scispace - formally typeset
Search or ask a question

How do I increase the voltage on my ATX power supply? 

Answers from top 8 papers

More filters
Papers (8)Insight
Results demonstrate CSs with more capacity farther power supply may increase loss and voltage degradation.
The concept of this power supply is readily extendable to more or fewer voltage levels or higher or lower output currents or voltages than the particular supply described.
Our study shows that the key factor in achieving power saving is including the most comfortable supply voltage in the scaling process.
This paper presents a novel photovoltaic system suitable for disparate population centers in the developing world utilising the ubiquitous ATX computer power supply.
In addition, its twofold increase in SOA extension can improve the performance of circuit designs for switching power supply applications.
The design provides a high-performance solution to on-chip adaptive power supply designs.
The analysis demonstrates system-level limitations of supply voltage reduction.
Proceedings ArticleDOI
A. Siddique, G.S. Yadava, Bhim Singh 
19 Sep 2004
91 Citations
The analysis shows the effects of resulting losses during unbalanced voltage supply.

See what other people are reading

What are existing transformer variations?
4 answers
Existing transformer variations include Longformer, Transformer-XL, Big Bird, Star transformer, and various X-formers. Longformer is suitable for document-level tasks due to its lower complexity. Transformer-XL addresses context fragmentation to enhance accuracy. Big Bird replaces the whole attention mechanism with Block Sparse Attention, excelling in tasks with long text sequences like summaries and Q&A. Star-Transformer reduces time complexity and performs well in synthetic datasets. Additionally, recent works have proposed variations in positional encodings, such as relative position encodings, to improve performance on tasks like GLUE, XTREME, and WMT benchmarks while saving computation costs. A comprehensive review of various X-formers has been conducted, categorizing them based on architectural modifications, pre-training, and applications.
What are the methods to speed up training/inference for diffusion models?
5 answers
To accelerate training and inference for diffusion models, several methods have been proposed. One approach involves introducing momentum into the diffusion process to enhance both training and sampling speed, resembling the momentum SGD optimization technique. Another method focuses on utilizing masked transformers during training to reduce the training cost significantly, achieving comparable performance with less training time. Additionally, a strategy involves partitioning the learning task by employing independent networks to model the score evolution at different time points, leading to faster training and improved density estimation performance. These methods collectively aim to address the slow and costly training and sampling issues typically associated with diffusion models.
What are papers on end to end spoken language understanding?
5 answers
Papers on end-to-end spoken language understanding (SLU) include studies on contextual biasing techniques to enhance speech recognition of rare words, such as the tree-constrained pointer generator (TCPGen) and slot probability biasing (SPB) mechanisms. Additionally, non-autoregressive (NAR) approaches have been explored for joint automatic speech recognition (ASR) and SLU tasks, utilizing Conformer encoders and connectionist temporal classification (CTC) for transcription, along with bidirectional Transformers for refinement. These NAR systems have shown improved SLU metrics and competitive ASR accuracy compared to autoregressive models, with significantly faster decoding speeds, making them promising for efficient SLU systems.
What are the types of electric transformer in power systems?
5 answers
Electric transformers in power systems can be categorized into various types based on their functionality and design. These types include large scale power transformers crucial for interconnections, transformers used for voltage and current level modifications without changing frequency, single-stage AC-AC converters for direct conversion, transformers essential for AC power distribution systems with the ability to adjust voltages for long-range transmission, and variable frequency transformers (VFT) designed to control power flow and dampen inter-area oscillations in interconnected power systems. Each type serves a specific purpose in power systems, ranging from basic voltage transformation to advanced control mechanisms for enhancing system stability and efficiency.
What ist GAN, RNN and CNN?
5 answers
Generative Adversarial Networks (GANs) are a type of neural network consisting of a generator and a discriminator that compete to generate new instances resembling the training data distribution. Recurrent Neural Networks (RNNs) are designed for sequential data processing, allowing information to persist. Convolutional Neural Networks (CNNs) are specialized in processing grid-like data, such as images, by using convolutional layers to extract features and make them ideal for tasks like image classification and object detection. In summary, GANs focus on generating new data instances, RNNs are suitable for sequential data analysis, and CNNs excel in processing grid-like data, particularly in image-related tasks.
How does mathematical optimization modelling contribute to minimizing power losses in electricity distribution ?
5 answers
