Dynamic simulation of calibrated energy model by controlling indoor temperature to maximize the energy efficiency

Abstract

The residential sector is responsible for a large part of energy consumption. With the world's growing population, the need for buildings is increasing day by day. As a result, the amount of energy demanded by the building sector is also increasing. For these reasons, energy efficiency studies in the building sector, which has a large share in total energy consumption, are of great importance. Therefore, there is a need extensive research that identify the energy consumption profile of building in terms of heating, cooling, lighting, and their sources. The aim of the study is dynamically simulating a building energy model to maximise the energy efficiency. This can be achieved through a calibrated energy model with fifteen-minute measured data by eliminating over-heated and over-cooled periods. It is also to provide additional energy savings by managing the energy consumption of the different functional zones of the building in terms of occupational scheduling, building systems scheduling and sensible temperature comfort conditions. The case study building was selected as Energy Institute building located in Ayazağa Campus of Istanbul Technical University. The Energy Institute building was built in 1963 with 3 floors. It has a total area of 4272 m2 . The building is heated with a central radiator system using natural gas as fuel. As an educational building, it consists of classrooms, offices, and laboratories. A systematic methodology consisting of five main parts as is proposed in the thesis. The proposed steps are followed in order and their applications are realized on the building selected as a case study. These steps are the creation of a three-dimensional model of the building, the creation of the energy model of the building, the collection of measured data from the building, the calibration of the building energy model and the improvements on the existing energy profile of the building. In the first chapter defined in the methodology of the thesis, the three-dimensional model of the building was created using Autodesk Revit software using the floor plans of the building obtained from the university administration. Since the information in the floor plans was not sufficient, observations were made in the building and visuals of the building were utilized to detail the model. Then, in order to define the areas inside the building according to their functional uses, observations were made again, and each area was transferred to the model according to its usage function. There are seven different areas in the building with different usage functions. These are offices, classrooms, laboratories, conditioned corridors, unconditioned corridors, restrooms, and storage areas. These zones were modelled separately according to heating, cooling, lighting systems, equipment, and usage profiles. In the second part, Design Builder software was preferred to convert 3D model to building energy model because it allows easy transfer of the three-dimensional model, zone-based parameter description, easy modelling of building systems and the opportunity to obtain simulation results in desired periods. To provide data input to the energy model, the climate data of the region where the building is located was accessed from Energy Plus software. The thermal properties of the building envelope elements (exterior walls, roof, foundation, doors, and windows) were determined from the building drawings and observations made in the building. The properties of all systems used in the building were also obtained from technical documents and observations. The heating of the building is provided by a central radiator system, and there are split air conditioners only in the offices as a cooling system. Lighting throughout the building is provided by fluorescent lamps. There are also equipment and computers in classrooms, laboratories, and offices. In addition to these, the usage profiles of all zones in the building have been determined and the time intervals and how many hours these areas are actively used are defined in the model. In the third section, periodic measurements were made to be used both in the calibration of the energy model and in the improvement of the existing energy profile. Monthly natural gas and electricity consumption of building for the base year 2022 was measured. In the fourth step, the calibration of the energy model, the energy model described in detail in chapter two was first simulated. This simulation is called base line scenario in the thesis. The monthly natural gas and electricity consumptions obtained from the simulation were analysed and compared with the actual consumption of the building. As a result of the comparison, it was seen that although the simulation and real results are relatively close to each other in the summer months, the consumption values obtained from the simulation are much higher than the real values, especially in periods other than summer months. While the annual energy consumption is 307 kWh/m2 in the simulation results, the actual consumption result is 159 kWh/m2 . In order to reduce the differences between the model and the actual data, first of all, Scenario 2 energy model was created by fine-tuning the some model parameters. The parameters modified in base line scenario are the thermal properties of the roof, windows and doors, infiltration rate and power density of the equipment in the building. With the changes applied in Scenario 2, the energy consumption is reduced to 222 kWh/m2 . Since the difference between the measured and simulated consumption is still not acceptable, Scenario 3 energy model was created and usage profiles were created separately for functional zones. The planning of heating, cooling, lighting systems and equipment were taken into consideration in determining the working hours based on zones. With the changes made with Scenario 3, energy consumption was reduced to 152 kWh/m2 . Finally, MBE and CVRMSE values were calculated as a performance indicators to see how much the results of each scenario differ from the actual results. For the model to be considered calibrated, the MBE and CVRMSE values should be within certain ranges recommended by different institutions. Since the calibration process is performed monthly in the study, the MBE and CVRMSE values calculated for all months are shown. In the calculations made over total consumption, the MBE values was calculated as -92.60% in base line scenario, -39.11% in Scenario 2 and 4.67% in Scenario 3. Also, the CVRMSE values was calculated as 125.26% in base line scenario, 48.39% in Scenario 2 and 14.67% in Scenario 3. Based on these data, the energy model created in Scenario 3 is calibrated. In the fifth section, 15-minute measured temperature and humidity data in different zones were investigated to illustrate the effect of dynamic simulation on building energy consumption. Minimum, average and maximum temperatures of each zone during the day in a month during the heating season were determined. In addition, since the desired temperature value in the zones is also known, these four different temperatures values are defined as the heating system setpoint temperatures and the energy consumption profiles of the building at different temperatures are created. Finally, these energy consumption profiles were used to identify the periods of overheating and periods when comfort conditions were not met. The potential energy savings that can be achieved by largely eliminating these periods are calculated. Thus, it has been shown that 17% energy efficiency can be achieved with an automated heating system that controls the indoor temperatures in the zones and ensures that the temperature is always kept at the desired level. In addition, by defining the maximum temperatures recorded in the zones inside the building as the heating set temperature in those zones, the maximum amount of energy that the building can potentially consume is 95126 kWh. By eliminating the overheating time periods, the maximum energy efficiency that can potentially be achieved has been determined as 23%. Then, zone-based scheduling and zone-based comfort conditions were determined in order to achieve further improvements in the current energy model. Accordingly, with reference to ASHRAE and TS825 standards, the comfort temperature was changed to 19 °C for laboratories, offices and corridors, and the comfort condition of 20 °C was kept constant for classrooms. In addition, it has been understood that the usage profiles of the zones with different functional properties in the building differ from each other. In this case, the scheduling of the lighting system, heating system and interior equipment in the classrooms, corridors, laboratories and offices of the lecturers has been updated and determined as 9:30-16:30. The working periods for the other offices, which are actively used throughout the day, have been kept constant as 08:00-17:00. The energy model simulation was repeated and the total energy consumption in March was calculated as 59525.41 kWh. As a result of the changes made, 18% energy efficiency was achieved according to the scenario where the comfort conditions of all zones were determined as 20°C. Compared to the scenario where the average temperatures measured from the zones are defined as the heating system set temperature, it is possible to achieve 32% energy efficiency. Finally, by defining the maximum temperatures recorded in the zones as the heating set temperature in those zones, it was stated that the maximum amount of energy that the building could potentially consume was 95126 kWh. Thus, it has been proven that by determining zone-based scheduling and determining zone-based comfort conditions, periods of overheating can be eliminated and potentially a maximum of 37% energy efficiency can be achieved.