Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

The retraining process resulted in two new models, one for PD and one for 7day price forecast. After experimentation we retrained the price forecasting models using existing features and hyper parameter settings. The increase in accuracy (using Absolute Mean Error, AME) was approximately 5%. (see validation results table below comparing the previous models with the new models).

Training period

Both PD and 7day price forecast models were trained on data between 01-

...

01-

...

2021 and 15-01-

...

2024.

The market suspension period and a period after the Callide explosion were removed from the training data.

Validation set

The model was validated by predicting the price between 16-01-2024 and 30-01-2024

New features (input data) exceptions

...

Predispatch model

p7day model

Model_id

20231108_142510_CB_2

20231108_140809_CB_7D_2

Previous model release date

8th August 9th November 20238th August

9th November 2023

Latest release date (after retraining)

9th November 2023

9th November 20235th February 2024

5th February 2024

Machine Learning Algorithm

Ensemble predictor: gradient boosting

Ensemble predictor: gradient boosting

Training Period

1st October 2020 to 31st October 20231st October 2020 to 31st October 2023January 2021 to 15th January 2024

1st January 2021 to 15th January 2024

Hyper-parameter type

A

A

Hyper-parameter notes

No weighting across time periods

No weighting across time periods

Target

Dispatchprice (30min resolution)

Dispatchprice (30min resolution)

Input (and training) variables

PREDISPATCHPRICE

rrp

DISPATCHPRICE

last 3 day average actual RRP for same time

PREDISPATCHREGIONSUM (intervention = 0)

hours_out

time_of_day

day_type (weekday_or_weekend)

dispatchableload

netinterchage

(availablegeneration - uigf)

(dispatchablegeneration - scheduled_clearedmw)

ss_solar_clearedmw

ss_solar_uigf

ss_wind_clearedmw

ss_wind_uigf

totaldemand

totalintermittentgeneration

STPASA_REGIONSOLUTION

hours_out

time_of_day

day_type (weekday_or_weekend)

aggregatepasaavailability

aggregatescheduledload

netinterchangeunderscarcity

aggregatecapacityavailable

ss_solar_cleared

ss_wind_uigf

ss_wind_cleared

ss_solar_uigf

demand10

demand50

totalintermittentgeneration

demand_and_nonschedgen

...

These are the results of the best model generated through our experimentation process and compares the results with the previous model across our validation set (our test period).

Note that errors are considerably higher because the validation set was unusually volatile.

Average error is equal to Predication minus real prices.

All error metrics are in $/MWhr.

PD Results

Metric/Region

Actual Mean Price

Mean Prediction New Model

Mean Prediction Previous Model

Absolute Mean Error New Model

Absolute Mean Error Previous Model

Average Error New Model

Average Error Previous Model

NSW1

1892.8302822

1992.5768937

091.8939237970

183.7061902187

VIC1

2385.4639402

240.862815

-0.865354636

-2.249786354

QLD1

19.29253

19.63574

-2.421301777

-0.030032336

SA1

26.06217

27.25936

0.876947057

-1.019642282

TAS1

12.37019

13.42848

-1.026951774

2.36082158552

QLD1

214.73

238.09

202.77

225.16

206.22

23.36

-11.95

SA1

36.86

51.16

46.38

46.37

45.49

14.30

9.52

TAS1

37.43

42.49

39.79

19.24

19.92

5.06

2.36

VIC1

17.34

25.25

21.75

25.92

27.50

7.91

4.41

7 Day Results

20458700122208112427-894232067945-2.97437

Metric/

Region

Actual Mean Price

Mean Prediction New Model

Mean Prediction Previous Model

Absolute Mean Error New Model

Absolute Mean Error Previous Model

Average Error New Model

Average Error Previous Model

NSW1

89.0

70.

8

60.

6

72.

2

74.

VIC1

28.11448622

29.9156995

-3.37011

1.569229

QLD1

22.74275858

30.79893723

-13.7703

15.93677

SA1

31.05123392

69.518694

5.905758

48.31805

TAS1

18.28810034

21.29027034

-2.81489

6

-18.2

-28.4

QLD1

203.4

175.6

424.8

202.6

391.5

-27.8

221.3

SA1

31.1

45.4

79.0

51.8

83.0

14.3

47.9

TAS1

35.4

44.5

32.2

22.6

23.3

9.1

-3.1

VIC1

13.0

30.5

24.8

37.1

38.8

17.4

11.7

Info

Variables that imply binding interconnectors (such as limits and marginal values) are currently not included as features.

...