201101209 六、發明說明： 【發明所屬之技術領域】 本發明係關於-種灰色人機介面預測系統，特別是指 -種便於使时執行函數求得最佳結果的灰色人機介面曰 預測系統。 【先前技術】 在知識爆炸的今天，如何準確有效率地進行資訊管 理、分類、比對及整合，並且依照大量的數據求取出最符 合實際的答案，已經成為一種現代學者、生產業者都十分 重視的技術。 即使在科學技術高度發展的今天，人類面臨的眾多問 題也還包含有大量未知的、不確定的訊息及不完全的自然 因素和社會因素；這些因素中有些目前還無法被人完全掌 握，從而很難甚至還不能被人控制，因此，人們不得不在 『部分已知、部分未知』、『部分確定、部分不確定』及 『部分完全、部分不完全』等狀況下，儘量充分應用可得 之各種自然訊息和社會訊息進行預測與決策，也就是在 『灰色（Grey)』、『朦朧（Hazy)』及『模糊（Fuzzy)』的 曩士兄下，依據『不完全（not enough)』，『不確定（n〇t certain)』及『不明確（⑽土 ciear)』的訊息確定自己的 目標與行為。 現有預測方法將近有三百多種，通常採用的有迴歸 201101209 (regressi on)分析法、德爾菲（De丨ph i)法大意、統計趨勢 預測、馬爾可夫（Markov)預測法及模型法等。而灰色預測 方法為模型法的的一種，可以分成下列幾種： 1·數列預測：這是對系統行為特徵值大小的發展變化 進行預測，稱為系統行為數據列的變化預測，簡稱數列預 測，譬如商品銷售發展變化的預測、人口的預測、刑事案 件年發案率的預測、銀行存款的預測、貨運量的預測、全 〇 國大學畢業生每年人數變化的制及外諸發展變化預 測。 2.災變關：這是對系統行為特徵值超出某個閨值 (threshGld)的預測，顯示異常值將在何時再出現的預測 稱為災變預測。譬如年平均降雨量小於某個閣值是旱災。 環境中某種物質含量超出某個閑值是汗染。人體中某個參 數（如體溫、血壓、血中成分等）超出一定範圍，就發生病 3. 季節災變預測：例如春雨是在春天出現。而早霜在 秋末冬初的九、十及十一月出現。 4. 拓樸賴：這是對-段_岐為賴數據波形的 5.系統綜合預測··„-系統各_素之間的動態關 係找出，並建立系統動態框架圖。 201101209 *二十世、紀末迄今，經由國内外廣大灰色系統理論的研 九者之努力’使仔此一理論體系愈加地完善，並已成功地 應用於各種領域之中；而『灰色系統理論』在台灣也被應 用於資訊、電子、電機、機械、自動化、航太、土木、水 ^建築工業工知、工業教育、商業、國際貿易、財務 管理、交通運輸、企業管理、體育及軍事評估等方面均有 相當多的相關研究成果報告，並且快速的發展與成長中。 【發明内容】 本發明係一種灰色人機介面預測系統，其目的在於便 於使用者執行函數求得最佳結果。 _為達上述目的，本發明灰色人機介面預測系統包含一 丄由*構於-主機之Matlab軟體製作而成的灰色⑽模 型生成圖形使用者界面，該灰色_型生成圖形使用者界 面包3 用於帶入原始資料的資料輸入單元；一連結一 灰色GM拉型生成運算單元的資料預測輸出單元；一執行單 一 μ用於將D亥資料輸入單元的資料連結該灰色⑽模型生成 1算單7L進仃運算且帶人該資料制輸出單元輸出預測 L果’且该執行單元連結一比較單元，該比較單元對應該 資料預/1」輸出單元的輸出預測結果形成—圖型化資料區。 _為達上述目的，本發明灰色人機介面預測系統包含-經由-架構於一主機之Matlab軟體製作而成的灰色 GM(l’l)基本模型圖形使用者界面，該灰色即，^基本模 201101209 型圖形使用者界面包含：一用於帶入原始資料的資料輸入 單元；一連結一灰色GM(1，1)基本模型運算單元的資料預 測輸出單元；一執行單元，用於將該資料輸入單元的資料 連結該灰色GM(1，1)基本模型運算單元進行運算且帶入該 資料預測輸出單元輸出預測結果，且該執行單元連結一比 較單元，該比較單元對應該資料預測輸出單元的輸出預測 結果形成至少一圖型化資料區。 為達上述目的，本發明灰色人機介面預測系統包含一 經由一架構於一主機之Mat 1 ab軟體製作而成的灰色 GM(1，1)費爾哈斯特模型圖形使用者界面，該灰色GM(1，1) 費爾哈斯特模型圖形使用者界面：一用於帶入原始資料的 資料輸入單元；一連結一灰色GM(1，1)費爾哈斯特模型運 算單元的資料預測輸出單元；一執行單元，用於將該資料 輸入單元的資料連結該灰色GM(1，1)費爾哈斯特模型運算 單元進行運算且帶入該資料預測輸出單元輸出預測結 果，且該執行單元連結一比較單元，該比較單元38對應該 資料預測輸出單元32的輸出預測結果形成至少一圖型化 資料區。 藉由前述進一步分析將可獲得下述功效：本發明灰色 模型生成圖形使用者界面、灰色GM(1，1)基本模型圖形使 用者界面及灰色GM(1，1)費爾哈斯特模型圖形使用者界面 201101209 分別為理論、公式及方法化作錢形式之執行檔，方便使 用者清楚得知執行之聽，切最佳絲，再做進一步處 理時’均可以使資料傳輸的捨棄誤差達到最小，而將最正 確的結果加以呈現。 【實施方式】 有關本發明所採用之技術、手段及其他之功效，兹舉 車乂佳實施例’ its己合圖示詳細說明如下，相信本發明上 述之目的、特徵及其他優點，當可由之得1人而具體瞭 解； 本發明實施例請參閱第丄圖至第7圖所示： 請參閱第1圖所示，本發明灰色人機介面預測系統包 含-灰色GM模型生成圖形使用者界面（G卿^ ace) 10 Mat lab軟體A、一作業系統^、一主機c、 一輪入器D及一輪出器E ; 。亥灰色GM模型生成圖形使用者界面1〇(請參閱第i、2 及3圖所不）經由MatUb軟體A將理論、公式及方法化為函 =开^式1作成執行檔，製作方式如下：該Matlab軟體A係 个構於作業系_上’前述作業系統b係透過主機c運作， 透過4主機C電性連接輸入器D(鍵盤或滑鼠）與輸出器 E(勞幕）用於操作架構於作業系統B的Matlab軟體A，進而 作》玄灰色GM模型生成圖形使用者界面1〇，該輸入器D與 ~輪出器E係可整合成—體的輸人暨輸出器（觸控榮幕）， 201101209 一旦灰色GM模型生成圖形使用者界面10以執行檔呈現，使 用者僅要將該灰色GM模型生成圖形使用者界面1〇存放於 作業系統B即可執行達成分析之目的。 該灰色GM模型生成圖形使用者界面1〇(請參閱第！、2 及3圖所示）包含一資料輸入單元u、一資料預測輸出單元 12及一灰色GM模型生成運算單元13 ; 該資料輸入單元11用於帶入原始資料，前述帶入資料 〇 可為EXCEL檔案’使用者可先於EXcEL檔案建構參數（參 數為符號、文字或圖形），前提是帶入資料必須與該資料 輸入單元11相容。 該資料輸入單元11透過該灰色GM模型生成運算單元 13進行運算且傳送至該資料預測輸出單元12輸出預測結 果。 5亥灰色GM模型生成圖形使用者界面丨〇更包含一開啟 ❹ 單元14、一執行單元丨5、一儲存單元16及一清除單元17; 該開啟單元14用於指定資料位址載人f料至該資料 輸入單元11。 該執行單元15用於將該資料輸入單元丨1的資料連結 ’火色GM模型生成運异單元13進行運算且帶入該資料預 測輸出單το 12輸出預測結果，且該執行單元14另連結一比 較單疋18’該比較單元18對應該資料預測輸出單元的輸 201101209 出預測結果形成一顯示原始數據及累加生成數據之比較 曲線的圖型化資料區181。 該儲存單元16連結一指定單元161，該指定單元161 用於定義資料儲存類型、資料儲存位址及資料儲存名稱。 5亥清除單元17連結一確認單元171，該確認單元m 的執行將清除前述各單元的資料。 