80 100
020
4060
8010
0
p = Forecast percentage exceedance
G(p
) =
Act
ual p
erce
ntag
e ex
ceed
ance
Area = MAEP: Mean Absolute Excess Probability
MAEP =∫ 10 |G(p)− p|dp
![Page 18: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/18.jpg)
Evaluating forecast distributions
MAEP more sensitive and less variable than KS.
Weekly or monthly maximums are betterbecause there are more of them to evaluate.
What is a good value of KS or MAEP?
We could restrict the range of p to “interesting”values such as p > 0.5.
These only measure whether the PoEs wereexceeded, not how much they were exceeded.
Challenges in forecasting peak electricity demand How to evaluate forecast distributions 11
![Page 19: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/19.jpg)
Evaluating forecast distributions
MAEP more sensitive and less variable than KS.
Weekly or monthly maximums are betterbecause there are more of them to evaluate.
What is a good value of KS or MAEP?
We could restrict the range of p to “interesting”values such as p > 0.5.
These only measure whether the PoEs wereexceeded, not how much they were exceeded.
Challenges in forecasting peak electricity demand How to evaluate forecast distributions 11
![Page 20: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/20.jpg)
Evaluating forecast distributions
MAEP more sensitive and less variable than KS.
Weekly or monthly maximums are betterbecause there are more of them to evaluate.
What is a good value of KS or MAEP?
We could restrict the range of p to “interesting”values such as p > 0.5.
These only measure whether the PoEs wereexceeded, not how much they were exceeded.
Challenges in forecasting peak electricity demand How to evaluate forecast distributions 11
![Page 21: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/21.jpg)
Evaluating forecast distributions
MAEP more sensitive and less variable than KS.
Weekly or monthly maximums are betterbecause there are more of them to evaluate.
What is a good value of KS or MAEP?
We could restrict the range of p to “interesting”values such as p > 0.5.
These only measure whether the PoEs wereexceeded, not how much they were exceeded.
Challenges in forecasting peak electricity demand How to evaluate forecast distributions 11
![Page 22: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/22.jpg)
Evaluating forecast distributions
MAEP more sensitive and less variable than KS.
Weekly or monthly maximums are betterbecause there are more of them to evaluate.
What is a good value of KS or MAEP?
We could restrict the range of p to “interesting”values such as p > 0.5.
These only measure whether the PoEs wereexceeded, not how much they were exceeded.
Challenges in forecasting peak electricity demand How to evaluate forecast distributions 11
![Page 23: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/23.jpg)
Forecast scoring
Challenges in forecasting peak electricity demand How to evaluate forecast distributions 12
0 1 2 3 4 5 6
Demand distribution
Demand (GWh)
![Page 24: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/24.jpg)
Forecast scoring
Challenges in forecasting peak electricity demand How to evaluate forecast distributions 12
0 1 2 3 4 5 6
Demand distribution
Demand (GWh)
50%
50% PoE
![Page 25: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/25.jpg)
Forecast scoring
Challenges in forecasting peak electricity demand How to evaluate forecast distributions 12
0 1 2 3 4 5 6
Demand distribution
Demand (GWh)
50%
50% PoE
Score for 50% PoE
![Page 26: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/26.jpg)
Forecast scoring
Challenges in forecasting peak electricity demand How to evaluate forecast distributions 12
0 1 2 3 4 5 6
Demand distribution
Demand (GWh)
50%
50% PoE
Score for 50% PoEEquivalent toabsolute error
![Page 27: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/27.jpg)
Forecast scoring
Challenges in forecasting peak electricity demand How to evaluate forecast distributions 13
0 1 2 3 4 5 6
Demand distribution
Demand (GWh)
![Page 28: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/28.jpg)
Forecast scoring
Challenges in forecasting peak electricity demand How to evaluate forecast distributions 13
0 1 2 3 4 5 6
Demand distribution
Demand (GWh)
10%
10% PoE
![Page 29: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/29.jpg)
Forecast scoring
Challenges in forecasting peak electricity demand How to evaluate forecast distributions 13
0 1 2 3 4 5 6
Demand distribution
Demand (GWh)
10%
10% PoE
Score for 10% PoE
![Page 30: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/30.jpg)
Forecast scoring
Challenges in forecasting peak electricity demand How to evaluate forecast distributions 14
0 1 2 3 4 5 6
Demand distribution
Demand (GWh)
![Page 31: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/31.jpg)
Forecast scoring
Challenges in forecasting peak electricity demand How to evaluate forecast distributions 14
0 1 2 3 4 5 6
Demand distribution
Demand (GWh)
