Chung Cheng University Institutional Repository
Not a member yet
889 research outputs found
Sort by
考慮複循環機組等效運轉時數之機組排程資訊系統設計與實現;Development of Information System for Unit Commitment Considering Equivalent Operating Hours of Combined-cycle Units
No full text[[abstract]]近年來，提升營運效率與降低電能生產成本已成為各國電力市場改革的主流，如何決定發電機組最佳排程，且考慮等效運轉時數之限制，並找出最佳發電成本策略，成為一項重要的研究課題。本論文所建構之短期機組排程資訊系統，後端語言採用Python，網路應用程式框架採用Django，資料庫則使用MySQL資料庫來存放相關資料，核心機組調度程式由解法器GUROBI進行最佳化分析。本文重點在於設計人性化之使用者介面，並將其與資料庫跟系統核心調度程式整合，提高資訊系統效能。而建構完成之資訊系統，可以讓使用者不管在何時何處皆可上網使用，提供線上的資料分析服務，讓使用者能獲得最佳之發電成本策略。
In recent years, improving operational efficiency and reducing the cost of electricity production are the main goals of the electric market reform. In this research, unit commitment is used to determine optimal power generation cost and equivalent operating hours constraints are also considered. The purpose of this thesis is developing an efficient information system to provide a tool for estimating the economic benefits of the generation. The graphical user interfaces is developed for the system, connecting with unit commitment programs and MySQL database. This system is implemented by Python programming language. The web application framework is based on Django. The unit commitment problem is solved by GUROBI optimizer. The system provides online service which allows the UC program can be executed on internet and presents the simulation results on the browser. Therefore, the UC program can be executed more efficiently
空間規劃與自動家具擺設;Space Planning and Automatic Furniture Furnishing
No full text[[abstract]]公寓是大部分現代都市人居住的首選之一，因此如何在有限的空間之下，對公寓進行房間規劃與隔間，並在各個房間之中進行家具布置是非常重要的問題。大多數人委託室內設計師來解決此問題，但卻需花費時間與金錢。因此本論文提出空間規劃與自動家具擺放系統，讓一般使用者不需要具備相關室內設計知識，就可以得到空間規劃與家具擺設的方案。為此本論文提出結合空間規劃與自動家具擺設的系統。首先該系統會根據輸入公寓大小進行隔間，並使用SVM訓練得到的房間分類模型對各房間進行功能分配，確保本論文之房間功能分配結果符合「大多數設計師之推薦配置」。接著根據各房間類型與使用者輸入之額外家具建立家具清單，並且以「確保房間功能及其動線的流暢性」為目的設計成本函數，並使用模擬退火法尋找成本函數最佳解。最後根據最佳解進行家具擺設，取得具有「大多數設計師之推薦配置」之房間規劃與「確保房間功能及其動線的流暢性」之家具擺設方案。
The apartment is one of the first choice for most modern urban people to live. Room planning and compartments for apartments, in a limited space and furniture layout for each room are very important. Most people commission an interior designer to solve the problem, but it takes time and money. Therefore, this paper puts forward the space planning and automatic furniture furnishing system. So that the general users can get space planning and furnishings of the program.do not have the relevant interior design knowledge. Users only need to enter the space, doors and windows, room type three basic information This paper proposes a system combining space planning and automatic furniture furnishing. First of all, the system compartment the input apartment based on the input size. Second use the SVM training the room model to class the room, and use the model to make the room allocation. Third according to the type of room and the user input make the furniture list. Then design the cost function for the furniture furnishing base on the" ensuring room function and fluency of moving", and use the simulated annealing method to find the best solve of cost function. Finally, according to the best solution of cost function make the furniture furnishings
巨量多波束主動天線陣列發射機之射基頻整合設計;RF and Baseband Integrated Design of Massive Multibeam Phased-Array Transmitter
