دانلود مقاله ISI انگلیسی شماره 26366
عنوان فارسی مقاله

اعتبار سنجی تجربی و تجزیه و تحلیل حساسیت از مدل BES-HAM توأم

کد مقاله سال انتشار مقاله انگلیسی ترجمه فارسی تعداد کلمات
26366 2010 17 صفحه PDF سفارش دهید محاسبه نشده
خرید مقاله
پس از پرداخت، فوراً می توانید مقاله را دانلود فرمایید.
عنوان انگلیسی
Experimental validation and sensitivity analysis of a coupled BES–HAM model
منبع

Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)

Journal : Building and Environment, Volume 45, Issue 10, October 2010, Pages 2202–2217

کلمات کلیدی
مدل - مطالعه اعتبار سنجی - تجزیه و تحلیل حساسیت - بافر رطوبت -
پیش نمایش مقاله
پیش نمایش مقاله اعتبار سنجی تجربی و تجزیه و تحلیل حساسیت از مدل BES-HAM توأم

چکیده انگلیسی

In this paper the ability of a coupled BES–HAM model to reproduce realistic data is evaluated by comparing numerical results with measured data from a climatic chamber experiment. Calcium silicate plates are introduced into a test room and a small calcium silicate sample is installed in one of the walls. The response of the test room to relative humidity variations of the supply air is evaluated, while the supply air temperature is kept constant. The measurements confirm that due to the presence of hygroscopic materials in the test room, the relative humidity variations in the room are damped. The calculated temperature and relative humidity in the middle of the test room are well within the uncertainty interval of the measurements. On the other hand the coupled model predicts a larger damping and phase shift of the relative humidity variations inside the sample, yet the agreement between the calculated and the measured temperatures in the sample proves to be good. Finally, a sensitivity analysis is performed to evaluate the dependence of the numerical results on the uncertainty of the input parameters. It is demonstrated that by using a lower vapour resistance factor for the calcium silicate material, the agreement between the measured and calculated data is improved.

مقدمه انگلیسی

Hygroscopic materials are able to absorb water vapour as the relative humidity increases and release water vapour if the relative humidity drops. As a result the indoor relative humidity peaks are damped and therefore these materials may contribute to a more stable indoor climate [1] and [2]. Apart from the building envelope, the hygroscopic content of a room or a building also includes furniture and furnishings which often consist of porous materials. While the envelope is mainly characterized by materials such as uncovered concrete, wooden floors, brick walls, etc, furnishings usually consist of lightweight materials such as papers, books, textiles, untreated wood, which are part of the interior of most buildings (e.g. dwellings, libraries…). Since hygroscopic materials reduce relative humidity variations, not including them when predicting the indoor relative humidity will lead to incorrect estimations. Yet an accurate prediction of the indoor relative humidity is indispensable for numerous building applications, ranging from the assessment of the indoor climate to guarantee the conservation of valuable objects, e.g. in museums and libraries [3], [4] and [5], to preventing moisture-induced damage to envelopes (e.g. by mould and condensation). Furthermore, literature proves that a correct prediction of the indoor relative humidity is important to evaluate and size humidity-controlled HVAC systems such as indirect evaporative cooling or humidity-controlled ventilation [6] and [7]. Currently available multizone Building Energy Simulation (BES) tools, e.g. TRNSYS, EnergyPlus [8] and [9], generally focus on the prediction of thermal comfort and energy use and simplify moisture buffering in hygroscopic materials. They enable to calculate the temperature in buildings with respect to the outside climate, occupancy and the interaction with the air handling unit. So far simplified models, such as the effective capacitance model and the effective moisture penetration depth model in TRNSYS, are not suited to describe moisture buffering in a detailed way since they assume isothermal conditions, periodic moisture loads and constant material properties [10]. On the other hand transient HAM (Heat, Air and Moisture) models allow to describe the combined heat and moisture transfer processes in complex porous building structures in detail and are appropriate to account for the hygrothermal interaction between the building air and the porous surfaces. For applications on building scale, a coupled BES–HAM model is interesting since it allows for long term calculations and complex building geometries. In contrast, on object scale, a coupled CFD–HAM model allows the assessment of the microclimate around valuable artefacts [11]. Previously, a transient 1D HAM model was integrated into the multizone BES model TRNSYS [12] and [13]. The coupled model enables to predict the boundary conditions for HAM in detail, while on the other hand the effect of the latent heat on the room conditions is taken into account. To evaluate whether this coupled TRNSYS–HAM model is able to reproduce realistic data, a validation with measured data from a climatic chamber experiment is required. Therefore, the input parameters, e.g. boundary conditions, initial conditions, exposed hygroscopic surface…, necessary in the numerical model have to be precisely measured during the experiments. Additionally, material properties of hygroscopic materials have to be well known. Previous experiments in literature sometimes suffer from the problem that some of the measured data are insufficient or lacking. For instance Svennberg reported moisture buffering experiments in a room-size test cell in which furniture and furnishings were successively introduced [2]. While the room temperature was kept constant the humidity response of the test cell was measured. Uncertainties were due to some unexpected moisture buffering in the test room, a not well registered exposed hygroscopic area (i.e. of paper and books), uncertain material properties…. These uncertainties must be excluded in order to obtain a good agreement between the measurements and the numerical results. Another single room buffering experiment was reported by Fazio et al. In this test campaign the moisture buffering behaviour of two internal finishing materials (i.e. uncoated gypsum and pine paneling) and of furniture (i.e. bookshelf with books and full furnished room) was investigated [14]. Since a general assumption of BES models is that state variables (i.e. temperature and vapour pressure) are equal in every point of the room, a well-mixed air condition in the test chamber has to be guaranteed in order to validate the coupled BES-HAM model. This paper focuses on an experimental validation of a coupled TRNSYS–HAM model by means of climatic chamber experiments. The paper starts with a brief description of the coupled BES–HAM model. Next, a description of the test facility and the experiments carried out in the climatic chamber is given. The experimental results are discussed. Finally, the sensitivity of the numerical results to the uncertainty of different parameters in the experiment is investigated.

