Chung Cheng University Institutional Repository
Not a member yet
889 research outputs found
Sort by
數位咬合板建構系統應用於顳顎障礙症治療;The Digital Occlusal Splint Design System Applied to Temporomandibular Disorders.
No full text[[abstract]]在傳統的CAD軟體環境中，咬合板的建構會因對咬牙需全部接觸以及犬齒導引之設計需求難以達成等因素令使用者操作不易，本研究透過結合傳統手工製作咬合板之流程，提出咬合板的數位設計流程，並應用於顳顎障礙症治療。本論文目的為開發一套快速設計數位咬合板(Digital Occlusal Split Design System)的CAD系統。此系統包含齒模處理系統、咬合板主體建構系統以及數位咬合模擬系統；可透過三角網格修補、新增及移除網格點進行齒模倒凹區修補，利用曲線偏移與曲面建構製成咬合板之主體，接著使用布林操作與網格變形技術將咬合板成品製作完成，再應用咬合模擬技術進行成品模擬驗證，最後套用積層製造(Additive Manufacturing, AM)技術快速輸出成品；此外，搭配數位咬合器系統可提供建構咬合板所需之參數，如齒模開口大小、真實病患咬合路徑與犬齒導引位置及所需高度，並在成品設計完成後進行咬合模擬與驗證，以降低成品出錯之可能性，進而達成快速設計與建構咬合板之目的。
When it comes to the design of occlusal splint, particular in the analysis of cusp contact between the opposing teeth and the canine guidance, traditional CAD software has serious limitations and missing functions. Therefore, we follow the design principles of the traditional hand-made occlusal splint, create the digital design flow, and apply on the temporomandibular disorders.The purpose of this thesis is to develop a quickly design CAD system for digital occlusal splint. This system includes tooth mold repair system, occlusal splint construction and occlusion simulate system. To block out the tooth mold by the triangle mesh repair function, curve offset and surface construction function can manufacture the principle part of occlusal splint and use mesh deform, Boolean operation to finished the occlusal splint. At last, apply on the additive manufacturing to archive the purpose of construct quickly. In addition, the digital occlusion simulate system can provide the parameters which the occlusal splint needed, like the distance between upper and lower jaw, the real occlusal path and the location and require height of canine guidance. When the digital occlusal splint has finished, the occlusion simulate system used to check if the interference happened between tooth and splint or not. Finally obtain a good result and archive the purpose of quickly design the occlusal splint
菲律賓南部民答那峨島「摩洛國」之研究;The Study of Mindanao's Moro State in the Southern Philippines
No full text[[abstract]]菲律賓南部民答那峨島長期與菲律賓中央政府爆發衝突，至1946年到2016年之間歷經多任總統，歷任總統對於民答那峨島的治理態度從激烈的手段轉於採用和平的手段。在2010總統艾奎諾三世上任後，總統艾奎諾三世便積極地希望能夠改善民答那峨島的狀況，了解決民答那峨島的狀況，與MILF討論過後建立一個位階低於菲律賓中央政府的「摩洛國」。於是，總統艾奎諾三世於2012年與MILF簽訂「和談原則要點」文件，並在2014年再度與MILF擬制屬於民答那峨島「摩洛國」的「摩洛國基本法」草案。 然而，「摩洛國基本法」草案並沒有順利通過，因為總統艾奎諾三世任期結束，而參議院也以「摩洛國基本法」法案本身的複雜性而擱置「摩洛國基本法」。此外，新上任的菲律賓總統羅德里戈?杜特蒂較傾向採用「聯邦制」來治理民答那峨島，因此可以清楚地了解「摩洛國」已步入歷史。 本研究嘗試從文件、文獻裡取得資料，蒐集菲律賓南部民答那峨島相關官方網站、文獻、新聞探討菲律賓南部民答那峨島的背景、菲南政府與菲律賓中央政府各時期之互動關係及艾奎諾三世主政時期與「摩洛國」之關聯。藉由探討菲律賓南部民答那峨島建立「摩洛國」的有利條件與不利條件。關鍵字：菲律賓民答那峨島、亞國、摩洛國、艾奎諾三世
In southern Philippines, the island of Mindanao has been in conflict with the Philippine Central Government for a long time. The president's attitude towards the administration of the island of Mindanao shifted from a violent means to peaceful means. In 2010, President Aquinas III actively hope to improve the situation in Mindanao Island, to resolve the situation of the island of Mindanao, and MILF discussed after the establishment of a lower rank than the Philippines ─Moro State. In 