Study with the several resources on Docsity
Earn points by helping other students or get them with a premium plan
Prepare for your exams
Study with the several resources on Docsity
Earn points to download
Earn points by helping other students or get them with a premium plan
Community
Ask the community for help and clear up your study doubts
Discover the best universities in your country according to Docsity users
Free resources
Download our free guides on studying techniques, anxiety management strategies, and thesis advice from Docsity tutors
An in-depth analysis of the Seasonal Peak and On-Peak Coincidence Factors for Residential and Commercial & Industrial Lighting. The study compares the performance of logger data sources, sector types, and occupancy sensors during Summer and Winter seasons. The results show significant differences between Summer On-Peak and Seasonal Peak CFs, as well as between Large and Small C&I Lighting. The document also includes interactive tables and charts for a better understanding of the data.
Typology: Lecture notes
1 / 23
This page cannot be seen from the preview
Don't miss anything!
Coincidence Factor Study Residential and Commercial Industrial Lighting Measures
Acknowledgement:
The authors wish to thank all of the people at the State Program Working Group (SPWG), Northeast Energy Efficiency Partnership (NEEP) who took the time to support and help with this study. Regrettably, we cannot thank everyone individually, but we do want to acknowledge the contributions made by Julie Michaels, Jeff Schlegel Tom Belair, Carol White, Chris Neme and Ralph Prahl. The data, insight, and support provided by these individuals helped to establish the foundation for this report. RLW assumes sole responsibility for any errors or omissions in this report.
Coincidence Factors (CFs) are defined in this study as the fractions of the connected (or rated) load (based on actual lighting Watts, motor nameplate horsepower and efficiency, AC rated capacity and efficiency, etc.) reductions that actually occur during each of the seasonal demand windows. They are the ratio of the actual demand reductions during the coincident windows to the maximum connected load reductions. Under this definition other issues such as diversity and load factor are automatically accounted for, and only the coincidence factor will be necessary to determine coincident demand reductions from readily observable equipment nameplate (rated) information. In other words, coincident demand reduction will simply be the product of the coincidence factor and the connected equipment load kW reduction.
Table i - 1 and Table i - 2 provide the un-weighted and weighted, Summer On-Peak and Winter On-Peak CFs as well as the associated relative precisions for all residential lighting. The CFs were developed using only metered data that were acquired during the winter (December and January) or summer (June, July and August) peak months and the number of loggers used in the analysis is provided in the tables. The weighted CFs were developed by weighting the logger files based upon the connected load that the logger represents and in most cases the weighted results are slightly higher than the un-weighted results. The CFs for the summer range from a low of 0.06 for June to a high of 0.094 for August, with the average summer CF between 0. un-weighted and 0.082 weighted. If the average is carried to only two decimal places than the result is a summer average CF of 0.08 for both methodologies. The relative precision for the average summer on-peak period is ±6.1% at the 80% confidence interval.
Sample Size Un-weighted Weighted Un-weighted Data Period n CF CF Rel Precision June 210 0.060 0.069 ±11.6% July 102 0.081 0.086 ±12.5% August 189 0.094 0.092 ±8.7% Average Summer 501 0.076 0.082 ±6.1%
Summer On-Peak Hours 1PM - 5PM
Table i - 1: Summer On-Peak CFs and Relative Precisions Residential Lighting
Sample Size Un-weighted Weighted Un-weighted Data Period n CF CF Rel Precision December 282 0.263 0.281 ±6.5% January 264 0.301 0.320 ±6.5% Average Winter 546 0.286 0.298 ±4.5%
Winter On-Peak Hours 5PM - 7PM
Table i - 2: Winter On-Peak CFs and Relative Precisions Residential Lighting The winter CFs as expected are higher than the summer CFs ranging from 0.263 for December to 0.320 for January with the average winter CF for all lighting at 0.286 un-weighted and 0. weighted. The relative precisions is better during the winter peak periods primarily because the CFs are higher and there is less variation in the data, i.e. the Coefficient of Variation (Cv) is lower. The relative precision of the average winter un-weighted CF is ±4.5% at the 80% confidence interval and the December and January relative precisions are both better than ±10% at the 80% confidence interval.
The Seasonal Summer and Winter Peak performance hours were calculated using historical load data and the 50/50 Seasonal Peak Forecasts from the most recent Capacity Energy Loads and Transmission (CELT) reports. The seasonal peak performance hours were weighted based upon the frequency distribution of the hours observed where the load met or exceeded 90% of the 50/50 seasonal peak forecast and these values were used to calculate a weighted CF for each of the measure types. Table i - 3 and Table i - 4 provide the Summer Seasonal Peak and Winter Seasonal Peak CFs for all residential lighting. The CFs during the summer months range from a low of about 0.08 for June to a high of 0.10 for August, with an Average Summer CF of about 0.09. The relative precision during each of the summer months is within the range of ±10% at the 80% confidence interval. The Winter Seasonal Peak CFs as expected, are higher than the Summer Seasonal Peak CFs ranging from 0.25 in December to 0.28 in January with an Average Winter Seasonal Peak CF for all lighting at 0.26.