75%
75% PoE
![Page 32: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/32.jpg)
Forecast scoring
Challenges in forecasting peak electricity demand How to evaluate forecast distributions 14
0 1 2 3 4 5 6
Demand distribution
Demand (GWh)
75%
75% PoE
Score for 75% PoE
![Page 33: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/33.jpg)
Forecast scoring
Let Qt(1), . . . ,Qt(99) be the PoEs of the forecastdistribution for probabilities 1%,. . . ,99%. Then thescore for observation y is
S(Qt(i), yt) =
{1
100 i(Qt(i)− yt) if yt < Qt(i)1
100(100− i)(yt − Qt(i)) if yt ≥ Qt(i)
Scores are averaged over all observed data foreach i to measure the accuracy of the forecastsfor each percentile.Average score over all percentiles gives thebest distribution forecast.Takes account of how far PoEs are exceeded.
Challenges in forecasting peak electricity demand How to evaluate forecast distributions 15
![Page 34: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/34.jpg)
Forecast scoring
Let Qt(1), . . . ,Qt(99) be the PoEs of the forecastdistribution for probabilities 1%,. . . ,99%. Then thescore for observation y is
S(Qt(i), yt) =
{1
100 i(Qt(i)− yt) if yt < Qt(i)1
100(100− i)(yt − Qt(i)) if yt ≥ Qt(i)
Scores are averaged over all observed data foreach i to measure the accuracy of the forecastsfor each percentile.Average score over all percentiles gives thebest distribution forecast.Takes account of how far PoEs are exceeded.
Challenges in forecasting peak electricity demand How to evaluate forecast distributions 15
![Page 35: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/35.jpg)
Forecast scoring
Let Qt(1), . . . ,Qt(99) be the PoEs of the forecastdistribution for probabilities 1%,. . . ,99%. Then thescore for observation y is
S(Qt(i), yt) =
{1
100 i(Qt(i)− yt) if yt < Qt(i)1
100(100− i)(yt − Qt(i)) if yt ≥ Qt(i)
Scores are averaged over all observed data foreach i to measure the accuracy of the forecastsfor each percentile.Average score over all percentiles gives thebest distribution forecast.Takes account of how far PoEs are exceeded.
Challenges in forecasting peak electricity demand How to evaluate forecast distributions 15
![Page 36: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/36.jpg)
Challenges
1 How to evaluate forecast distributions
2 How to select the best forecasting method
3 How to account for off-grid generation
4 How to use smart metre data and networkdata when forecasting
5 How to improve forecasts
Challenges in forecasting peak electricity demand How to select the best forecasting method 16
![Page 37: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/37.jpg)
Forecasting the past
Traditional evaluation
Challenges in forecasting peak electricity demand How to select the best forecasting method 17
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● timeTraining data Test data
![Page 38: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/38.jpg)
Forecasting the past
Traditional evaluation
Time series cross-validation
Challenges in forecasting peak electricity demand How to select the best forecasting method 17
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● timeTraining data Test data
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
![Page 39: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/39.jpg)
Forecasting the past
Traditional evaluation
Time series cross-validation
Challenges in forecasting peak electricity demand How to select the best forecasting method 17
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● timeTraining data Test data
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
![Page 40: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/40.jpg)
Forecasting the past
Traditional evaluation
Time series cross-validation
Challenges in forecasting peak electricity demand How to select the best forecasting method 17
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● timeTraining data Test data
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
Also known as “Evaluation ona rolling forecast origin”
![Page 41: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/41.jpg)
Forecasting the past
Traditional evaluation
Time series cross-validation
GEFCOM 2012 competition
Challenges in forecasting peak electricity demand How to select the best forecasting method 17
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● timeTraining data Test data
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●
Also known as “Evaluation ona rolling forecast origin”
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● time
![Page 42: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/42.jpg)
GEFCom2012
Challenges in forecasting peak electricity demand How to select the best forecasting method 18
![Page 43: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/43.jpg)
GEFCom2012
Challenges in forecasting peak electricity demand How to select the best forecasting method 18
Accuracy measuredby weighted root MSE.
