Electricity Market Price Forecasting Based on Weighted Nearest Neighbors Techniques
Abstract
This paper presents a simple technique to forecast next-day electricity market prices based on the weighted nearest neighbors methodology. First, it is explained how the relevant parameters defining the adopted model are obtained. Such parameters have to do with the window length of the time series and with the number of neighbors chosen for the prediction. Then, results corresponding to the Spanish electricity market during 2002 are presented and discussed. Finally, the performance of the proposed method is compared with that of recently published techniques.
Full text
! ! " # ! ! $ ! % &'($)* ! + ! ,! % - ! -! . /# ( (- &/((* ( (!% &((* ' ! &'/* 0 1 2% 3 ! 45%! ! " # ! ! 6!% ,! -! ! ,! ! ! % ! " # $%%& ! # !' " ()*+ ,! " , #! " - .-/ ' # # .# " 0 ! ,! , 12 0 0 34 0 0, !/5 - "', 0 # 0 ! 5 " 0 6 0 ! #5 " 0 , 5 - ,! " ! - # 0 , . ! , 2 / 0 0! 0 0 34 0 0, !5 " , , , # 0 , ! 0! #! , ,7 # 8 # .2/ .12/5 " 0 ' ,! . # $) 0 0 / 05
9 0, ,! ! 5 9 : # 0 ,! " #, ,5 ;! ' ! 0, ! ,! < 0 , ' .=;-/ " 0 1, # #! , ,!5 0 ,! 0, + : = =" .==/ # = =,# ."==/ , ./5 > '! ? 34' ,! 0, # ' :5 ,# , , 2,' 5 , #! , , ,5 == # # 0 0 0 0, # ($ 3*5 , @ 0 " , ! # 0 ! ' 0 , "5 ! # ! " 0 ! 3AA$ , 3AA$5 " ! ! , ", ! # !5 >, $ ! , 0 0 ", ! 0 3AA$ 0 "25 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 1 3 5 7 9 11 13 15 17 19 21 23 Hours c ent/kW h Mean Mean + S.D. Mean - S.D. 0 1 2 3 4 5 6 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Hours c ent/kW h Thursday-1st Friday-30th 7 2! 4889 - 2! 4889 , , 3 1"2 " , , , " .$A'3 &'$A !/5 0 0 ! 55 0 3AB 0 , 4AB & %5
>, $# ,! 0 ! 0 3AA$5 ,: 8 0 " # #! , : ##! , " 5 #0 0 == " . == #", 0 (C* (4*/5 == 0 # 0 , #!!5 , ' 0 0 5 7 8 #!,, 0 # , , 5 D! == 0 ! . / 0 ! 0 0 !5 0 == + E, ,! #! 0 :, # 0 !5 -#, ,, 0 0 , ' 0 .' , /5 , 7 0 == 0, 0 , 34 5 ! 0 34 5 0 # 0 # # 5 / , #! 34 5 D ! 0, 0 5 #/ ! 0 , 0 ! #! 0 !5 0 34 5 # ! 0 #/ ! 5 9 0! : == ! # 0 ! # 0 ! 0 , 5 >, 3 0 $3 34 CF ! , 0, ,! , ", ! 0 >#' ! 3AA$ #0! ==5 # 3A ! Æ == # 0 # , 5 > # 34 ! # 0 " ,! 5
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 0 5 10 15 20 25 30 35 40 45 Days E rror (cent/kWh ) n12 n24 n36 / ! # 0 # .! ' >#! 3AA$/ : # == ,! , .', 3AA$/ " 0, 5 >, C # 0 0 0 " 0 ! 0 3AA$ , , , 5 0 0.5 1 1.5 2 2.5 3 1 3 5 7 9 11 13 15 17 19 21 23 Hours E rror (cent/kWh ) Friday 23 Monday 12 0 0.5 1 1.5 2 2.5 3 3.5 1 4 7 10 13 16 19 22 Hours c ent/kW h Errors Forecasted / ! ! & - * 7 ! &2! 4889* 3 &+6!* / / 2" 2!%2 44:;; 8<;89 8=5>5 4><9 ?%/ =:5;4 98:@< 854; 48<=> 0 ! >, C# ! , 0 0 ,! ' , 3AA$ , .# #! 8 0 :, $/5 = 0, , , " 5 >! # $ , 55 0 , # 0 0, 0 ,