2012, President Aquino III and MILF signed the "Essential Points of Agreement" document, and in 2014 again with the MILF to be made in Mindanao "Moro State" "Moro Basic Law." However, " Bangsamoro Basic Law " bill did not pass because the end of the term of President Aquino III, and the Senate also pointed out that "Moro Basic Law" bill itself is complex and shelved the " Bangsamo Basic Law ." In addition, the newly appointed President of the Philippines, Rodrigo Duttetti, is more inclined to adopt a "federal system" to govern Mindanao, so it is clear that the Moro State has entered history. This paper attempts to obtain information from documents and literatures, and collects relevant official websites, documents and news on the background of the Mindanao Island in the southern Philippines, the interactive relationship between the Philippine Government and the Philippine Central Government in various periods, The Connection between Aquino III and the Moro States. By exploring the advantages and disadvantages of establishing "Moro State" in Mindanao.Key Words: The Island of Mindanao, Sub-state, Moro State, Aquino II
職能建構與分析-以彰化縣政府中階主管核心職能為例;The Construction and Analysis of Competency: A Case of Core Competency of Middle-rank Officials in Changhua County Government
No full text[[abstract]]本研究以建構、分析彰化縣政府中階主管人員之職能為目的，分析策略上採用行政院2016年公務人員學習地圖之「薦任官等人員管理核心能力」能力模式為參考基礎（下稱行政院2016年版）來發展，首先透過文件分析法，論述分析策略的合理性後，接續透過職能分析方法中的「蝶勘法」及「行為事例訪談法」進行適用性、重要度及行為指標的分析。 在「蝶勘法」的分析上，以彰化縣政府各局處績優中階主管人員為對象，集中進行分組操作，並針對操作成果彙整分析，結論發現行政院2016年版能力模式對於彰化縣政府中階主管而言，具備適用性但卻呈現重要度落差，高度集中於少數能力模式的態樣。 在「行為事例訪談法」的分析上，係依據上開「蝶勘法」分析結果，選定最高重要度的行為指標，並以該指標相關的中階主管人員為對象進行訪談，逐步分析建構出符合彰化縣政府中階主管人員實務需求的行為指標。 本研究透過簡易的職能分析方式，呈現將中央訂定的能力模式（行政院2016年版）轉化為地方政府（彰化縣政府中階主管）適用職能模式之歷程，雖然囿於時間與人力成本因素考量，僅建構出單一行為指標，然而本研究的分析邏輯，操作過程之諸多流程、表單表件等設計，應能供提他機關或相關研究者之參考，達到拋磚引玉的效果。
This paper constructs and analyzes the Core Competency of the middle-rank officials of Changhua County government. It uses the competence model specified by the “Core Competences for the Staff,” which is part of the learning roadmap for government functionaries issued by the Executive Yuan in 2016 (hereafter “the EY Competence Model 2016”). Based on documentary analysis, this paper first discusses the rationality and appropriateness of the research design. It follows by using the Developing a Curriculum (DACUM) and Behavioral Event Interview (BEI) to analyze the content, suitability, and importance of several behavior indexes. In applying the DACUM, highly efficient middle-rank officials coming from different bureaus/departments of Changhua County government are grouped and analyzed. According to the analysis results, the EY Competence Model 2016 is fit in general for the middle-rank officials of Changhua County Government, but the importance of the indexes for them varies. What are thought to be important are highly concentrated on only a handful of patterns in the competence model.Using the BEI method, this paper selects the behavior index with the highest importance, and uses the middle-rank officials associated with this index as the interviewees. It thus offers a step-by-step manuel for the construction of the behavior indexes