Topic:"Energy Efficiency Resource Assessment Model" in M12861

.. 27 15 Figure 15: Yearly Change in Customers, Population, and Housing Completions........................ 29 16 Figure 16: Yearly Change in Residential Customers.................................................................... 30 17 F...

AI summary The text lists figures analyzing customer trends, economic drivers (residential, commercial, industrial), energy forecasts (heat pumps, EVs, hybrid heating), and load modeling scenarios. It focuses on data visualization for regulatory proceedings, including residential and commercial energy demand projections.

.............................................................. 97 30 Figure 72: Commercial End-Use Peak Shares .............................................................................. 97 31 Figure 73: Comparison of bottom-up and top-...

AI summary The 2026 Load Forecast Report includes figures analyzing energy demand patterns, peak load forecasts, and system sensitivity. Attachments detail residential and commercial intensity models, demand forecasting methodologies, and peak load inputs, supporting the Integrated Resource Plan (IRP) scenarios discussed in Figure 77.

D 1 Figure 9: HDD Trend 2 3 4 Figure 10: CDD Trend 5 6 DATE: May 15, 2026 Page 24 of 105 REDACTED (CONFIDENTIAL INFORMATION REMOVED) 2026 Load Forecast Report REDACTED 1 These trends are reduced over time (approximately 40 years) such that...

AI summary The 2026 Load Forecast Report discusses trends in Heating Degree Days (HDD) and Cooling Degree Days (CDD) over a 10-year period, showing a decrease in HDD and an increase in CDD. The report also details updates to the peak temperature formula used in the load forecasting model, incorporating a 24-hour lagged temperature for improved accuracy.

Page 25 of 105 REDACTED (CONFIDENTIAL INFORMATION REMOVED) 2026 Load Forecast Report REDACTED 1 Consistent with the updates made in the 2025 model, the 12-hour lagged temperature component 2 in this calculation is based on a rolling 10-yea...

AI summary The 2026 Load Forecast Report incorporates updated temperature data, including a 12-hour and 24-hour lagged temperature component, based on rolling 10-year averages. These adjustments aim to reduce year-to-year fluctuations and better reflect long-term warming trends in Nova Scotia. A climate change trend is also included, showing a gradual increase in peak day temperatures over time.

1 The Climate Atlas of Canada 8 provides estimates for annual HDD as well as number of “Winter 2 Days” that are -15°C or colder under different climate change scenarios between now and 2050. 3 Annual HDD for Halifax is forecast to decline...

AI summary The text discusses climate change impacts on heating and cooling degree days in Halifax, referencing the Climate Atlas of Canada and Signal49's 20-year economic forecast. It highlights declining winter temperatures and increasing summer temperatures, as well as the influence of economic data on load forecasting.

IAL INFORMATION REMOVED) 2026 Load Forecast Report REDACTED 1 Figure 16: Yearly Change in Residential Customers 2 3 4 Regarding household size, population is currently used in combination with customer count to 5 estimate average household...

AI summary The document discusses the estimation of average household size in the SAE model using population and customer count data, and how changes in household size affect usage variables. It references figures illustrating yearly changes in residential customers and household size over time.

1 Figure 21: Economic Forecast Comparison GDP Employment Housing Starts 2026 (%) 2027 (%) 2026 (%) 2027 (%) 2026 2027 Signal49 9 1.3 1.8 0.6 0.2 8181 6771 BMO 10 1.4 1.8 0.5 0.6 8500 8000 RBC 11 1.5 1.6 0.4 0.4 7800 6000 TD 12 1.6 1.2 0.3...

AI summary The document discusses economic forecasts for GDP, employment, and housing starts through 2027, as well as the use of end-use data from NRCan and the EIA to develop load forecasts. Historical data and efficiency estimates are used to model residential and commercial energy consumption trends.

e consistent with 17 NRCan reported end-use consumption and actual average use derived from NS Power billing data. 18 19 In the case of end uses where there is little historical activity or where future behaviour is expected 20 to vary sig...