![Page 44: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/44.jpg)
GEFCom2012
Challenges in forecasting peak electricity demand How to select the best forecasting method 18
Accuracy measuredby weighted root MSE.
Methodspublishedin IJF,April 2014
![Page 45: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/45.jpg)
GEFCom2014
Probabilistic forecasting of demand, price,wind, and solar.
Forecasts to be submitted in the form ofpercentiles of future distributions.
Accuracy measured by scoring.
Rolling forecasts with incremental data updateon a weekly basis.
Prizes for student teams, and for best methods.
Winning methods to be published in the IJF.
Challenges in forecasting peak electricity demand How to select the best forecasting method 19
![Page 46: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/46.jpg)
GEFCom2014
Probabilistic forecasting of demand, price,wind, and solar.
Forecasts to be submitted in the form ofpercentiles of future distributions.
Accuracy measured by scoring.
Rolling forecasts with incremental data updateon a weekly basis.
Prizes for student teams, and for best methods.
Winning methods to be published in the IJF.
Challenges in forecasting peak electricity demand How to select the best forecasting method 19
![Page 47: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/47.jpg)
GEFCom2014
Probabilistic forecasting of demand, price,wind, and solar.
Forecasts to be submitted in the form ofpercentiles of future distributions.
Accuracy measured by scoring.
Rolling forecasts with incremental data updateon a weekly basis.
Prizes for student teams, and for best methods.
Winning methods to be published in the IJF.
Challenges in forecasting peak electricity demand How to select the best forecasting method 19
![Page 48: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/48.jpg)
GEFCom2014
Probabilistic forecasting of demand, price,wind, and solar.
Forecasts to be submitted in the form ofpercentiles of future distributions.
Accuracy measured by scoring.
Rolling forecasts with incremental data updateon a weekly basis.
Prizes for student teams, and for best methods.
Winning methods to be published in the IJF.
Challenges in forecasting peak electricity demand How to select the best forecasting method 19
![Page 49: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/49.jpg)
GEFCom2014
Probabilistic forecasting of demand, price,wind, and solar.
Forecasts to be submitted in the form ofpercentiles of future distributions.
Accuracy measured by scoring.
Rolling forecasts with incremental data updateon a weekly basis.
Prizes for student teams, and for best methods.
Winning methods to be published in the IJF.
Challenges in forecasting peak electricity demand How to select the best forecasting method 19
![Page 50: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/50.jpg)
GEFCom2014
Probabilistic forecasting of demand, price,wind, and solar.
Forecasts to be submitted in the form ofpercentiles of future distributions.
Accuracy measured by scoring.
Rolling forecasts with incremental data updateon a weekly basis.
Prizes for student teams, and for best methods.
Winning methods to be published in the IJF.
Challenges in forecasting peak electricity demand How to select the best forecasting method 19
![Page 51: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/51.jpg)
Challenges
1 How to evaluate forecast distributions
2 How to select the best forecasting method
3 How to account for off-grid generation
4 How to use smart metre data and networkdata when forecasting
5 How to improve forecasts
Challenges in forecasting peak electricity demand How to account for off-grid generation 20
![Page 52: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/52.jpg)
How to account for off-grid generation?
Locally generated power may not be recorded.But forecasts better if they are total demand.Need a model for PV generation that is linked tothe model for demand.
Challenges in forecasting peak electricity demand How to account for off-grid generation 21
![Page 53: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/53.jpg)
How to account for off-grid generation?