that can meet the practical needs of the middle-rank officials.Through a simplified approach to competency contruction, this paper transforms the EY Competence Model 2016 into a model that fits a local government. Under the constraints of time and human cost, this paper could only focus on the construction of one behavior index. However, the analytic logic and operation process taken, together with various forms designed for the present research, can also be used for reference by other governmental agencies or researchers
群集資料/長期追蹤資料下重覆抽樣之模型選擇方法;Resampling Model Selection for Cluster/Longitudinal Data
No full text[[abstract]]在分析長期追蹤資料與群集資料時，常使用的方法之一為Liang and Zeger(1986) 所提出的廣義估計方程式。在群集資料中，群集大小不同是很常見的情況，因此群集大小是否有資訊就變成研究者感興趣的一個問題。而因為在群集大小有資訊時，廣義估計方程式會有偏差。所以Hoffman et al.(2001) 提出群內重複抽樣解決這個問題。但需要大量時間運算，因此 Williamson et al. 提出群加權廣義估計方程式，並證明群加權廣義估計方程式與群內重複抽樣是一致的，且需要的時間更短。選模指標一直以來是個重要的議題。在群集大小沒有資訊時，常用的有Shen and Chen(2012) 提出的 MLIC (missing longitudinal information criterion)和 Pan(2001) 所提出的 QIC， 然而在群集大小有資訊時，並沒有一個好的選模指標。因此在本文中，我們將利用群內重複抽樣，與誤差平方和的概念來建構新的選模指標 Resampling Longitudinal lnformation Criterion (RLIC)。並且利用模擬研究來佐證RLIC不管在群集大小有沒有資訊都能有不錯的結果，最後利用新的選模指標RLIC對實際資料作分析。
The generalized estimating equation (GEE) has been a popular tool for marginal regression analysis with longitudinal data, and its extension, the Cluster weighted GEE approach (CWGEE, Williamson et al., 2003), can further accommodate data when cluster size is informative. The model selection issue in the GEE framework has been discussed in previous literature when cluster size is non-informative. However, it is still unknown how to perform model selection in GEE analysis when the cluster size is informative. In this work, we have proposed a resampling model selection criterion RLIC for selection of the mean model in the GEE model. The proposed method is shown to perform well when cluster size is non-informative/informative. We also examine the performance of a naive procedure namely to apply the QIC method (Pan, 2001) with informative cluster size treated as non-informative. We find that such a naive application of QIC do not perform as well as the RLIC method. The proposed method is further applied to data from the Taiwan longitudinal study on aging to assess the marginal relationship of frailty status with health and social status in the elderly, when cluster size is informative
Trend-Gamma Process 模型下之衰變分析;Degradation Analysis on Trend-Gamma Process
No full text[[abstract]]對於高可靠度的產品，壽命往往難以觀察，此時我們常使用衰變分析 (degradation analysis) 來解決失效樣本不足的問題。透過產品品質特徵值之衰變路徑，建構衰變模型來配適衰變路徑，完成壽命之預測。其中當特徵值之增量恆正時，gamma process (GP) 模型為常見之衰變模型。然而，有時候 GP 模型亦無法確切捕捉衰變路徑。因此本文進一步提出 trend-gamma process (TGP) 模型以改善 GP 模型不足之處。透過模擬與實例分析結果，我們發現 TGP 模型能廣義的處理增量恆正之衰變資料
長期追蹤資料下模型選擇之不同估計法的比較;
No full text[[abstract]]群集長期追蹤資料 (cluster longitudinal data) 是一種在很多領域當中常看見的資料類型，但是這種資料常常存在複雜的相關結構，故分析資料時，常使用 Liang and Zeger (1986) 提出的廣義估計方程式 (Generalized estimating equations ，簡稱GEE )。 如何選擇一個正確的模型，是研究者需要重視的一個課題。由於群集長期追蹤資料會出現有遺失值的情況，所以我們利用 Shen and Chen (2012) 提出的選模指標 MLIC (Missing Longitudinal Information Criterion) 來處理反應變數有遺失之群集長期追蹤資料的模型選擇問題。在本文中我們探討在反應變數有缺失時，三種不同估計法 (泰勒展開式、資料擾動法、拔靴法) 下的 MLIC 之選模表現。根據模擬結果發現，我們建議當反應變數有缺失時，利用拔靴法估計 MLIC 之選模結果，較為穩健並且我們將這個方法應用到青年關於哮喘的實例分析上
