电测与仪表
電測與儀錶
전측여의표
ELECTRICAL MEASUREMENT & INSTRUMENTATION
2014年
15期
1-7
,共7页
郭晓利%张玉萍%曲朝阳%任有学%辛鹏
郭曉利%張玉萍%麯朝暘%任有學%辛鵬
곽효리%장옥평%곡조양%임유학%신붕
风电功率短期预测%FKNN 算法%相似数据%K - means 聚类算法
風電功率短期預測%FKNN 算法%相似數據%K - means 聚類算法
풍전공솔단기예측%FKNN 산법%상사수거%K - means 취류산법
short - term wind power prediction%FKNN algorithm%similar data%K - means clustering algorithm
风电场输出功率预测精度的提高能够极大的减轻风力发电对电网的冲击,提高风电并网的安全性和可靠性。针对 KNN(K - Nearest Neighbor algorithm)算法存在的不足进行改进,提出了 FKNN(Fast K - Nearest Neigh-bor algorithm)算法并将其应用到风电短期功率预测当中。首先,FKNN 算法基于相似数据原理,针对每个预测样本,只需遍历一次训练样本集,得出 K 值最大时的相似历史样本优先级队列。然后,通过逐渐缩减优先级队列的长度,产生其他 K 值对应的相似样本优先级队列。其次,从产生的优先级队列中获取多数类样本,并应用其输出功率的平均值对预测样本的输出功率进行预测。最后,通过对吉林省某风电场的大量历史数据进行预测分析,充分证明该算法的简单性和实用性。
風電場輸齣功率預測精度的提高能夠極大的減輕風力髮電對電網的遲擊,提高風電併網的安全性和可靠性。針對 KNN(K - Nearest Neighbor algorithm)算法存在的不足進行改進,提齣瞭 FKNN(Fast K - Nearest Neigh-bor algorithm)算法併將其應用到風電短期功率預測噹中。首先,FKNN 算法基于相似數據原理,針對每箇預測樣本,隻需遍歷一次訓練樣本集,得齣 K 值最大時的相似歷史樣本優先級隊列。然後,通過逐漸縮減優先級隊列的長度,產生其他 K 值對應的相似樣本優先級隊列。其次,從產生的優先級隊列中穫取多數類樣本,併應用其輸齣功率的平均值對預測樣本的輸齣功率進行預測。最後,通過對吉林省某風電場的大量歷史數據進行預測分析,充分證明該算法的簡單性和實用性。
풍전장수출공솔예측정도적제고능구겁대적감경풍력발전대전망적충격,제고풍전병망적안전성화가고성。침대 KNN(K - Nearest Neighbor algorithm)산법존재적불족진행개진,제출료 FKNN(Fast K - Nearest Neigh-bor algorithm)산법병장기응용도풍전단기공솔예측당중。수선,FKNN 산법기우상사수거원리,침대매개예측양본,지수편력일차훈련양본집,득출 K 치최대시적상사역사양본우선급대렬。연후,통과축점축감우선급대렬적장도,산생기타 K 치대응적상사양본우선급대렬。기차,종산생적우선급대렬중획취다수류양본,병응용기수출공솔적평균치대예측양본적수출공솔진행예측。최후,통과대길림성모풍전장적대량역사수거진행예측분석,충분증명해산법적간단성화실용성。
The improvement of wind farm’s output power prediction accuracy can greatly reduce the impact of wind power on the grid and improve the security and reliability of wind power integration. In this paper,the FKNN(Fast K- Nearest Neighbor algorithm)algorithm is proposed to improve the shortcomings of KNN(K - Nearest Neighbor algo-rithm)algorithm and is used for short - term wind power prediction. First,for each prediction sample,by using FKNN algorithm,which is based on the principle of similarity data,you can obtain the maximum priority queue of similar sample through traversing the set of training sample only one time. Then,gradually reduce the length of the priority queue to produce different size priority sub - queues of similar sample in which the majority class samples can be obtained and its average is used to predict the output power of prediction sample. Finally,the algorithm’s simplici-ty and practicality was fully proved through the prediction of a large amount historical data of a wind farm in Jilin Prov-ince.