No full text[[abstract]]本論文設計一適用於巨量主動天線雛形系統之射頻發射機模組，可同時支援類比與數位波束成形，並以多波束掃描為目標，可應用於下世代行動通訊波束追蹤的概念。此巨量主動天線雛形系統有64支天線以及16路射頻通道，以具備波束形成的功能，其以4通道為次矩陣的整合方式進行系統整合。其中，各組發射機通道包含：功率放大器、可調式衰減器、正交調變器、反射式相移器，以及注入式鎖定振盪器等。並整合微處理機控制單元，進行一對多主從裝置的溝通，包含健康狀況回報以及波束形成控制命令上傳。在系統整合上，解決多項議題，包含多項控制溝通協定的干擾、截割接地面的射頻效應，以及射頻互耦合干擾效應等。最終量測結果與波束形成預測模型的比較下：單波束的平均主波束指向角度誤差為1.4°，平均旁波瓣比為5.8 dB，3-dB波辦寬為7°，等效全向幅射功率(EIRP)約40 dBm；多波束的平均主波束指向角度誤差為1.3°，平均旁波瓣比為3.5 dB，3-dB波辦寬為8°。
A 64-element active antenna transmitter array system is presented in this thesis, which is aimed for the fifth-generation mobile communication. There are 16 RF Transmitting chain, which chain is composed of RF circuits including power amplifier, variable attenuator, quadrature modulator, reflection-type phase shifter, and injection-lock oscillator, and microcontroller for monitoring circuit health states and beamforming controlled codes. The measurement results show the distinct angle and multiple beam generation in a 120° sector. For the simple beam case, the beam has 7° of HPBW and 40 dBm of EIRP, with 1.4° beam direction error and 5 dB SLL. For multiple beam patterns, there are 1.3° of main beam directioin error, 3.5 dB of average SLL, and 8° of HPBW
以視覺資訊預估天氣屬性;Visual Weather Property Prediction
No full text[[abstract]]在本論文中，我們嘗試單純使用圖片的資訊，利用卷積遞迴神經網路來預估天氣屬性。我們研究了兩種情況下的天氣屬性預估:1)預估單張室外的照片溫度2)預測一序列照片中，最後一張照片的天氣屬性。第一種情況下，使用以大規模影像數據庫訓練的卷積神經網路提取視覺特徵，我們證明可獲得相當好的性能，並分析訓練數據量如何影響性能。在第二種情況下，我們考慮視覺外觀隨著時間的變化，建構出一個遞迴神經網路來預測一序列圖片中最後一張圖片的天氣屬性。與目前最先進的模型相比，我們獲得了更好的預測準確度。此外，我們研究從場景不同區域提取視覺資訊以及在不同的白天時間拍攝的照片對性能的影響。我們的方法進一步的強化僅使用圖片中視覺資訊來進行有效天氣屬性預測的這個想法。
In thesis paper, we attempt to employ convolutional recurrent neural networks for weather weather properties estimation using only image data. We study ambient weather properties estimation based on deep neural networks in two scenarios: a) estimating temperature of a single outdoor image, and b) predicting weather properties of the last image in an image sequence. In the first scenario, visual features are extracted by a convolutional neural network trained on a large-scale image dataset. We demonstrate that promising performance can be obtained, and analyze how volume of training data influences performance. In the second scenario, we consider the temporal evolution of visual appearance, and construct a recurrent neural network to predict weather properties of the last image in a given image sequence. We obtain better prediction accuracy compared to the state-of-the-art models. Further, we investigate how performance varies when information is extracted from different scene regions, and when images are captured in different daytime hours. Our approach further reinforces the idea of using only visual information for cost efficient weather prediction in the future
抗電壓與製程飄移之全數位溫度感測器;An All-Digital Temperature Sensor with Process and Voltage Variation Tolerance for IoT Applications
No full text[[abstract]]現今的晶片設計都以高整合度為主，晶片處理的問題日益複雜，但晶片的包裝卻沒有突破性的革命，也因此在設計晶片的過程中也需要考慮到散熱的問題，在目前的系統晶片上，會內嵌許多溫度感測器來監控晶片有無過熱的情形。 而一般的溫度感測器都未考量到電壓漂移的影響，但電壓飄移可能會帶給溫度感測器有上百度的誤差影響，而要消除電壓飄移的影響，通常都是以類比的穩壓電路或是開發低電壓敏感度的電路來克服，但因兩者都需要客製化設計，設計的複雜度也較高，本論文提出的全數位溫度感測器可估計目前的工作電壓，並根據工作的電壓不同，來補償溫度感測器的輸出，並透過校正來消除製程變異的影響，讓溫度感測器可以同時抵抗電壓、製程飄移。本論文中的低電壓敏感度全數位溫度感測器，利用三個環形震盪器來建立相對參考模型，並開發出1. 低溫度敏感度的工作電壓估測電路 2. 低電壓敏感度的溫度感測電路，並利用多點校正來將建立輸出與實際溫度的關係。並改善先前建立相對參考模型中需要大量時間來找尋標準元件庫中適合的元件，改以一較有理論根據且較省時的方法。 在這篇論文中，我們提出的低電壓敏感度全數位溫度感測器，採用90奈米製程實現，並可在0.9 ~ 1.1伏特的電壓及 20 – 80度的溫度範圍內工作，在下線版本中，最大誤差為-4.3℃~4.3℃.，在低成本版本中，我們會犧牲12倍的轉換率來降低電路所需的參考時脈速度以及功耗和成本，最大誤差為-1.6℃~1.8℃，並適合內嵌在電池供電或物聯網裝置等較易受到電壓飄移的應用上。
