The Happold Foundation and BuroHappold Engineering hosted an evening debate on 22nd February on the topic of District Cooling. This was the first of a series of thought-leadership discussions called ‘Energy Conversations’, focusing on broad issues surrounding the supply and use of energy. Aimed at planners, developers, designers and building occupiers, these debates explore key political, planning and technical issues that influence the sustainability of buildings and the built environment.
Attendees were treated to the following speakers (in order of presentation)
- Graeme Sims – Dubai RSB
- Niel Coertse – Galadari, Advocates & Legal Consultants
- Saeed Alabbar – Emirates Green Building Council
- Abeer Manneh – Woods Bagot
- Dr. Peter Armstrong – Masdar Institute
The chair was Alasdair Young Head of Energy for BuroHappold Engineering
Who summarized the initial 45 minutes of presentations and provided a series of probing questions. These were followed by direct questions from the audience. The discussions delved into a wide range of areas, from policies, regulations, prioritization of actions, business models, urban design implications, operational issues, to new innovations and consumer payments.
The summary of the main points of discussion were as follows:
Currently Dubai is seen to be the single largest DC market in the world and has a market share of 20% of all cooling demand, supplying 600,000TR at peak capacity. With Global and local policy shifts towards energy saving, all eyes are on options for improvements. Dubai is targeting a 30% reduction in energy and water use by 2030 and DC is part of 9 programs under an umbrella topic of Demand Side Management. DC is targeted to double in % market share and deliver 14% of the 2.7TWh of annual targeted savings in power demand. Through a program of regulatory framework development, Dubai RSB are trying to encourage private sector involvement in energy supply and efficiency savings.
A current challenge for the RSB is understanding how DC fits in to the DSM strategy and how policy and regulatory changes can help to ensure benefits of DC are maximized.
Historically in Dubai the cooling technology is the choice of the site developer. Recent regulation changes impact on that choice. In 2014 the Dubai Green Building Regulations were enforced and they set minimum efficiency requirements for cooling. But this is not enough to enable the targets to be met. To do this it is not viable to leave it to developers to decide.
The typical market approach is to minimize capital costs incurred by the developer and push the majority of costs onto tenants, which may not deliver lowest cost long term cost or efficiency savings. The developer will take the savings from avoided cost (and space) of building chillers on each plot, but then the tenant often pays high standing charges in their bills to repay the investment cost from the DC company.
How do we encourage appropriate DC while recognizing that it is not appropriate for all times in all places? The calculated benefit and optimum use of DC is complicated. The benefits are largely seen over air-cooled cooling systems, with little or no improvement over water cooled systems in buildings, whereas DC is not cost effective in low density (which is where air-cooled is commonly used). The relationship the consumer has with DC is different to a standard building system where they are only metered for electricity and water. Having cooling as a separate item can provide scope for more energy concious behaviour. However current charging mechanisms make the perception that DC is more expensive. With building cooling systems there is no fixed capacity charge applied to cover the cost of the infrastructure investment, whereas in DC contracts there often is, making it unpopular. Many systems also operate on a fixed cost basis and not a charge per use, which disincentives energy efficiency.
From the operational side the regulations need to encourage better explanation to customers of the relative costs, with clear and transparent bills and fair billing. DC providers need to treat customers like customers, rather than an incumbent demand. To increase efficiency they should also consider smarter cooling demand tariffs such as the use of time of use charging.
From a design perspective it is also recognized that there are inefficiencies in operation that stem from the current approach to sizing and design of DC systems. Cooling load demand estimations must be improved and modeling completed to encourage the design of systems that offer better operational efficiency. This includes Thermal Energy Storage to be required for all new DC plant, use of desalinated water for heat rejection being discouraged and system improvements to help maintain higher temperature differentials between supply and demand (referred to as Delta T).
