題目：Short-term load forecasting based on multivariate adaptive step FOA optimized GRNN

時間：5月5號（星期六）下午 3：00-4：00

地點：X2511

摘要：Short-term load forecasting plays a significant role in power system. In this paper, we propose multivariate adaptive step fruit fly optimization algorithm (MAFOA) to optimize the smoothing parameter of generalized regression neural network (GRNN) in short-term power load forecasting. In addition, due to the great impact of some external factors including temperature, weather types and date types on short-term power load, we take these factors into account. Moreover, we propose an efficient interval partition technique to handle with the structured and unstructured data. The empirical results demonstrate that convergence speed and forecasting accuracy of the proposed model are superior to BP neural network, GRNN and fruit fly algorithm optimized GRNN.

作者簡介： 蔣鋒，博士，教授，文瀾學者。中南財經(jīng)政法大學統(tǒng)計與kaiyun開云官方網(wǎng)站數(shù)理與金融統(tǒng)計系主任，應用統(tǒng)計專業(yè)碩士大數(shù)據(jù)導師組組長。澳大利亞Monash University訪問學者。

主持國家自然科學基金面上項目一項；主持湖北省科研項目2項；主持完成國家自然科學基金青年項目一項；主持完成湖北省自然科學基金一項，主持完成中央高校科研基金項目兩項，主持完成中國博士后基金項目一項；曾獲湖北省優(yōu)秀博士學位論文獎；曾參與國家杰出青年基金項目、省杰出青年科學基金項目和多項國家自然科學基金面上項目。擔任國際期刊“Neural Comput Appl”、“IEEE Trans Neural Netw Learn Syst”、“Comput Appl Math.”等和國際會議的評審人，擔任國際學術(shù)會議“ICACI2018”、“ICICIP2016”等的PC Member。目前出版學術(shù)專著一部，已經(jīng)發(fā)表SCI或EI論文50余篇，其中30多篇論文被SCI收錄。現(xiàn)為TCCT隨機控制分委員會委員（SCSSC）和美國數(shù)學評論評論員。