Due to more circuits integrated on a single chip, but the IC package technology does not catch up the speed of chip growing. Therefore, the cooling problem should be considered during the design process. As a result, embedded temperature sensors on a system-on-a-chip (SoC) to monitor the hot spots becomes more popular.Most of the temperature sensors can’t resist the influence of voltage variations and will cause hundreds of degrees Celsius errors. Some of temperature sensors may use a bandgap reference or low voltage sensitivity circuits to eliminate the influence of voltage variations which need to full-custom design and can’t be portable to different process in a short time. In this thesis, the proposed all-digital temperature sensor can estimate the current supply voltage and compensates for the temperature sensor’s output based on the supply voltage information to resist the voltage variations. Addition, the process variations can be removed by three-point calibration.The proposed temperature sensor uses three ring oscillators to build up the relative reference modeling (RRM) and proposed 1. Low-temperature sensitivity voltage classifier. 2. Low voltage sensitivity PTAT (Proportional to Absolute Temperature). The relationship between the sensor output and the actual temperature value are determined after three-point calibration. The proposed negative voltage coefficient and zero temperature difference ratio design is theoretical basis and save the time in searching cell combinations in previous method. In this thesis, the proposed all-digital temperature sensor with process and voltage variation tolerance is implemented in TSMC 90nm CMOS technology. The temperature sensor can operate at 0.9 ~ 1.1V supply voltage and 20 – 80 Celsius with a maximum temperature error -4.3℃~4.3℃ in the tape-out version. In the cost-down version, the proposed temperature sensor will reduce the conversion to one-twelfth to lower the area , power consumption and the speed of reference clock. The maximum temperature error -1.6℃~1.8℃ in the cost-down version
基於以太坊區塊鏈ＤAPP研究和開發;Case Study on DAPP Based on Ethereum and Blockchain
No full text[[abstract]]從比特幣開始後來慢慢發展出了區塊鏈這個技術的概念,區塊鏈主要的特色為去中心化和資料不可篡改特性,然而本篇論文是用以太坊來開發,以太坊是一個開源區塊鏈平台，其交易貨幣為以太幣,但其不同之處是具有智能合約的功能，開發可以撰寫智能合約並部署到區塊鏈上，部署後的合約可以被其他人存取，然而被部署到以太坊的智能合約都具有一個地址，之後又從智能合約衍伸出Ｄapp。DApp為decentralized application的簡寫，意思為去中心化的應用程式，本論文想解決投票系統眾多問題中的身份認證問題所以希望設計出一個將來能夠應用到區塊鏈ＤApp的會員系統，未來希望能擴充到投票系統上應用。關鍵詞：以太坊、區塊鏈、ＤApp
AbstractStarting from Bitcoin, the concept of blockchain technology was slowly developed. The main features of the blockchain are decentralization and data non-tamperability. However, this paper is developed by Taifang, and Ethereum is an open source. The blockchain platform, whose trading currency is Ethereum, differs in that it has the function of a smart contract. Development can write smart contracts and deploy them on the blockchain. The deployed contracts can be accessed by others, but Smart contracts deployed to Ethereum have an address, and then Dapp is derived from smart contracts. DApp is a shorthand for decentralized application, meaning decentralized application. This paper wants to solve the identity authentication problem in many problems of the voting system. Therefore, I hope to design a membership system that can be applied to the blockchain DApp in the future. I hope to expand it in the future. Apply to the voting system.Keyword - Ethereum, blockchain, DAp
高溫質子交換膜燃料電池之氣體管理策略之研究;
No full text[[abstract]]高溫質子交換膜燃料電池具有啟動快、能量密度高、低汙染和不用加濕器等特點，為了提高燃料電池的燃料使用率，高溫質子交換膜燃料電池會使用陽極封閉式的方式，使用方法是在陽極出口處加裝電磁閥來控制。在陽極封閉式下，電池性能會逐漸下降，因此需要定時排出積累氣體使性能回復。性能下降的速率主要受到操作參數影響，包含陰極計量比、陽極提供的氫氣壓力與操作電流密度。為了開發一套高溫質子交換膜燃料電池氣體管理的策略，本篇論文將主要目標鎖定在調查操作參數對於性能下降的影響。在本篇論文中，首先會建立一套數學模型來預測在陽極封閉式下，局部電流密度沿著陽極流道的分布情形，此模型是使用單電池的實驗數據進行擬合；此外，在實際局部電流密度分布的量測上是使用特別設計過的單電池進行實驗。實驗結果顯示電流密度對於性能下降的影響非常顯著，而陰極計量比對性能下降的影響較小，另外在局部電流密度的分布上則會由陽極入口至陽極出口處逐漸遞減，此外建議電磁閥在電池性能下降0.01 V時排出雜質。關鍵詞：高溫質子交換膜燃料電池、陽極封閉式、性能下降、燃料使用率，排出。