微機電感測器資料手套建構與手部復健姿態偵測實作;Hand Motion Detection with MEMS Sensor Data Glove for Rehabilitation
No full text[[abstract]]關於資料手套的研究，近年來成為熱門的研究領域，雖然實際的資料手套相關產品目前並不普遍，但資料手套被大量的研究且結合影音娛樂、工業設計，以及醫療。在醫療方面，目前資料手套多為手部復健器材，作為復健治療執行的操作介面，因量測結果為大略近似值，鮮少應用於協助病情評估中。我們希望資料手套，可在使用復健器材時，偵測手部姿態，量測關節角度。協助醫療人員診斷患者手部能力，這對中風以及手部手術後的患者有相當的幫助。本論文將透過姿態資料手套，取得手部姿態。在手套上手指與手掌的背面位置安裝十六組重量輕且體積小的姿態感測器(Motion Sensor)，其為含有加速計、角速度計以及磁力計的微機電感測器(Micro Electro Mechanical Systems Sensor)，姿態感測器量測的數據被收集後，經過校正，利用航姿參考系統演算法融合數據，這類姿態偵測以地磁為參考，取得手套上的十六組感測器姿態。所有感測器姿態以Wi-Fi傳輸至電腦，在電腦上藉由Vpython執行模擬運算以及透過向量運算關節彎曲角度，呈現手部的姿態模擬，並可算得各個關節的彎曲角，成為診斷時的參考。協助醫療人員判斷患者手部能力。
In recent years, data gloves have become one of the popular researching topics, although data glove products are not common now. Data gloves can provide easier control by direct hand actions. Many researchers use it in their research areas which include media, industrial design, and medical area. In the medical area, most of data gloves are used just as control interface of rehabilitation systems. The measuring results are rough approximations. It is rarely applied as a diagnosis supporting device in a rehabilitation process. We want to develop a data glove as a diagnosis device for rehabilitation. The new data glove can detect hand attitude and hand joint bend angles. Doctor and therapist can use the data glove to diagnose hands ability of patients. This is helpful for patients after stroke and hand surgery.In the thesis, we get the hand motion by the attitude detecting data glove. Sixteen motion sensors are set on the back of the glove. The motion sensor includes an accelerometer, a gyroscope, and a magnetometer. Those are MEMS (Micro Electro Mechanical Systems) sensors. Data of the inertial sensor and the magnetic sensor are fused to get the attitude by AHRS algorithm. The attitude detection refers the magnetic field of the earth. The attitudes of 16 sensors are gathered into a socket and the socket is sent to the computer by Wi-Fi. The hand attitude simulation and the joint angles are computed with Vpython on the computer. Finally, the hand attitude is presented on screen with the limitation for impossible hand action. The bend angles of joints can be fetched by the vector computing. The result can be a reference for diagnosis
應用於穿戴式/可攜式多媒體顯示裝置之節能及抗錯高畫質影像縮放器設計;AN ENERGY-EFFICIENT AND ERROR-RESILIENT HIGH QUALITY SCALER FOR WEARABLE/PORTABLE MULTIMEDIA DISPLAY
No full text[[abstract]]近年來，顯示技術與半導體製程技術的蓬勃發展，穿戴式及可攜式多媒體裝置(如頭戴式顯示器、智慧型手機、微型投影機及智慧眼鏡等)在人類生活中扮演越來越重要的角色，同時為了傳達資訊給使用者，絕大多數穿戴式及可攜式多媒體裝置配有尺寸及解析度不一的顯示面板。由於輸入影像的多元化，每個顯示裝置皆需要一個影像縮放器，來將各種輸入影像做解析度的縮放，並於面板顯示。然而，在多媒體裝置中，影像縮放品質提供人類一個深刻且直接的印象，因此，提高影像縮放品質一直是多媒體裝置的研究重點。所謂的影像縮放品質，指的是減少影像失真。例如，減少在影像縮放時，所造成的模糊現象，即為失去原始影像細節，或是減少因為影像縮放的長寬比不同，而使得影像中重要物件扭曲或變形。另一方面，為了延長電池的使用時間，低功率消耗以及節能一直是影像縮放器設計的重要目標，特別是穿戴式及可攜式多媒體裝置(其電源大多來自於電池)。而電壓調整機制已被證實能有效降低功率消耗，且部份具效能感知的電壓調整機制可偵測晶片對於先進製程變異或操作環境變異(如溫度及電壓)所造成的時序錯誤並更正，以進一步提高晶片良率。因此為了提升穿戴式或可攜式裝置的顯示品質與延長電池的使用時間，節能及抗錯高畫質影像縮放器(可用於電壓調整機制)是迫切需要的。因此本論文首先從演算法層面中提出兩個應用於不同場景的演算法，分別是具有高影像品質的可調式參數基於多項式影像縮放技術及低成本基於加法器的步階式線性內插技術。可調式參數基於多項式影像縮放技術可以同時保留影像細節及維持重要物件長寬比。因為保留影像細節，因此相較於傳統的雙線性內插提升大約5%到15%的影像品質，而這些影像品質是根據峰值訊號雜訊比來評比。另一方面，基於加法器的步階式線性內插技術維持傳統的雙線性的影像品質，並且在設計成本及功率消耗分別節省45%和44%。在架構方面，本論文提出一個縮放方向可交換式的架構搭配本論文所提出的一個節省記憶體使用量策略，相較於傳統的兩種架構分別可以節省101%和26%的記憶體使用量以及只124%和33%的功率消耗。另外，搭配本論文針對影像縮放架構所提出的記憶體存取方法，記憶體使用率可以在任何情況下達到接近100%的結果。另外，本論文考慮即時影像資料的特性，提出一個新的錯誤回復方法。首先忽略部分影像資料的影響，針對會嚴重影響品質的資料，利用周期延長的概念，搭配影像內插來校正因為電壓調整所造成的影像資料錯誤。在維持一定的影像品質下，系統因為電壓調整而在邏輯電路及整個晶片分別節省29%及10%的功率消耗。本論文中提出一個完整的設計流程，可因應設計應用於穿戴式/可攜式多媒體裝置之節能及抗錯高畫質影像縮放器，並且根據實驗結果，成功驗證此技術的可行性。