Niel Coertse outlined the Global context and how this relates to Dubai. This presentation outlined the long-standing criticism of inefficient central provision of fossil fuel and nuclear power as a chainsaw to cut butter, by the energy efficiency guru Amory Lovins. Lovins has been voicing his concerns since the 1970’s, but what was once a lone-voice is now considered popular opinion, with a 2015 UNEP report showing that DE is a proven technique to make energy efficiency improvements, encouraging governments to use DE. DE is presented as a best practice proven measure for reducing GHG emissions in cities. With the uptake of DE and renewable energy the largest proportion of potential for GHG emission reductions of all measures proposed.
The UNEP report outlines 45 DE ‘champion cities’ worldwide including Dubai, which provides more cooling from DC than any other city.
The benefits of DE including enabling rapid shifts away from fossil fuel energy through renewable energy technology integration, air quality improvements, local security of supply benefits and efficiency savings. To put these savings into perspective, it is estimated that if Kuwait were to employ a comprehensive DE plan, they could reduce energy use by 46%.
The plans for increasing DC in Dubai will be challenging and Niel put into focus the relevance of regulation and ownership. Of the 45 champion cities public ownership is the most dominant in 18, with 22 owned through Public-Private Partnerships and other hybrid systems.
Saeed Alabbar from the Emirates Green Building Council re-iterated the commitment from the UAE Government to reducing energy use. He reflected on experience from the COP21 negotiations with the UAE delegation. 2 key take-aways from this were, firstly, the commitment to a maximum 2degC global temperature rise (with a target to decarbonise economies within 40 years). Secondly, increasing focus on cities to act as enablers. With this focus on cities, measures such as Building Regulations and DC can come to the forefront.
To achieve the Global target of decarbonisation cities need to enable and embrace DE, and in particular DE fed by renewable energy sources.
Saeed outlined that the problems with DC in Dubai are the same as the issues with DH in Denmark and other cities. The technology exists and the economics can be proven. What is required are the right mechanisms that ensure DE is employed and in the most effective manner. The RSB are working hard on regulatory side focusing on how to encourage concessionary agreements rather than mandating DE which is key success factor for to achieve the expansion in Dubai that is desired.
Through economies of scale, DC can be more efficient and be cheaper to maintain. Iremain with existing systems that are less efficient and less attractive to consumers than they have potential for, due to historic over-sizing of systems and (low) ‘deltaT’ issues. This was felt to be a consequence of systems being built too large, too soon (before the cooling demand is in place), which has led to inefficiencies and high standing charge costs for users.
With a history of sub-ideal performance and a push for much greater take-up there is a need to underline the issues of the past and understand the ‘true’ potential benefits of DC and how they can be maximized. With this in mind, the RSB have commissioned a study into the delivered efficiency benefits of DC. The first stage of this work has been to explore the average efficiency of the cooling technologies applied across Dubai. This has included broad estimations of the quantum of each system type, combined with in situ measurements to create a weighted average efficiency level. Using a hybrid of SI/IP values the average cooling system in Dubai was deemed to consume 1.5 kW of power/ton of refrigeration generated. The average DC system was estimated to consume 0.8-0.9kW/ton. This would suggest a 40% improvement but in isolation this is misleading. Many of the systems in Dubai are old and poorly maintained, therefore replacing these with modern air-cooled or water-cooled systems and applying a regular maintenance regime would make a similar saving. Therefore, though DC can be seen as a potential smart solution, it is not the answer to everything and needs to be looked at carefully for each development/retrofit case, in particular design integration and phasing needs to be considered in detail.
Abeer Manneh from Woods Bagot gave an urban design and spatial planning perspective, looking at the considerations during the design process and decision making into construction. These considerations can be summarized as:
- Density assessment
- Space requirements for DC plant and networks
- Space opportunities
- Scale and aesthetic
The density of the development needs to be explored from a technical perspective, as this will be critical to the business case of DC. The overall density can give an early indication of practicality but the location and connectivity of this density is key. This will influence the optimum location for the DC plant. The urban planning point of view is often at odds with the business case. Often plants are placed on the periphery of a development but within proximity so as not to lose efficiency. Proximity and development density should be evaluated from a phasing perspective, as the system needs to be efficient and viable from phase 1.