The advantages of high-temperature proton exchange membrane fuel cells (HT-PEMFCs) include fast reaction kinetics, high energy density, less pollution, and no need of humidifiers. In order to increase hydrogen utilization, the HT-PEMFC is operated with a dead-ended anode, in which a solenoid valve is installed at the anode exit. In the dead-end mode, impurities gradually accumulate within the anode channel, causing performance drop of the HT-PEMFC. The performance drop rate is affected by operating parameters, including cathode stoichiometric ratio, supplied hydrogen pressure in the anode, and operating current density. In order to develop a gas management strategy for a HT-PEMFC, this study is aimed for investigating the effects of these factors on the performance drop. In this study a mathematical model is developed to predict the gases and the current density distributions along the anode channel during a dead-ended anode mode. The model is calibrated by experimental data using a single cell. In addition, the current density distribution within the active area at various operating conditions is measured by a specially designed single cell. The results suggest that operating current density significantly influences the performance drop, whereas the cathode stoichiometric ratio shows less impact on the performance drop. The local current density gradually decreases from the anode inlet toward the outlet. The solenoid valve is suggested to purge impurities when the cell voltage drops by 0.01 V.Keyword: High temperature proton exchange membrane fuel cell (HT-PEMFC), dead-ended anode, performance drop, hydrogen utilization, purge
球化鐵酸洗汙泥熱裂解與防鏽處理機制研究;Study on Mechanism of Pyrolysis Pickling Sludge and Antirust Treatment of Spheroidized Iron
No full text[[abstract]]本研究將廢水廠汙泥及磷酸鋅汙泥以無氧裂解方式進行處理，實驗中由常溫加熱到目標溫度450℃，並從常溫到150℃蒐集冷凝液，之後每個100℃蒐集到目標溫度，在100℃時汙泥會進行脫水，裂解至450℃汙泥以碳化，失重率能夠到達80%，剩餘固體含有氧化鐵0.5%及2.5%的鋅能回收處理，以處理費用之觀點加熱到350℃為最合適，剩餘固體含有6.58%的鋅能夠回收處理。在鐵線浸泡磷酸鹽液形成磷酸鹽薄膜，在浸泡1分鐘之後就會開始生成，到達5分鐘時能夠形成較完整的薄膜，結晶狀態也較良好，到達9分鐘時雖然結晶狀態已碎裂，但表面薄膜更加的密集，抗繡程度與表面薄膜密度成正比
5自由度三極磁浮軸承系統之實驗驗證;Experimental Validation of a 5-DOF Three-pole Active Magnetic Bearing System
No full text[[abstract]]隨著科技的發展，學界和業界一直不斷致力於提高渦輪機械的可靠性和性能，例如壓縮機，主動磁浮軸承是近年來一直看漲的發展技術之一，但是由於位置感測器和功率放大器價格昂貴，因此主動磁浮軸承的整體成本很高。本文提出一個只需使用六組功率放大器的5自由度之主動磁浮軸承系統。本文中，磁力、線圈電流和氣隙之間的關係將透過實驗來鑑別，其系統模型和控制器也將根據鑑別結果進行修正。數值模擬結果顯示此系統理論上可將轉子穩定控制在平衡位置。實驗結果發現利用有效質量的觀點以解耦並分兩階段的方式也可穩定控制轉子。
There have been continual efforts by the academic and industrial communities to improve the reliability and performance of turbomachinery, such as compressors, as new technologies become available. Active magnetic bearing (AMB) is one such enhancing technology that has been gaining strong momentum in recent years. However, the overall cost of AMB is high since position sensors and power amplifiers are expensive. This paper proposes a 5-DOF AMB system using just six power amplifiers. In this study, the relationships between the magnetic force, coil current, and air gap will be identified experimentally. The system model and controller will be modified accordingly. Numerical simulation results show that the rotor can be theoretically stabilized at the equilibrium position. Experiment show that using a two stage decoupling control method with view of effective mass can also stabilize the rotor