On-Peak Seasonal % Change Un-weighted Un-weighted Seasonal/ Data Period CF CF On-Peak December 0.263 0.249 95% January 0.301 0.279 93% Average Winter 0.286 0.264 92% Table i - 6: Comparison of Winter On-Peak and Seasonal Peak CFs Residential Lighting
A similar Coincidence Factor analysis was also conducted for Commercial and Industrial Lighting and Occupancy Sensor measures. The logger data were analyzed by sector so that results could be applied to multiple programs with different participation rates among the different sectors. Table i - 7and Table i - 8 provide the On-Peak CFs for the ten C&I sectors along with the associated relative precisions and total estimated CFs based on a logger weighted strategy and weighting each sector equally. The Summer On-Peak CFs indicates that the Grocery sector has the highest CF of about 0.95, while the Other sector has the lowest CF of about 0.54. All of the sectors have relative precisions that are within ± 5% at the 80% confidence interval. The Grocery sector also had the highest Winter On-Peak CF of about 0.78, while the School sector had the lowest CF of about 0.34. Once again the relative precisions were all quite good with each sector exceeding ± 10% at the 80% confidence interval. As expected the Winter On- Peak CFs were lower than the Summer On-Peak CFs for all of the C&I lighting sectors, because the performance hours occur later in the day as C&I facilities are shutting down and lighting is being switched off.
Sample Size Calculated Logger Calculated Calculated Sector Type n CF Weight CV Rel Precision Grocery 37 0.948 0.026 0.179 ±1.9% Manufacturing 169 0.729 0.119 0.488 ±2.4% Medical (Hospital) 58 0.769 0.041 0.425 ±3.6% Office 259 0.750 0.183 0.438 ±1.7% Other 192 0.543 0.136 0.675 ±3.1% Restaurant 43 0.811 0.030 0.347 ±3.4% Retail 166 0.824 0.117 0.342 ±1.7% University/College 70 0.680 0.049 0.483 ±3.7% Warehouse 59 0.781 0.042 0.359 ±3.0% School 362 0.633 0.256 0.503 ±1.7% 0.704 1.
Summer On-Peak Hours 1PM - 5PM
Total Weighted by Logger Total Equal Weight by Sector Table i - 7: Summer On-Peak CFs and Relative Precision C&I Lighting
Sample Size Calculated Logger Calculated Calculated Sector Type n CF Weight CV Rel Precision Grocery 37 0.776 0.026 0.474 ±7.1% Manufacturing 169 0.399 0.119 0.983 ±6.9% Medical (Hospital) 58 0.603 0.041 0.593 ±7.1% Office 259 0.537 0.183 0.725 ±4.1% Other 192 0.426 0.136 0.804 ±5.3% Restaurant 43 0.663 0.030 0.557 ±7.7% Retail 166 0.655 0.117 0.592 ±4.2% University/College 70 0.523 0.049 0.679 ±7.4% Warehouse 59 0.496 0.042 0.787 ±9.3% School 362 0.343 0.256 1.010 ±4.8% 0.480 1.
Winter On-Peak Hours 5PM - 7PM
Total Weighted by Logger Total Equal Weight by Sector Table i - 8: Winter On-Peak CFs and Relative Precision C&I Lighting
Table i - 9and Table i - 10 provide the Summer and Winter Seasonal-Peak CFs for the ten C&I sectors along with the associated relative precisions and total estimated CFs based on a logger weighted strategy and weighting each sector equally (which is the simple average of the CFs across all sectors. The Seasonal Peak Performance Hours were determined by analysis of historic ISO-NE Load Data and Forecast Data to determine the frequency distribution for each hour where the demand was greater than or equal to 90% of the seasonal forecast. A simple probabilistic weighting scheme was applied based upon the number of observation during each hour as described in section
Sample Size Calculated Logger Calculated Calculated Sector Type n CF Weight CV Rel Precision Grocery 37 0.770 0.026 0.44 ±4.6% Manufacturing 169 0.432 0.119 0.91 ±4.2% Medical (Hospital) 58 0.618 0.041 0.58 ±4.5% Office 259 0.539 0.183 0.71 ±2.6% Other 192 0.428 0.136 0.80 ±4.4% Restaurant 43 0.644 0.030 0.59 ±5.3% Retail 166 0.647 0.117 0.59 ±2.7% University/College 70 0.528 0.049 0.60 ±4.2% Warehouse 59 0.535 0.042 0.70 ±5.6% School 362 0.388 0.256 0.85 ±2.7% 0.497 1.