請參閱第1、4及5圖所示，該灰色GM(l，1)基本模型圖 形使用者界面20製作方式與該灰色GM模型生成圖形使用 者界面10相同，該灰色GM(M)基本模型圖形使用者界面 20包含一資料輸入單元21、一資料預測輸出單元以及一灰 色GM(1，1)基本模型運算單元23 ; 該資料輸入單元21用於帶入原始資料，前述帶入的資 料可為EXCEL檔案，使用者可先利用EXCEL檔案建構參 數（參數為符號、文字或圖形）’但是帶入的資料必須與該 資料輸入單元21相容。 該資料輸入單元21透過該灰色GM(1，1)基本模型運算 單元23進行運算且傳送至該資料預測輸出單元22輸出預 測結果，該資料預測輸出單元2 2依照預測結果分為一累加 生成區221、一平滑常數區222、一預測區223及一誤差區 224。 該灰色GM(1，1)基本模型圖形使用者界面2〇更包含一 10 201101209 開啟單元24、一執行單元25、一儲存單元26及一清除單元 27 ； 該開啟單元24用於指定資料位址載入資料至該資料 輸入單元21。 該執行單元25用於將該資料輸入單元21的資料連結 該灰色GM(1，1)基本模型運算單元23進行運算且帶入該資 料預測輸出單元22輸出預測結果，且該執行單元25另連結 一比較單元2 8，該比較單元2 8對應該資料預測輸出單元2 2 的輸出預測結果形成一顯示原始數據及累加生成數據之 比較曲線的第一圖型化資料區281、一顯示評定誤差曲線 的第二圖型化資料區282及一顯示預測數據及原始數據之 比較曲線的第三圖型化資料區283。 該儲存單元26連結一指定單元，該指定單元用於定義 資料儲存類型、資料儲存位址及資料儲存名稱。 該清除單元27連結一確認單元，該確認單元的執行將 清除前述各單元的資料。 請參閱第1、6及7圖所示，該灰色GM(1，1)費爾哈斯特 模型圖形使用者界面30製作方式與該灰色GM模型生成圖 形使用者界面10相同，該灰色GM(1,1)費爾哈斯特模型圖 形使用者界面3 0包含一資料輸入單元31、一資料預測輸出 單元32及一灰色GM(1, 1)費爾哈斯特模型運算單元33 ; 11 201101209 該資料輸入單元31用於帶入原始資料，前述帶入的資 料可為EXCEL檔案，使用者可先利用EXCEL檔案建構參 數（參數為符號、文字或圖形），,但是帶入的資料必須與該 資料輸入單元31相容。 該資料輸入單元31透過該灰色GM(1, 1)費爾哈斯特模 型運算單元33進行運算且傳送至該資料預測輸出單元32 輸出預測結果，該資料預測輸出單元3 2依照預測結果分為 一第一參數區321、一第二參數區322及一結果區323。 該灰色GM(1，1)費爾哈斯特模型圖形使用者界面加更 包含-開啟單元34、_執行單元35、—儲存單元36及一清 除單元37; 旧 。亥開啟單το34用於指定資料位址載入資料至該資料 輸入單元31。 ' ▲為執订單tl35用於將該資料輸人單元31的資料連結 錢色GM(1，1)費爾哈斯龍㈣算單元⑽進行運算且帶 入及資料制輸^單元32輸出預縣果，且該執行單元％ Ϊ連較單元38,該比較單元38對應該資料預測輸出 、4預測結果形成—顯示原始數據及累加 數據之比較曲線的筮—jg^丨_ 生成 、、 51化資料區381及一顯示原始數 據及費爾哈斯特餐輔夕丄h 382。 特數據之比較曲線的第二圖型化資料區 12 201101209 5亥儲存單元36連結一指定單元，該指定單元用於定義 資料儲存類型、資料儲存位址及資料儲存名稱。 及h除早元3 7連結一確認早元’該確認單元的執行將 清除前述各單元的資料。 【圖式簡單說明】 第1圖本發明之架構示意圖。 第2圖本發明灰色GM模型生成圖形使用者界面之系統示意 圖。 第3圖本發明灰色GM模型生成圖形使用者界面之局部示音 圖。 第4圖本發明灰色gm(1，1)基本模型圖形使用者界面之系 統示意圖。 第5圖本發明灰色GM(1，1)基本模型圖形使用者界面之局 部示意圖。 第6圖本發明灰色WU)費爾哈斯特模型圖形使用者界 面之系統示意圖。 第7圖本發明灰色GM(U)費爾哈斯特模型圖形使用者界 面之局部示意圖。 【主要元件符號說明】201101209 VI. Description of the Invention: [Technical Field] The present invention relates to a gray human-machine interface prediction system, and more particularly to a gray human-machine interface prediction system that facilitates the time-performance function to obtain the best result. [Prior Art] In today's knowledge explosion, how to accurately and efficiently conduct information management, classification, comparison and integration, and to extract the most realistic answers according to a large amount of data has become a modern scholar and producer. Technology. Even today, with the rapid development of science and technology, many problems faced by mankind also contain a large number of unknown, uncertain messages and incomplete natural and social factors; some of these factors are currently not fully grasped, so It is difficult or even impossible to be controlled by people. Therefore, people have to apply as much as possible under the conditions of "partially known, partially unknown", "partially determined, partially uncertain" and "partially complete, partially incomplete". Natural information and social information for prediction and decision-making, that is, under the gentlemen of "Grey", "Hazy" and "Fuzzy", according to "not enough (not enough)" Uncertain (n〇t certain) and "unclear ((10) soil ciear" messages determine their goals and behaviors. There are nearly 300 kinds of existing prediction methods, and the regression