Locally generated power may not be recorded.But forecasts better if they are total demand.Need a model for PV generation that is linked tothe model for demand.
Challenges in forecasting peak electricity demand How to account for off-grid generation 21
![Page 54: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/54.jpg)
How to account for off-grid generation?
Locally generated power may not be recorded.But forecasts better if they are total demand.Need a model for PV generation that is linked tothe model for demand.
Challenges in forecasting peak electricity demand How to account for off-grid generation 21
![Page 55: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/55.jpg)
How to account for off-grid generation?
Locally generated power may not be recorded.But forecasts better if they are total demand.Need a model for PV generation that is linked tothe model for demand.
Challenges in forecasting peak electricity demand How to account for off-grid generation 21
Better: measure the off-grid generation viasmart metres.
![Page 56: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/56.jpg)
Challenges
1 How to evaluate forecast distributions
2 How to select the best forecasting method
3 How to account for off-grid generation
4 How to use smart metre data and networkdata when forecasting
5 How to improve forecasts
Challenges in forecasting peak electricity demand How to use smart metre data 22
![Page 57: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/57.jpg)
How to use smart metre data?
Challenges in forecasting peak electricity demand How to use smart metre data 23
Smart metre data allow predictionof usage at household level basedon household characteristics:number of occupants, ages, etc.
So we could build a model forindividual usage, and scale it upfor the entire network.
How to allow for demandresponse?
Need aggregate information onhousehold characteristics for thenetwork.
Combine network and smartforecasts to improve accuracy
![Page 58: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/58.jpg)
How to use smart metre data?
Challenges in forecasting peak electricity demand How to use smart metre data 23
Smart metre data allow predictionof usage at household level basedon household characteristics:number of occupants, ages, etc.
So we could build a model forindividual usage, and scale it upfor the entire network.
How to allow for demandresponse?
Need aggregate information onhousehold characteristics for thenetwork.
Combine network and smartforecasts to improve accuracy
![Page 59: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/59.jpg)
How to use smart metre data?
Challenges in forecasting peak electricity demand How to use smart metre data 23
Smart metre data allow predictionof usage at household level basedon household characteristics:number of occupants, ages, etc.
So we could build a model forindividual usage, and scale it upfor the entire network.
How to allow for demandresponse?
Need aggregate information onhousehold characteristics for thenetwork.
Combine network and smartforecasts to improve accuracy
![Page 60: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/60.jpg)
How to use smart metre data?
Challenges in forecasting peak electricity demand How to use smart metre data 23
Smart metre data allow predictionof usage at household level basedon household characteristics:number of occupants, ages, etc.
So we could build a model forindividual usage, and scale it upfor the entire network.
How to allow for demandresponse?
Need aggregate information onhousehold characteristics for thenetwork.
Combine network and smartforecasts to improve accuracy
![Page 61: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/61.jpg)
How to use smart metre data?
Challenges in forecasting peak electricity demand How to use smart metre data 23
Smart metre data allow predictionof usage at household level basedon household characteristics:number of occupants, ages, etc.
So we could build a model forindividual usage, and scale it upfor the entire network.
How to allow for demandresponse?
Need aggregate information onhousehold characteristics for thenetwork.