Winter Seasonal Peak Hours (90% of 50/50 Peak)
Total Weighted by Logger Total Equal Weight by Sector Table i - 10: Winter Seasonal Peak CFs and Relative Precision C&I Lighting Table i - 11 provides a comparison of the Summer On-Peak and Seasonal Peak CFs for each of the C&I sectors, which shows that for every sector the Summer Seasonal CFs are lower than the Summer On-Peak CFs. This means that if the C&I lighting were classified as Summer Seasonal Peak assets the demand reductions would be lower.
Table i - 11: Comparison of Summer On-Peak and Seasonal Peak CFs C&I Lighting
Table i - 12 provides a similar comparison of the Winter On-Peak and Seasonal Peak CFs for each of the C&I Lighting sectors. In this case the results are mixed, with 7 of the 10 sectors showing an
increase in the Winter Seasonal Peak CFs compared to the Winter On-Peak CF. This seems to indicate that in general for the winter, C&I lighting would have more demand reduction if classified as a Seasonal Peak asset.
Table i - 12: Comparison of Winter On-Peak and Seasonal Peak CFs C&I Lighting
Table i - 13 and Table i - 14 present the Summer On-Peak and Winter On-Peak CFs for occupancy sensors for seven of the ten C&I sectors as well as the total CFs for all seven sectors on a logger weighted basis and by weighting each sector equally. During the Summer On-Peak Period the occupancy sensors installed in the University/College sector had the highest CF of about 0.30, while the Other sector had the lowest CF of about 0.02. The Summer On-Peak CF for the remaining sectors ranged from about 0.21 for Manufacturing to 0.27 for the Office Sector. During the Winter On-Peak the Office sector had the highest CF of about 0.31 and the Other sector had the lowest CF of 0.09. The CFs for the remaining sectors ranged from a low of about 0.17 for the Warehouse sector to a high of about 0.23 for the University/College sector. The relative precision for all of the CFs were estimated by calculating the relative precision of the occupancy sensors profiles, since only aggregate savings profiles were developed for the analysis. In this case we would recommend using the logger weighted Total CFs since the relative precision for individual sector results are not that good particularly during the Winter period.
Sample Size Calculated Logger Estimated Estimated Data Period n CF Weight CV Rel Precision Manufacturing 12 0.198 0.035 0.712 ±8.9% Medical 59 0.239 0.170 0.649 ±3.6% Office 69 0.274 0.199 0.606 ±3.2% Other 56 0.024 0.161 0.808 ±4.6% University/College 16 0.283 0.046 0.720 ±7.6% Warehouse 77 0.246 0.222 0.700 ±3.3% School 58 0.209 0.167 0.739 ±4.2% 0.208 1.
Total Weighted by Logger Total Equal Weight by Sector
Summer Seasonal Peak Hours (90% of 50/50 Peak)
Table i - 15: Summer Seasonal-Peak CFs and Relative Precision C&I Occupancy Sensors
Sample Size Calculated Logger Estimated Estimated Data Period n CF Weight CV Rel Precision Manufacturing 12 0.172 0.035 1.063 ±17.3% Medical 59 0.221 0.170 0.827 ±6.3% Office 69 0.296 0.199 0.966 ±6.9% Other 56 0.066 0.161 0.990 ±7.7% University/College 16 0.231 0.046 0.819 ±11.9% Warehouse 77 0.183 0.222 0.986 ±6.6% School 58 0.159 0.167 1.140 ±8.7% 0.191 1.
Total Weighted by Logger Total Equal Weight by Sector
Winter Seasonal Peak Hours (90% of 50/50 Peak)
Table i - 16: Winter Seasonal-Peak CFs and Relative Precision C&I Occupancy Sensors
Table i - 17 and Table i - 18 provide a comparison of the Summer and Winter On-Peak and Seasonal Peak CFs for occupancy sensors for seven C&I sectors as well as the totals for all seven sectors calculated on a logger weighted and sector weighted basis. The results for the Summer period show that the Summer Seasonal CFs are lower than the On-Peak CFs for four of the seven sectors and for the total CF using both calculation methods. The results for the Winter period are similar, with five of the sectors having lower Seasonal Peak CFs and lower Total CFs using both calculation methods. Classifying the occupancy sensors as Seasonal Peak assets would result in a slight reduction in demand savings during both periods.
