Author: HongGic Oh

Indicative Heating and Cooling Costs of High and Low Performing Homes and Appliances

Homes are given an Energy Efficiency Rating (EER) based on the annual energy demand (heat required to be added in winter and removed in summer to keep the home comfortable) and we have interpreted that into running costs of a property in terms of energy consumption for cooling and heating. We offer the calculation of annual gas and electricity cost for heating and cooling energy consumption for a user-selected single-storey house size in Canberra on our website; including the option of all-electric home conditioning since July 2022.

The calculations are based on the prices published by ActewAGL, and users are able to input a preferred home size between 75 and 500m2 to compare the estimated energy costs of heating and cooling a home corresponding to an EER using appliances of varying energy rating.

Until now, we have only covered the Canberra region (and been indicative of other locations in the NCC Climate Zone 7 – Cool Temperate) and are about to update the matrix to reflect current gas and electricity prices; but we will then be working on the other 7 major cities’ calculations, so they will be added soon. We will announce when the other cities are ready with Melbourne being our first target.

Exemplary Weather and Energy (EWE) Index June 2024

The Exemplary Real Time Year weather files (RTYs), current Reference Meteorological Year files (RMYs) and Ersatz Future Meteorological Years (EFMYs) used for these monthly simulations are available for purchase. This will allow clients to simulate their own designs for energy budgeting and monitoring rather than rely on analogy with the performance of these archetypical buildings and systems. Especially in mild months, small differences in energy consumptions can result in large percentage differences. Solar irradiation data courtesy of Solcast.

Archetypical buildings and systems

10-storey office

3-storey office

Supermarket

5 kW domestic
PV system

Get the Best out of our Interactive Features

This monthly report has been interactive since April 2023.  Once you have scrolled to your city of interest, check out those interactive features and how they work.  Click here to read about the introduction.

  • 1. Choose the energy or peak demand graph to best match your building or system of interest.
  • 2. Choose the weather element graph to best match the sensitivity of your building or system of interest.
  • 3. Mix and match to learn about their relative importance or sensitivity

ADELAIDE

Energy Index (%)

WordPress Data Table Plugin

The solar PV simulation output results were 21.8% lower than the long-term average. The heating peak load was higher than the long-term average for all archetypes (8%, 6.6% and 13.1% for 3-storey office, 10-storey office and supermarket respectively). It should be noted that peak load results are highly sensitive to the particular building and HVAC design and settings – it is more appropriate to evaluate those results from a bespoke building model using our RTY data.

Adelaide experienced a more humid and cooler June compared to the long-term average. The GHI and wind speed were much lower than the long-term average.

Weather Index

WordPress Data Table Plugin
WordPress Data Table Plugin
WordPress Data Table Plugin

BRISBANE

Energy Index (%)

WordPress Data Table Plugin

The solar PV simulation output results were 12.7% higher than the long-term average. Both Cooling and Heating peakload was lower than the long-term average. The cooling peak load was 30.8%, 33.1% and 11.7% lower and the heating peak load was 39.4%, 37.4%, and 29.3% lower for the 3-storey office, 10 storey office and supermarket respectively. It should be noted that peak load results are highly sensitive to the particular building and HVAC design and settings – it is more appropriate to evaluate those results from a bespoke building model using our RTY data.

Brisbane experienced comparably warmer and less humid June than the long-term average. The GHI was much higher while wind speed was similar to the long-term average.

Weather Index

WordPress Data Table Plugin
WordPress Data Table Plugin
WordPress Data Table Plugin

CANBERRA

Energy Index (%)

WordPress Data Table Plugin

The solar PV simulation output results were 6.5% higher than the long-term average. The heating peak load was 4.5%, and 32.5% higher for the 3-storey office and 10 storey office while 0.3% lower for supermarket respectively. It should be noted that peak load results are highly sensitive to the particular building and HVAC design and settings – it is more appropriate to evaluate those results from a bespoke building model using our RTY data.

Canberra experienced a more humid and cooler June compared to the long-term average. The GHI was similar but wind speed was higher than the long-term average.

Weather Index

WordPress Data Table Plugin
WordPress Data Table Plugin
WordPress Data Table Plugin

DARWIN

Energy Index (%)

WordPress Data Table Plugin

The solar PV simulation output results were 4.6% lower than the long-term average. The cooling peak load was higher than the long-term average for all archetypes (4.4%, 6.8% and 0.8% for 3-storey office, 10-storey office and supermarket respectively). It should be noted that peak load results are highly sensitive to the particular building and HVAC design and settings – it is more appropriate to evaluate those results from a bespoke building model using our RTY data.