AI summary The text discusses the modeling of end-use consumption, noting that heat pumps and rooftop solar PV are modeled outside average use intensities due to limited historical data and the potential for significant future behavior changes. This approach allows for better tracking and fine-tuning of future forecasts.

0 56 44 71 3,232 87 283 2036 170,636 55 45 73 3,317 89 290 2 3 4.5.2 Hybrid Heating 4 Residential Hybrid Heating 5 6 Residential hybrid heating assumptions have been updated for the 2026 Load Forecast, consistent 7 with the Board’s 2025 Lo...

AI summary The 2026 Load Forecast Report updates residential hybrid heating assumptions, aligning with the 2025 Load Forecast Decision. The report includes a regression model that incorporates potential energy and peak reductions from hybrid programs, modeled by NS Power and used by a Department of Energy-led working group with Net Zero Atlantic and other stakeholders.

1 Figure 24: Range of Modelled Hybrid Peak and Energy Reductions Peak Reduction Energy Reduction (MW) (GWh) Year Min Max Min Max 2028 -2 -18 -3 -23 2029 -4 -36 -5 -47 2030 -7 -54 -8 -70 2031 -9 -71 -10 -92 2032 -11 -87 -12 -114 2033 -13 -1...

AI summary The document discusses the 2026 Load Forecast, which adopts updated modelling results showing a peak reduction of −48 MW and an energy reduction of 83 GWh by 2036. The forecast incorporates residential hybrid heating as a distinct end-use and splits the NAE heat-pump category into hybrid and non-hybrid segments based on a 50% participation rate.

1 to achieve benefits to the system (see Section 10.4). Figure 27 shows the expected changes in 2 electric water heater saturation and overall intensity over the forecast period. 3 4 Consistent with the Board’s direction in the 2025 Load F...

AI summary The document discusses the modeling of heat pump water heaters as a separate end use in Nova Scotia, their current and projected saturation rates, and factors influencing their growth, including U.S. efficiency standards and a market transformation pilot by E1. Efficiency improvements are modeled with heat pump water heaters at 40% of standard electric water heaters.

https://ecologyaction.ca/sites/default/files/2023-05/RegionalZEVAdoptionOptions_Dunsky_March2023.pdf, 24 prepared by Dunksy Energy+Climate Advisors for the Ecology Action Center DATE: May 15, 2026 Page 45 of 105 REDACTED (CONFIDENTIAL INFO...

AI summary The 2026 Load Forecast Report discusses the impact of electric vehicles (EVs) on residential energy sales and peak demand based on analysis of customer-level AMI data. It estimates a per-customer load increase of 3820 kWh per year and a coincident peak impact of 0.39 kW for at-home charging. Commercial and MDV/HDV charging impacts are estimated using E3's EV Load Shaping Tool.

1 Figure 29: EV Mileage Assumptions and Load/Peak Modeling Results Vehicle Avg kW/vehicle Avg kWh/year Type on Peak LDV 4,202 0.6 MDV 8,205 1.6 HDV 113,890 7.3 2 3 Figure 30 provides the estimated energy and peak impacts that correspond to...

AI summary The text provides figures analyzing the impact of electric vehicles (EVs) on energy and peak load forecasts, including average kilowatt usage per vehicle type and projected cumulative energy and peak load impacts for the years 2026 to 2029.

Page 47 of 105 REDACTED (CONFIDENTIAL INFORMATION REMOVED) 2026 Load Forecast Report REDACTED

AI summary The 2026 Load Forecast Report provides an analysis of projected electricity demand in Nova Scotia for the year 2026. The report includes key factors influencing load forecasts, such as weather patterns, economic trends, and the impact of energy efficiency programs.

lowance. The forecast is for total installed capacity of 655 21 MW by 2036. This assumes no change to the incentive landscape in the coming years. The overall 22 impact is shown in Figure 31. There is no forecast reduction in NS Power peak...

