רҵÌṩÂÛÎÄ·¢±í×Éѯ£¬¹ú¼Ê³ö°æ£¬Ó¢ÎÄÆÚ¿¯Í¶¸å×Éѯ£¬SCI£¬SSCI£¬EIÔ´¿¯£¬A&HCI£¬SCOPUSµÈ¸ß¶ËѧÊõ×Éѯ·þÎñ
ɨÂë¹Ø×¢¹«ÖÚºÅ
΢ÐÅɨÂë¼ÓºÃÓÑ
ѧÊõ³ö°æ£¬¹ú¼Ê½ÌÖø£¬¹ú¼ÊÆÚ¿¯£¬SCI£¬SSCI£¬EI£¬SCOPUS£¬A&HCIµÈ¸ß¶ËѧÊõ×Éѯ
À´Ô´£ºÖ°³ÆæäÕ¾ËùÊô·ÖÀࣺÊг¡ÓªÏúÂÛÎÄ·¢²¼Ê±¼ä£º2021-01-07 08:47:23ä¯ÀÀ£º´Î
¡¡¡¡Õª Òª£ºËæ×Å´¿µç¶¯Æû³µÊг¡µÄ²»¶ÏÀ©´ó£¬¶ÔÏúÊÛÁ¿µÄ¾«È·Ô¤²â³ÉΪÈËÃǵ±Ç°¹Ø×¢µÄÈȵ㡣ÌáÈ¡Ó°ÏìÏúÊÛÁ¿µÄ7¸ö¹Ø¼üÒòËØÒÔ¼°2017¡«2019Äê9ÔµĴ¿µç¶¯Æû³µÏúÊÛÁ¿£¬Ê×ÏÈÀûÓÃBPÉñ¾ÍøÂçÄ£ÐͶÔ33¸öÔµÄÊý¾Ý½øÐÐœyÊÔ£¬²¢ÓÃѵÁ·ºÃµÄÄ£ÐÍÔ¤²â2019Äê1¡«9ÔÂÏúÊÛÁ¿£¬ÔÙÀûÓÃÂí¶û¿Æ·ò£¨Markov£©Ä£Ðͽ«BPÉñ¾ÍøÂçÄ£ÐÍÔ¤²âµÄÏà¶ÔÎó²î»®·ÖΪ6ÖÖ״̬£¬¶ÔÔ¤²â½á¹û½øÐÐÐÞÕý¡£Í¨¹ý¶ÔBPÉñ¾ÍøÂçÄ£ÐÍÓëMarkov-BPÉñ¾ÍøÂçÄ£ÐÍÔ¤²â½á¹û½øÐжԱȼìÑ飬·¢ÏÖMarkov-BPÉñ¾ÍøÂçµÄÔ¤²â׼ȷ¶È¸ü¸ß£¬±íÃ÷²ÉÓÃMarkov-BPÉñ¾ÍøÂçÄ£ÐͶԴ¿µç¶¯Æû³µÔ¶ÈÏúÊÛÁ¿½øÐÐÔ¤²â¾ßÓÐÒ»¶¨ÏÖʵÒâÒå¡£
¡¡¡¡¹Ø¼ü´Ê£ºBPÉñ¾ÍøÂç;Âí¶û¿Æ·òÁ´;´¿µç¶¯Æû³µ;ÏúÊÛÁ¿Ô¤²â;MATLAB
¡¡¡¡DOI£º10. 11907/rjdk. 201180 ¿ª·Å¿Æѧ£¨×ÊÔ´·þÎñ£©±êʶÂ루OSID£©£º
¡¡¡¡ÖÐͼ·ÖÀàºÅ£ºTP306 ÎÄÏ×±êʶÂ룺A ÎÄÕ±àºÅ£º1672-7800£¨2020£©011-0050-04
¡¡¡¡¡¶³É¹¦ÓªÏú¡·ÔÓÖ¾´´¿¯ÓÚ2000Ä꣬Á¥ÊôÓÚÏã¸ÛÉÏÊй«Ë¾²ÆѶ´«Ã½¼¯ÍÅ(SEEC)¡£¼¯ÍÅÓµÓеÄýÌ帲¸ÇÁ˽ðÈÚ¡¢µØ²ú¡¢ÓªÏú¡¢Æû³µ¡¢µçÄÔ¡¢ÌåÓý¡¢Ê±ÉС¢¼Ò¾ÓµÈÖî¶àÁìÓò¡£
¡¡¡¡Pure Electric Vehicle Sales Forecast Research Based on
¡¡¡¡Markov and BP Neural Network
¡¡¡¡SHI Cui-cui£¬ LIU Yuan-hua
¡¡¡¡£¨Business School£¬ University of Shanghai for Science & Technology£¬ Shanghai 200093£¬ China£©
¡¡¡¡Abstract£ºWith the continuous expansion of the pure electric vehicle market£¬ the accurate prediction of sales volume has become the focus of attention. Seven key factors affecting the sales volume and the sales volume of pure electric vehicles in the 33 months from 2017 to 2019 are extracted. Firstly£¬ the data of these 33 months are tested by BP neural network model£¬ and the sales volume from January to September in 2019 is predicted by the trained model. Then£¬ the relative error of BP neural network model is divided into six states by using Markov model. The result is modified to get the forecast of pure electric vehicle sales volume based on Markov modified BP neural network £¨BPMC£©. Through the comparative test£¬ it is verified that the prediction accuracy of Markov correction is higher£¬ which shows that the BPMC model has a certain practical significance for the monthly sales forecast of pure electric vehicles.
