ASHENDEN WIND TURBINE TRIAL: PHASE I RESULTS

Size: px

Start display at page:

Download "ASHENDEN WIND TURBINE TRIAL: PHASE I RESULTS"

Madeleine George
6 years ago
Views:

1 ASHENDEN WIND TURBINE TRIAL: PHASE I RESULTS

2 PROJECT TEAM London Borough of Southwark: Tony Moseley, Sustainable Development and Infrastructure Manager; Client; data dissemination. Brian Dunlop Associates and Gas Dynamics Ltd: Planning application; technical co-ordination; instrumentation selection; commissioning; data acquisition software; quality assurance and data processing. London South Bank University: Prof. Tony Day, Dr. Steve Dance and students. Acoustic and vibration monitoring; analysis and reporting; wind and energy monitoring research programme. KCCC Ltd: Site survey; design and construction of footings and mounting frame; installation of anemometer mast. Photon Ltd: Lead installer construction programme; site manager; turbine assembly; erection and electrical installation

3 CONTENTS Introduction Background Trial objectives Target site description Target site yield predictions Installation Results Yield Anemometry Power Curve Yield reconciliation Annual yield estimate Carbon abatement costs Noise and vibration results Discharging of planning conditions

4 The purpose of this pilot wind turbine installation is to assess the viability of deploying small scale roof top turbines across the Elephant and Castle core development area with a view to generating a significant proportion of the Mayor's 10% renewable energy requirement. INTRODUCTION

CORE DEVELOPMENT AREA SHOWING LOCATION OF ASHENDEN Gas Dynamics Ltd and London South Bank University will continue to monitor the performance of both turbines.

5 CORE DEVELOPMENT AREA SHOWING LOCATION OF ASHENDEN Gas Dynamics Ltd and London South Bank University will continue to monitor the performance of both turbines. The results of Phase 2 will be posted on the Elephant and Castle website at a later date. gt on Bu tt s New d Roa NOTE: Phase 2 of the trial commenced in June The Proven HAWT has been moved to a new location further along the building and a second turbine, the Quiet Revolution QR5 (a vertical axis turbine), installed in its place. org er oad rth lwo Wa The performance of the turbine, based on 12 months of monitoring, is assessed in the presentation that follows. Ge Ne w in Phase I of the Ashenden wind turbine trial commenced in June A Proven 6kW horizontal axis wind turbine (HAWT) was installed on the roof of Ashenden House, an 11 storey residential block within the Heygate estate. This initial phase is now complete. Kent Road es ga t y e H et tre

6 TURBINE TRIAL OBJECTIVES Monitor site specific energy yield Compare with current prediction methods and manufacturer s data Use measured data to refine prediction of energy yield from desk-top studies Identify impacts of constraints e.g. size, weight, maintenance requirements etc Obtain acoustic data: pre- and post-installation Gauge community reaction to deployment of wind turbines on buildings e.g. visual impact Compare performance of turbine technologies and responses from local community Dissemination of site specific advice to planners and design teams

7 Planning TIMELINE Planning permission was granted in December The following files can be downloaded from the website: Planning application supporting documentation Conditions attached to planning consent Roofworks Carried out during March/ April Construction of concrete footings/ making good roof membrane. Design and fabrication of steel mounting frames. Turbine Trial The trial is being conducted in two phases. Phase 1 (June 2007 June 2008): COMPLETED Install the Proven WT6000 turbine above a vacant flat. Monitor wind speed, turbine power, noise and vibration. Analyse data, discharge planning conditions and release preliminary results. A series of photographs documenting the full construction/ turbine erection sequence can be downloaded from the website. Phase 2 (commenced June 2008) Relocate the Proven turbine above an occupied flat to make way for the installation of the Quiet Revolution QR5 6kW turbine. Ongoing monitoring of wind speed, power output from both turbines, noise and vibration emanating from the QR5 turbine. Analyse data and release final results. It is anticipated that at the end of this period the turbines will be demounted and relocated to London South Bank University for ongoing study.

8 TEST SITE DESCRIPTION 3-axis ultrasonic anemometer Proven 6kW HAWT Turbine(s) mounted on a 9m mast on top of an 11 storey tower block. 3-axis ultrasonic anemometer mounted at hub height ~10m from the turbine.