AI summary The document discusses a forecast of total installed capacity reaching 655 MW by 2036, assuming no changes to current incentive structures. It also notes that solar generation does not reduce NS Power's peak demand due to non-coincidence with system peak times.

ters and printers 26 • Misc: other loads including motors, servers, escalators, medical equipment, etc. Small 27 scale solar and EV load have also been included in this category. 28 DATE: May 15, 2026 Page 53 of 105 REDACTED (CONFIDENTIAL...

AI summary The 2026 Load Forecast Report discusses historical and projected end-use intensities for commercial sectors, noting updated baseline data from the EIA 2025 Annual Energy Outlook. The report highlights a reduction in residential energy usage due to new appliance standards, while commercial energy intensity has increased.

2026 Load Forecast Report REDACTED 1 Figure 39: Historical and projected real electricity prices (real dollars per kWh) 2 3 4 Prices impact the class sales through imposed price elasticities. The SAE models are estimated 5 using a price el...

AI summary The 2026 Load Forecast Report discusses the impact of electricity prices on sales, using a price elasticity of -0.15. It also outlines the role of Demand Side Management (DSM) in the forecast, citing specific DSM plans and applications currently under review by the NSEB.

1 5. RESIDENTIAL SECTOR 2 3 The Residential sales forecast is generated as the product of a residential average use forecast and 4 a customer count forecast. The residential average use model is specified using a SAE model 5 structure and...

AI summary The residential sector sales forecast is based on average use and customer count projections, with EVs and solar contributing to load. The forecast accounts for billing issues from a cyber incident, using accrued sales for 2025. Weather-normalized sales grew by 0.9% in 2025, driven by new customers and heat pumps.

Page 67 of 105 REDACTED (CONFIDENTIAL INFORMATION REMOVED) 2026 Load Forecast Report REDACTED

AI summary The 2026 Load Forecast Report provides an analysis of expected electricity demand in Nova Scotia for the year 2026. The report includes projections based on various factors such as population growth, economic development, and energy efficiency initiatives.

increase in the energy purchases under the BUTU Tariff (expected to 23 increase from around 25 GWh to 79 GWh). 24 25 Because NS Power is required to provide back-up capacity, including reserve margin, to the 26 municipal electric utilities...

AI summary The text discusses an increase in energy purchases under the BUTU Tariff, from 25 GWh to 79 GWh, due to NS Power's obligation to provide backup capacity to municipal electric utilities. It also mentions system losses and unbilled sales, with system losses forecasted to remain between 6.0% and 7.0% over the next 10 years.

1 9. NET SYSTEM REQUIREMENT 2 3 The NSR is the energy required to supply the sum of residential, commercial, and industrial 4 electricity sales, plus the associated system losses, within the province of Nova Scotia. Loads 5 served by indus...

AI summary The Net System Requirement (NSR) in Nova Scotia is calculated based on residential, commercial, and industrial electricity sales, plus system losses. The 2025 NSR was slightly lower than forecast due to colder weather and a large customer variance. From 2026 to 2036, NSR is expected to grow at 0.4% annually, driven by new customers, heating, and EV adoption, partially offset by solar, DSM, and RTR initiatives.

-25 -411 models 2 37 This corresponds to the portion of energy provided by NS Power as top-up under the Energy Balancing Service tariff (as outlined in Section 4.8). DATE: May 15, 2026 Page 82 of 105 REDACTED (CONFIDENTIAL INFORMATION REMO...

AI summary The document refers to a portion of energy provided by NS Power under the Energy Balancing Service tariff, as outlined in Section 4.8. It also mentions the 2026 Load Forecast Report, which has been redacted.

on are described in Figure 62. 10 DATE: May 15, 2026 Page 85 of 105 REDACTED (CONFIDENTIAL INFORMATION REMOVED) 2026 Load Forecast Report REDACTED 1 Figure 61: Peak Normalization Regression Model Details SUMMARY OUTPUT Regression Statistic...

AI summary The document presents a regression analysis from the 2026 Load Forecast Report, highlighting a model with a high R-square value of 0.669, indicating strong explanatory power. Key variables include wind, 12-hour lag, 24-hour lag, and weekdays, with coefficients showing significant impacts on load forecasting.