¡¡¡¡Key Words£ºBP neural network; Markov chain; pure electric vehicle; sales forecast; MATLAB
¡¡¡¡0 ÒýÑÔ
¡¡¡¡´¿µç¶¯Æû³µµÄÓ¦ÓÿÉÒÔ»º½âÄÜÔ´¶Ìȱ¡¢»·¾³ÎÛȾµÈÎÊÌ⣬Òò´Ë¸÷ÆóÒµÔ½À´Ô½ÖØÊÓ´¿µç¶¯Æû³µÉú²ú¡¢¼¼ÊõºÍÏúÊÛ·½ÃæµÄ´´Ð¡£´¿µç¶¯Æû³µÏúÊÛÁ¿Ò²ÈÕÒæÔö³¤£¬¸ù¾ÝÇ°Õ°²úÒµÑо¿Ôº±¨¸æÏÔʾ£¬2019ÄêÉÏ°ëÄê´¿µç¶¯Æû³µÈ«ÇòÏúÊÛÁ¿Ôö³¤ÁË92%£¬´ïµ½76.5ÍòÁ¾£¬ÖйúÊг¡ÏúÊÛÁ¿Îª43.07ÍòÁ¾£¬Í¬±ÈÔö³¤ÁË111%¡£ÓÉÏúÁ¿Êý¾Ý¿ÉÒÔ¿´³ö£¬´¿µç¶¯Æû³µÊг¡·¢Õ¹Ç°¾°Êǿɹ۵ġ£Òò´Ë£¬´¿µç¶¯Æû³µÑз¢ÓëÍƹãÒÀÈ»ÊÇÐÂÄÜÔ´Æû³µÖØÒªµÄÑо¿·½Ïò¡£½üÄêÀ´ÔÚÕþ¸®ÓëÆóÒµµÄ¹²Í¬Å¬Á¦Ï£¬µç¶¯Æû³µÔÚÉú²ú³É±¾¡¢ÐÐÊ»Àï³Ì¡¢×ʽðͶÈëÓëʹÓõȹؼüÎÊÌâÉÏ»ñµÃÁËÖØ´óÍ»ÆÆ[1]¡£¶Ô´¿µç¶¯Æû³µÔ¶ÈÏúÊÛÁ¿½øÐÐÔ¤²â£¬ÄÜΪÆóÒµ¿ÆѧÖƶ¨Æû³µ²úÁ¿ÌṩÒÀ¾Ý¡£
¡¡¡¡Ó°Ïìµç¶¯Æû³µÏúÊÛÁ¿µÄÒòËØÖڶ࣬ÇÒ¸´ÔÓ¶à±ä£¬Í¨³£¾ßÓзÇÏßÐԱ仯ÌØÐÔ£¬µ¼Ö´«Í³ÏßÐÔÓ³ÉäÄ£ÐͶԴ¿µç¶¯Æû³µÏúÊÛÁ¿µÄÔ¤²â¾«¶È½ÏµÍ[1]¡£Òò´Ë£¬ºÜ¶àѧÕß²ÉÓûùÓÚ·ÇÏßÐÔÄ£Ð͵ÄÔ¤²â·½·¨¡£Èç¶ÀöÀöµÈ[2]ÀûÓÃLSTMÉñ½›ÍøÂç¶Ôú¿óͻˮÇé¿ö½øÐÐÔ¤²â;Áõè´µÈ[3]ÀûÓÃSVM¶ÔÑ̲ÝÏúÊÛÁ¿½øÐÐÔ¤²âµÈ¡£ÎªÁËʵÏÖ¸üºÃµÄÔ¤²âЧ¹û£¬½«¶àÖÖÔ¤²â·½·¨Ïà½áºÏÒ²³ÉΪµ±Ç°µÄÑо¿Èȵã[4-5]¡£ÈçÀî¾ü»³µÈ[6]Ìá³ö½èÖúÂí¶û¿É·òÁ´ÐÞÕýÖ¸Êýƽ»¬µÄ·½·¨Ô¤²â¶Ìʱ½»Í¨Á÷Á¿£¬ÒÔÌá¸ßÔ¤²â¾«¶È;»Æ¸þǽµÈ[7]Ê×ÏÈÀûÓÃBPÉñ¾ÍøÂçÄ£ÐͶԶÌÆÚµç¼Û½øÐгõ²½Ô¤²â£¬ÔÙÓÃÂí¶û¿É·òÄ£ÐͶÔÔ¤²âÎó²î½øÐÐÐÞÕý£¬¿ÉÌá¸ß¶ÌÆÚµç¼ÛÔ¤²â¾«¶È;ʯƼµÈ[8]¿¼Âǵ綯Æû³µÏúÁ¿µÄ¼¾½ÚÐԱ仯ÒòËØ£¬ÔËÓÃ×éºÏÄ£ÐÍÔ¤²âÐÂÄÜÔ´µç¶¯Æû³µÏúÊÛÁ¿;ÖÜÑ帣µÈ[9]ÏÈÀûÓûÒÉ«¹ØÁª·ÖÎö8ÖÖÓ°ÏìÏúÊÛÁ¿µÄÒòËØ£¬ÔÙ¹¹½¨¹ûÓ¬Ëã·¨ÓÅ»¯»ÒÉ«Éñ¾ÍøÂ磨FOA-GNN£©Ô¤²âÄ£ÐͽøÐÐÐÂÄÜÔ´Æû³µÔ¶ÈÏúÊÛÁ¿Ô¤²â£¬Í¨¹ý¶Ô±ÈÔ¤²â½á¹û£¬ÏÔʾ¸ÃÄ£Ð;ßÓнϸߵÄÔ¤²â¾«¶È¡£