5 , , 0 $3B ./ $)B .,/ 0 ! , 5 , 55 0 , , 0, 5 9 # "== ($$ $C* ($4* 0 ! " ,! 0, #5 , , , ,: 5 "== : ! " ,! G H 0 !5 , 0 ! 0 # 0 ,! 0 ! 0 ,# !5 0 ", 0 + ! #, # 0 ,#5 0 # ! (I &*5 ,! ! ! 5 > ' 0, # , 0, 0 , 34 ! # , 345 0 # 0 ! #! # + ! 0 , 0 ! #! 0 !5 #! 0 34 5 # ' 5 ! 0 , # ' 5 0 !# ! : 0 0 5 ' 0 , # (%*5 9 ! " ,! 2,' 0 # 0 ! @ 0 0, !5 : #! & *A & * &9* J ,. / (A $*5
0 " ,! 0 ! K $ , ", , 0 0 !+ $5 ; # 0 ! , , .$/5 # 555 ! ! #! 5 35 + A 9 B B &4* A & * & * & * & ½ * &=* 0 , ! 0 0, 5 = A $ 55 , ! ! 5 0 ,# 0 ( ½ * ,# 0 ( ½ * ! # 5 , ./ # 0 # ! , 0 #, ,5 # # : 0 , # 0 # , , : 05 #! 5 9 : #! ,0 , ! : 0 0 , ./ 0 0 , 5 # , # 0 ($3*+ $5 ! 5 35 : 0 0 5 C5 ? ! , 5 45 ? # 5 )5 5 F5 # : 0 5 I5 90 # 0 , # , 35
= : 0 0 = =,# , 5 9 ! #! "== 5 9 #! , 34 0 !5 # 0 ! 0 0 # 0 ! 5 90 #0! # 0 , ' 0 ! ! 5 9 0 #0! ! ,5 - 0 0,+ ! ! ! 5 " : 0 # # , ' 0 !0 15 "==K?E9;9-= ' ,! , .3/5 : 0 : #! 0 A 9 & * &5* # 0 ! ,5 0 ,# 0 ! # "5 # , ! , # 0 5 ##! 0 #, : 0 ", # , 5 ,, ##! # 55 # # ! # # 5 ; # #! , , 0 0 5 >, , , ': ##!5 , # #! ! ! 5 , # ! : ##!5 $ % "==K # # 0 0 ! " ,! 5 >#! 3AA$ # ' ,0 ! 0 ! , # 0 # 35 # 0 ', 3AA$ # " 0, 5 # 0 ,# # @ 0 ($A*5
0.3 0.35 0.4 1 501 1001 1501 2001 2501 Number of Generations S quare Erro r ! 0, # #! , 0 == # : 0 5 >, 4 0 0 : , 5 3 1 988 3 9 3 89 2" :888 '/ 0 0.5 1 1.5 2 2.5 3 3.5 4 1 3 5 7 9 11 13 15 17 19 21 23 Hours P rices (Cent/kWh ) 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 Weights March 21 March 22 0 0.5 1 1.5 2 2.5 3 3.5 4 1 3 5 7 9 11 13 15 17 19 21 23 Hours P rices (Cent/kWh ) February 28 March 22 ( ! . ! 6!% ! >, ) , 0 ! , ,! 0 ! 33 ,# , 2,' .$/5 ,# 3$5 9 # , , .$ I $$ $$ $3/ ,! 5 L C 4 ) F I 8 , , ' ! 5 >, )# ,! 0 33 , ' ,# , 5 9 ,# >#! 3&5 = @ 0 ' 0 ,# 5
0 0.5 1 1.5 2 2.5 3 1 3 5 7 9 11 13 15 17 19 21 23 Hours E rror (Cent/kWh ) With Euclidean distance With WeightedEuclidean distance / ! ! &2! 4= * 3 &+6!* / / 2" 2!%2 44:;; 849;> 84:> 49= ?%/ =:5;4 84> 8== 4=> 0 >, F # 0 0 0 " 0 3C , 2,' 5 0, , 0 ! ,! @ #! 0 ,# 5 >! # C , 55 0 , # 0 0, 0 , 5 = , , == , 0 $3B ./ $)B .,/ 0 ! , , # #! 0 "== # , 0 %B ./ $$B .,/5 # 4 # , # 5 - #, , " 0, 0 : # , 0 , 5 9 0, + : = =" # = =,# ,5 M , # 34' ' ,! 0, # , 0 ,! " 0 # 5 2,' "== 0, 0 ,! == 5 2!%2 ?%/ ((B'/ /(( ((B'/ /(( 849;> 8<;89 84> 98:@< / 84:> 8=5>5 8== 854; 2" 49= 4><9 4=> 48<=> / C &D* 99 9: @ 94 $ &$+6!*