The space for and location of distribution routes are important from an urban design perspective. Chilled water distribution typically requires the largest corridor width in the Right of Way (ROW) and so makes streets wider (and hence more expensive and inefficient). Some cities overcome this through vertical stacking of utilities but this is not currently allowed in the UAE due to access and maintenance concerns. In the example provided there was a 3 metre increase in the street width where DC distribution was provided. Urban designers must consider how the impact of this can be reduced, such as by distributing through sikkas and green corridors.
Despite these space and design issues, the urban design implications are not all bad, with an opportunity in the use of DC to remove rooftop equipment and use rooftops for gardens. The noise from heat rejection plant is also removed, thus providing a nicer environment. The issue of scale and aesthetic of DC plant to fit into the development was discussed with examples shown how DC plants have been disguised with fake frontages and careful massing.
Dr. Peter Armstrong presented research from Masdar Institute into the performance of cooling technologies in the UAE. His emphasis was that to understand the wider picture it was important to dig deep into the technology. The presentation focused on detailed studies of in-situ cooling efficiencies, by looking at whole system picture. This approach included isolating and modeling each single component of the cooling and heat rejection system and creating a calibrated model from first principles based on operational measurements. Through demonstrating the understanding of component performance relating to load, temperature and other variables it is possible to challenge generalisations of system performance and optimize system design. A particular generalization used in the cooling industry for quoting and analyzing cooling performance is the International standard Coefficient of Performance (COP) ratings, which are calculated at 35degC under standard control conditions, and thus are not an appropriate reference in the UAE market.
The electricity use charts for Abu Dhabi island show a 600MW winter baseload and 1.4GW summer peak load. Therefore around 50% of peak electricity usage can be attributed to space cooling. Much of the reason for this is that, prior to recent regulation changes, typical UAE building construction included only 40mm of foam insulation on the roof and no other insulation in walls, floors, etc. combined with a poor quality and leaky building envelope. In addition, minimum performance standards for air conditioning systems were not mandated or, more recently, have been set at a relatively low level.
Therefore, looking at cooling efficiencies is critical for reducing peak and annual energy demands. This would include enforcement of building fabric performance standards through qualified building inspectors, and improved equipment and appliance rating standards.
Peter presented detailed breakdowns of building cooling loads across the year, demonstrating that annual cooling demand is dominated by (dry-bulb) temperatures (60% of load) followed by specific humidity cooling response (20%) (conditioning of outside air brought in through mechanical ventilation or infiltration . The remaining cooling demand is due to diffuse and direct solar gains and latent cooling.
Prior to the application of district cooling or other supply-side measures, Peter recommended the use of Dedicated Outdoor Air Systems (DOAS) to work in parallel with a smaller system meeting latent cooling demands. Using this system around 90% of the annual cooling load could be met using a high temperature chilled water circuit with a return temperature of around 20oC. The remaining 10% load would use a standard low-temperature system. The use of the higher temperature system significantly increases the operational performance of the chiller plant and with a higher deltaT also reduces pumping energy and other parasitic system losses.
The analysis completed by the Masdar Institute have included detailed monitoring of in-situ air-conditioning systems. This has included compartmentalizing each part of the system, so as to understand what is going on in the equipment and thus validate engineering models of the components to solve energy models and create chiller performance maps.
Through creating these detailed performance map of different chiller types and systems, MI established that key aspects to improving chiller efficiency in the Abu Dhabi climate for air-cooled chillers in particular are larger evaporators and the use of variable speed fans. In addition it was shown that a reciprocating chiller offered the best in-use performance for this hot and humid climate.
