基因型模糊推論機制於能源管理決策支援系統之研究

傳送到

題名:
基因型模糊推論機制於能源管理決策支援系統之研究
著者: 蔡育娣; Tsai, Yu-ti
李健興; 劉哲宏; Chang-Shing Lee; Che-Hung Liu; 科技管理碩士班
主題: 模糊推論; 基因學習; 決策支援系統; 知識本體; 能源管理; Fuzzy Inference; Genetic Learning; Decision Support System; Ontology; Energy Management
描述: 能源的運用有正反兩面影響，雖然對社會的進步與科技的演進有莫大的貢獻，但能源的開發及使用對環境卻造成重大影響，例如：空氣污染及因溫室氣體排放造成全球暖化及臭氧層破壞等。除了節約能源及提高能源效率以直接改進外，更要積極尋找替代能源的可能性，以促使台灣逐步往「綠色矽島」方向邁進。本論文建置一套適用於已裝置太陽能光電系統用電戶的太陽能供需分析決策支援系統，此系統根據太陽能動態評估本體論、模糊理論及決策支援系統等所建置而成，接著基於本論文研發之太陽光電供需分析決策支援系統及專家知識，結合模糊邏輯理論及基因學習演算法，進行推論家電耗電程度及太陽光電系統發電程度，再由家電耗電程度與太陽光電系統發電程度決定用電戶是否售電或購電。若為售電，用電戶可選擇全額躉售或餘額躉售，相反的，若為購電則進行用電戶購買能源類型可能性之推論，讓用電戶進行決策購買能源類型。藉由FML System進行實驗分析後可知：(1)第二型模糊推論系統於基因學習前產生的均方差、正確性及適應值皆優於第一型模糊推論系統於基因學習前的結果；(2)無論是第一型模糊推論系統或第二型模糊推論系統於基因學習後的結果，皆優於基因學習前結果及(3)本論文所提出方法，應用於能源管理決策支援是可行的。未來，將收集不同區域資料，並考慮較多影響因素，以建置一個智慧型能源管理決策支援系統。
Energy use has its positive and negative influence. Although energy has a great contribution to the social progress and technical evolution, the development and usage of energy also make a great influence on the environment, such as air pollution and greenhouse gas emissions to cause global warming and ozone depletion. In addition to conserving energy and increasing energy efficiency, it is necessary to positively look for the possibilities of alternative energy to promote Taiwan to move gradually toward Green Silicon Island. This thesis constructs one solar energy supply and demand analysis (SESDA) for PV-installed household according to the adaptive solar energy evaluation ontology, fuzzy theory, and decision support system. Based on SESDA and the knowledge of the domain experts, this thesis combines fuzzy logic theory and genetic learning algorithm to infer the appliances power consumption level (APCL) and power generation level (PGL) and then to make a decision for the household to sell electricity or buy electricity. If it is to sell electricity, the household can choose full wholesale or balance wholesale. On the contrary, this proposed approach infers the possibility of purchasing the type of energy. In this thesis, the fuzzy markup language (FML) system, developed by OASE Lab., is used to analyze the experiments. The experimental results show that (1) before-learning mean square error, accuracy, and fitness from type-2 fuzzy logic system (T2FLS) perform better than the ones from type-1 fuzzy logic system (T1FLS), (2) no matter T1FLS or T2FLS, after-learning results are better than before-learning ones, and (3) the proposed method is feasible for applying to decision support of the energy management. In the future, more data from different areas in Taiwan will be collected and more factors will be considered to construct an intelligent energy management decision support system.
碩士
建立日期: 2013
格式: 121 bytes
text/html
語言: 中文
識別號: http://nutnr.lib.nutn.edu.tw/handle/987654321/307
資源來源: NUTN IR

連結

連回原始介面

基因型模糊推論機制於能源管理決策支援系統之研究

蔡育娣; Tsai, Yu-ti 李健興;劉哲宏; Chang-Shing Lee;Che-Hung Liu; 科技管理碩士班 2013

正在檢索遠程資料庫，請稍等