Printer Friendly

Performance evaluation of the centre for efficient renewable energy in buildings.


London South Bank University, established in 1892, is one of London's largest and oldest universities. The main campus, based the Elephant and Castle in the Borough of Southwark, houses over twenty buildings and has excellent access via the local transport infrastructure. The K2 building is the newest development on campus (completed construction in November 2009) and is situated on Keyworth Street. K2 is an eight storey building with a central atrium from ground to fourth floor. The K2 building consists mainly of offices, simulated healthcare environments, and teaching spaces. The facility is used by the Faculty of Health and Social Care and the Department of Education (from the Faculty of Arts and Human Sciences). Located on the 8th floor (roof level), surrounded by mechanical plant and built from sustainable timber panels is CEREB's seminar room as shown in Figure 1. In addition to housing a number of low and zero carbon (LZC) technologies

CEREB is a hub for research and data collection. The centre's administration sits in the Urban Engineering department in another part of the campus remotely monitoring the performance of K2/CEREB and other buildings in and around Southwark. All the technologies at K2/CEREB are extensively monitored with the data stored in the cloud, and therefore accessible from anywhere in the world.



The design and construction of the K2 base building and the CEREB space was undertaken in two phases by two different sets of design teams.

The shape and orientation of K2 building is designed to maximize daylight reducing the need for artificial lighting. The building envelope is composed of a primary wall structure clad in zinc and terracotta rain screen. The building has recessed double glazed window units that are only openable for cleaning. Vents are provided at the top of the atrium, for both temperature control and as smoke vent in the event of fire. The southwest elevation has a layered curtain wall arrangement with stainless steel mesh providing solar shading thus minimize summer heat gains. The shape, form and external facade from the southeast elevation is as shown in Figure 2. The thermal transmittance (u-value) of each building element has been designed to meet the UK Building Regulations 2006 [DCLG 2006]. The development was designed to achieve a BREEAM1 rating of Good but during construction the university took the opportunity to improve the environmental performance of the design and construction of the building to achieve a BREEAM rating of Very Good on Post Constrictions Review stage [AECOM].


The heating, cooling and ventilation for K2 is provided through seven air handling units (AHU). The AHUs each have a thermal wheel designed to recover up to 80% of heat from return air thus pre-heating the supply air during winter months reducing the heating load. The heating and cooling for K2 is met by four ground sourced heat pumps (GHSP) at rated capacity of 120 kW heating and 125 kW cooling each. The heat pump installation comprises four heat pumps, circulation pumps, ground loop system and four dry coolers, the last of these being for use in the event that the ground store overheats. The GSHP system produces low temperature hot water (LTHW) at 50[degrees]C flow and 43[degrees]C return, designed to provide space heating through underfloor air distribution system, trench heaters and radiators with thermostatic radiator valves (TRV) controls.

The domestic hot water demand is met through 19 direct flow evacuated solar collectors installed on the roof of CEREB. The system comprises of three storage cylinders and two standby gas boilers which provide top up to the required hot water temperature when necessary. Water wastage on the user side is reduced by the installation of electronic taps.

The CEREB seminar room showcases two AHUs within its floor. The AHUs incorporate phase change material which provides pre-cooling during summer months thus reducing the cooling load. CEREB is partly lit by solar fiber optic lighting, and heating, cooling and ventilation provided through a displacement system. The seminar room has two active glass windows, as shown in Figure 1, through which the plant areas can be viewed. The heating and cooling load for CEREB is met by a 6-30 kW capacity modulating gas fired boiler and a 12 kW capacity absorption chiller respectively. CEREB also has 1 kW of poly-crystalline photovoltaic (PV) cells integrated into the rooflight and 9 kW of roof mounted poly-crystalline PV system, which are grid connected.

Other control measures include the installation of presence detectors throughout the building which will automatically switch off artificial lighting when the space is not occupied. Energy meters have been installed on every equipment, each zone and floor of the building in order to capture maximum data for effective monitoring and building performance evaluation. Currently there are no occupant controls for room temperature provided in any part of the building.