Table i - 17: Comparison of Summer On-Peak and Seasonal Peak CFs Occupancy Sensors
Table i - 18: Comparison of Winter On-Peak and Seasonal Peak CFs Occupancy Sensors
Several of the study sponsors wanted to calculate the C&I Lighting CFs using an alternative method that grouped the logger data into two categories, Large C&I and Small C&I since this provides them with results that are more in-line with their tracking systems, which track results at the program level. The sponsors also wanted to estimate the electrical demand impacts attributable to the interaction between the lighting and the HVAC systems and those results are presented in the following sections as well. The new C&I lighting CFs and interactive effects were only developed for C&I lighting measures, occupancy sensor measures were not included as part of the analysis.
for both the Large and Small C&I CFs are better than ± 3% at the 80% confidence interval. For the Winter, the results are similar to the On-Peak results however the Seasonal Peak values are slightly higher for both Large and Small C&I lighting. This is due to the fact that there are more morning and afternoon hours included in the CF calculation where the lighting operates at a higher percent on. Once again note that the relative precision for each of the C&I sector CFs is better than ± 5% at the 80% confidence interval.
Sample Size Calculated Calculated Calculated Program Type n CF CV Rel Precision Large C&I 408 0.714 0.416 ±2.6% Small C&I 496 0.613 0.493 ±2.8%
Summer Seasonal Peak Hours (90% of 50/50 Peak)
Table i - 21: Summer Seasonal Peak Coincidence Factors Large and Small C&I Lighting
Sample Size Calculated Calculated Calculated Program Type n CF CV Rel Precision Large C&I 408 0.595 0.590 ±3.7% Small C&I 496 0.431 0.738 ±4.2%
Winter Seasonal Peak Hours (90% of 50/50 Peak)
Table i - 22: Winter Seasonal Peak Coincidence Factors Large and Small C&I Lighting
Table i - 23 provides a comparison of the Summer On-Peak and Seasonal Peak CF for the Large and Small C&I Lighting, which shows that the On-Peak CF is higher than the Seasonal Peak CF. This is due to inclusion of more evening hours in the Seasonal Peak CF calculation when the percent on for the lighting is lower. This means that if the C&I Lighting measures were classified as Summer Seasonal Peak assets instead of Summer On-Peak assets the demand reduction would be lower for both Large and Small C&I Lighting. Table i - 24 provides the same comparison for the Winter On-Peak and Seasonal Peak CFs for C&I Lighting. In this case the Winter Seasonal Peak CFs are higher than the Winter On-Peak CFs. This indicates that for the Winter, both Large and Small C&I Lighting would have more demand reduction if classified as a Seasonal Peak asset.
% Change On-Peak Seasonal Seasonal/ Program Type CF CF On-Peak Large C&I 0.736 0.714 97% Small C&I 0.661 0.613 93%
Summer
Table i - 23: Comparison of Summer On-Peak and Seasonal Peak CFs C&I Lighting % Change On-Peak Seasonal Seasonal/ Program Type CF CF On-Peak Large C&I 0.576 0.595 103% Small C&I 0.418 0.431 103%
Winter
Table i - 24: Comparison of Winter On-Peak and Seasonal Peak CFs C&I Lighting
Table i - 25 and Table i - 26 provides the Interactive Summer and Winter On-Peak CFs for the Large and Small C&I sectors. For the Summer, the Large C&I has the higher Interactive On-Peak CF of about 0.14, while the Small C&I had an Interactive Summer On-Peak CF of 0.13. The Coefficient of Variation (CV) and relative precision are also provided and the relative precisions for both the Large and Small C&I CFs are better than ± 6% at the 80% confidence interval. For the Winter, the Small C&I sector has the higher Interactive On-Peak CF of about -0.05, while the Large C&I sector had an Interactive Winter On-Peak CF of -0.004. The Coefficient of Variation (CV) and relative precision are also provided and the relative precisions for both the Large C&I CF was ± 72% and the Small C&I CF was ± 27% at the 80% confidence interval. This was because CFs are so small and the coefficients of variation are so large.
Sample Size Calculated Calculated Calculated Program Type n CF CV Rel Precision Large C&I 376 0.139 0.718 ±4.7% Small C&I 425 0.125 0.907 ±5.6%
Interactive Summer On-Peak Hours 1PM - 5PM
Table i - 25: Interactive Summer On-Peak Coincidence Factors C&I Lighting^2
(^2) The number of log files (n) is lower than in the previous tables because we were unable to determine the maximum lighting demand reduction and were therefore unable to calculate the interactive demand reduction. In order to calculate the interactive CF it was necessary to calculate the interactive demand