method is usually used in 201101209 (regressi on) analysis method, Delphi (De丨ph i) method, statistical trend prediction, Markov prediction method and model method, etc. . The gray prediction method is a kind of model method, which can be divided into the following types: 1. Sequence prediction: This is the prediction of the development and change of the system behavior characteristic value, which is called the change prediction of the system behavior data column, referred to as the series prediction. For example, predictions of changes in the development of merchandise sales, population projections, predictions of the annual incidence of criminal cases, predictions of bank deposits, forecasting of freight volume, changes in the number of graduates of all universities in the country, and predictions of changes in development. 2. Cataclysm Off: This is a prediction that the system behavior characteristic value exceeds a certain threshold (threshGld), and the prediction that the outlier will reappear is called the disaster prediction. For example, if the annual average rainfall is less than a certain value, it is a drought. A certain substance in the environment exceeds a certain idle value and is sweaty. When a certain parameter in the human body (such as body temperature, blood pressure, blood components, etc.) exceeds a certain range, it will occur. 3. Seasonal disaster prediction: For example, spring rain occurs in spring. Early frost appeared in September, October and November at the end of autumn and winter. 4. Topography: This is the comparison of the dynamic relationship between the system and the __ system-based prediction of the data waveform of the segment- _ _ _ _ _ _ system _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ At the end of the world and the end of the dynasty, through the efforts of the research and development of the vast number of gray system theories at home and abroad, the theoretical system has been improved and applied successfully in various fields. The "grey system theory" is also in Taiwan. It is used in information, electronics, electrical machinery, machinery, automation, aerospace, civil engineering, water, construction industry, industrial education, commerce, international trade, financial management, transportation, business management, sports and military evaluation. A considerable number of relevant research results are reported and rapidly developed and growing. SUMMARY OF THE INVENTION The present invention is a gray human-machine interface prediction system, which aims to facilitate a user to perform a function to obtain an optimal result. The gray human-machine interface prediction system of the present invention comprises a gray (10) model generated graphical user interface made by the Matlab software of the host-host, the gray Color_type generation graphical user interface package 3 is used to bring in the data input unit of the original data; a data prediction output unit connected to a gray GM pull type generating operation unit; a single μ is used to input the D Hai data into the unit The data is linked to the gray (10) model to generate 1 calculation 7L operation and the data output unit outputs