Combine network and smartforecasts to improve accuracy
![Page 62: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/62.jpg)
Challenges
1 How to evaluate forecast distributions
2 How to select the best forecasting method
3 How to account for off-grid generation
4 How to use smart metre data and networkdata when forecasting
5 How to improve forecasts
Challenges in forecasting peak electricity demand How to improve forecasts 24
![Page 63: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/63.jpg)
Ten steps to improving your forecasts
1 Look after your data
2 Understand how your forecasts will be used
3 Find the right forecasting tools
4 Use appropriate accuracy measures
5 Do not set targets
6 Do not adjust dishonestly
7 Keep it simple, stupid
8 Combine forecasts
9 Share ideas and mix with other forecasters
10 Adopt a process of continuous improvement
Challenges in forecasting peak electricity demand How to improve forecasts 25
![Page 64: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/64.jpg)
Ten steps to improving your forecasts
1 Look after your data
2 Understand how your forecasts will be used
3 Find the right forecasting tools
4 Use appropriate accuracy measures
5 Do not set targets
6 Do not adjust dishonestly
7 Keep it simple, stupid
8 Combine forecasts
9 Share ideas and mix with other forecasters
10 Adopt a process of continuous improvement
Challenges in forecasting peak electricity demand How to improve forecasts 25
![Page 65: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/65.jpg)
Ten steps to improving your forecasts
1 Look after your data
2 Understand how your forecasts will be used
3 Find the right forecasting tools
4 Use appropriate accuracy measures
5 Do not set targets
6 Do not adjust dishonestly
7 Keep it simple, stupid
8 Combine forecasts
9 Share ideas and mix with other forecasters
10 Adopt a process of continuous improvement
Challenges in forecasting peak electricity demand How to improve forecasts 25
![Page 66: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/66.jpg)
Ten steps to improving your forecasts
1 Look after your data
2 Understand how your forecasts will be used
3 Find the right forecasting tools
4 Use appropriate accuracy measures
5 Do not set targets
6 Do not adjust dishonestly
7 Keep it simple, stupid
8 Combine forecasts
9 Share ideas and mix with other forecasters
10 Adopt a process of continuous improvement
Challenges in forecasting peak electricity demand How to improve forecasts 25
![Page 67: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/67.jpg)
Ten steps to improving your forecasts
1 Look after your data
2 Understand how your forecasts will be used
3 Find the right forecasting tools
4 Use appropriate accuracy measures
5 Do not set targets
6 Do not adjust dishonestly
7 Keep it simple, stupid
8 Combine forecasts
9 Share ideas and mix with other forecasters
10 Adopt a process of continuous improvement
Challenges in forecasting peak electricity demand How to improve forecasts 25
![Page 68: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/68.jpg)
Ten steps to improving your forecasts
1 Look after your data
2 Understand how your forecasts will be used
3 Find the right forecasting tools
4 Use appropriate accuracy measures
5 Do not set targets
6 Do not adjust dishonestly
7 Keep it simple, stupid
8 Combine forecasts
9 Share ideas and mix with other forecasters
10 Adopt a process of continuous improvement
Challenges in forecasting peak electricity demand How to improve forecasts 25
![Page 69: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/69.jpg)
Ten steps to improving your forecasts
1 Look after your data
2 Understand how your forecasts will be used
3 Find the right forecasting tools
4 Use appropriate accuracy measures
5 Do not set targets
6 Do not adjust dishonestly
7 Keep it simple, stupid
8 Combine forecasts
9 Share ideas and mix with other forecasters
10 Adopt a process of continuous improvement
Challenges in forecasting peak electricity demand How to improve forecasts 25
![Page 70: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/70.jpg)
Ten steps to improving your forecasts
1 Look after your data
2 Understand how your forecasts will be used
3 Find the right forecasting tools
4 Use appropriate accuracy measures
5 Do not set targets
6 Do not adjust dishonestly
7 Keep it simple, stupid
8 Combine forecasts
9 Share ideas and mix with other forecasters
10 Adopt a process of continuous improvement
Challenges in forecasting peak electricity demand How to improve forecasts 25
![Page 71: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/71.jpg)
