34254 research outputs found
Sort by
Analysis of Matrix Completion State Estimation in Distribution Systems under Insufficient Measurements Conditions
No full textThe increased integrations of distributed energy resources (DER) in the distribution network have greatly affected the distribution system operations. To handle the increasingly complex bi-directional power flows in the distribution system, the distribution system state estimation (DSSE) has become one of the important tools for system operators to maintain system situation awareness, and to mitigate problems encountered in system operations.
It was claimed that a complete DSSE can be provided by matrix completion (MC) methods with only a small number of actual measurements. This thesis discusses the applicability of a matrix completion (MC) based DSSE method for distribution system operations. The method is formulated with the incomplete measurement matrix augmented by power network physical constraints. It is tested with different amount of measurement data in three test systems. Numerical results indicate that, even with the provision of a complete state estimation, there exist non-ignorable errors in the estimated bus power injections and branch flows in areas with insufficient measurement data. To address this problem, this study investigates the properties of the MC based DSSE method and explores several measurement placement methods, including greedy algorithm, vertex cover, centrality and Chebyshev bound methods, to enhance the applicability of the tested MC-based DSSE method in distribution system operations
The construction about Class System of Legal Reservation for epidemic prevention in Taiwan (R.O.C.).
No full textAfter the SARS epidemic , Taiwan\ue2s medical epidemic prevention system has gradually become institutionalized and professionalized , but the COVID-19 epidemic still reveals its omissions and shortcomings. This study focuses on the major epidemic prevention measures in Taiwan involving people's rights to body and health , freedom , information privacy , and property , and establishes the regulated intensity of the aforementioned fundamental rights during the epidemic with reference to the spirit of the \ue2 Class System of Legal Reservation \ue2 and the speed of variation and contagiousness of the epidemic
Peripheral Irisin Gene Delivery Ameliorates the Inflammation and Ferroptosis in the Brain of Type 1 Diabetes Akita Mice
No full textDiabetic neuropathy (DN) is a major complication of diabetes mellitus (DM). DN is usually considered a disease of the peripheral nervous system in the past. Nevertheless, more and more evidence has demonstrated that the central nervous system (CNS) is also involved in DN. Currently, there has no effective therapeutic method for preventing or treating DN. Management of blood glucose is regarded as the best approach for preventing the disease onset of DN. Irisin is a myokine through proteolytical cleavage from its precursor fibronectin type III domain containing 5 (FNDC5). Irisin is highly expressed in response to exercise, stimulates mitochondria increased energy expenditure, promotes glucose homeostasis, and exerts anti-inflammatory effects. We want to test whether Irisin gene delivery can ameliorate neuropathy through its anti-inflammation effect against hyperglycemia-induced DN. In this study, we utilized Insulin2Akita mice as our animal model of DM to investigate DN. Insulin2Akita mice have a spontaneous mutation in the insulin 2 gene leads to incorrect folding of the insulin protein producing toxicity in pancreatic \uce\ub2 cells, reduced \uce\ub2 cell mass, and reduced insulin secretion. It was observed that the oxidative stress marker 8-OHdG immunostaining and Fe3+ accumulation was significantly higher in the substantia nigra of Insulin2Akita mice than that in C57BL/6 mice. This was accompanied by elevated pro-inflammatory IL-6 and IL-1\uce\ub2 expression and I-Ba1+ /CD11c+ microglia activation. Oxidative stress increase has been reported to link to ferroptosis. Ferroptosis is a novel form of nonapoptotic regulated cell death. Glutathione peroxidase 4 (GPX4) inactivation is the main characteristic of ferroptosis. Reduced GPX4 level leads to the accumulation of lipid peroxides and activates ferroptosis. It was found that Irisin gene delivery restored the GPX4 expression and reduced Fe3+ accumulation in the substantia nigra of Insulin2Akita mice. Furthermore, the data had shown that neuroinflammation was involved in DN. Irisin gene therapy alleviates microglia activation and descends pro-inflammation cytokines expression in the substantia nigra of Insulin2Akita mice. Excessively activated microglia produce several pro-inflammatory mediators, leading to neuroinflammation and damage to the surrounding neurons. In conclusion, we report that Irisin gene delivery in Insulin2Akita mice alleviates oxidative stress, thereby alleviating the ferroptosis and neuroinflammation in the substantia nigra of diabetic mice. These findings support that Irisin could be developed as a therapeutic method to improve DN through decreased activation of ferroptosis and anti-inflammation in substantia nigra in DM patients
Reinforcement Learning-Based Rate Maximization for Energy-Constrained NOMA Networks