the prediction L fruit ' and the execution unit is coupled to a comparison unit, the comparison unit corresponds to the output of the data pre/1" output unit The prediction results form a graphical data area. For the above purposes, the gray human-machine interface prediction system of the present invention comprises a gray GM (l'l) basic model graphical user interface fabricated via a Matlab software of a host, the gray is, the basic mode The 201101209 type graphical user interface includes: a data input unit for bringing in the original data; a data prediction output unit connected to a gray GM (1, 1) basic model operation unit; and an execution unit for inputting the data The data of the unit is connected to the gray GM (1, 1) basic model operation unit for calculation and brought into the data prediction output unit to output a prediction result, and the execution unit is coupled to a comparison unit, and the comparison unit corresponds to the output of the data prediction output unit. The predicted result forms at least one graphical data area. To achieve the above objective, the gray human-machine interface prediction system of the present invention comprises a gray GM (1, 1) Fairhas model graphical user interface made by a Mat 1 ab software architecture built on a host, the gray GM(1,1) Fairhas model graphical user interface: a data input unit for bringing in raw data; a data prediction of a gray GM (1,1) Fairhard model operation unit An output unit, configured to link the data of the data input unit to the gray GM (1, 1) Felhurst model operation unit, and bring the data prediction output unit to output a prediction result, and the execution is performed. The unit is coupled to a comparison unit that forms at least one patterned data area corresponding to the output prediction result of the data prediction output unit 32. By the above further analysis, the following effects can be obtained: the gray model of the present invention generates a graphical user interface, the gray GM (1, 1) basic model graphical user interface, and the gray GM (1, 1) Fairhas model graphic The user interface 201101209 is an execution file for the theory, formula and method, respectively, which is convenient for the user to know the execution of the hearing, cut the optimal silk, and then further process the data to minimize the discarding error. And present the most correct results. [Embodiment] The techniques, means and other functions of the present invention are described in detail below with reference to the preferred embodiments of the present invention. It is believed that the above objects, features and other advantages of the present invention are For the purpose of the present invention, please refer to FIG. 1 to FIG. 7 : Referring to FIG. 1 , the gray human-machine interface prediction system of the present invention includes a gray GM model to generate a graphical user interface ( G Qing ^ ace) 10 Mat lab software A, an operating system ^, a host c, a wheel D and a wheel E; The Hi-Gray GM model generates a graphical user interface (ie, see Figures i, 2, and 3). The theory, formula, and method are translated into a function by the MatUb software A. The format