1 |
Extending and formalizing the energy signature method for calibrating simulations and illustrating with application for three California climatesBensouda, Nabil 15 November 2004 (has links)
This thesis extends and formalizes the energy signature method developed by Wei et al. (1998) for the rapid calibration of cooling and heating energy consumption simulations for commercial buildings. This method is based on the use of "calibration signatures" which characterize the difference between measured and simulated performance.
By creating a library of shapes for certain known errors, clues can be provided to the analyst to use in identifying what simulation input errors may be causing the discrepancies. These are referred to as "characteristic signatures". In this thesis, sets of characteristic signatures are produced for the climates typified by Pasadena, Sacramento and Oakland, California for each of the four major system types: single-duct variable-air-volume, single-duct constant-volume, dual-duct variable-air-volume and dual-duct constant-volume.
A detailed step-by-step description is given for the proposed methodology, and two examples and a real-world case study serve to illustrate the use of the signature method.
|
2 |
Kalibrering och validering av en IDA ICE modell : Ett flerbostadshus från 1970-talets miljonprogramÖstlin, Olof, Sjödén Havik, Mikaela January 2020 (has links)
Aktuellt examensarbete är en fallstudie som utförts på en miljonprogramsbyggnad i Andersberg ägd av AB Galvegårdarna vilka även är uppdragsgivarna. Då miljonprogramsbyggnader är dåligt värmeisolerade och har stora värmeläckage är det idag av stort intresse att se över eventuella förbättringsåtgärder då dessa byggnader har en potential att minska energianvändningen med 50 procent. Syftet med detta projekt är att få en kalibrerad och validerad modell med hjälp av den BES-modell (Building Energy System) som kommer att tas fram i detta examensarbete. Genom litteraturstudie, platsbesök samt inhämtning av protokoll, ritningar och uppmätta data för byggnaden kunde modellen skapas och kalibreras i simuleringsprogrammet IDA Indoor Climate and Energy. Ritningar och data tillhandahölls från AB Gavlegårdarna och platsbesök gjordes för att komplettera dessa genom att göra mätningar av temperaturer i de allmänna utrymmena. På plats kunde även byggnadens mått mätas för att säkerställa att byggnaden inte hade uppdaterats sedan tilldelade ritningarna skapats. När samtlig information ansågs ha införskaffats lades all data in i IDA ICE där även en modell av byggnaden byggdes upp. För köldbryggorna användes simuleringsverktyget COMSOL Multiphysics för att ta fram de enskilda köldbryggornas psi-värden vilka därefter användes som input i byggnadsmodellen i IDA ICE. Den kalibrerade modellen framtagen i detta projekt visade sig stämma med uppmätta värden så när som på +- 10% då den ställdes mot det uppmätta energibehovet för byggnaden. Mot en nyutvecklad energisignatursmodells byggnadsförlustkoefficient blev skillnaden 19.6% vilket kan bero på att fel från simuleringsverktygen samt osäkerheter angående omätbara parametrar. Slutsastsen utav detta arbete var att ”performance gap” även inträffade på den framtagna modellen i detta arbete. Vilket verkar vara svårt att undvika. På platsbesöket upptäcktes vattensamlingar på taket på byggnaden vilket var en förvåning för författarna då det fanns dokument som sade att ytskiktet var bytt 2015 och att det fanns indikeringar på att detta kunde få omfattande konsekvenser om det inte åtgärdas vilket tas upp under diskussion Framtida arbete om varför boendes bettendemönster underskattas vore något att gå vidare med i framtida studier för att kunna minska ”performance gap” på BES modeller. / This thesis is a case study carried out on a Million Homes Program (MHP) building in Andersberg owned by AB Galvegårdarna, whom are also the clients. Since MHPbuildings are poorly insulated and have major heat leaks, it is of great interest today to investigate any improvement measures as these buildings have a potential to reduce their energy use by 50 percent. This is possible with the help of the calibrated model in a building energy performance simulation (BEPS) tool, which is the purpose of developing in this thesis. Through a literature study, visit in the building and gathering protocols, drawings and measured data, a model could be built and calibrated in IDA Indoor Climate and Energy was started. Drawings and data were provided from AB Gavlegårdarna and site visits were made to supplement these by taking measurements of temperatures in the common areas. On site, the dimensions of the building were also measured to ensure that the building had not been upgraded since the assigned drawings were created. When all the information was considered to have been obtained, all data was entered into IDA ICE where a model of the building was also built up. For the thermal bridges, the COMSOL Multiphysics simulation tool was used to generate their individual linear heat loss coefficient which were used as input in the building model of IDA ICE. The calibrated model developed in this project turned out to have a deviation of 10 % against annual district heating energy. The simulated building heat loss coefficient differed with 19.6 % compared to the one produced with a newly developed energy signature method for the corresponding year which may be caused by errors in the simulation tools and uncertainty concerning immeasurable parameters. The final conclusion of this work was that the performance gap also occurred on this model developed in this work, which seems to be hard to avoid. During the site visit, water collections on the roof of the building were discovered which was a surprise to the authors as there were documents that said that the surface layer had been changed in 2015 and that there were indications that this could have significant consequences if not addressed which is mentioned in the chapter of discussion. Future work on why residents’ behavioral patterns are underestimated would be something to continue with in future studies in order to reduce the “performance gap” in BES models.
|
3 |
Quantitative Methods of Statistical ArbitrageBoming Ning (18414465) 22 April 2024 (has links)
<p dir="ltr">Statistical arbitrage is a prevalent trading strategy which takes advantage of mean reverse property of spreads constructed from pairs or portfolios of assets. Utilizing statistical models and algorithms, statistical arbitrage exploits and capitalizes on the pricing inefficiencies between securities or within asset portfolios. </p><p dir="ltr">In chapter 2, We propose a framework for constructing diversified portfolios with multiple pairs trading strategies. In our approach, several pairs of co-moving assets are traded simultaneously, and capital is dynamically allocated among different pairs based on the statistical characteristics of the historical spreads. This allows us to further consider various portfolio designs and rebalancing strategies. Working with empirical data, our experiments suggest the significant benefits of diversification within our proposed framework.</p><p dir="ltr">In chapter 3, we explore an optimal timing strategy for the trading of price spreads exhibiting mean-reverting characteristics. A sequential optimal stopping framework is formulated to analyze the optimal timings for both entering and subsequently liquidating positions, all while considering the impact of transaction costs. Then we leverages a refined signature optimal stopping method to resolve this sequential optimal stopping problem, thereby unveiling the precise entry and exit timings that maximize gains. Our framework operates without any predefined assumptions regarding the dynamics of the underlying mean-reverting spreads, offering adaptability to diverse scenarios. Numerical results are provided to demonstrate its superior performance when comparing with conventional mean reversion trading rules.</p><p dir="ltr">In chapter 4, we introduce an innovative model-free and reinforcement learning based framework for statistical arbitrage. For the construction of mean reversion spreads, we establish an empirical reversion time metric and optimize asset coefficients by minimizing this empirical mean reversion time. In the trading phase, we employ a reinforcement learning framework to identify the optimal mean reversion strategy. Diverging from traditional mean reversion strategies that primarily focus on price deviations from a long-term mean, our methodology creatively constructs the state space to encapsulate the recent trends in price movements. Additionally, the reward function is carefully tailored to reflect the unique characteristics of mean reversion trading.</p>
|
Page generated in 0.0556 seconds