9 TURBINE DESCRIPTION PHASE 1 Proven WT 6000 rated at 6kW with a wind speed of 12 m/s 1 Cut-in wind speed of 2.5 m/s 1 Rotor diameter = 5.5m Power, W Wind speed (m/s) 1 Power curve and data extracted from Catalogue of European Urban Wind Turbine Manufacturers

10 MOUNTING FRAME

11 CONSTRUCTION SEQUENCE

12 INSTALLATION COSTS Roof works (exclusive of VAT) Common Crane Hire (one visits) 2, Handrail 8, Plinths 6, Sarnafil roofing 6, Lightning protection Walkway Total 26, Turbine Installation (exclusive of VAT) Common Proven QR5 Turbine, mast, inverter, controller & delivery 19, , Steel mounting frames (2 nr) Crane Hire (two visits) 6, Installation & commissioning 8, Meeting, site organization, health and safety Electrical items 4, Total 20, , ,925.00

TARGET SITE YIELD PREDICTION WIND ROSE TRANSPOSITION Essentially a technique for extrapolating wind resource data from one site to another, by correcting

13 TARGET SITE YIELD PREDICTION WIND ROSE TRANSPOSITION Essentially a technique for extrapolating wind resource data from one site to another, by correcting for differing terrain and altitude. 1. Using known data at a reference site (Heathrow) 2. Wind speed data is transposed to target site (Elephant & Castle)

14 YIELD PREDICTIONS REFERENCE SITE DATA FOR HEATHROW Wind speed distribution based on data supplied by the Met Office for Heathrow for the period Dec 97 to Nov % 0 15% Average wind speed = 3.96m/s % A = 4.46m/s, k = 1.9 5% < V (knots) < Probability < V (knots) < 8 8 < V (knots) < < V (knots) < < V (knots) Wind speed (knots) Wind speed distribution modelled using a 2 parameter Weibull function in the form: f(u) = k u k-1 A A k exp - u A Where A is the scale factor and k is the shape parameter

15 ESTIMATED WIND SPEED AND ENERGY YIELD PROFILES FROM DESKTOP STUDY Wind speed distribution based on data supplied by the Met Office for Heathrow for the period Dec 97 to Nov 07 Height (m) Wind speed (m/s) Yield (MWh) Upper and lower estimates of wind speed (left plot) and yield (right plot) for the target site. Estimated average wind speed m/s, giving an estimated annual yield figure of between 8,000 and 8,900 kwh.

1 Hz on a PC. Sunny-boy controller: Gross yield readings at ~15 min intervals, downloaded to a PC.

16 INSTRUMENTATION Anemometer: 3-axis Gill ultrasonic sensor logging continuously at 1 Hz on a PC. Multicube multifunction current transformer (CT) meter: Instantaneous power readings logging continuously at 1 Hz on a PC. Sunny-boy controller: Gross yield readings at ~15 min intervals, downloaded to a PC. Data analysis performed using SCILAB CT Meter Rectifier Inverter Data Acquisition PC Anemometer comms interface Sunny-boy controller

17 RESULTS TURBINE YIELD Yield (kwh) /Jun 27/Jul 15/Sep 4/Nov 24/Dec 12/Feb 2/Apr 22/May 11/Jul CF (%) time /Jun 27/Jul 15/Sep 4/Nov 24/Dec 12/Feb 2/Apr 22/May 11/Jul Turbine Yield kwh (blue curve) and cumulative capacity factor CF (red curve) for data between June 2007 and June The recorded annual yield from 07th June 2007 was 4,200kWh. time

18 RESULTS - ANEMOMETRY < V (knots) < 4 4 < V (knots) < < V (knots) < < V (knots) < < V (knots) % 0 20% 15% 10% 5% Probability Vmean = 7.4knots A = 8.4knots, k = Wind speed, knots Wind rose and wind speed distribution evaluated from all data collected between June 2007 to May 2008 inclusive. Distribution based on hourly averages. Average wind speed of 3.82 m/s, significantly less than the desktop estimate of m/s. Shape parameter k of 2.24 also much higher than the reference site (Heathrow).

19 RESULTS - ANEMOMETRY Ashenden House Heathrow Short Term Annual Data 0 < V (knots) < % 0 20% 15% 10% 5% < V (knots) < % 0 15% 10% 5% Wind rose comparison between the target site at Ashenden (left) and Heathrow (right). 4 < V (knots) < 8 8 < V (knots) < < V (knots) < < V (knots) < V (knots) < 8 8 < V (knots) < < V (knots) < < V (knots) Note: monthly wind roses for the test site are available to download from the website Jul % 35% 30% 25% 20% 15% 10% 5% % 0 25% 20% 15% 10% 5% < V (knots) < 4 0 < V (knots) < 4 4 < V (knots) < < V (knots) < < V (knots) < < V (knots) < V (knots) < 8 8 < V (knots) < < V (knots) < < V (knots) % 20% % 15% % % 60 Aug % 5% 90 0 < V (knots) < % 90 0 < V (knots) < 4 4 < V (knots) < < V (knots) < 8 8 < V (knots) < < V (knots) < < V (knots) < < V (knots) < V (knots) < < V (knots)

20 RESULTS POWER CURVES One minute average data Black dots denote raw data; Green curve denotes published power curve; Red curve denotes mean power in 0.5m/s intervals 0.6 Average power coefficient curve Black curve denotes published power curve; Red dots denote one hour average in 0.5m/s intervals (limited data points above 8 m/s); Blue dots denotes one minute average in 0.5m/s intervals; Green curve represents power curve used to reconstruct yield. Power Coefficient Wind speed (m/s)

21 V (m/s) V (m/s) V (m/s) V (m/s) dprob E (kwh) Cumulative yield % P (kw) RESULTS YIELD RECONCILIATION Annual yield calculation using annual Weibull parameters (A = 8.4 knots, k= 2.24) and the measured average power curve. Yield using Weibull distribution = 4,051kWh, compared with actual reading from inverters of 4,200kWh First plot shows the wind speed distribution in 0.25m/s intervals; Second plot shows the published power curve (red) and measured average power curve (blue); Third plot shows the yield in each wind speed interval using the published and measured power curve; Fourth plot shows the cumulative yield as a function of wind speed using the measured power curve note that by 8m/s the turbine has generated 90% of the total yield. Note also that the yield found using the published power curve is around 36% higher than the yield calculated using the measured average power curve.