The baseline air-cooled (screw-compressor) chiller operated with an annual COP 3.4 (3.4 units of cooling per unit of electricity), a 50% increase in condenser size improved this to 3.75, better controls and VSD takes this to 3.9, meanwhile an equivalent optimised reciprocating chiller could provide a COP of up to 5.2. in theory, this suggests that individual air-cooled chiller efficiency improvements may be a smarter approach than District Cooling with water-cooled chillers.
However it was noted that these efficiency levels for any system will only be delivered if there are the right incentives for developers, users and operators to employ them in the most effective manner.
Alasdair Young’s summary
Dubai has ambitious targets for penetration of DC in the market and overall contribution to energy savings. There are different models for ownership, which bring out thoughts of risk allocation and energy pricing. In the field measurements in Dubai suggest that DC is more efficient than the average system. For integration into projects, considerations need to be made for space requirements, balancing building space provision versus the urban realm, and the challenge of creating beautiful urban design objects. We finished with very technical analysis trying to answer the question is DC better than an optimised building chiller system. This highlighted the point that it can be in the right circumstances, but not de facto, and the issue needs to be considered as a whole broad system including controls, behavior, price incentives, etc.
Key questions that stem from these opening presentations are:
- The technical question is the elephant in the room – there is still a lack of independent peer reviewed evidence of proof of improved performance and analysis needs to be specific to the climate and demand profile.
- ‘Is District Cooling Smart?’ needs to be considered from the consumer perspective; for instance, what is the fair price for cooling? Should consumers be paying the full price for capital recovery of assets, while developers benefit from having more saleable land. Should consumer prices be controlled and pegged against alternatives?
- Can buildings be designed to work properly with district cooling? What are the variables and key design or contractual requirements? Can the industry provide proof of performance for developers?
- How can the Government incentivise the benefits of DC and make the most of the efficiency benefits offered. How can this be enforced and what standards are required?
- What is the right business model – state-owned, independent or hybrid, and how does this impact on the balance of efficiency, development impact and operational cost?
Q&A with the panel
Questions from the floor and respective answers.
Q. We have a lot of free energy from the sun, vision to create a lot more clean energy. When we desalinate we create brine and raise salt levels in the gulf. So is the energy as much of an issue as the water?
Graeme. It is too soon to say that we have infinite amount of free solar power to run things how we like. I do not personally know what the impact is on the salinity of the gulf. Would the 40% DC target be jeopardized if water is not available? Probably addressable by looking at the amount of gas producing water and power and calculating the best allocation between gas use, but in general it is believed that improvements in cooling are not offset by the increase in water use. All DC plants in Dubai bar one include RO plants to treat TSE for use in cooling towers. With desalination, water is not the issue that perhaps people think it is.
Q. It looks like technology is here already. For every new development should DC be embedded in design from the very beginning? How difficult to legislate to make mandatory?
One good thing to have the vision and targets but how do you actually enforce it. How do you punish a developer for not complying?
Abeer . There is an issue from a design perspective of knowing when considerations of DC come into play. If the intention to look into DC it must come from the beginning due to the necessary changes in infrastructure design, building design and phasing. All stakeholders must buy in.
Saeed. It is generally a more efficient way of delivering cooling though there are also a lot of projects where problems have been faced. It is difficult to enforce the use of DC. Based on current evidence the comparable performance difference against best practice building systems is not significant.
Niel. Perhaps the Government can apply the ‘polluter pays principle’, where DC is not enforced but payments are made for using less efficient systems. It would need to be established who is responsible for paying for poor performance?
Q. If you compared a typical apartment building with DC and traditional systems, who is paying more on consumer bills? People with DC are surprised by how much they pay.
Graeme. In a traditional system you pay for the system when you bought the apartment, the purchaser of the unit in the first instance pays, this includes the cost to the developer of upgrading the power supply system. In a DC market the consumer pays for the costs of the DC infrastructure. So the comparison needs to compare what people pay DEWA plus capital cost already paid.
The unit price of cooling for DC systems is also increased because the purchase price of electricity that the DC provider pays is double the price of a normal consumer, as their high demand levels place them in the higher price slab.