K2 has a total gross floor area of 8,500[m.sup.2] with occupancy ranging from 40 people in the CEREB seminar room, 85 people in meeting rooms to 1,060 people in teaching space. The building is designed to operate from 7:30 to approximately 20:00 weekdays and 7:30 to 13:00 Saturday. Taking all the above energy efficient measures into consideration and applying conventional technologies (non renewables), the predicted annual energy consumption and carbon emission of the building as estimated by building services consultants is represented in Table 1 [AECOM 2006].
Tablel: Predicted Annual Consumption and Carbon Emissions

                Hot Water  Heating  Cooling  Electricity

Energy           411,268*  391,317   49,545      459,754

Carbon dioxide     79,786   75,911   20,197      194,016


Energy                               -

Carbon dioxide                      44


The energy performance of K2/CEREB is managed through the Building Management System (BMS) and monitored through the Workplace Footprint Tracker (WFT) which draws data from the installed energy meters. The metering strategy for K2 is extensive and includes sub metering of each floor, equipment and technology. A simple schematic representation is shown in Figure 4 below. The web enabled interactive WFT dashboard sits above the BMS with its ergonomically designed navigation system providing extensive information, easily accessible through a web interface. The WFT displays real time half hourly information on the energy consumption and generation (where applicable) for each technology at K2/CEREB. The data for the graphical and tabular displays are drawn from the recordings of the BMS. The BMS displays real time operational schematic diagrams of the mechanical systems, and temperature data for all installed sensors. The BMS monitors the ground temperature through 22 thermocouples placed in the structural piles, to study ground behavior with the use of GSHP for heating and cooling.

CEREB was completed on the 23rd of June 2010 therefore the data collection period to study the performance of K2/CEREB was chosen between 27th June 2010 to 26th June 2011 to give a complete calendar year. However, the available data from the WFT was inconsistent over this period for each equipment and technology, as shown in Figure 4 below. With reference to Figure 4 and Table 2 below an explanation of the makeup of each energy use, the meters identified for data collection and the annual consumption and generation is as follows:
Table 2: Actual annual Generation, Consumption and Carbon

               Hot     Heating  Cooling  Electricity

LZC Energy   66,049  190,171   13,294        8,223

Energy       54,888            16,773    1,534,462

Carbon       10,648             3,254      647,543


LZC Energy                        -

Energy                            -

Carbon                           78

1. The LZC generation for hot water is accounted for by readings from the primary energy meter (P1) for the solar thermal loop, and the energy consumption for hot water is accounted for by readings from the dedicated gas meter (GM3) for the two gas boilers.

2. The LZC generation for heating is accounted for by readings from the main load-side heat meter (M1) for the four GSHPs.

3. The LZC generation for electricity is accounted for by readings from the energy meter (E) for the roof mounted and rooflight integrated PV panels, and the energy consumption for electricity is accounted for by readings from the MEM mains electricity meter recording consumption from small power, lighting, equipment and heat pumps.

4. The LZC generation for cooling is accounted for by readings from the main load-side meter (M1CHW) for the four GSHPs, and the energy consumption for cooling is accounted for by readings from the dedicated gas meter (GM2) for the CEREB gas boiler providing heat for the absorption chiller.

A summary of the actual annual energy generation from LZC technologies, energy consumption from conventional technologies and the carbon emissions from the use of conventional technologies, against each energy use is given in Table 2 below. Despite having installed LZC technologies the actual building's carbon footprint is 78 kg[CO.sub.2]/[m.sup.2] which is significantly higher than the design carbon footprint of 44 kg[CO.sub.2]/[m.sup.2], therefore understanding the differences between the design principles and actual operation of the building is fundamental in identifying why the discrepancies exist.



The predicted energy consumption figures of K2 building as shown in Table 1, assumes the minimum input design requirements stipulated by the Building Regulations to produce the energy model. The design parameters stipulated by the Building Regulations are for the purpose of showing compliance and not for predicting the actual energy consumption of a building. Additionally the Energy Performance Certificate (EPC), an output of the Building Regulation assessment, is generally used within the industry as a benchmark to assess the performance of buildings. As the EPC is a compliance assessment it does not take into account building specific design criteria such as actual occupancy patters, lighting and small power, and non conventional use of building [DCLG]. These building specific energy loads referred to as unregulated loads are not accounted for in Building Regulation assessment. Therefore inclusion of such loads in the prediction model would reflect a building's actual energy usage better.