is as follows: The Matlab software A is configured in the operating system. The operating system b is operated by the host c, and is electrically connected to the input device D (keyboard or mouse) and the output device E (the screen) through the 4 host C. The Matlab software A is constructed in the operating system B, and then the "black" GM model generates a graphical user interface. The input device D and the wheeled device E can be integrated into a body input and output device (touch Honor), 201101209 Once the gray GM model generates the graphical user interface 10 to perform the file presentation, the user only needs to store the gray GM model generating graphical user interface 1 in the operating system B to perform the analysis. The gray GM model generates a graphical user interface 1 (shown in Figures!, 2 and 3) including a data input unit u, a data prediction output unit 12, and a gray GM model generation operation unit 13; The unit 11 is used to bring in the original data, and the aforementioned data can be an EXCEL file. The user can construct parameters (the parameters are symbols, characters or graphics) before the EXcEL file, provided that the data is entered with the data input unit 11 Compatible. The data input unit 11 performs an operation through the gray GM model generation operation unit 13 and transmits it to the material prediction output unit 12 to output a prediction result. The 5th gray GM model generating graphical user interface further includes an opening unit 14, an executing unit 丨5, a storage unit 16, and a clearing unit 17; the opening unit 14 is configured to specify a data address to carry the material To the data input unit 11. The execution unit 15 is configured to perform the operation of the data link 'fire color GM model generation and transfer unit 13 of the data input unit 且1 and bring the data to the data prediction output list το 12 to output the prediction result, and the execution unit 14 is further connected to the The comparison unit 18' compares the prediction result of the data prediction output unit with the prediction result of the data prediction output unit to form a pattern data area 181 which displays the comparison curve of the original data and the accumulated generation data. The storage unit 16 is coupled to a designation unit 161 for defining a data storage type, a data storage address, and a data storage name. The 5H clearing unit 17 is connected to a confirmation unit 171, and the execution of the confirmation unit m will clear the data of each unit. Referring to Figures 1, 4 and 5, the gray GM (l, 1) basic model graphical user interface 20 is created in the same manner as the gray GM model generating graphical user interface 10, the gray GM (M) basic model The graphical user interface 20 includes a data input unit 21, a data prediction output unit, and a gray GM (1, 1) basic model operation unit 23; the data input unit 21 is used to bring in the original data, and the aforementioned information can be For the EXCEL file, the user can first construct the parameters (the parameters are symbols, text or graphics) using the EXCEL file's but the data to be imported must be compatible with the data input unit 21. The data input unit 21 performs an operation through the gray GM (1, 1) basic model operation unit 23 and transmits the result to the data prediction output unit 22 to output a prediction result, and the data prediction output unit 2 2 is divided into an accumulation generation area according to the prediction result. 