Challenges – the many

The original intent of the project was to implement the infrastructure required to produce real-world data from which to draw conclusions regarding the viability of the various technologies and configurations. Given the delays in installing this infrastructure, due to a number of complications including the onset of the Covid-19 pandemic, the course of this project had veered to rely on a modelled/simulated output.

Preliminary financial modelling was undertaken by consultant Aeris Capital, under the direction of project partners. The preliminary modelling was undertaken using spreadsheets and consequently was limited to modelling step tariffs only. This was still extremely useful in 2020 to obtain a broad understanding of the impacts of decision making but was not able to answer the primary question posed in the hypothesis underlying this project: that exposure to the wholesale spot price, or a wholesale price pass-through electricity retailer, such as Amber Electric – in conjunction with supporting technology – produces economic benefits that provide income to improve the educational outcomes of students in NSW.

Early in 2021, Buildings Evolved in negotiation with Aeris Capital formed a resolution that more needed to be done in order to fully test the hypothesis – namely a requirement for interval by interval dynamic pricing from the wholesale spot market. The CSIRO battery control simulation was optimised around the wholesale spot price, so therefore the financial modelling needed to be able to handle wholesale spot in order to properly model the benefits of the battery control system.

The next steps include proceeding with the installation of the cabling and control systems to enable the collection of actual data for a period of 12 months. This data will then be analysed and compared with the modelling to provide insights and inform the direction SINSW will take with regard to the technology and future viability. It should be noted that the inclusion of wholesale spot pricing (without FCAS) found in some modelling scenarios can be implemented through procuring electricity from wholesale price pass-through retailers such as Amber Electric or PowerShop – however, this may conflict with whole-of-government electricity procurement contracts.

The other option: participation directly in the wholesale market, requires a virtual power plant to enable access to the FCAS markets (as modelled in scenarios labelled with FCAS); adding an additional layer of complexity for the Department of Education, and the NSW government more broadly, requiring additional consideration around the practicalities of implementation including resourcing.

Advanced modelling – applied

Buildings Evolved hired software development staff and reallocated existing resources towards agile software development with the aim of solving the hypothesis. This was undertaken only after market research proved existing modelling tools to either be targeted at completely different electricity markets (EnergyPlus, SAM), or likewise had similar limitations in only being able to model conventional retail step tariffs, or were not able to capture the complexity of the Australian energy market.

Extensive background IP was drawn upon to create a tool that could model an extensive array of what-if scenarios and deliver it in a method compatible with NSW Treasury business case guidelines. Major components of work include:

• extract, load transform scripts for NEM12, EDI retail data, BOM, AEMO & other data sources
• tariff engine to model and normalise the extreme complexity of network and retail stepped tariffs
• emissions factor calculations and repository of variables based on state, year and emissions source
• scenario and assumption generator/editor
• 50-year NPV & BCR report outputs
• software is written in Python using a PostgreSQL database back-end
• web user interface written using the React.JS framework

Undertaking this project now allows infinitely more sophisticated modelling to be undertaken, principally in being able to process each interval of data through a unique (wholesale) price. Central to the modelling effort is the ability to create infinite numbers of load profile assumptions. Buildings Evolved in conjunction with CSIRO produced over 10,000 days of unique load profiles for simulation and results generation in the advanced modelling tool. More info about the tool here.

Advanced energy management – proof of concept

The living lab configuration functions to replicate controls technologies and connection methods used for simulating a typical NSW air conditioning controls scenario used for testing controls strategies in the Buildings Evolved living lab.

The purpose of testing control strategies is to provide proof-of-concept testing to:

1. develop full vertical proof-of-concept of the Operational Technology and Information Technology to be used at NSW Schools in DCH 6
2. recreate and test SINSW control logic to determine baseline/BAU controls operation
3. manipulate air-conditioning set-points in response to internal and external conditions, such as:
   1. internal/external temperature, humidity and internal C02 levels,
   2. in context of ASHRAE Comfort Bands, and
   3. control signals from AEMO (simulated).
4. demonstrate control functionality in concert with solar inverter operation, energy generation and onsite consumption. Noting that testing of batteries operation and control are not available as batteries are not installed in the laboratory.
5. integrate the living lab with DCH to demonstrate operational capability and response times through various IoT integration layers, on-premise server Node RED – DCH response and latency testing.
6. for Buildings Evolved consultants and staff to “Live” the control system, meaning effective debugging of functionality and logic
7. reduce risk in the specification, tendering and commissioning phases of works
8. providing a local historian functionality with more detailed information than sent to DCH
9. allowing CSIRO and Buildings Evolved teams to rapidly develop control algorithms

Key benefits – realised

The modelling demonstrates HVAC controls without battery consistently return the best benefit-cost ratio (BCR) for all three sites in DCH 6.1, of above 2x BCR. These results mirror the preliminary modelling work, undertaken prior to procurement of the batteries and additional solar PV on the three school sites covered in this report. By comparison, the modelled business case for battery storage systems for the purpose of energy arbitrage shows a lower financial return, although this is improved by utilising batteries within the FCAS market (and wholesale spot pricing). The scenario with larger solar PV, without batteries or controls and using retail energy accounts, was also notable, showing an impressive BCR in its own right for two of the sites.

The results indicate that where batteries are introduced into the mix of technologies, the BCR and NPV are often reduced. Singleton HS, the largest energy consumer, produced the best financial returns of the three schools studied, and the best BCR – when using the CCP7A scenario (solar PV + HVAC controls). Over 15 years, the modelling indicates a saving of $1.19m can be made with a spend of $0.47m at this one school location.

Additionally, HVAC controls have the ability to improve financial returns while simultaneously improving educational outcomes through strategies including the ability to:

• Set, achieve and measure thermal comfort (e.g. ASHRAE 55 );
• Use night purge to provide fresh air to students at the start of the day;
• Provide thermal comfort models to adjust set points based on external conditions;
• Understand the thermal properties of each building and optimise against it (e.g. thermal mass of demountable buildings vs a triple brick building);
• Improve ventilation to maintain a good indoor environment quality and reduce the risk of infection;
• Making sensor data available to students studying STEM subjects;
• Altering HVAC modes automatically based on external conditions & forecast, and favour fan and dry modes where possible; and
• Reduce HVAC system downtime through better maintenance methods.

Based on the outcomes of this project, there may be opportunities for further modelling focusing on the HVAC controls opportunity in concert with the demand flexibility market opportunities that are emerging adjacent to the wholesale spot price of electricity (WSP) and the frequency control and ancillary service markets (FCAS). Now that the financial modelling tool has been built, adjusting assumptions or adding scenarios can be done with relative ease.

The project has wider implications for society, namely, energy efficiency, security and reliability. Smart controlled operation of assets can provide rapid response in a dynamic fashion according to the needs of the electricity network and market. This will contribute to improved power quality in the network, mitigate the threat of negative energy pricing and ensure cheaper, more reliable electricity with reduced emissions.