透過您的圖書館登入
IP:54.224.90.25
  • 學位論文

類神經系統應用導電混凝土改善高壓輸電塔擊接地之配比最佳化

Optimal Design of Conductive Concrete to Improve the Anti-lightning Grounding of ElectricityTransmission Tower by using Artificial Neural Networks

指導教授 : 段永定
共同指導教授 : 葉怡成

摘要


造成輸電系統跳電故障的原因很多,由於台灣地區雷擊密度相當高,根據台灣電力公司統計,輸電系統由雷擊原因造成的機電事故已接近全部事故的50%,在整體上佔有相當的嚴重性。 近年來,國內許多地區連續發生多起因接地網未滿足要求而引起的設備損壞事故,同時雷擊是主要天然災害,因此良好的接地裝置一定是防雷擊的重要措施。在防雷接地裝置中,接地電阻阻值越小,則瞬間衝擊接地電壓就降越低,遭受雷擊的危險性就越低,因此足夠小的接地電阻值和安全可靠的防雷接地裝置是防雷的重要保證。 以往工程上主要採用的降低電阻方法為增加接地網的面積、引外接地、或化學降阻劑,但容易造成接地導體的腐蝕。大地是一不良導電物體,雖然大地之電阻率甚高,但電流一進入大地之後,因大地的廣大斷面積而使其電阻近似於零,本計畫提出高導電性混凝土基礎結構,成為大面積接地電阻,與接地網及接地棒並聯共同分擔雷擊電流,將雷擊電流導向地底以降低斷電、跳電而造成的損失。 類神經網路法有別於以往傳統方法,是一種平行分散式處理計算模式。其基本的運作原理乃以大量、簡單的處理單元,或稱神經細胞互相連接,藉由整體處理單元對外界輸入訊號的簡單運算來處理資訊,擁有類似於人腦的許多特性及優點。本計畫將類神經網路法配合交叉驗證法與訓練測試法運用高導電性混凝土的成本、電阻及抗壓的最佳化。

並列摘要


Due to high domestic lightning density, there are many reasons to cause electricity transmission system failure. According to Taipower (Taiwan Power Company, TPC, also known as Taipower) statistics, in the transmission system, the electrical accident came from lightning strikes were close to 50% of all accidents. In recent years, due to insufficient of grounding web, domestic equipment damage happened continuously. Lightning strikes are major natural disaster. Therefore, good grounding device should be an important measure in lightning protection. In the lightning protection and grounding device, the smaller the grounding resistivity is, the lower the instantaneous voltage drops. Therefore, the risk of being struck by lightning becomes less with smaller grounding resistance value and reliable anti-lightning grounding equipment. This study proposes the use of highly conductive concrete basic structure to be a large area of grounding resistance with grounding grid and grounding rods sharing lightning current in parallel,then to drive the lightning strike current underground and reduce the power loss caused by power failure. Previous works on the main drag reduction methods including increasing the area of the grounding network, citing external grounding and using chemical reducing agent were likely to cause corrosion of the grounding conductor. The earth is a poor conductive object due to the high resistivity of the earth. Once the electric current flows into the earth, the resistance of earth would be close to zero because of its large cross-section. Artificial Neural Networks method is different from the traditional one which is a kind of parallel distributed processing computing model. The basic principle of operation is based on a large but simple processing unit, or called Neuron connected to each other; by using the whole processing unit by the simple arithmetic of the external input signal to process information, which is similar to many features and benefits of human brains.This study proposes the use of the Artificial Neural Networks training with cross-validation method and Train-and-Test method used in optimizing the cost,resistance, and compressive of highly conductive concrete.

參考文獻


64. 陳建志,「實驗計畫法應用於太陽能光電製程改善:以線鋸切片機台製程最佳化為例」,碩士論文,國立成功大學,2011。
49. 林旺春,「建築結構強震紀錄之類神經網路參數識別」,碩士論文,中原大學土木工程研究所, 2006。
26. 王湘仁,高導電性混凝土之電磁及接地特性分析,中原大學電機工程系碩士論文,民國99年7月
7. 洪得峻,“架空輸電線路之防雷策略及其接地系統之雷擊特性研究”,中原大學電機工程系碩士論文,民國93年6月。
51. 黃廷堅,「以類神經網路預估建築工程造價之研究」,碩士論文,國立臺灣大學土木工程研究所,2010。

延伸閱讀