Two main issues identified that would explain the increased electrical consumption in the actual building are:

Occupancy pattern. The predicted model assumes operational period of 9:00-5:30 Monday to Saturday for offices, 9:00-9:00 Monday to Friday and 9:00-5:30 Saturday for lecture theatres and standard for all other ancillary areas. But in reality is there control strategy in place dictating the design operational periods for the different spaces and equipment? The annual data for energy generation and consumption shows that some systems are working 24-7, constantly responding to demand. The lack of time control increases operative hours of the plant, increasing the electrical and gas consumption far beyond design. It has been found that in the laboratory floors which have simulated hospital environments, with no automatic or manual switching to turn off lighting during out of office hours, the PIR sensors are triggered for every movement consuming substantial electrical energy at night.

Lighting and small power. The use of each space is assigned a "type" from a list of pre-described design conditions within the model therefore spaces like laboratory and cafe have not been assigned the actual space loads, as these options are not available for a University type building. Correcting this in the actual model will increase the cooling, lighting, small power requirements thus increasing the electrical consumption of the building's predicted energy/carbon footprint.

Hot water

The actual hot water consumption (as seen in Table 2) is far less than predicted (as seen in Table 1); demonstrating a well performing building. The gas consumption data is the readings of GM3 gas meter therefore it is assumed that this data is relatively accurate. On the other hand the solar hot water has missing data from the end June 2010 to the end July 2010, during which period as the solar thermal system had not been properly installed and therefore the meter (P1) failed to record any data. Secondly the issue of the wrong glycol mix in the solution and that specified for the meter also adds to inaccuracy of data. The [P.sub.1] meter is designed for a system with glycol mix of 25% to water; this gives an inbuilt [c.sub.p] of 1.09-1.14 kWh/[m.sup.3]/K depending on the temperature of heat transfer fluid. However the actual glycol mix in the heat transfer fluid is 40% glycol to water, which gives a cp of 1.08-1.13 kWh/[m.sup.3]/K [Metrima]. Though the variation in cp is not substantial this will still affect the output energy reading from the P1 meter. The designed metering strategy for the hot water system was to enable calculation of energy output of each equipment on the system. However one of the energy meter specified to be installed on the return pipe just before the boiler, is incorrectly installed on the supply pipe going from the cylinder to the boiler but before mixing from the secondary system. This incorrect installation leads to calculated boiler efficiencies of over 100%.


Comparison of the predicted consumption to actual consumption once again shows a well performing building. The GHSPs have no doubt been working throughout the year providing good space heating to K2 but why then is there a discrepancy between predicted and design? The only explanation that could be sought is that the GSHP meter, M1, could be reading less energy than actually generated. The M1, as shown in Figure 5 below on the load side displays the total heat generated by the heat pump; this is calculated using mcp[DELTA] [T.sup.4] where the flow rate (m) from the flow rate meter and temperatures ([DELTA]T) from the temperature sensor recordings are utilized. This dependency on secondary readings adds to the risk of inaccurate data. The specification of the current temperature sensors does not account for resistive losses in the wire between the sensor and the meter; therefore the temperatures registered by the energy meter [M.sub.1] are slightly off the actual. By replacing current sensors with the type that accounts for resistive losses within the wire will resolve this problem.

Some other issues with the GSHP heating data collected to date include, missing data for some parts of the June 2010 and February 2011 and for full month of July10-November 2010, and the source-side meter reading higher load than the load-side meter reading. A possible explanation at this stage to the source-side reading is that like the solar thermal P1 meter, the G1 meter is designed for a system with glycol mix of 25% to water; this gives an inbuilt cp of 1.09-1.14 kWh/[m.sup.3]/K depending on the temperature of heat transfer fluid [Metrima]. However the actual glycol mix in the heat transfer fluid is 33% glycol to water, which gives a cp of 1.08-1.13 kWh/[m.sup.3]K. Though the variation in cp is not substantial this will still affect the output energy reading from the G1 meter.



On the cooling side, there are a number of other issues with quality of data; mainly due to the metering installation. The [G1.sub.chw] meter is currently reading temperatures from the G1 as shown in figure 5 above. When reading data off G1, there is additional resistance in the system that is causing a temperature difference of 15-45% between the G1 reading and [G1.sub.chw] reading. Again this could be corrected by re-commissioning the sensor installation to feed directly into [G1.sub.chw]. The lack of data until March 2011 is yet to be explained and currently the data produced since March 2011 cannot be validated due to the meter/sensor installation issues. The cooling for CEREB is met by the absorption chiller power by the gas boiler. Here again the gas boilers consumption data is drawn directly from the gas meter reducing the risk of inaccuracy in data. To date the dry coolers have not had the need to operate reflecting that the GSHPs have been meeting the required cooling load.


