基于局部模糊推理的區域電力系統超短期負荷預算方法
武曉朦1,宋晨曦1,2
(1 西安石油大學 電子工程學院,陜西 西安 710005;2 陜西省油氣井測控技術重點實驗室,陜西 西安 710065)
摘 要:為了解決傳統超短期負荷預算方法誤差較大的問題,提出一種基于局部模糊推理的區域電力系統超短期負荷預算方法。使用局部模糊推理定義利用的電力系統負荷歷史數據,使用序偶法處理負荷數據的模糊集,計算數據的模糊度,得出時點負荷的偏離程度。規定特征時間尺度,依據特征時間尺度的數量關系,預測超短期隨機電荷分量,計算電荷分量的均值,形成新的負荷序列,平滑處理得到最終的超短期負荷預算表達式,完成對區域電力系統超短期負荷的預算。實驗表明:與傳統基于BP 神經網絡的短期負荷預算方法相比,基于局部模糊推理的區域電力系統超短期負荷預算方法誤差率只有0.07%,誤差更小,更適合預算區域電力系統的超短期負荷。
關鍵詞:局部模糊推理;區域電力系統;超短期負荷;誤差率
中圖分類號:TM715 文獻標識碼:A 文章編號:1007-3175(2020)03-0032-04
Ultra-Short Term Load Budget Method for Regional Power System Based on Local Fuzzy Reasoning
WU Xiao-meng1, SONG Chen-xi1,2
(1 School of Electronic Engineering, Xi’an Shiyou University, Xi’an 710005, China;
2 Shaanxi Key Laboratory of Oil and Gas Well Measurement and Control Technology, Xi’an 710065, China)
Abstract: In order to solve the problem of large errors in traditional ultra-short-term load budget method, a regional power system ultrashort-term load budget method based on local fuzzy reasoning is proposed. Local fuzzy reasoning is used to define the historical load data of the power system, and the fuzzy set of load data is processed using the method of sequence coupling. The sequential couple method is used to process the fuzzy sets of load data, calculate the fuzziness of the data, get the deviation degree of the load at the time point. By specifying the characteristic time scale and predicting the ultra-short-term random charge component according to the quantity relationship of the characteristic time scale, the average value of the charge component is calculated to form a new load sequence, and the smoothing process is used to obtain the final ultra-short-term load budget expression, thereby completing the regional budget for ultra short-term load of the power system. Experiments show that compared with the traditional short-term load budgeting method based on BP neural network, the ultra-short-term load budgeting method of regional power systems based on local fuzzy reasoning has an error rate of only 0.07%, and the error is smaller, which is more suitable for the ultra-short-term load of the budgeted regional power system.
Key words: local blur reasoning; regional power system; ultra-short term load; error rate
參考文獻
[1] 張宇帆,艾芊,林琳,等. 基于深度長短時記憶網絡的區域級超短期負荷預測方法[J]. 電網技術,2019,43(6):1884-1891.
[2] 曲正偉,張坤,王云靜,等. 基于PSO-OMP優化的WD-ASD超短期負荷預測[J]. 電工電能新技術,2017,36(12):39-45.
[3] 張麗,張濤,王福忠,等. 基于柔性負荷響應特性的超短期預測方法[J]. 電力系統保護與控制,2019,47(9):27-34.
[4] 許剛, 吳舜裕. 區域負荷趨勢特征分析與金字塔模型超短期預測方法[J]. 計算機工程,2018,44(2):287-293.
[5] 熊軍華,牛珂,張春歌,等. 基于小波變異果蠅優化支持向量機短期負荷預測方法研究[J]. 電力系統保護與控制,2017,45(13):71-77.
[6] 孔祥玉,鄭鋒,鄂志君,等. 基于深度信念網絡的短期負荷預測方法[J]. 電力系統自動化,2018,42(5):133-139.
[7] 于惠鳴,張智晟,龔文杰,等. 基于深度遞歸神經網絡的電力系統短期負荷預測模型[J]. 電力系統及其自動化學報,2019,31(1):112-116.
[8] 謝真楨,楊秀,張鵬,等. 基于信息論與混合聚類分析的短期負荷預測方法研究[J]. 電測與儀表,2017,54(19):67-72.
[9] 孫海蓉,謝碧霞,田瑤,等. 基于數據的SecRPSOSVM短期電力負荷預測[J]. 系統仿真學報,2017,29(8):1829-1836.
[10] 朱祥和. 基于小波變換和改進螢火蟲算法優化極限學習機的短期負荷預測[J]. 數學的實踐與認識,2017(3):136-144.
[11] 王為國,竇震海,申晉,等. 基于逆推理論改進模糊均生函數的短期負荷預測[J]. 水電能源科學,2017(12):208-211.
[12] 史佳琪,譚濤,郭經,等. 基于深度結構多任務學習的園區型綜合能源系統多元負荷預測[J]. 電網技術,2018,42(3):698-706.
[13] 林芳, 林焱, 呂憲龍, 等. 基于均衡KNN算法的電力負荷短期并行預測[J]. 中國電力,2018,51(10):88-94.
[14] 劉明順,龍志君,趙立進,等. 基于模糊推理的電網風險評估及線路重要度辨識[J]. 電力科學與技術學報,2017,32(3):131-138.