Over the past decades, rapid development of display technology boosts demands for diverse multimedia displays such as smart phones, smart glasses, televisions, head-mount-display (HMD) and head-up-display (HUD). Since multimedia displays have a wide range of information sources and display panels, a scalar is indispensable for conversion of image sources from input resolution to output resolution (i.e. image scaling). To improve the experience perceived by human observers, multimedia displays are required to have superb quality. Such as less image content distortion, loss of detail and deformed salient object. Additionally, energy efficiency has become more and more important nowadays. Therefore, multimedia displays often have built in voltage scaling techniques to reduce power consumption and effects of PVT variations.Two image scaling algorithms are proposed in this thesis for different claims, including an adjustable coefficient polynomial-based interpolation (ACPI) and adder-based stepwise linear interpolation (ABSI). With ACPI, it is able to simultaneously maintain image details and aspect ratio. 5%-15% peak signal-to-noise ratio (PSNR) improvement is observed as compared with conventional polynomial-based image scaling algorithms. The ABSI has a similar scaling quality in terms of PSNR but also enables reduction of chip area and power consumption by 45% and 44%, respectively. Furthermore, proposed storage requirement reduction strategy and interchangeable two-step scaler architecture achieve 101% and 26% storage requirement reduction and 124% and 33% lower of power consumption as compared with two types of conventional inflexible scaler. Moreover, linebuffer utilization of the scalar is close to 100% with proposed pixel-based memory access method.Lastly, a new error resilient policy considering characteristics of linebuffer-based real-time scaler is proposed. It ignores errors of a nibble (i.e. 4bits) from LSB to MSB and uses a cycle extension and interpolation technique to correct timing error of the rest data bits while maintaining the image quality in terms of PSNR. And power consumption of logic and total chip are reduced by about 29% and 10%, respectively.In summary, a complete design procedure of an energy-efficiency and error-resilient high quality scaler for wearable/portable multimedia display is demonstrated in this thesis. And experimental results also verify the feasibility of proposed techniques
考慮頻率調節限制條件之交流微電網電能調度;Frequency-Regulation Constrained Energy Dispatch for an AC Microgrid
No full text[[abstract]]交流微電網的電能調度為一最佳化的問題，必須在所設定的限制條件下最小成本的運轉。本文使用混合整數線性規劃法(Mixed Integer Linear Programming, MILP)的方式來描述問題模型，然後採用最佳化求解軟體CPLEX來建立微型電網系統模型並求解。本文以國內核能研究所建構之微型電網示範系統為參考系統，考慮風力發電與太陽能發電系統之發電量，來最佳化微型渦輪機與儲能設備之運轉調度。 為了提高微電網運轉時頻率的穩定度，本文使用下垂控制及虛擬慣性來做為最佳化中的限制條件來調節微電網運轉時的頻率。最後再各種案例顯示，在孤島及併網模式下，得以用頻率調節限制及卸載機制來符合台電規定的標準。
