• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 68
  • 20
  • 12
  • 8
  • 3
  • 2
  • 2
  • 1
  • Tagged with
  • 143
  • 16
  • 14
  • 14
  • 11
  • 11
  • 11
  • 10
  • 10
  • 10
  • 10
  • 10
  • 9
  • 9
  • 9
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
141

Deep Reinforcement Learning for Building Control : A comparative study for applying Deep Reinforcement Learning to Building Energy Management / Djup förstärkningsinlärning för byggnadskontroll : En jämförande studie för att tillämpa djup förstärkningsinlärning på byggnadsenergihushållning

Zheng, Wanfu January 2022 (has links)
Energy and environment have become hot topics in the world. The building sector accounts for a high proportion of energy consumption, with over one-third of energy use globally. A variety of optimization methods have been proposed for building energy management, which are mainly divided into two types: model-based and model-free. Model Predictive Control is a model-based method but is not widely adopted by the building industry as it requires too much expertise and time to develop a model. Model-free Deep Reinforcement Learning(DRL) has successful applications in game-playing and robotics control. Therefore, we explored the effectiveness of the DRL algorithms applied to building control and investigated which DRL algorithm performs best. Three DRL algorithms were implemented, namely, Deep Deterministic Policy Gradient(DDPG), Double Deep Q learning(DDQN) and Soft Actor Critic(SAC). We used the building optimization testing (BOPTEST) framework, a standardized virtual testbed, to test the DRL algorithms. The performance is evaluated by two Key Performance Indicators(KPIs): thermal discomfort and operational cost. The results show that the DDPG agent performs best, and outperforms the baseline with the saving of thermal discomfort by 91.5% and 18.3%, and the saving of the operational cost by 11.0% and 14.6% during the peak and typical heating periods, respectively. DDQN and SAC agents do not show a clear advantage of performance over the baseline. This research highlights the excellent control performance of the DDPG agent, suggesting that the application of DRL in building control can achieve a better performance than the conventional control method. / Energi och miljö blir heta ämnen i världen. Byggsektorn står för en hög andel av energiförbrukningen, med över en tredjedel av energianvändningen globalt. En mängd olika optimeringsmetoder har föreslagits för Building Energy Management, vilka huvudsakligen är uppdelade i två typer: modellbaserade och modellfria. Model Predictive Control är en modellbaserad metod men är inte allmänt antagen av byggbranschen eftersom det kräver för mycket expertis och tid för att utveckla en modell. Modellfri Deep Reinforcement Learning (DRL) har framgångsrika tillämpningar inom spel och robotstyrning. Därför undersökte vi effektiviteten av DRL-algoritmerna som tillämpas på byggnadskontroll och undersökte vilken DRL-algoritm som presterar bäst. Tre DRL-algoritmer implementerades, nämligen Deep Deterministic Policy Gradient (DDPG), Double Deep Q Learning (DDQN) och Soft Actor Critic (SAC). Vi använde ramverket Building Optimization Testing (BOPTEST), en standardiserad virtuell testbädd, för att testa DRL-algoritmerna. Prestandan utvärderas av två Key Performance Indicators (KPIs): termiskt obehag och driftskostnad. Resultaten visar att DDPG-medlet presterar bäst och överträffar baslinjen med besparingen av termiskt obehag med 91.5% och 18.3%, och besparingen av driftskostnaden med 11.0% och 14.6% under topp och typisk uppvärmning perioder, respektive. DDQN- och SAC-agenter visar inte en klar fördel i prestanda jämfört med baslinjen. Denna forskning belyser DDPG-medlets utmärkta prestanda, vilket tyder på att tillämpningen av DRL i byggnadskontroll kan uppnå bättre prestanda än den konventionella metoden.
142

Investigating the Relationship between Binocular Disparity, Viewer Discomfort, and Depth Task Performance on Stereoscopic 3D Displays

McIntire, John Paul 04 June 2014 (has links)
No description available.
143

Health for community dwelling older people : trends, inequalities, needs and care in rural Vietnam