The electrical consumption data extracted from the Mains Electricity Meter (MEM) as shown in Figure 4 above, accounts for all electrical consumption in the buildings; inclusive of regulated and unregulated loads. As explained in Modeling section, the design did not include unregulated loads and secondly those consumptions that were included are operating out of office hours causing a greater discrepancy between predicted (as shown in Table 1) and actual (as shown in Table 2). The PV panels have performed well to date and have been feeding electricity back to grid.


Despite these irregularities, how is K2/CEREB performing? The answerer is in two fold. The general day to day operation of the building and the comfort conditions of the internal space experienced by the building users and visitors appear to be satisfactory with very few complaints. This is further evident with data from the BMS and the WPT which shows that the comfort condition of the internal spaces are hardly compromised; thus on the surface reflecting a well performing building. However dwelling deep into actual consumption data and the benchmark prediction figures, there is a fundamental difference between design input parameters and actual performance. Therefore, firstly a strategy for assessing building operation against design is urgently needed, and secondly the quality of data and a strategy for validating the data is required.

Our experience at K2/CEREB has raised a number of important questions mainly on the quality of the measuring equipment, and on the installation and commissioning, and therefore the quality of the data itself. The answer to high performance buildings depends on a good design and construction from start to operation to post occupancy and requires:

1. Measuring tools that are standard throughout the lifecycle of project.

2. Measuring equipment that are reliable, and installed and commissioned by experts.

3. Good data management and monitoring strategies that are fit for purpose and comparable to other buildings

4. Most importantly a skilled team that understand the intricacies of designing, specifying, installing, commissioning and operating buildings with LZC technologies.


I would like to thank Maxima Blanquet and Maija Krizmane of CEREB for data collection.

The base building was architecturally designed by Grimshaw Architects and building services designed by AECOM. The project was managed by Gardiner & Theobald LLP, constructed by Kier London and commissioned by CML Ltd. The CEREB space was architecturally designed by Shepheard Epstein Hunter and building services designed by TGA Ltd. The project was constructed by Lakehouse Ltd and services installed and commissioned by Gratte Bothers Ltd.


AHU = Air Handling Units

BMS = Building Management System

CEREB = Centre for Efficient and Renewables in Buildings

GSHP = Ground Source Heat Pumps

LTHW = Low Temperature Hot Water

LZC = Low and Zero Carbon technologies

MEM = Mains Electricity Meter

PV = Photovoltaic

TRV = Thermostatic Radiator Valves

WFT = Workplace Footprint Tracker

(1.) BREEAM is Building Research Establishment's Environmental Assessment Method, akin to the LEED assessment.

(2.) Assumed efficiencies - 82% for domestic hot water gas boiler, 86% space heating gas boiler and COP 2.3 for chiller [AECOM 2006]

(3.) Carbon factors of 0.192 kg[CO.sub.2]/kWh for gas and 4.22 kg[CO.sub.2]/kWh for electricity [DCLG 2006]

(4.) [c.sub.p] is the specific heat capacity of the heat transfer fluid.


DCLG. April 2006. Building Regulations. Approved Document L2A: Conservation of fuel and power (New buildings other than dwellings)

AECOM. May 2010. Keyworth II Bespoke BREEAM Assessment certification report, London South Bank University AECOM. June 2006. Keyworth II Site Wide Energy Statement, London South Bank University

DCLG. March 2010. NCM Database, NCM_db_activity

Metrima. Manufacture's specification - SVM MF4 Calculator for Mixed Fluid (-20Cto 150C).

Jeya Bavan, BEng, PhD, CEng


Tony Day, BEng, PhD, CEng

Jeya Bavan (MCIBSE) is a Principal ESD Engineer, AECOM, Perth, Australia. Tony Day (MCIBSE, FEI) is a professor in the Department of Urban Engineering, London South Bank University, London, UK.
COPYRIGHT 2012 American Society of Heating, Refrigerating, and Air-Conditioning Engineers, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2012 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Bavan, Jeya; Day, Tony
Publication:ASHRAE Transactions
Article Type:Report
Geographic Code:4EUUK
Date:Jan 1, 2012
Previous Article:Developing advanced virtual models of combined buildings and systems for energy reduction and integrating renewables.
Next Article:An analysis of photovoltaic performance on fire station buildings in London.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters