Final report on heat pump developments in WP 4 - MacSheep Deliverable 4.4 : MacSheep -New Materials and Control for a next generation of compact combined Solar and heat pump systems with boosted energetic and exergetic performance

FP7- MacSheep

Discrepancies in solar irradiation data for Stockholm and Athens

The aim of this study is to evaluate the variation of solar radiation data between different data sources that will be free and available at the Solar Energy Research Center (SERC). The comparison between data sources will be carried out for two locations: Stockholm, Sweden and Athens, Greece. For the desired locations, data is gathered for different tilt angles: 0°, 30°, 45°, 60° facing south. The full dataset is available in two excel files: “Stockholm annual irradiation” and “Athens annual irradiation”. The World Radiation Data Center (WRDC) is defined as a reference for the comparison with other dtaasets, because it has the highest time span recorded for Stockholm (1964–2010) and Athens (1964–1986), in form of average monthly irradiation, expressed in kWh/m2. The indicator defined for the data comparison is the estimated standard deviation. The mean biased error (MBE) and the root mean square error (RMSE) were also used as statistical indicators for the horizontal solar irradiation data. The variation in solar irradiation data is categorized in two categories: natural or inter-annual variability, due to different data sources and lastly due to different calculation models. The inter-annual variation for Stockholm is 140.4kWh/m2 or 14.4% and 124.3kWh/m2 or 8.0% for Athens. The estimated deviation for horizontal solar irradiation is 3.7% for Stockholm and 4.4% Athens. This estimated deviation is respectively equal to 4.5% and 3.6% for Stockholm and Athens at 30° tilt, 5.2% and 4.5% at 45° tilt, 5.9% and 7.0% at 60°. NASA’s SSE, SAM and RETScreen (respectively Satel-light) exhibited the highest deviation from WRDC’s data for Stockholm (respectively Athens). The essential source for variation is notably the difference in horizontal solar irradiation. The variation increases by 1-2% per degree of tilt, using different calculation models, as used in PVSYST and Meteonorm. The location and altitude of the data source did not directly influence the variation with the WRDC data. Further examination is suggested in order to improve the methodology of selecting the location; Examining the functional dependence of ground reflected radiation with ambient temperature; variation of ambient temperature and its impact on different solar energy systems; Im pact of variation in solar irradiation and ambient temperature on system output.

Influence of hydraulics and control of thermal storage in solar assisted heat pump combisystems

This paper studies the influence of hydraulics and control of thermal storage in systems combined with solar thermal and heat pump for the production of warm water and space heating in dwellings. A reference air source heat pump system with flat plate collectors connected to a combistore was defined and modeled together with the IEA SHC Task 44 / HPP Annex 38 (T44A38) “Solar and Heat Pump Systems” boundary conditions of Strasbourg climate and SFH45 building. Three and four pipe connections as well as use of internal and external heat exchangers for DHW preparation were investigated as well as sensor height for charging of the DHW zone in the store. The temperature in this zone was varied to ensure the same DHW comfort was achieved in all cases. The results show that the four pipe connection results in 9% improvement in SPF compared to three pipe and that the external heat exchanger for DHW preparation leads to a 2% improvement compared to the reference case. Additionally the sensor height for charging the DHW zone of the store should not be too low, otherwise system performance is adversely affected

Emissions Characterisation of residential pellet boilers during start-up and stop periods

In this study, gaseous emissions and particles are measured during start-up and stop periods for an over-fed boiler and an under-fed boiler. Both gaseous and particulate matter emissions are continuously measured in the laboratory. The measurement of gaseous emissions includes oxygen (O2), carbon dioxide (CO2), carbon monoxide (CO), nitrogen oxide and (NO). The emissions rates are calculated from measured emissions concentrations and flue gas flow. The behaviours of the boilers during start-up and stop periods are analysed and the emissions are characterised in terms of CO, NO, TOC and particles (PM2.5 mass and number). The duration of the characterised periods vary between two boilers due to the difference in type of ignition and combustion control. The under-fed boiler B produces higher emissions during start-up periods than the over-fed boiler A. More hydrocarbon and particles are emitted by the under-fed boiler during stop periods. Accumulated mass of CO and TOC during start-up and stop periods contribute a major portion of the total mass emitted during whole operation. However, accumulated mass of NO and PM during start-up and stop periods are not significant as the duration of emission peak is relatively short.

Energy performance comparison of three innovative HVAC systems for renovation through dynamic simulation

In this paper, dynamic simulation was used to compare the energy performance of three innovativeHVAC systems: (A) mechanical ventilation with heat recovery (MVHR) and micro heat pump, (B) exhaustventilation with exhaust air-to-water heat pump and ventilation radiators, and (C) exhaust ventilationwith air-to-water heat pump and ventilation radiators, to a reference system: (D) exhaust ventilation withair-to-water heat pump and panel radiators. System A was modelled in MATLAB Simulink and systems Band C in TRNSYS 17. The reference system was modelled in both tools, for comparison between the two.All systems were tested with a model of a renovated single family house for varying U-values, climates,infiltration and ventilation rates.It was found that A was the best system for lower heating demand, while for higher heating demandsystem B would be preferable. System C was better than the reference system, but not as good as A or B.The difference in energy consumption of the reference system was less than 2 kWh/(m2a) betweenSimulink and TRNSYS. This could be explained by the different ways of handling solar gains, but also bythe fact that the TRNSYS systems supplied slightly more than the ideal heating demand.