Mathematical optimization modeling plays a crucial role in minimizing power losses in electricity distribution systems. Various studies have proposed optimization models to address this issue. For instance, research has focused on utilizing techniques like Genetic Algorithms (GA) and Newton-Raphson (NR) to minimize line losses by implementing grid strategies such as real power sharing, reactive power injection, and transformer tap changing. Additionally, the application of Second Order Cone Program (SOCP) relaxation has been suggested to solve the nonconvex and NP-hard optimization problem of the Alternating Current Optimal Power Flow (ACOPF), contributing to reducing active power losses in distribution systems. Moreover, the use of mixed-integer convex formulations and optimization models based on electrical momentum concepts have shown effectiveness in minimizing power losses in asymmetric distribution grids. These optimization techniques aid in enhancing system reliability and performance by reducing power losses and improving voltage profiles.
How effective are AI-prompted frameworks in improving the quality and efficiency of research paper writing?
5 answers
AI-prompted frameworks have shown significant potential in enhancing the quality and efficiency of research paper writing. These frameworks, such as GPT-3, can assist researchers in organizing their thoughts, generating drafts, and improving the overall caliber of scientific work. By utilizing AI technologies like machine learning and natural language processing, researchers can benefit from tools that streamline the writing process and enhance critical analysis, particularly in literature reviews and language style. Despite the advancements, it is crucial to acknowledge the limitations and challenges posed by AI, such as biases, ethical concerns, and the need for human creativity. Integrating AI in research writing has the potential to revolutionize academic work, but it is essential to ensure transparency, ethics, and reliability in AI-driven technologies for scholarly endeavors.
Are there examples of unsupervised machine learning used in gait analysis?
5 answers
Unsupervised machine learning has been applied in gait analysis, particularly in the context of Unsupervised Gait Recognition (UGR). Research has introduced methods like cluster-based contrastive learning and Selective Fusion techniques to address challenges in UGR. Additionally, studies have explored Unsupervised Domain Adaptive Gait Recognition (UDA-GR) using uncertainty estimation and regularization methods to adapt gait identification models from indoor to outdoor scenes. Furthermore, the use of self-supervised training regimes with transformer models like ViT, CaiT, CrossFormer, Token2Token, and TwinsSVT has shown promise in gait recognition tasks, enabling learning without costly manual annotations. These approaches showcase the effectiveness of unsupervised machine learning in enhancing gait analysis methodologies.
What are the limitations of using spatial or temporal deepfake detectors for deepfake videos?
5 answers
Spatial or temporal deepfake detectors face limitations due to various factors. Current methods struggle with open data, re-compression susceptibility, and reliance on pixel-level abnormalities. Additionally, the spatio-temporal modeling of existing detectors lacks in capturing subtle temporal changes, concealing minor inconsistencies, and addressing transient inconsistencies alongside persistent motions. Moreover, detectors are vulnerable to attacks from advanced generative models like Denoising Diffusion Models, leading to decreased accuracy without visible changes. Furthermore, detecting low-quality deepfake videos remains a challenge, as many models degrade in performance with heavily compressed or low-resolution videos. These limitations highlight the need for more robust and comprehensive approaches to effectively detect deepfake videos.
How do distribution stations contribute to the efficiency and reliability of the power system?
5 answers
Distribution stations play a crucial role in enhancing the efficiency and reliability of the power system. They are essential components responsible for voltage transformation and power distribution within the distribution network. By implementing strategies like spare transformer sharing policies, distribution stations can reduce the number of spare transformers required while maintaining high reliability levels. Moreover, integrating distributed generation (DG) units at optimal locations within the distribution system can significantly enhance reliability by reducing outage durations and improving system performance. Additionally, the incorporation of energy storage systems can help balance power supply reliability with economic considerations, especially in situations where locally generated power exceeds demand, thus preventing voltage issues and network losses. These measures collectively contribute to ensuring a stable and reliable operation of the power system.
What are the current trends in on-device AI?
5 answers
Current trends in on-device AI involve optimizing AI models for edge devices to enhance sustainability, focusing on deploying Transformer models for time-series tasks on embedded hardware like FPGAs. Additionally, there is a shift towards moving intelligence services from cloud servers to on-device systems, offering benefits like privacy preservation, reduced latency, and cost savings. To address the challenges posed by varying hardware resources, a stream pipeline framework called NNStreamer is being utilized for on-device AI systems, aiming to expand device types and applications while ensuring atomic and re-deployable AI services shared among devices from different vendors. These trends highlight the ongoing efforts to make on-device AI more efficient, sustainable, and widely accessible across diverse hardware platforms.