¡¡¡¡ÔÚ֮ǰµÄÑо¿ÖУ¬ÌáÈ¡´¿µç¶¯Æû³µÏúÊÛÁ¿Ó°ÏìÒòËØʱ²¢Î´¿¼ÂÇÏû·ÑÕßÔ¤ÆÚÖ¸Êý¡¢Ïû·ÑÕßÂúÒâÖ¸Êý¡¢ÆóÒµ¼ÒÐÅÐÄÖ¸ÊýºÍúÓ͵ç¼Û¸ñÖ¸ÊýµÈÒòËØ£¬±¾ÎĽ«ÕâЩÒòËØ¿¼ÂÇÔÚÄÚ£¬ÀûÓÃBPÉñ¾ÍøÂçÔ¤²â´¿µç¶¯Æû³µÏúÊÛÁ¿£¬ÔÙ¸ù¾ÝÂí¶û¿Æ·òÁ´ÐÞÕýÔ¤²â½á¹û£¬ÒÔ½øÒ»²½Ìá¸ßÔ¤²â¾«¶È¡£
¡¡¡¡1 BPÉñ¾ÍøÂçÓëÂí¶û¿Æ·òÄ£ÐÍ
¡¡¡¡1.1 BPÉñ¾ÍøÂçÄ£ÐÍ
¡¡¡¡BPÉñ¾ÍøÂçÊÇ1986ÄêÓÉÒÔRumelhart&McCellandΪÊ׵ĿÆѧ¼ÒС×éÌá³öµÄ£¬ÊÇÒ»ÖÖÀûÓÃÎó²îÄæ´«²¥Ë㷨ѵÁ·µÄ¶à²ãÇ°À¡ÍøÂ磬ÒòÆä¾ßÓмòµ¥Ò×ÐС¢¼ÆËãÁ¿Ð¡¡¢²¢ÐÐÐÔÇ¿µÈÓŵ㣬³ÉΪĿǰӦÓñȽϹ㷺µÄÒ»ÖÖÉñ¾ÍøÂçÄ£ÐÍ[10-11]¡£BPÉñ¾ÍøÂçÓÉÊäÈë²ã¡¢Òþº¬²ãºÍÊä³ö²ã×é³É£¨¼ûͼ1£©¡£ÔÚÍøÂçѵÁ·¹ý³ÌÖУ¬Í¨¹ý½«Êä³ö²ãÓëʵ¼ÊÊä³öÑù±¾Ö®¼äµÄÎó²î¾¹ý¸÷Òþº¬²ã·´Ïò´«µÝµ½ÊäÈë²ã£¬Öð²ã¼ÆËã¸÷²ãÉñ¾ÔªÎó²î£¬È»ºó¸ù¾ÝÎó²îÌݶÈϽµ·¨µ÷½Ú¸÷²ãȨֵºÍãÐÖµ£¬Ê¹Ð޸ĺóÍøÂç×îÖÕÊä³öÄܽӽüÆÚÍûÖµ[12-14]¡£
¡¡¡¡1.1.1 ÐźÅÇ°Ïò´«²¥¹ý³Ì
¡¡¡¡ÉèÍøÂçÊäÈëģʽΪxx1£¬x2£¬…£¬xnT£¬Òþº¬²ãÓÐh¸öµ¥Ôª£¬ÆäÊä³öΪy=£¨y1£¬y2£¬…£¬yh£©T£¬Êä³ö²ãÓÐm¸öµ¥Ôª£¬ÆäÊä³öΪz=£¨z1£¬z2£¬…£¬zm£©T£¬Ä¿±êÊä³öΪt=£¨t1£¬t2£¬…£¬tm£©T£¬Òþº¬²ãµ½Êä³ö²ãµÄ´«µÝº¯ÊýΪf£¬Êä³ö²ã´«µÝº¯ÊýΪg¡£¼ÆË㹫ʽÈçÏ£º
¡¡¡¡yj=fi=1nwijxi-θ=fi=0nwijxi £¨1£©
¡¡¡¡Ê½ÖУ¬yj±íʾÒþº¬²ãµÚj¸öÉñ¾ÔªµÄÊä³ö£¬wΪȨÖØÏòÁ¿£¬w0j=裬x0=-1¡£fΪ¼¤»îº¯Êý£¬Ò»°ãÑ¡È¡sÐͺ¯Êý¡£
¡¡¡¡fx=11+e-ax£¬£¨0
¡¡¡¡zk=g£¨j=0hwjkyj£© £¨3£©
¡¡¡¡Ê½ÖУ¬zk±íʾÊä³ö²ãµÚk¸öÉñ¾ÔªµÄÊä³ö¡£´ËʱÍøÂçÊä³öÓëÄ¿±êÊä³öµÄÎó²îΪ£º