Darwin experienced a slighlty more humid of normal temperatures June compare to the long-term average. The GHI and wind speed were slightly lower than the long-term average.

Weather Index

WordPress Data Table Plugin
WordPress Data Table Plugin
WordPress Data Table Plugin

HOBART

Energy Index (%)

WordPress Data Table Plugin

The solar PV simulation output results were 12.9% lower than the long-term average. The heating peak load was lower than the long-term average for all archetypes (25.9%, 21.5% and 26.6% for 3-storey office, 10-storey office and supermarket respectively). It should be noted that peak load results are highly sensitive to the particular building and HVAC design and settings – it is more appropriate to evaluate those results from a bespoke building model using our RTY data.

Hobart experienced a slightly more humid and warmer June compared to the long-term average. The GHI was slightly higher while wind speed was lower than the long-term average.

Weather Index

WordPress Data Table Plugin
WordPress Data Table Plugin
WordPress Data Table Plugin

MELBOURNE

Energy Index (%)

WordPress Data Table Plugin

The solar PV simulation output results were 11.1% lower than the long-term average. The heating peak load was higher than the long-term average for all archetypes (25.9%, 29.4% and 13.0% for 3-storey office, 10-storey office and supermarket respectively). It should be noted that peak load results are highly sensitive to the particular building and HVAC design and settings – it is more appropriate to evaluate those results from a bespoke building model using our RTY data.

Melbourne experienced a more humid and slightly cooler June compared to the long-term average. The GHI was almost similar to the long-term average.

Weather Index

WordPress Data Table Plugin
WordPress Data Table Plugin
WordPress Data Table Plugin

PERTH

Energy Index (%)

WordPress Data Table Plugin

The solar PV simulation output results were 0.3% lower than the long-term average. Compare to long term average, the cooling peak load was 9.1% and 6.3 % higher for 3-storey office and 10-storey office while 6.2% lower for supermarket. It should be noted that peak load results are highly sensitive to the particular building and HVAC design and settings – it is more appropriate to evaluate those results from a bespoke building model using our RTY data.

Perth experienced a more humid and warmer June compared to the long-term average. The GHI was slightly higher while wind speed was lower than the long-term average.

Weather Index

WordPress Data Table Plugin
WordPress Data Table Plugin
WordPress Data Table Plugin

SYDNEY

Energy Index (%)

WordPress Data Table Plugin

The solar PV simulation output results were 5.7% lower than the long-term average. Compare to long term average, the heating peak load was 11.9% and 26.0 % higher for 3-storey office and 10-storey office while 5.0% lower for supermarket.. It should be noted that peak load results are highly sensitive to the particular building and HVAC design and settings – it is more appropriate to evaluate those results from a bespoke building model using our RTY data.

Sydney experienced a more humid and cool temperatures June compared to the long-term average. The GHI was much lower while wind speed was higher to the long-term average.

Weather Index

WordPress Data Table Plugin
WordPress Data Table Plugin
WordPress Data Table Plugin

Examining Hobart’s Climate Trends: Implications for PV System and Building Performance

As part of the ongoing investigation into our evolving climate, we routinely process and analyse meteorological data from successive years, conducting comparative assessments to reveal emerging trends and patterns.

Our previous temporal analyses only focused on examining variations in various weather elements. However, in this latest iteration, we have incorporated the results of EnergyPlus simulations, specifically targeting HVAC systems and heating and cooling dynamics within buildings. We have also added the results of System Advisor Model (SAM) photovoltaic (PV) system simulations to enhance the comprehensiveness of our investigation.

The most recent temporal analysis was carried out for all eight capital cities, though this issue of Exemplary Advances will focus on the city of Hobart. The findings for Darwin were previously discussed in the June issue of Exemplary Advances. The analysis for other capital cities can be viewed here.

For the analysis of weather elements, we examined the temporal variations in dry bulb temperature, humidity, wind speed, global horizontal irradiation (GHI), direct normal irradiation (DNI), and total precipitation. The analysis involved averaging these elements over three 15-year periods—1990-2004, 2005-2019, and the latest 15-year period from 2009 to 2023—and then comparing the results. A comparison between data from the latest 15 years, the data corresponding to the years and months specified in Industry Standard Meteorological Year (ISMY) files, and the data exclusively from 2023 was also undertaken. ISMYs were originally developed for application in house energy rating software used in NatHERS and derive from historical Bureau of Meteorology (BOM) weather data spanning from 1990 to 2015. Over time, they have become the industry’s de facto standard. It is therefore important to compare against ISMY data, as it provides a reference to gauge alignment with established benchmarks and understand the significance of temporal variations in weather elements.