نتیجه گیری انگلیسی

In order to check whether a coupled BES–HAM model is able to reproduce realistic results, the model is validated with measured data from a climatic chamber experiment. The response of a test room to humidity variations of the supply air is evaluated when calcium silicate plates are added to the room. During the experiments the temperature of the supply air is kept constant. Temperature and relative humidity both in the test room and on different depths in a calcium silicate sample are registered during the experiment. Preliminary measurements have demonstrated that the well-mixed air assumption is ensured during the tests. Consequently the experiments are appropriate for the validation of a coupled BES–HAM model. A comparison between a hygroscopic and a non-hygroscopic experiment confirms that the calcium silicate plates are able to damp the relative humidity variations inside the test room. The agreement between the measured and predicted temperature and relative humidity in the test room is found to be satisfactory. The amplitude of the relative humidity variations is slightly underestimated by the calculations in the hygroscopic case. At 10 mm inside the sample, the measured and calculated relative humidity and temperature agree well. Deeper in the sample the predicted relative humidity reduction is overestimated by the model. On the other hand, the predicted and measured temperature agree well. A sensitivity analysis allows to determine which parameters have the largest influence on the outcome of the BES–HAM model. It reveals that the water vapour resistance factor used in the calculations may be overestimated. A lower μ-value is found to enhance the agreement between measurements and calculations, both in the test room and inside the sample. Still the measured RH variation inside the material sample is larger than the RH variation predicted by the simulations. The currently implemented HAM model neglects the effect of hysteresis and consequently the vapour permeability is determined by the moisture content associated to the absorption isotherm. As a result, during the desorption cycle, the model underpredicts the moisture content leading to an overestimation of the vapour resistance and hence smaller RH variations inside the material sample. In Fig. 11, some of the results of the sensitivity analysis are summarized. A differentiation is made between the RH in the test room (Fig. 11a) and the RH inside the sample (at 10 mm, Fig. 11b). In these figures the size of the bars is a measure for the relative impact of the tested parameters on the predicted RH amplitude compared to the reference simulation. Fig. 11 shows that the moisture related parameters (i.e. mass transfer coefficient β, sorption isotherm and vapour resistance factor μ) are clearly dominant over the thermal parameters when predicting the RH course in the test room and inside the test sample. Full-size image (49 K) Fig. 11. Relative difference of predicted RH amplitude (a) in the test room and (b) at 10 mm inside the sample compared to the reference simulation (light gray bar: lower value and dark gray bar: upper value used in the sensitivity analysis).

خرید مقاله
پس از پرداخت، فوراً می توانید مقاله را دانلود فرمایید.