【摘要】：隨著我國水電在電力系統(tǒng)中比重的不斷加大,以及工農(nóng)業(yè)用水、城鄉(xiāng)居民用水需求的不斷增加,水庫在自身防洪的前提下,面臨著供電和供水的雙重壓力。加之近年來干旱等極端天氣頻發(fā),水庫的運行調(diào)度面臨巨大挑戰(zhàn)。 水庫優(yōu)化調(diào)度通過運用一定的優(yōu)化方法和技術,尋求水庫及其水電站在不同需求下的最優(yōu)運行調(diào)度方式和最優(yōu)控制策略�，F(xiàn)行水庫調(diào)度方式主要考慮其生產(chǎn)電力的可靠性和經(jīng)濟性,以期最大化利用水庫所控制的水能資源。對具有發(fā)電和供水等多重利用要求的水庫,統(tǒng)籌考慮發(fā)電可靠性和供水可靠性有著重要的現(xiàn)實意義。如何通過制定合理調(diào)度規(guī)則,實現(xiàn)水庫水電站運行期內(nèi)復雜的目標需求是關鍵問題之一。 本文在總結水庫優(yōu)化調(diào)度相關研究的基礎上,針對徑流隨機性和水庫功能轉變后的優(yōu)化調(diào)度等問題,以廣東省境內(nèi)的新豐江水庫為實際工程背景,首先研究考慮棄水風險的水庫水位控制問題,而后研究發(fā)電供水雙重利用水庫的優(yōu)化調(diào)度模型,最后,基于遺傳算法,對發(fā)電供水優(yōu)化調(diào)度模型生成的的調(diào)度規(guī)則進行進一步優(yōu)化。本文研究的主要內(nèi)容具體如下： (1)針對實際調(diào)度中的最優(yōu)水位控制問題,提出一種以棄水概率為約束條件,以多年平均發(fā)電量最大為目標的最優(yōu)軌跡優(yōu)化模型,并通過模擬優(yōu)化方法求解。以新豐江水庫為工程背景的實際應用表明,該方法在尋求水庫最優(yōu)水位控制線時,考慮了實際調(diào)度中棄水風險控制和提高電量的協(xié)調(diào)問題,有利于實際應用。 (2)針對具有發(fā)電、城市供水雙重功能的水庫優(yōu)化調(diào)度問題,提出具有發(fā)電保證率、供水保證率和最小供水流量三重約束條件的隨機動態(tài)規(guī)劃模型。以拉格朗日松弛法處理兩個保證率約束,以懲罰函數(shù)法處理最小供水流量約束,通過模擬調(diào)度結果指導松弛因子和懲罰系數(shù)的更新。以新豐江水庫為工程背景的實例研究表明,提出的方法能夠在有效協(xié)調(diào)發(fā)電供水功能的基礎上,緩解枯水時段供水流量的極端破壞,可應用于具有城市供水任務的水庫的優(yōu)化調(diào)度中。 (3)基于一種改進的遺傳算法,以多年平均發(fā)電量最大為目標,對上一章的發(fā)電供水調(diào)度規(guī)則進行進一步優(yōu)化。編碼采用十進制浮點數(shù),初始種群生成采用一種特定的策略,適應度函數(shù)采用變量懲罰函數(shù)法,遺傳操作采用錦標賽選擇,兩點交叉、非均勻變異。以新豐江水庫為工程背景的計算結果表明,這種方法能夠在上一章結果的基礎上,得到發(fā)電量指標更優(yōu)的調(diào)度規(guī)則。 最后,對本文研究的內(nèi)容進行總結,同時對研究存在的問題以及前景進行展望。
[Abstract]:With the increasing proportion of hydropower in the electric power system and the water use in industry and agriculture, and the increasing demand of urban and rural residents, reservoirs are faced with the double pressure of power supply and water supply under the premise of flood control. Coupled with the frequent occurrence of drought and other extreme weather in recent years, the operation and operation of reservoirs are facing great challenges. By using certain optimization methods and techniques, the optimal operation mode and optimal control strategy of the reservoir and its hydropower station under different requirements are sought. The current reservoir operation mode mainly considers the reliability and economy of producing electric power in order to maximize the utilization of the hydropower resources controlled by the reservoir. It is of great practical significance to consider power generation reliability and water supply reliability as a whole for reservoirs with multiple utilization requirements such as power generation and water supply. It is one of the key problems how to realize the complex target demand in the operation period of reservoir hydropower station by making reasonable operation rules. On the basis of summarizing the relevant research of reservoir optimal operation, aiming at the randomness of runoff and the optimal operation after reservoir function transformation, this paper takes Xinfengjiang Reservoir in Guangdong Province as the actual engineering background. Firstly, the problem of reservoir water level control considering abandoned water risk is studied, and then the optimal operation model of dual-use reservoir is studied. Finally, based on genetic algorithm, The regulation rules generated by the optimal dispatching model for power generation water supply are further optimized. The main contents of this paper are as follows: (1) aiming at the problem of optimal water level control in actual dispatching, an optimal trajectory optimization model is proposed, which is constrained by the probability of abandoning water and aims at the maximum of multi-year average power generation. The simulation optimization method is used to solve the problem. The practical application of Xinfengjiang Reservoir shows that this method takes into account the risk control of abandoned water and the coordination of increasing electricity quantity in practical operation when searching for the optimal water level control line of the reservoir, which is beneficial to the practical application. (2) for the optimal operation of reservoir with double functions of power generation and urban water supply, a stochastic dynamic programming model with triple constraints of power generation guarantee rate, water supply guarantee rate and minimum water supply discharge is proposed. The Lagrangian relaxation method is used to deal with the two guaranteed rate constraints, and the penalty function method is used to deal with the minimum water supply discharge constraints. The relaxation factor and penalty coefficient are updated by simulating the operation results. The case study of Xinfengjiang Reservoir shows that the proposed method can alleviate the extreme damage of water flow in low water period on the basis of coordinating the water supply function of power generation effectively. It can be applied to the optimal operation of reservoirs with urban water supply tasks. (3) based on an improved genetic algorithm, with the goal of maximum generation capacity for many years, the regulation of generation water supply scheduling is further optimized in the last chapter. The coding adopts decimal floating point number, the initial population is generated by a special strategy, the adaptation function is based on variable penalty function, the genetic operation is selected by tournament, the two points are crossed, and the variation is non-uniform. Based on the engineering background of Xinfengjiang Reservoir, the results show that this method can obtain better dispatching rules of power generation index on the basis of the results of the previous chapter. Finally, the research contents are summarized, and the existing problems and prospects are prospected.
【學位授予單位】：大連理工大學
【學位級別】：碩士
【學位授予年份】：2014
【分類號】：TV737

【參考文獻】

相關期刊論文 前10條

1 王金文,石琦,伍永剛,樂玉華,張世欽;水電系統(tǒng)長期發(fā)電優(yōu)化調(diào)度模型及其求解[J];電力系統(tǒng)自動化;2002年24期

2 紀昌明;喻杉;周婷;楊子俊;劉方;;蟻群算法在水電站調(diào)度函數(shù)優(yōu)化中的應用[J];電力系統(tǒng)自動化;2011年20期

3 湯斌,劉健民,仲偉俊;水電站水庫優(yōu)化調(diào)度的隨機動態(tài)規(guī)劃方法[J];東南大學學報;1998年02期

4 曾勇紅,姜鐵兵,張勇傳;三峽梯級水電站蓄能最大長期優(yōu)化調(diào)度模型及分解算法[J];電網(wǎng)技術;2004年10期

5 馮U，

本文編號：2455519