Comparing 1990-2004 with 2009-2023 showed an increase in Hobart’s mean temperature of 0.57°C, an increase in moisture of 2.40%, and a significant increase in wind speed of 9.45%. GHI had a decrease of 1.54%, and DNI had a significant decrease of 11.28%. Meanwhile, comparing 2005-2019 with 2009-2023 showed a decrease in the mean temperature of 0.09°C, an increase in moisture of 1.31%, a decrease in wind speed of 0.35%, and a decrease in GHI and DNI of 0.96% and 4.65%, respectively. The small increase in mean temperature and decrease for GHI and DNI for 2005-2019 vs 2009-2023 is likely a result of 2020-2023 experiencing comparatively higher annual average dry bulb temperatures and lower GHI and DNI when compared to other years.

Total precipitation in 2009-2023 averaged 3.04% higher than in 1990-2004, and 6.18% higher than the 2005-2019 period.

The annual trends of energy consumption reveal intriguing patterns across various building archetypes. All archetypes had increasing trends for cooling energy consumption from 1990 to 2023, as well as in the 15-year periods of 1990-2004, 2005-2019 and 2009-2023, while heating energy consumption had decreasing trends for all archetypes in all periods. These trends are indicative of warming climate, and highlight the importance of using relevant climate files from the more recent 2009-2023 period in building energy simulations rather than the older ISMY data.

Progress on Precipitation Data – Disaggregation of 1990s Daily Data

In the May edition of Exemplary Advances, we introduced our research into precipitation disaggregation from daily to half-hourly using a long short-term memory (LSTM) model. Since then, we’ve experimented with a number of other architectures, including foundation models specifically designed for time-series applications. Our most promising results so far have come from a modified LSTM.

High temporal resolution of precipitation is necessary for the design and simulation of building components. Due to tipping bucket rain gauges only being installed from the 1990s and the early 2000s in most localities, climate files created by concatenating the twelve most typical months selected from three decades often include months where that site only has daily precipitation data, measured at 9am local time. Our research aims to provide high-quality, high-resolution half-hourly precipitation data that is consistent with the measured daily value and the simultaneous hourly values of the other weather elements.

LSTM is a recurrent neural network that can handle long-term dependencies using special memory cells which allow the network to learn important details from a sequence of data while discarding irrelevant information. The ‘short-term’ in its name refers to the network’s ability to capture data dependencies and patterns over short sequences that are then extended to longer sequences using the memory mechanism. A diagram of our new architecture can be seen below.

The three major changes we undertook were:

1) Adding a differentiable normalization layer to enforce the daily total constraint from the model itself, rather than later during post-processing. Previously, we used the SoftMax activation function to ensure outputs summed to 1 were multiplied by the daily total when generating the results. However, this has two main issues:

  • First is that SoftMax is non-linear, meaning that it distorts the contributions of time periods to the total.
  • Second is that the daily total is not accounted for at all during training.

Our new layer enforces this constraint effectively during training, leading to more accurate results.

2) Re-shaping the inputs and outputs. We now estimate each half-hour period one-at-a-time and squeeze these into a single tensor, rather than producing 24-hour estimations at each timestep and taking the last estimated value as the output tensor.

3) Further feature engineering. We have some more input variable combinations to integrate into the new model, but each new feature adds significant ‘training’ time for the software. Currently, we are testing and trying to find the best combination based on correlation with precipitation to find the pros of adding more features.

We’ve seen significant improvement in several key metrics.

We are now exceeding the performance of Exemplary’s previous hourly precipitation disaggregation method using a Markov Chain Monte Carlo (MCMC) approach despite estimating to a finer resolution. For example, if we compare absolute error in the total number of rainfall periods, the LSTM scores ~13%, which is within the range recorded by our colleagues’ previous work despite being a much finer time series. Further, the LSTM correctly detects 64.8% of rainfall periods with no temporal error – this is a 2x improvement over the previous model, and still five percentage points more accurate than the previous model with a ± 2-hour window. Even if our new model is precisely predicting daily precipitation, we still need to improve the model to translate it to half hourly resolution.

The graph above shows our current model’s synthesised half hourly precipitation against real data. The model works well on low precipitation while there are big gaps on high precipitation values. It performs 57% of precision and 67% of recall but we are aiming to get these two factors above 80 % in our next iteration of the code.

We are working on hyperparameter fine-tuning to get better results and a testing setup to evaluate the model’s performance for a number of Australian cities. The updated results will be announced in future editions of “Exemplary Advances”.

Project Team: Hong Gic Oh (leader), Nayan Aroroa (graduate) and Harrison Oates (intern)