Search

Ground-coupled heat pumps are increasingly being utilized to heat and cool buildings. Although it is difficult to size and to predict their behavior and performance, their design can be optimized via simulations. EnergyPlus is a popular energy simulation program for modeling building heating and other energy flows and, since it is organized to consider borehole heat exchangers via the well-known g-functions approach, it can be used advantageously for that purpose. The Capacity Resistance Model is another recent numerical simulation tool devoted to ground and borehole heat exchangers. In this work, two methods to calculate the g-fucntions were analyzed, using as case-study a real office building, whose imbalance between the heat extracted and injected into the ground was found to be appreciable. The energy imbalance involves a ground temperature drift affecting the system efficiency. The results of the EnergyPlus g-functions and the Capacity Resistance Model model approaches were compared. The capacity of the two methodologies to accurately simulate this phenomenon were analysed also with reference to the available building's long-term monitoring data. The analysis showed the importance of using g-functions suitable to reflect the layout of the borehole field, in order to correctly evaluate the energy performance of the entire ground source heat pump system. [ABSTRACT FROM AUTHOR]

This article deals with the problem of classifying and ranking several multivariate populations of interests using the permutation and combination approach providing also an inferential validity of the procedure. The need to define an appropriate classification of populations, i.e., products, services, teaching courses, degree programs, and so on, is very common within many areas of applied research. Many times the populations of interest are multivariate in nature meaning that many aspects of that populations can be simultaneously observed on the same unit/subject. From a statistical point of view, when the response variable of interest is multivariate in nature, the problem may become quite difficult to cope with especially in case of ordered categorical responses, due to the large dimensionality of the parametric space. Nonparametric inference based on the NPC methodology however, allows us to overcome these limitations, without the need of referring to assume any specified random distribution. [ABSTRACT FROM PUBLISHER]