No full textThe paper focuses on optimizing the transmission strategy of a non orthogonal multiple access network (NOMA) using deep reinforcement learning algorithms. The NOMA system employs time division multiple access (TDMA) to schedule transmissions for multiple primary users and one secondary user, all of which face energy constraints similar to IoT devices. The secondary user has two main tasks in each time slot: data transmission and energy harvesting based on the signal from the primary user. The primary objective is to maximize the long-term throughput of the secondary user by optimizing transmit power and time-sharing coefficient for these tasks. However, achieving this goal is challenging due to the need for robust strategies that balance short-term losses with long-term gains. For instance, when the primary user transmits in a strong channel, intuition might suggest that the secondary user should focus solely on energy harvesting to avoid interference, leading to zero data transmission during those slots, but with potential long-term benefits. To address this, the researchers employ deep reinforcement learning algorithms (PPO and DDPG) along with convex optimization techniques. Simulations show the proposed approach outperforms two baseline strategies significantly. This research offers a promising solution for optimizing transmission strategies in non-orthogonal multiple access networks with energy-constrained secondary user, leading to enhanced long-term throughput and overall system efficiency
Control Method of O/U Tube: Integrating Hydrodynamic Modeling with Genetic Algorithm
No full textThis study conducted experiments in a O/U-Tube Oscillatory Water Tunnel (O/U-tube) equipped to measure velocity profiles at different locations using particle image velocimetry (PIV) technique. To achieve specific flow rates, a dynamic model for the water in O/U-tube was proposed. The parameters of the dynamic model were determined through experimental fitting, and numerical simulations based on the dynamic model. Using Genetic Algorithms Combined with Numerical Simulations to find the optimal control parameters for the flume. Various flow patterns, such as small amplitude waves, inclined waves, and Stokes waves, were generated in the flume as examples of controlled flows.
The experimental results revealed that the velocity profiles in the flume test section did not exhibit similar development across the entire cross-section. Instead, the velocity profiles varied with the change in position, indicating the need to consider three-dimensional effects when studying the velocity profiles in the flume. These three-dimensional characteristics reflect the complexity of the flow field inside the flume, requiring detailed analysis at different positions and directions. It was observed that the accuracy of the dynamic model parameters and the application of the genetic algorithm for flume control were affected by significant errors when the wave period was 12 seconds. Reducing the errors in the dynamic model parameters would enhance the precision of the flume control method.
The findings of this study contribute to a deeper understanding of the characteristics of velocity profiles in the flume and the feasibility and limitations of using a dynamic model and genetic algorithm for flume control. This has significant implications for research and applications in related fields
Philippines' Security Issues:The Impact of the New Cold War on National and Regional Defense Decisions
No full textThe international community has experienced a relatively peaceful atmosphere for nearly a century, with the last truly global war ending in 1945. Due to the invention of unconventional weapons, governments were equipped with a tool that would allow them to deter external threats, with the drawback of having an adversary being equipped with the same capabilities. This deterrence promised total annihilation once triggered, and would end the current world order. With the international structure becoming an ultimate bi-product of World War II, it was definitely bound by theory that the existing leading power would have its time. Riding on this, China, a revisionist power, aspires to take a portion of that power to be exerted in Asia. The Philippines, an archipelagic country that has robust ties with both China and the United States, is caught in the middle of a timely shift of power. The archipelago has shared its fair share of bad sentiments towards both powers, but its defense foreign policy depends on the geopolitical environment surrounding the state-actors. With the group of islands becoming entangled with countries that have now engaged in a New Cold War, it has been forced to take action and proceed to where the best interests of the country lie in. This thesis talks about the implication that the New Cold War between China and the United States of America bring to the table of the Philippines in terms of its decision-making processes and its foreign defense policy. The researcher talks about how the past administrations of the Philippines approach this dilemma and how the country moves forward
Direct Drive Ultrasonic Vibration-Assisted Machining Control System
No full textThe purpose of this thesis is to improve CNC machining by adding a
piezoelectric transducer to provide vibration on the cutting tool, which provides axial
vibration during the machining process to reduce processing time and enhance
machining quality. To improve driving efficiency, driving circuit is installed inside
the ultrasonic machining tool holder to directly drive the transducer and sense its
resonant frequency. This enables the driving frequency to adapt to the load during
machining to track the resonant frequency during machining under different loads,
thereby achieving higher driving efficiency and resulting in vibration amplitudes of
the transducer. In the design of this system, the core technology involves the use of
phase-locked loops and minimum phase seeking methods to ensure resonant
operations of both wireless power transfer and ultrasonic driving that the electrical
frequency and driving frequency are both tuned to the resonant frequency. The control
circuit, serving as the system's core, is implemented in FPGA and CPLD. The
wireless power transmission frequency is approximately 72 kHz, while the operating
frequency of the transducer is around 34 kHz. This system has undergone preliminary
testing on machining alumina ceramic, and the initial results confirm a reduction in
tool wear. Future experiments will be conducted on a wider range of materials to