Ten steps to improving your forecasts
1 Look after your data
2 Understand how your forecasts will be used
3 Find the right forecasting tools
4 Use appropriate accuracy measures
5 Do not set targets
6 Do not adjust dishonestly
7 Keep it simple, stupid
8 Combine forecasts
9 Share ideas and mix with other forecasters
10 Adopt a process of continuous improvement
Challenges in forecasting peak electricity demand How to improve forecasts 25
![Page 72: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/72.jpg)
Ten steps to improving your forecasts
1 Look after your data
2 Understand how your forecasts will be used
3 Find the right forecasting tools
4 Use appropriate accuracy measures
5 Do not set targets
6 Do not adjust dishonestly
7 Keep it simple, stupid
8 Combine forecasts
9 Share ideas and mix with other forecasters
10 Adopt a process of continuous improvement
Challenges in forecasting peak electricity demand How to improve forecasts 25
![Page 73: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/73.jpg)
Some resourcesBlogs
robjhyndman.com/hyndsight/blog.drhongtao.com/
OrganizationsInternational Institute of Forecasters:forecasters.orgIEEE Working Group on Energy Forecasting:linkedin.com/groups/IEEE-Working-Group-on-Energy-4148276
BooksDickey and Hong (2014) Electric loadforecasting: fundamentals and best practices,OTexts. www.otexts.org/book/elf
Challenges in forecasting peak electricity demand How to improve forecasts 26
![Page 74: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/74.jpg)
Some resourcesBlogs
robjhyndman.com/hyndsight/blog.drhongtao.com/
OrganizationsInternational Institute of Forecasters:forecasters.orgIEEE Working Group on Energy Forecasting:linkedin.com/groups/IEEE-Working-Group-on-Energy-4148276
BooksDickey and Hong (2014) Electric loadforecasting: fundamentals and best practices,OTexts. www.otexts.org/book/elf
Challenges in forecasting peak electricity demand How to improve forecasts 26
![Page 75: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/75.jpg)
Some resourcesBlogs
robjhyndman.com/hyndsight/blog.drhongtao.com/
OrganizationsInternational Institute of Forecasters:forecasters.orgIEEE Working Group on Energy Forecasting:linkedin.com/groups/IEEE-Working-Group-on-Energy-4148276
BooksDickey and Hong (2014) Electric loadforecasting: fundamentals and best practices,OTexts. www.otexts.org/book/elf
Challenges in forecasting peak electricity demand How to improve forecasts 26
![Page 76: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/76.jpg)
Some resourcesBlogs
robjhyndman.com/hyndsight/blog.drhongtao.com/
OrganizationsInternational Institute of Forecasters:forecasters.orgIEEE Working Group on Energy Forecasting:linkedin.com/groups/IEEE-Working-Group-on-Energy-4148276
BooksDickey and Hong (2014) Electric loadforecasting: fundamentals and best practices,OTexts. www.otexts.org/book/elf
Challenges in forecasting peak electricity demand How to improve forecasts 26
![Page 77: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/77.jpg)
Some resourcesBlogs
robjhyndman.com/hyndsight/blog.drhongtao.com/
OrganizationsInternational Institute of Forecasters:forecasters.orgIEEE Working Group on Energy Forecasting:linkedin.com/groups/IEEE-Working-Group-on-Energy-4148276
BooksDickey and Hong (2014) Electric loadforecasting: fundamentals and best practices,OTexts. www.otexts.org/book/elf
Challenges in forecasting peak electricity demand How to improve forecasts 26
![Page 78: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/78.jpg)
Some resourcesBlogs
robjhyndman.com/hyndsight/blog.drhongtao.com/
OrganizationsInternational Institute of Forecasters:forecasters.orgIEEE Working Group on Energy Forecasting:linkedin.com/groups/IEEE-Working-Group-on-Energy-4148276
BooksDickey and Hong (2014) Electric loadforecasting: fundamentals and best practices,OTexts. www.otexts.org/book/elf
Challenges in forecasting peak electricity demand How to improve forecasts 26
![Page 79: Challenges in forecasting peak electricity demand](https://reader030.fdocuments.net/reader030/viewer/2022012719/61b1920aa3ae50541f421639/html5/thumbnails/79.jpg)
Some resourcesBlogs
robjhyndman.com/hyndsight/blog.drhongtao.com/
OrganizationsInternational Institute of Forecasters:forecasters.orgIEEE Working Group on Energy Forecasting:linkedin.com/groups/IEEE-Working-Group-on-Energy-4148276
BooksDickey and Hong (2014) Electric loadforecasting: fundamentals and best practices,OTexts. www.otexts.org/book/elf
Challenges in forecasting peak electricity demand How to improve forecasts 26