¡¡¡¡ε=12k=1m£¨tk-zk£©2 £¨4£©
¡¡¡¡ÓÉÓÚ¸ºÌݶȷ½ÏòÊǺ¯ÊýÖµ¼õС×î¿ìµÄ·½Ïò[11]£¬Òò´Ë¿ÉÉ趨һ¸ö²½³¤η£¬Ã¿´ÎÑظºÌݶȷ½Ïòµ÷Õûη¸öµ¥Î»£¬¼´Ã¿´ÎȨֵµÄµ÷ÕûΪ£º
¡¡¡¡Δwpq=-η£¿ε£¿wpq £¨5£©
¡¡¡¡ÆäÖУ¬ηΪѧϰËÙÂÊ¡£
¡¡¡¡1.1.2 Îó²î·´Ïò´«²¥
¡¡¡¡Òþº¬²ãµ½Êä³ö²ãµÄȨֵµ÷Õûµü´ú¹«Ê½Îª£º
¡¡¡¡wjkt+1=wjkt+ηδkyj £¨6£©
¡¡¡¡´ÓÊäÈë²ãµ½Òþº¬²ãµÄȨֵµ÷Õûµü´ú¹«Ê½Îª£º
¡¡¡¡wijt+1=wijt+ηδjxj £¨7£©
¡¡¡¡1.2 Âí¶û¿Æ·òÁ´Ô¤²âÄ£ÐÍ
¡¡¡¡Âí¶û¿Æ·ò¹ý³ÌÓÃÓÚÑо¿Ò»¸öϵͳµÄ×´¿ö¼°ÆäתÒÆ£¬ÊÇÒ»ÖÖ³£¼ûµÄ¼òµ¥Ëæ»ú¹ý³Ì¡£Æäͨ¹ý¶Ô²»Í¬×´Ì¬³õʼ¸ÅÂÊÒÔ¼°×´Ì¬Ö®¼äתÒƸÅÂʵÄÑо¿£¬È·¶¨×´Ì¬±ä»¯Ç÷ÊÆ£¬´Ó¶ø´ïµ½Ô¤²âδÀ´µÄÄ¿µÄ[10]£¬¾ßÓÐÎÞºóЧÐÔ¡¢±éÀúÐԺ͹ý³ÌËæ»úÐÔ¡£Ó¦ÓÃÂí¶û¿Æ·ò£¨Markov£©Ä£ÐͶÔBPÉñ¾ÍøÂçÄ£ÐÍÔ¤²âÊý¾Ý½øÐд¦Àí£¬¿ÉÌá¸ßÔ¤²â¾«¶È¡£
¡¡¡¡1.2.1 Âí¶û¿Æ·òÁ´
¡¡¡¡Ê±¼äºÍ״̬¶¼ÊÇÀëÉ¢µÄÂí¶û¿Æ·ò¹ý³Ì³ÆΪÂí¶û¿Æ·òÁ´[10]¡£¼ÙÉèÂí¶û¿Æ·òÁ´ÓÐn¸ö״̬£¬Ôò״̬¸ÅÂÊÏòÁ¿Îªα=α1£¬α2£¬…£¬αn£¬¹«Ê½Îª£º
¡¡¡¡αn=α0pn £¨8£©
¡¡¡¡Ê½ÖУ¬αn¡¢α0·Ö±ð±íʾnʱ¿ÌºÍ³õʼʱ¿ÌµÄ״̬¸ÅÂÊÏòÁ¿£¬p±íʾ״̬תÒƸÅÂʾØÕó¡£
¡¡¡¡1.2.2 ³õʼ״̬¸ÅÂÊÓë״̬תÒƸÅÂʾØÕó
¡¡¡¡£¨1£©³õʼ״̬¸ÅÂÊ¡£ÓÃMi±íʾEjÐÍ״̬³öÏÖµÄ×Ü´ÎÊý£¬Ôò³õʼ¸ÅÂÊΪ£º
¡¡¡¡α£¨0£©i≈Fi=Mii=1nMi£¨i=1£¬2£¬…£¬n£© £¨9£©
¡¡¡¡£¨2£©×´Ì¬×ªÒƸÅÂÊ¡£Èç¹ûÑо¿¶ÔÏó´¦ÓÚEi£¬ÏÂÒ»²½×ªÒƵ½Ej£¬´Ë¹ý³Ì¹²·¢Éúnij´Î£¬Ôò£º
¡¡¡¡pij=nijni£¬i£¬j=1£¬2£¬…£¬n
¡¡¡¡0≤pij≤1£¬j=1npij=1 £¨10£©
¡¡¡¡£¨3£©×´Ì¬×ªÒƸÅÂʾØÕ󡣸ù¾Ý״̬תÒƸÅÂʵóö״̬תÒƸÅÂʾØÕóΪ£º
¡¡¡¡P=P11P12£¿P1nP21P22£¿P2n£¿Pn1£¿Pn2£¿£¿£¿Pnn£¬
¡¡¡¡0≤pij≤1£¬j=1npij=1 £¨11£©