221. A smooth constant region 222, a prediction region 223, and an error region 224. The gray GM (1, 1) basic model graphical user interface 2 further includes a 10 201101209 open unit 24, an execution unit 25, a storage unit 26 and a clearing unit 27; the open unit 24 is used to specify a data address The data is loaded to the material input unit 21. The execution unit 25 is configured to connect the data of the data input unit 21 to the gray GM (1, 1) basic model operation unit 23 for calculation and bring the data prediction output unit 22 to output the prediction result, and the execution unit 25 is further connected. a comparing unit 2 8, the comparing unit 28 corresponding to the output prediction result of the data prediction output unit 2 2 forms a first graphical data area 281 displaying a comparison curve of the original data and the accumulated generated data, and a display evaluation error curve The second patterned data area 282 and a third graphical data area 283 displaying a comparison curve of the predicted data and the original data. The storage unit 26 is coupled to a designated unit for defining a data storage type, a data storage address, and a data storage name. The clearing unit 27 is coupled to a confirmation unit whose execution will clear the data of each of the aforementioned units. Referring to Figures 1, 6 and 7, the gray GM (1, 1) Fairhard model graphical user interface 30 is created in the same manner as the gray GM model generating graphical user interface 10, the gray GM ( 1, 1) The Fairhas model graphical user interface 30 includes a data input unit 31, a data prediction output unit 32, and a gray GM (1, 1) Fairhard model operation unit 33; 11 201101209 The data input unit 31 is used to bring in the original data, and the data to be imported can be an EXCEL file. The user can first construct the parameters (the parameters are symbols, characters or graphics) by using the EXCEL file, but the data to be imported must be The data input unit 31 is compatible. The data input unit 31 performs an operation through the gray GM (1, 1) Felhurst model operation unit 33 and transmits the result to the data prediction output unit 32 to output a prediction result, and the data prediction output unit 32 is classified according to the prediction result. A first parameter area 321, a second parameter area 322, and a result area 323. The gray GM (1, 1) Fairhas model graphical user interface plus more includes - open unit 34, _ execution unit 35, - storage unit 36 and a clear unit 37; The Hai open single το34 is used to specify the data address loading data to the data input unit 31. ' ▲ is the order tl35 for the data input unit 31 data link chromatic color GM (1,1) Fairhasron (four) calculation unit (10) to calculate and bring in and data system output unit 32 output pre- The county unit, and the execution