Van Hoi, Le January 2011 (has links)
Background InVietnam, the proportion of people aged 60 and above has increased rapidly in recent decades. The majority live in rural areas where socioeconomic status is more disadvantaged than in urban areas.Vietnam’s economic status is improving but disparities in income and living conditions are widening between groups and regions. A consistent and emerging danger of communicable diseases and an increase of non-communicable diseases exist concurrently. The emigration of young people and the impact of other socioeconomic changes leave more elderly on their own and with less family support. Introduction of user fees and development of a private sector improve the coverage and quality of health care but increase household health expenditures and inequalities in health care. Life expectancy at birth has increased, but not much is known about changes during old age. There is a lack of evidence, particularly in rural settings, about health-related quality of life (HRQoL) among older people within the context of socioeconomic changes and health-sector reform. Knowledge of long-term elderly care needs in the community and the relevant models are still limited. To provide evidence for developing new policies and models of care, this thesis aimed to assess general health status, health care needs, and perspectives on future health care options for community-dwelling older people. Methods An abridged life table was used to estimate cohort life expectancies at old age from longitudinal data collected by FilaBavi DSS during 1999-2006. This covered 7,668 people aged 60 and above with 43,272 person-years. A 2007 cross-sectional survey was conducted among people aged 60 and over living in 2,240 households that were randomly selected from the FilaBavi DSS. Interviews used a structured questionnaire to assess HRQoL, daily care needs, and willingness to use and to pay for models of care. Participant and household socioeconomic characteristics were extracted from the 2007 DSS re-census. Differences in life expectancy are examined by socioeconomic factors. The EQ-5D index is calculated based on the time trade-off tariff. Distributions of study subjects by study variables are described with 95% confidence intervals. Multivariate analyses are performed to identify socioeconomic determinants of HRQoL, need of support, ADL index, and willingness to use and pay for models of care. In addition, four focus group discussions with the elderly, their household members, and community association representatives were conducted to explore perspectives on the use of services by applying content analysis. Results Life expectancy at age 60 increased by approximately one year from 1999-2002 to 2003-2006, but tended to decrease in the most vulnerable groups. There is a wide gap in life expectancy by poverty status and living arrangement. The sex gap in life expectancy is consistent across all socioeconomic groups and is wider among the more disadvantaged populations.  The EQ-5D index at old age is 0.876. Younger age groups, position as household head, working, literacy, and belonging to better wealth quintiles are determinants of higher HRQoL. Ageing has a primary influence on HRQoL that is mainly due to reduction in physical (rather than mental) functions. Being a household head and working at old age are advantageous for attaining better HRQoL in physical rather than psychological terms. Economic conditions affect HRQoL through sensory rather than physical functions. Long-term living conditions are more likely to affect HRQoL than short-term economic conditions. Dependence in instrumental or intellectual activities of daily living (ADLs) is more common than in basic ADLs. People who need complete help are fewer than those who need some help in almost all ADLs. Over two-fifths of people who needed help received enough support in all ADL dimensions. Children and grand-children are confirmed to be the main caregivers. Presence of chronic illness, age groups, sex, educational level, marital status, household membership, working status, household size, living arrangement, residential area, household wealth, and poverty status are determinants of the need for care. Use of mobile teams is the most requested service; the fewest respondents intend to use a nursing centre. Households expect to use services for their elderly to a greater extent than did the elderly themselves. Willingness to use services decreases when potential fees increase. The proportion of respondents who require free services is 2 to 3 times higher than those willing to pay full cost. Households are willing to pay more for day care and nursing centres than are the elderly. The elderly are more willing to pay for mobile teams than are their households. ADL index, age group, sex, literacy, marital status, living arrangement, head of household status, living area, working status, poverty and household wealth are factors related to willingness to use services.   Conclusions                                                                                         There is a trend of increasing life expectancy at older ages in ruralVietnam. Inequalities in life expectancy exist between socioeconomic groups. HRQoL at old age is at a high level, but varies substantially according to socioeconomic factors. An unmet need of daily care for older people remains. Family is the main source of support for care. Need for care is in more demand among disadvantaged groups.  Development of a social network for community-based long-term elderly care is needed. The network should focus on instrumental and intellectual ADLs rather than basic ADLs. Home-based care is more essential than institutionalized care. Community-based elderly care will be used and partly paid for if it is provided by the government or associations. The determinants of elderly health and care needs should be addressed by appropriate social and health policies with greater targeting of the poorest and most disadvantaged groups. Building capacity for health professionals and informal caregivers, as well as support for the most vulnerable elderly groups, is essential for providing and assessing the services. / Aging and Living Conditions Program / Vietnam-Sweden Collaborative Program in Health, SIDA/Sarec

Page generated in 0.0323 seconds