电网技术
電網技術
전망기술
POWER SYSTEM TECHNOLOGY
2014年
11期
3081-3090
,共10页
刘方%杨秀%时珊珊%张美霞%邓虹%郭鹏超
劉方%楊秀%時珊珊%張美霞%鄧虹%郭鵬超
류방%양수%시산산%장미하%산홍%곽붕초
微网%日前调度%实时调度%机会约束规划%混合储能系统%低通滤波%滑动平均滤波
微網%日前調度%實時調度%機會約束規劃%混閤儲能繫統%低通濾波%滑動平均濾波
미망%일전조도%실시조도%궤회약속규화%혼합저능계통%저통려파%활동평균려파
microgrid%day-ahead scheduling%real-time scheduling%chance-constrained programming%hybrid energy storage system%low-pass filtering%moving average filtering
针对混合储能系统中能量型和功率型储能的不同特点,建立了符合实际需求的微网经济优化模型。根据不同时间尺度提出了日前调度策略和实时调度策略。考虑到日前调度预测精度差的弊端,应用机会约束规划,以一定置信水平下满足备用需求作为可靠性约束,将最小运行成本作为优化目标,利用分时电价差异,优化协调各微源和储能系统出力。针对微网内风机、光伏和负荷等不确定性因素带来的功率波动,实时调度中应用1阶低通滤波算法平滑波动功率,并通过滑动平均滤波方法对滤波后的波动功率在混合储能中进行合理分配,使蓄电池储能承担更多低频功率波动,超级电容器承担高频波动。实时调度按照日前计划进行,实时监测各单元运行状态,并及时修正功率波动累积引起的日前计划偏差。
針對混閤儲能繫統中能量型和功率型儲能的不同特點,建立瞭符閤實際需求的微網經濟優化模型。根據不同時間呎度提齣瞭日前調度策略和實時調度策略。攷慮到日前調度預測精度差的弊耑,應用機會約束規劃,以一定置信水平下滿足備用需求作為可靠性約束,將最小運行成本作為優化目標,利用分時電價差異,優化協調各微源和儲能繫統齣力。針對微網內風機、光伏和負荷等不確定性因素帶來的功率波動,實時調度中應用1階低通濾波算法平滑波動功率,併通過滑動平均濾波方法對濾波後的波動功率在混閤儲能中進行閤理分配,使蓄電池儲能承擔更多低頻功率波動,超級電容器承擔高頻波動。實時調度按照日前計劃進行,實時鑑測各單元運行狀態,併及時脩正功率波動纍積引起的日前計劃偏差。
침대혼합저능계통중능량형화공솔형저능적불동특점,건립료부합실제수구적미망경제우화모형。근거불동시간척도제출료일전조도책략화실시조도책략。고필도일전조도예측정도차적폐단,응용궤회약속규화,이일정치신수평하만족비용수구작위가고성약속,장최소운행성본작위우화목표,이용분시전개차이,우화협조각미원화저능계통출력。침대미망내풍궤、광복화부하등불학정성인소대래적공솔파동,실시조도중응용1계저통려파산법평활파동공솔,병통과활동평균려파방법대려파후적파동공솔재혼합저능중진행합리분배,사축전지저능승담경다저빈공솔파동,초급전용기승담고빈파동。실시조도안조일전계화진행,실시감측각단원운행상태,병급시수정공솔파동루적인기적일전계화편차。
According to different features of energy-based energy storage and power-based energy storage in hybrid energy storage system an economic optimization model of microgrid, which corresponds with actual demand, is established. Based on different time scales the day-ahead and real-time scheduling strategies are proposed. Considering poor prediction accuracy of day-ahead scheduling, using chance constrained programming and taking satisfying the demand on reserve under a certain confidence level as reliability constraint and taking the minimum operation cost as the optimization objective and utilizing the difference of TOU price the optimal coordination of the output of micro energy sources and energy storage systems are performed. In allusion to power fluctuation brought about by such uncertain factors as wind turbine, photovoltaic generation and load, in real-time scheduling the first-order low-pass filtering algorithm is applied to smooth the power fluctuation, and the moving average filtering method is utilized to reasonably allocate the post-filtering fluctuating power among hybrid energy storage systems to make the battery energy storage system taking on more low-frequency power fluctuation and the supercapacitor taking on high-frequency power fluctuation. The real-time scheduling is carried out according to day-ahead plan to monitor operating states of all units in real time and to modify the deviation of day-ahead plan caused by the cumulation of power fluctuation in time.