unit % is connected to the unit 38, and the comparison unit 38 corresponds to the data prediction output, the 4 prediction result formation-displaying the comparison curve of the original data and the accumulated data 生成-jg^丨_ generation, 51 The data area 381 and one show the original data and the Fairhas meal supplement 丄 h 382. The second graphic data area of the comparison curve of the special data 12 201101209 The 5th storage unit 36 is connected to a designated unit for defining the data storage type, the data storage address and the data storage name. And h except for the early 3 7 link to confirm the early element 'The execution of the confirmation unit will clear the data of the above units. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a schematic view showing the structure of the present invention. Figure 2 is a schematic diagram of a system for generating a graphical user interface of the gray GM model of the present invention. Figure 3 shows a partial vocal diagram of the graphical user interface of the gray GM model of the present invention. Fig. 4 is a schematic diagram showing the system of the graphical user interface of the gray gm (1, 1) basic model of the present invention. Fig. 5 is a partial view showing the graphical user interface of the gray GM (1, 1) basic model of the present invention. Figure 6 is a schematic diagram of the system of the user interface of the gray WU) Fairhas model of the present invention. Figure 7 is a partial schematic view of the graphical user interface of the gray GM (U) Fairhas model of the present invention. [Main component symbol description]
Matlab 軟體 AMatlab software A
作業系統B 主機C 13 201101209 輸入器D 輪出器E 灰色GM模型生成圖形使用者界面1〇 資料輸入單元11 資料預測輸出單元12 灰色GM模型生成運算單元13 開啟單元14 執行單元15 儲存單元16 清除單元17 比較單元18 圖型化資料區181 灰色GM(1，1)基本模型圖形使用者界面2〇 資料輸入單元21 資料預測輸出單元22 累加生成區221 平滑常數區222 預測區223 誤差區224 灰色GM(1，1)基本模型 運算單元23 開啟單元24 執行單元25 儲存單元26 清除單元27 比較單元28 第一圖型化資料區281 第二圖型化資料區282 第三圖型化資料區283 灰色GM(1，1)費爾哈斯特模型圖形制者界面％ 資料輸入單元31 資料預測輸出單元32 第一參數區321 第二參數區322 結果區323 201101209 灰色GM(1,1)費爾哈斯特模型運算單元33 開啟單元34 儲存單元36 比較單元38 第一圖型化資料區381 執行單元35 清除單元37 第二圖型化資料區382Operating System B Host C 13 201101209 Input D Wheeler E Gray GM Model Generation Graphical User Interface 1 Data Entry Unit 11 Data Prediction Output Unit 12 Gray GM Model Generation Operation Unit 13 Open Unit 14 Execution Unit 15 Storage Unit 16 Clear Unit 17 Comparison unit 18 Graphical data area 181 Gray GM (1, 1) Basic model Graphic user interface 2 Data input unit 21 Data prediction output unit 22 Accumulation generation area 221 Smooth constant area 222 Prediction area 223 Error area 224 Gray GM (1, 1) basic model operation unit 23 on unit 24 execution unit 25 storage unit 26 clear unit 27 comparison unit 28 first patterning data area 281 second patterning data area 282 third patterning data area 283 Gray GM (1,1) Fairhas model graphic maker interface % data input unit 31 data prediction output unit 32 first parameter area 321 second parameter area 322 result area 323 201101209 gray GM (1, 1) Fair Haast model operation unit 33 On unit 34 Storage unit 36 Comparison unit 38 First pattern data area 381 Execution unit 35 Clear unit 37 Second Type of data region 382
❹ 15❹ 15