An AC Microgrid dispatch is an optimal problem, it must operate at the minimum cost under the set restriction conditions. This thesis uses mixed integer linear programming (MILP)-based model for microgrid power dispatch, and then use the commercial software CPLEX to create a microgrid system model and solve the optimal problem. The microgrid demonstration system implemental by Institute of Nuclear Energy Research (INER) is selected as a demonstration system. The operation scheduling optimization of microturbines and energy storage devices effectively coordinates with wind power and solar photovoltaic system. In order to improve the frequency stability in microgrid, the constraints of droop control and virtual inertia are used to regulate system frequency. Finally, a variety of cases show that in the islanded and the grid-connected mode, it can use constraints of the frequency regulation and shedding to meet the requirements of Taipower standards
連桿結構開發輕量化類人機器人雙足於RoboCup;Development of a lightweight biped adult-size humanoid robot with connecting lever structure in RoboCup
No full text[[abstract]]關於機器人的研究，在近幾年中是相當熱門的議題之一，尤其是大型類人機器人這塊領域。如何使大類人型機器人在外型或身體結構上與人類更相近，或是能作出更相似的動作，一直是主要的研究的方向。然而進行大型類人機器人的運動學或演算法開發，往往必須先投入大量的硬體成本，導致許多開發者僅能以模擬方式進行開發，而無法進行實體研究。吾人計畫建造成人尺寸類人型機器人之雙足(約 90公分高)，並利用輕量化的連桿結構仿造人類雙足上肌肉，用來提供雙腳運動時的能量，大部分的重量(馬達、齒輪)將集中在機器人的臀部、腰部位置。如同人類的核心肌群，人類利用核心肌群來支撐上半身、保持身體穩定、驅動雙腿，並減經下肢的負擔，因此並不需要過於強壯或笨重的四肢。我們將得到輕量化的四肢以及較好控制重心的重量配置，這使我們能減少馬達的數量以及使用較低成本的伺服馬達，來完成大型類人機器人雙足的建製。根據本研究的實驗結果，此大型類人機器人總重約10公斤，擁有12個關節自由度(DOF)，只需要10顆60 kgw-cm以及2顆40 kgw-cm 之直流伺服馬達，最終使機器人完成向前步行的動作。
Recently, the humanoid robot is a popular research topic in robotics. Especially, how to make the robot more like a human being not only behavior but also body structures always has been the main researching direction of all developers.However, before they start to research the kinematics and algorithm. Developers have to invest a huge amount of money in order to build the hardware. It leads to most developers without resources, can only work in simulation, not the real one.In this project, we are going to build the legs of the adult-size humanoid robot (About 90cm height). In our design, the legs use a connecting lever structure that works similarly to the human muscle to deliver the energy. In order to design the lightweight legs and feet, we also set up most of the heavy items compress around the hip part. The heavy hip approximate to human’s core muscle group, it can be used to support upper-body, keep balancing, drive the legs and reduce the load of the lower limbs. Therefore, our robot doesn’t need to have strong or heavy limbs and is more like a human. Eventually, we design and build the robot with lightweight legs and easy control center of mass. This helps us to reduce the strength of the servo motor and use low-cost servo motor only. We built humanoid robot legs around 10 kg with 12 degrees of freedom. It only requires ten 60 kgf-cm and two 40 kgf-cm servo motors. Finally, the robot can finish the walking motion forward