22 RESULTS ANNUAL YIELD ESTIMATION BASED ON 12 MONTHS DATA Can we predict the annual yield at the target site in a 'mean' year? i.e. is the wind speed distribution at the target site representative of an average year? Begin by comparing short-term wind data (from June 2007 to May 2008) with the long-term distribution at the reference site (Heathrow). Plots below (from left to right) show; annual wind speed distribution at Ashenden House; annual wind speed distribution at Heathrow; long-term average distribution at Heathrow. Note that the scale factor for the Heathrow short term data is considerably higher than the long term average. If the same trend was applied to the Ashenden dataset, the average wind speed would be 3.44m/s, resulting in a mean yield of around 2,800 to 3,050 kwh (cf 4,200kWh in the year to June 2008). Ashenden House Heathrow Short Term Heathrow Long Term Vmean = 7.4knots A = 8.4knots, k = Vmean = 8.7knots A = 9.72knots, k = Vmean = 7.7knots A = 8.67knots, k = Probability Probability Wind speed, knots Wind speed (knots) Wind speed (knots)

23 CARBON ABATEMENT COST Comparison of cost per kg of carbon abated for three renewable energy technologies based on detailed consideration of the available resource and the physical constraints of the Elephant and Castle development: roof mounted pv (limited to roof areas receiving kwh/m 2 ) MAGENTA LINE; 6kW wind turbines (spaced at 5 times diameter apart) RED LINE centrally located biomass boilers (i.e. operating as peak load boilers and part of a shared infrastructure feeding an area-wide district heating network) BLACK LINE 6kW Proven roof mounted turbine (extrapolated from measured data) AEC Cost ( /kgco2) Limited by available space in energy centre Limited by number of fuel deliveries during peak month (10 lorry loads) Limited by heat demand KYOCERA KC w panels on Solion Sunmount (data supplied by LSBU) BLUE SHADED AREA INDICATES UPPER AND LOWER LIMITS FROM RENEWABLES TOOLKIT %Carbon abatement vs Discounted Abatement cost

NOISE AND VIBRATION STUDY Background noise study The plot below shows the 15 minute equivalent noise levels at six sites on and around the target building.

24 NOISE AND VIBRATION STUDY Background noise study The plot below shows the 15 minute equivalent noise levels at six sites on and around the target building. Major source of noise at Ashenden House is the A2. The location is in central London and as such even in the middle of the night there are significant levels of noise. A B C D E F 11th floor bedroom of flat th floor balcony opposite flat 223 Roof top By the road side Garages behind target building 85m behind target building

NOISE AND VIBRATION STUDY Turbine noise levels 5minute equivalent noise levels presented as a function of wind speed over a three-day period from 23rd

25 NOISE AND VIBRATION STUDY Turbine noise levels 5minute equivalent noise levels presented as a function of wind speed over a three-day period from 23rd - 25th July There appears to be no correlation with wind speed, although further data is required to establish noise levels at higher wind speeds.

26 NOISE AND VIBRATION STUDY Vibration Study Plots show vibration levels for various wind speeds recorded on the turbine frame (bottom plot) and a wall in flat 223, located immediately below the turbine (top plot). The coloured lines represent the human body's ability to work efficiently under various vibration periods. In both cases the vibration levels are below the threshold tolerance for continuous working (24hrs).

27 PHASE I CONCLUSIONS Yield well below expectations based on desk top studies using transposed wind data from Heathrow and manufacturer's power curve Sources of this discrepancy are 1) the manufacturers power curve which indicated higher power outputs at any given wind speed in particular in the 3m/s to 8m/s range and 2) lower mean wind speed coupled with a tighter distribution around the mean (higher shape factor than Heathrow data) No measurable impact on noise levels both externally and within the flat due to operation of the wind turbine Vibration levels within the flat are well below the comfort criteria for 24hr working Residents supportive of the project. No adverse reaction from occupants or members of the public to date Proceed with Phase II

Ashenden House Wind Turbine Trial. CIBSE ASHRAE Seminar May 2009

Ashenden House Wind Turbine Trial CIBSE ASHRAE Seminar May 2009 PROJECT TEAM London Borough of Southwark: Tony Moseley, Sustainable Development and Infrastructure Manager; Client; data dissemination. Brian