¡¡¡¡1.2.3 Âí¶û¿Æ·òÁ´ÐÞÕýBPÉñ¾ÍøÂçÔ¤²âÖµ
¡¡¡¡½«BPÉñ¾ÍøÂçÔ¤²âÖµÓëʵ¼ÊÖµµÄÏà¶ÔÎó²î»®·ÖΪ²»Í¬×´Ì¬£¬αnΪʱ¿ÌnµÄ״̬¸ÅÂÊÏòÁ¿£¬αnÖÐ×î´ó¸ÅÂÊÖµ¶ÔÓ¦µÄ״̬EjΪԤ²âʱ¿ÌËù¶ÔÓ¦µÄ״̬£¬ÔòÐÞÕýºóµÄÔ¤²â½á¹ûΪ£º
¡¡¡¡e*=ea+eb2 £¨12£©
¡¡¡¡Q*=Qx£¨1+e*£© £¨13£©
¡¡¡¡Ê½ÖУ¬ea¡¢ebžé״̬Çø¼äÉÏϽ磬QxΪBPÉñ¾ÍøÂçÔ¤²âµÄ´¿µç¶¯Æû³µÏúÊÛÁ¿¡£
¡¡¡¡2 Êý¾ÝÀ´Ô´¼°Ô¤´¦Àí
¡¡¡¡Êý¾ÝÑù±¾µÄÑ¡È¡ÊÇÓ°ÏìBPÉñ¾ÍøÂçÄ£ÐÍÔ¤²â½á¹ûµÄ¹Ø¼üÒòËØ£¬Òò´ËÑ¡È¡´¿µç¶¯Æû³µÏúÊÛÁ¿µÄÓ°ÏìÒòËØÒªÓë´¿µç¶¯Æû³µÏúÊÛÁ¿¾ßÓиßÏà¹ØÐÔ¡£ÃºÓ͵ç¼Û¸ñÖ¸ÊýºÍÖйúÆûÓͼ۸ñ¿É·´Ó³´¿µç¶¯Æû³µÏúÊÛÇ÷ÊÆ£¬¾ÓÃñÏû·Ñ¼Û¸ñÖ¸Êý¿É·´Ó³¾ÓÃñÏû·ÑˮƽµÄ¸ßµÍ£¬Ïû·ÑÕßÂúÒâÖ¸ÊýÓëÔ¤ÆÚÖ¸Êý´ú±í×ÅÏû·ÑÕßµÄÂúÒâ¶ÈÓëÆÚÍû£¬ÆóÒµ¼ÒÐÅÐÄÖ¸ÊýÓ°ÏìÆóÒµ¹ÜÀíģʽÓëµç¶¯Æû³µ²úÏúģʽ£¬ÕâЩÒòËض¼ÄÜ×ۺϷ´Ó³´¿µç¶¯Æû³µÏúÁ¿Çé¿ö¡£Òò´Ë£¬±¾ÎÄÑ¡È¡Öйú¾ÓÃñÏû·Ñ¼Û¸ñÖ¸Êý¡¢ÃºÓ͵ç¼Û¸ñÖ¸Êý¡¢Ïû·ÑÕßÂúÒâÖ¸Êý¡¢Ïû·ÑÕßÔ¤ÆÚÖ¸Êý¡¢Ïû·ÑÕßÐÅÐÄÖ¸Êý¡¢ÖйúÆûÓͼ۸ñ¡¢ÆóÒµ¼ÒÐÅÐÄÖ¸Êý×÷ΪӰÏì´¿µç¶¯Æû³µÏúÊÛÁ¿µÄÖ÷ÒªÒòËØ¡£
¡¡¡¡±¾ÎÄͨ¹ý¶«·½²Æ¸»ÍøÊý¾ÝÖÐÐÄËѼ¯µ½2017Äê1ÔÂ-2019Äê9Ô¸÷Ó°ÏìÒòËØÏà¹ØÊý¾Ý£¬²¢ÔÚÇ°Õ°²úÒµÑо¿Ôº²éѯµ½33¸öÔ·ݵĴ¿µç¶¯Æû³µÏúÊÛÁ¿¡£²ÉÈ¡µÄÑù±¾Êý¾ÝÖи÷ÒòËØÆÀ¼Û±ê×¼²»Í¬£¬Í¨³£¾ßÓв»Í¬µÄÁ¿¸ÙºÍÊýÁ¿¼¶£¬ÐèÒª¶ÔÔʼÊý¾Ý½øÐбê×¼»¯´¦ÀíÒÔÈ·±£½á¹ûµÄ¿É¿¿ÐÔ¡£¶øÇÒ±¾ÎÄBPÉñ¾ÍøÂç²ÉÓÃsÐͺ¯Êý£¬Í¨¹ý¹éÒ»»¯Äܹ»·ÀÖ¹Éñ¾ÔªÊä³ö±¥ºÍÏÖÏó£¬Ê¹ÍøÂç¿ìËÙÊÕÁ²¡£Òò´Ë£¬Ê×ÏÈÔÚMATLABÖжÔÒÔÉÏ7¸öÓ°ÏìÒòËؽøÐÐÊý¾Ý¹éÒ»»¯´¦Àíµ½[0£¬1]Çø¼ä¡£
¡¡¡¡x'=x-xminxmax-xmin £¨14£©
¡¡¡¡3 Ä£ÐÍ·ÖÎöÓëÐÞÕý½á¹û
¡¡¡¡BPÉñ¾ÍøÂçΪËõ¶ÌÍøÂçѵÁ·Ê±¼äÓëÌá¸ßѵÁ·¾«¶È£¬Í¨³£Ñ¡Ôñ½ÏÉÙµÄÒþº¬²ã½áµãÊý[7]¡£±¾ÎÄѡȡѧϰÂÊlr=0.000 1£¬×î´óѵÁ·´ÎÊýepochs=10 000£¬ÊÕÁ²Îó²îgoal=1e-3£¬Òþº¬²ãÉñ¾Ôª½áµãÊýΪ9£¬ÑµÁ·º¯ÊýΪtrainlm¡£ÑµÁ·ÍøÂçÖн«ÊäÈëÊý¾ÝµÄ70%×÷ΪѵÁ·¼¯£¬15%×÷Ϊ²âÊÔ¼¯£¬15%ÓÃÀ´½øÐÐÄ£ÐÍÔ¤²â[15]£¬·ÂÕæ½á¹ûÈçͼ2Ëùʾ¡£ÔËÓÃBPÉñ¾ÍøÂç¶Ô2017-2018Äê24¸öÔµĴ¿µç¶¯Æû³µÏúÊÛÁ¿½øÐÐÔ¤²â£¬Æäʵ¼ÊÖµÓëÔ¤²âÖµÏà¶ÔÎó²îÈç±í1Ëùʾ¡£½«Ïà¶ÔÎó²î·ÖΪ6¸ö״̬£¬·Ö±ðΪ£ºE1£º-35%¡«-20%£¬E2£º-20%¡«-5%£¬E3£º-5%¡«10%£¬E4£º10%¡«25%£¬E5£º25%¡«40%£¬E6£º40%¡«55%¡£
¡¡¡¡¸ù¾ÝÏà¶ÔÎó²î»®·Ö״̬£¬ÀûÓÃÂí¶û¿Æ·ò½øÐÐÓÅ»¯£¬ÓÉÉÏÊöµÄÂí¶û¿Æ·òÄ£ÐÍ¿ÉÇóµÃÒ»´Î״̬תÒƸÅÂÊΪ£º
¡¡¡¡P1=03/401/92/94/91/30001/20011/6010 01/401/901/9000000000100 £¨15£©
¡¡¡¡½«2018Äê12ÔÂ×÷Ϊ³õʼ״̬α0=010000£¬¸ù¾ÝÂí¶û¿Æ·òÔÀí¿ÉÇóµÃ2019Äê1ÔÂ״̬Ϊα1=α0£¿P1=1/92/94/91/901/9£¬ÔòËù´¦×´Ì¬ÎªE3£¬¶ÔÓ¦Ïà¶ÔÎó²îΪ£¨-5%+10%£©/2=2.5%£¬Markov-BPÉñ¾ÍøÂçÄ£ÐÍÔ¤²âֵΪ6.35×1+2.5%=6.51¡£Í¬Àí£¬ÀûÓÃMarkov-BPÉñ¾ÍøÂçÄ£ÐͿɵóö2019Äê2ÔÂ-9ÔµĴ¿µç¶¯Æû³µÔ¤²âÖµ£¬Èç±í2Ëùʾ¡£
¡¡¡¡´ÓBPÉñ¾ÍøÂçÄ£ÐÍÓëMarkov-BPÉñ¾ÍøÂçÄ£ÐͶԴ¿µç¶¯Æû³µÏúÊÛÁ¿Ô¤²âµÄÏà¶ÔÎó²î¶Ô±È¿ÉÒÔ·¢ÏÖ£¬Markov-BPÉñ¾ÍøÂçÄ£ÐÍÔ¤²âÖµµÄÏà¶ÔÎó²î¸üС£¬Òò¶øÌá¸ßÁËÔ¤²â׼ȷ¶È¡£
¡¡¡¡4 ½áÓï
¡¡¡¡ÔÚÌᳫÂÌÉ«·¢Õ¹µÄ´ó±³¾°Ï£¬Õþ¸®Ö§³Ö´¿µç¶¯Æû³µµÄÑз¢ÓëÍƹ㣬µ«Ïֽ׶δ¿µç¶¯Æû³µµÄÏúÊÛÇé¿ö²¢²»ÀÖ¹Û¡£Òò´Ë£¬¶Ô´¿µç¶¯Æû³µÏúÊÛÁ¿½øÐÐ׼ȷԤ²â£¬¿ÉÒÔ¼õÉÙ×ÊÔ´ÀË·Ñ£¬½µµÍÆóÒµ³É±¾¡£±¾ÎÄÑ¡È¡Ó°Ïì´¿µç¶¯Æû³µÏúÊÛÁ¿µÄ7¸ö¹Ø¼üÒòËØ£¬ÀûÓÃBPÉñ¾ÍøÂçÄ£ÐͶÔÏÖÓÐÊý¾Ý½øÐÐѵÁ·£¬ÓòâÊԺõÄÄ£ÐÍÔ¤²â2019Äê1ÔÂ-9Ô´¿µç¶¯Æû³µÏúÊÛÁ¿£¬ÔÙÓÃÂí¶û¿Æ·ò£¨Markov£©Ä£ÐÍÐÞÕýÆäÔ¤²âÖµ¡£Í¨¹ý¶Ô±È·¢ÏÖ£¬Markov-BPÉñ¾ÍøÂçÄ£Ð͵ÄÔ¤²â¾«¶È¸ü¸ß£¬¿ÉÓÐЧÃÖ²¹BPÉñ¾ÍøÂçÔ¤²âµÄ²»×ã¡£µ«±¾ÎÄÑ¡È¡µÄ´¿µç¶¯Æû³µÏúÊÛÁ¿Ó°ÏìÒòËز»¹»È«Ã棬δ½áºÏ´¿µç¶¯Æû³µ¼Û¸ñ¼°ÏàÓ¦¼¼ÊõÖ¸±ê½øÐÐ×ۺϿ¼ÂÇ¡£Òò´Ë£¬³ä·Ö¿¼ÂÇÕâЩӰÏìÒòËØ£¬²¢ÕÒµ½¸üºÏÊʵĴ¿µç¶¯Æû³µÏúÊÛÁ¿Ô¤²âÄ£ÐÍ£¬ÊÇÏÂÒ»²½µÄÑо¿·½Ïò¡£
¡¡¡¡²Î¿¼ÎÄÏ×£º
¡¡¡¡[1] ËïæÃæã¬ÉòÒ㣬ÕÔÁÁ. Ò»ÖÖ»ùÓÚBPÉñ¾ÍøÂçµÄ·¿¼ÛÔ¤²âÄ£ÐÍ[J]. µçÄÔ֪ʶÓë¼¼Êõ£¬2019£¬15£¨28£©£º215-218.
¡¡¡¡[2] ¶ÀöÀö£¬·Ñ³Ç£¬ÕÅÏ裬µÈ. »ùÓÚLSTMÉñ¾ÍøÂçµÄú¿óͻˮԤ²â[J]. úÌïµØÖÊÓ뿱̽£¬2019£¬47£¨2£©£º137-143.
¡¡¡¡[3] Áõè´£¬¶¡¸£Àû£¬ËïÁ¢Ãñ. »ùÓÚSVMµÄÑ̲ÝÏúÊÛÁ¿Ô¤²â[J]. Èí¼þµ¼¿¯£¬2016£¬15£¨11£©£º134-137.
¡¡¡¡[4] ÕÅÜ磬·ëÃñȨ. »ùÓÚBPÉñ¾ÍøÂçÂí¶û¿Æ·òÄ£Ð͵ÄÕÄÔóË®¿âË®ÖÊÔ¤²â[J]. ºÚÁú½´óѧ¹¤³Ìѧ±¨£¬2018£¬9£¨2£©£º38-44.
¡¡¡¡[5] Íõ±£Ùt£¬ÁõÒã. »ùÓÚ»ÒÉ«BPÉñ¾ÍøÂçÄ£Ð͵ÄÈËÁ¦×ÊÔ´ÐèÇóÔ¤²â·½·¨[J]. ͳ¼ÆÓë¾ö²ß£¬2018£¬34£¨16£©£º181-184.
¡¡¡¡[6] Àî¾ü»³£¬¸ßÕ°£¬ÍõÖ¾Ïþ£¬µÈ. »ùÓÚÖ¸Êýƽ»¬ºÍÂí¶û¿É·òÁ´µÄ¶Ìʱ½»Í¨Á÷Á¿Ô¤²â[J]. ¼ÆËã»úϵͳӦÓã¬2013£¬22£¨12£©£º132-135.
¡¡¡¡[7] »Æ¸þǽ£¬Ñî¿¡½Ü. »ùÓÚBPÉñ¾ÍøÂçÓëÂí¶û¿É·òÁ´µÄ¶ÌÆÚµç¼ÛÔ¤²â[J]. ÉϺ£µçÁ¦Ñ§ÔºÑ§±¨£¬2017£¬33£¨1£©£º1-3£¬10.
¡¡¡¡[8] ʯƼ£¬Àè¿¡»Ô£¬Áõºé. ÐÂÄÜÔ´µç¶¯³µÏúÁ¿Ô¤²âÄ£Ð͵ÄÓ¦ÓÃ[J]. ÁÉÄþ¿Æ¼¼´óѧѧ±¨£¬2016£¬39£¨5£©£º395-400.
¡¡¡¡[9] ÖÜÑ帣£¬ÍõºìÀÙ. ÎÒ¹úÐÂÄÜÔ´Æû³µÔ¶ÈÏúÊÛÁ¿Ô¤²âÄ£ÐÍÑо¿[J]. Èí¼þµ¼¿¯£¬2019£¬18£¨8£©£º149-153.
