本論文提出應用基因演算法於電力系統有效─無效功率之最佳化排程。過去研究常將一天24小時的有效、無效功率排程之最佳化(即機組調派、虛功率與電壓控制)分開探討,本文將以智慧型控制方法來整合兩者之最佳化,期望達到更完整之排程。在滿足所有運轉的限制式情況下,使火力發電機組之總發電成本、傳輸線總實功率損失最小化。近年來風力發電逐漸受到重視,但因其具有不確定之特性,故本文利用機率之抽樣方法取出10種24小時之風力發電資料,使火力發電機組在考慮此風力情境下達到近似最佳化的24小時排程。 本文透過分析模擬IEEE 30-Bus系統,以輪流方式對有效─無效功率進行最佳化,經由測試後證實所提出方法之可行性,並可求得近似最低總發電成本、降低輸電線有效功率損失以及使負載匯流排電壓在合理範圍內,期望能幫助調度人員作更經濟、安全的排程。
The thesis indicates the application of genetic algorithms in the optimal dispatch of active-reactive power of electric system. Past studies usually explore the optimization of one-day active-reactive power dispatch separately(unit commitment、reactive Power and voltage control). The thesis integrates the optimization of the two with intelligent control method, hopefully making more complete dispatch as well as minimizing the total cost of thermal power turbines and total MW loss of transmission lines under all the operational restrictions. Though wind power was taken more and more seriously recently; due to its uncertainty, ten different “twenty four hour- wind power data” was randomly picked in this thesis to decide how thermal power turbines achieve the quasi optimal dispatch considering different wind levels. Through the analysis and simulation of IEEE 30-Bus system, we prove that it is feasible to acquire optimization by taking turns implementing active-reactive power experiments. This system also contributes to getting quasi minimal total electricity generation cost, decreasing the loss of active power of transmission lines and limiting voltage of load bus to a reasonable range, hopefully assisting dispatchers in making more economical and safer dispatch.
為了持續優化網站功能與使用者體驗,本網站將Cookies分析技術用於網站營運、分析和個人化服務之目的。
若您繼續瀏覽本網站,即表示您同意本網站使用Cookies。