Q. We have heard a lot about generation and transmission, but how can we operate the building for optimum DC efficiency? DC may have restrictiosn due to technology and distribution losses which prevents them being more efficient without action from the building users/operators. Can actions on FM be applied that encourages compliance?
Peter. The contract with the building operator needs to be evaluated based around a pricing algorithm that incentives the building operator to control internal building systems to have higher DeltaT and lower flow rates on the demand side.
Q. How smart can we make district cooling?
Saeed. There is lots of room for development on the DC side and potential to leverage on the economies of scale, need to invest in smarter DC systems and use more innovation.
Peter. Use of thermal storage, which will have the following three advantages: The power utility provider will be happier as it will help to reduce the peak loads. It allows chillers to operate at peak efficiency longer by modulating compressor load over the day. It allows for the system to take advantage of the temperature difference between day and night.
It should be noted that the future of controls will mean that thermal storage technologies need not be unique to DC plants. The intelligence technology of driverless cars can apply to building cooling systems thus optimizing control systems for efficiency.
Q. A major problem for the development rate of DC is that it is very dependent on the rate of new development. In the future I see a lot of retrofits. Areas not suitable for DC before. Is there a strategy to put in DC in these areas?
Abeer. The challenge with retrofitting DC in an existing urban context relates to space and coordination with existing utilities. Is there room in the Right of Way (ROW)? Are you going to disturb the whole area to do that? Perhaps a clustering of 4-5 buildings may be more feasible to look at on a wider scale.
Saeed. Large areas of high density buildings will have no allocated plot for DC and no space in the ROW. If old buildings in a cluster are due for replacement then it could be possible. The benefit offered by DC is that it can replace the capital cost of replacing individual chillers.
Graeme. The regulation is the tricky bit. In a new development the master developers make the decision that they want DC or not. Existing developments often do not have a master developer, so who is making that choice?
The most likely opportunity would be to investigate if there is a DC network nearby, then you can tap in.
Q. What is the one recommendation you would make to drive toward COP21 and who would you make it to?
Peter. Minimum equipment performance standards for all equipment based on demonstrated optimized life cycle cost would increase efficiency by 50%.
Graeme. Users to pay an appropriate economic price for what they use. Currently price not appropriate and people do not pay for what they use
Abeer . Energy efficiency of buildings
Saeed. Lowest hanging fruit: quality control on building envelopes, integrity of insulation. Project example of initial permeability test 18m3/m2 small changes and improvements took this to 4m3/m2 permeability.
Niel. Better enforcement and implications for non-compliance around energy efficiency.
Q. Stacking of chilled water pipe – has that been done and what are the risks of doing this?
Abeer. Project in Qatar uses an existing utilities tunnel, where the new phase connected to the tunnel which already had stacking of utilities. There are concerns about challenges in maintenance if something happens.
Q. Where could we push boundaries using innovative technologies?
Peter. I’ve seen so many people try to shoe-horn lots of technologies into DC plants, these are not necessary and there needs to be more focus on getting the basics right.
Alasdair. There are no silver bullets when it comes to DC. Basic improvements can reduce consumption by 50% through good design.
Graeme. We are interested in trying to define DC (scale, etc) and to write a law about it, it is very difficult to pin down. Talking about how the property market operates. If I buy a car I think I understand what the specified performance is it will deliver, the property market is different, there is no measure of environmental performance. Perceptions of buildings may change in the future.
Niel. I would like to see how the PPP business structures will be applied. Lets do the small things right. What are the penalties are non compliant and litigation from that.
Saeed. Is district cooling smart? Numbers show it can be. Dubai has highest installed capacity in the world. Take the lessons we have learnt and leverage economies of scale for innovation, commissioning, integration with buildings side, etc.. Need to change the perception of the DC and of the costs.
Abeer. From the urban planning side, the integration early in the project with the space requirements makes a smart approach, based on the type of project.
Peter. Transparency of incremental cost, true costs is essential.