validate the performance of this system on CNC machining
An Improved Search Economics for Multi-Objective Vehicle Routing Problem with Time Windows
No full textVehicle routing problem (VRP) has long been a popular research topic because the cost of transportation is often closely related to the planning of vehicle routing. In the past, planning the shortest routing distance of the vehicle was often considered as the only optimization objective. However, multiple aspects, beyond the consideration of the shortest route, should be considered in order to reflect the actual situation in the world. Therefore, the multi-objective vehicle routing problem with time windows (MOVRPTW), a well-known multi-objective problem, was developed that strives for minimizing the routing distance and dispatched number of vehicles. When optimizing one objective, the result of another objective may become worse. As a result, considering multiple objectives at the same time, improving the results, and providing a pareto optimal set have been the research topics of interest in recent years, and also the major basis of this thesis. This thesis proposes an improved version of the search economics algorithm which is applied to the multi-objective vehicle routing problem with time windows, creating SE-MOVRPTW. The improved mechanisms of the algorithm which different from the search economics algorithm are solution encoding, region division, and the objective function that create increased differences and lead to more diversity in each iteration during the convergence process. The core concept of the proposed algorithm remains the same as the original search economics algorithm, which portrays the solution space and searches potential areas for solutions. These features can prevent the algorithm from falling into the local optimum, and broaden the algorithm to search the solution space to find a good routing strategy. In order to evaluate the performance of the proposed algorithm SE-MOVRPTW, this thesis compares it with four state-of-the-art multi-objective algorithms. Experimental results show that SEMOVRPTW outperforms all the algorithms compared in this thesis for datasets that contain a wide variety of customers in the sense that the final pareto optimal set of SE-MOVRPTW is closer to the pareto front than all the algorithm compared in thesis study
Discussion on the Operation Mode of Taiwan's Logistics Industry: A Comparative Analysis of Self-Built Fleets and Relying on Driving Fleets
No full textThis study aims to conduct a comparative analysis of self-built fleets and contract-based fleets in Taiwan's logistics industry. Through a literature review, the research covers the development history of the logistics industry, technological innovation, policy environment and regulatory factors, green logistics, and sustainable development. Qualitative research methods are employed, including the design of interview questions and the analysis of customer responses from six Taiwanese logistics companies. The results show significant differences between self-built fleets and contract-based fleets in terms of cost-effectiveness, transport capacity, and service quality. Based on the SWOT analysis, corresponding strategic recommendations are proposed. In the analysis of market trends and industry development trends, we discuss the future development directions, such as green logistics and intelligent logistics. Finally, the study summarizes the research findings, and points out the research limitations and future research directions. This research hopes to provide a clearer direction and strategic choice for Taiwanese logistics companies when facing future competition
Analysis of Heat Transfer Performance on Different Coated Surfaces by Nanofluid Spray Cooling Using a Twin Fluid Nozzle
No full textAs technological advancements lead to more powerful, faster, and sleeker electronics, but the density of internal components raises heat, which can affect performance if not effectively dissipated. Therefore, developing and optimizing advanced cooling technologies like spray cooling is crucial. By adjusting cooling fluids or optimizing surfaces, especially using nanofluids and modifying surface materials and structures, cooling efficiency is significantly enhanced.
The primary objective of this study is to use nanofluid in a twin-fluid nozzle for spray cooling on a smooth copper surface and two types of nanocoatings (single-layer graphene transfer <5nm, graphene deposition <50nm). The experiment used three types of nanoparticles (Al2O3, SiO2, TiO2) mixed with deionized water to prepare the nanofluid, varying three volume concentrations (0.03vol%, 0.06vol%, 0.09vol%). We used a twin-fluid nozzle with an aperture of 1.6mm, and by changing three air to liquid ratios (0.185, 0.246, 0.308), and fixing the spray height at a distance of 50mm from the test surface, we conducted quenching cooling experiments after heating the test surface to 350\uc2\ub0C. The temperature field was measured using thermocouples to calculate the test surface temperature and heat transfer quantity, and to draw quenching cooling curves and transient boiling curves. Additionally, we used a \uce\ubcPIV system to measure the velocity field and the IPI system to measure the particle size of the spray. Through analysis of the thermal and flow fields, we understood the heat transfer enhancement effect.
Experimental results show that the critical heat flux of deionized water and nanofluids increases with the increase of the air to liquid ratio and volume percent concentration. Among all nanofluids, Al2O3 nanofluid demonstrates the best heat transfer enhancement performance, followed by SiO2 nanofluid, while TiO2 nanofluid has the poorest effect. The highest critical heat flux (CHF) occurs when the volume percent concentration of Al2O3 nanofluid is 0.09vol% and the air to liquid ratio is 0.308, reaching 1058.94 W/cm2 on the surface of the graphene transfer coating