¡¡¡¡[10] ³Â¾ê£¬²ÜÇßÓä. »ùÓÚÂí¶û¿Æ·òÁ´¶ÔBPÉñ¾ÍøÂç·¨²Ð²îÐÞÕýÄ£Ð͵ĶÌÆÚ·çµç¹¦ÂÊÔ¤²â[J]. µçÁ¦ÓëÄÜÔ´£¬2017£¬38£¨5£©£º567-570.
¡¡¡¡[11] ÍõÞÈÏ裬³Â¼ÃÓ±£¬ÍõêÉÈ«£¬µÈ. »ùÓڸĽøÐÍRF-BPÉñ¾ÍøÂçµÄ·¿µØ²ú¼Û¸ñÔ¤²â[J]. ¹¤Òµ¿ØÖƼÆËã»ú£¬2019£¬32£¨10£©£º122-124.
¡¡¡¡[12] ³Â¹ãÒø£¬²Ìå°¾¤£¬½ªÐÀ. Âí¶û¿É·òÐÞÕýµÄBPÉñ¾ÍøÂçÔÚPM2.5Ô¤²âÖеÄÓ¦ÓÃ[J]. ÄÜÔ´»·¾³±£»¤£¬2017£¬31£¨5£©£º8-11.
¡¡¡¡[13] WANG H£¬XU L£¬WANG X. Outage probability performance prediction for mobile cooperative communication networks based on artificial neural network[J]. Sensors£¬2019£¬19£¨21£©£º4789-4805.
¡¡¡¡[14] ÐìÉ. »ùÓÚBPÉñ¾ÍøÂçµÄ¸ÊËàÊ¡¹«Â·¿ÍÔËÁ¿Ô¤²â·½·¨[J]. ½»Í¨ÓëÔËÊ䣬2019£¬35£¨5£©£º28-31.
¡¡¡¡[15] ÍõóãÐÀ£¬¸ßÅÊ. »ùÓÚBPÉñ¾ÍøÂçµÄÖØÇìÊз¿¼ÛÑéÖ¤ÓëÔ¤²â[J]. ÖØÇìÀí¹¤´óѧѧ±¨£¨Éç»á¿Æѧ£©£¬2016£¬30£¨9£©£º49-53.
¡¡¡¡
¡¶ÈÚºÏMarkovÓëBPÉñ¾ÍøÂçµÄ´¿µç¶¯Æû³µÏúÊÛÁ¿Ô¤²âÑо¿¡·
±¾ÎÄÓÉÖ°³ÆæäÕ¾Ê×·¢£¬ÄúÉí±ßµÄ¸ß¶ËÂÛÎÄ·¢±íѧÊõ¹ËÎÊ
ÎÄÕÂÃû³Æ£º ÈÚºÏMarkovÓëBPÉñ¾ÍøÂçµÄ´¿µç¶¯Æû³µÏúÊÛÁ¿Ô¤²âÑо¿
ÎÄÕµØÖ·£º https://www.zhichengyz.com/lunwen/jingji/shichangyingxiao/43925.html
¿ìËÙÔ¤Éó¡¢Í¶¿¯Ç°Ö¸µ¼¡¢×¨ÒµÑ§ÊõÆÀÉ󣬶ÔÎÄÕ½øÐÐÆÀ¼Û
Á¢¼´×ÉѯУ¶Ô±à¼¡¢Éî¶ÈÈóÉ«,Èøå¼þ·ûºÏѧÊõ¹æ·¶£¬¸ñʽÌåÀýµÈ±ê×¼
Á¢¼´×ÉѯÊÊÓÃÓÚÓï¾äºÍ½á¹¹ÉÐÐèÍêÉƺ͵÷ÕûµÄÖÐÎÄÎÄÕ£¬È·±£¸å¼þ´ïµ½ÒªÇó
Á¢¼´×ÉѯÊý¾Ý¿â°üÀ¨£ºÆÚ¿¯¡¢ÎÄÕ¡¢Êé¼®¡¢»áÒé¡¢Ô¤Ó¡Êé¡¢°Ù¿ÆÈ«ÊéºÍÕªÒªµÈ
Á¢¼´×ÉѯÈÃ×÷ÕßÔÚÆÚ¿¯Ñ¡Ôñʱ±ÜÃâ×ßÍä·£¬Ëõ¶Ì¸å¼þ±»½ÓÊÕµÄÖÜÆÚ
Á¢¼´×Éѯ¸ù¾ÝÄ¿±êÆÚ¿¯¸ñʽҪÇó¶Ô×÷ÕßÎÄÕ½øÐÐÈ«ÃæµÄ¸ñʽÐ޸ĺ͵÷Õû
Á¢¼´×Éѯ°ïÖú×÷Õß½«¸å¼þÌá½»ÖÁÄ¿±êÆÚ¿¯Í¶¸åϵͳ£¬½µµÍÍ˸å»ò¾Ü¸åÂÊ
Á¢¼´×Éѯ°´ÕÕÄúÌṩµÄ¸å¼þÄÚÈÝ£¬Ö¸µ¼Íê³ÉͶ¸å¸½ÐÅ£¨cover letter£©
Á¢¼´×Éѯְ³ÆæäÕ¾ www.zhichengyz.com °æȨËùÓÐ ·Âð±Ø¾¿ ¼½ICP±¸16002873ºÅ-3