-2.86370325 0.353258208 -8.106544 5.6E-16 -3.55613436 -2.17127215 -3.55613436 -2.17127215 2 Weekdays 41.4298405 2.545256524 16.277275 4.8E-59 36.4408134 46.4188677 36.4408134 46.4188677 3 4 Figure 62: Peak Normalization Regression Coeffici...

AI summary The document discusses peak normalization regression coefficients, including the impact of weekdays, wind speed, and temperature on electricity demand peaks. It outlines how large customer contributions to the system peak are calculated using historical load factors and forecasts.

1 km/h. The recorded peak was 2,267 MW with a firm peak of 2180 MW. Figure 63 provides a 2 breakdown of actual system peak compared to the forecast for 2025, using the variables from the 3 previous load forecast. 4 5 Figure 63: Forecast Pe...

AI summary The document discusses the 2025 forecast peak variance compared to actual system peak, highlighting factors such as interruptible load, weather, wind, and morning peak impact. The actual peak was 2,267 MW, lower than the forecast of 2,403 MW, with various contributing factors identified.

1 contribution to peak is expected to be partially mitigated via utility managed charging. The firm 2 peak assuming the current non-coincident residential EV peak value of 0.5kW/vehicle and that 3 commercial charging does not include peak...

AI summary The text discusses the impact of electric vehicle (EV) charging and hybrid heating on peak demand, noting that utility-managed charging and DR programs are expected to mitigate some of the growth in peak demand. It also references the 2026 Load Forecast and the inclusion of DR programs from the 2022 Evergreen IRP, with future DR amounts based on achievable potential.

Page 93 of 105 REDACTED (CONFIDENTIAL INFORMATION REMOVED) 2026 Load Forecast Report REDACTED

AI summary The 2026 Load Forecast Report has been redacted, with confidential information removed. The report likely contains projections and analysis related to electricity demand in Nova Scotia for the year 2026.

ariation in weather conditions at time of system peak, as well as the time of 5 peak occurrence. 6 7 Figure 74: Comparison of annual sales and coincident peak for non-large customer classes 8 9 10 Empirical evidence suggests the assumed re...

AI summary The document discusses the relationship between peak load and sales for residential and non-residential customer classes. It highlights that the correlation between energy use and peak load is strong for residential customers but weak for non-residential classes, leading to potential overestimation of coincident peak growth when using a bottom-up scaling approach.

GWh during the 10-year period, which is explained mainly by 21 the impact of weather variation (HDD) and economic impact in the long term. The black line 22 represents actual system totals. 23 DATE: May 15, 2026 Page 102 of 105 REDACTED (C...

AI summary The document discusses the 2026 Load Forecast Report, highlighting the impact of weather variation (HDD) and economic factors on energy demand over a 10-year period. It also presents a P10/P90 scenario for peak demand, showing a range of 352 – 391 MW, and includes adjustments to the peak end-use model based on wind and temperature averages.

Appendix A – Forecast Values 1.1 Table A1: Energy Requirement – 2026 NS Power Forecast Energy Forecast

AI summary Appendix A presents Table A1, which outlines the 2026 energy requirement forecast by NS Power. This table provides a forecast of energy needs for the year 2026, likely used for planning and regulatory purposes.

Residential Commercial Industrial Municipal Total Year Sector Growth Sector Growth Sector Growth and Other Growth Losses Energy Growth GWh % GWh % GWh % GWh % GWh GWh % 2016 4,264 -5.3% 3,133 -3.7 2,444 -0.5 179 -9.2 789 10,809 -2.6 2017 4...

AI summary The table presents energy consumption growth data across various sectors (residential, commercial, industrial, municipal, and other) from 2016 to 2028, showing fluctuating trends in energy usage, including both increases and decreases in different years.

3 of 3 Appendix A – Forecast Values Table A2: Coincident Peak Demand - 2026 NS Power Forecast Peak Forecast

AI summary The document presents a table titled 'Table A2: Coincident Peak Demand - 2026 NS Power Forecast Peak Forecast' which provides forecast values for peak demand in Nova Scotia for the year 2026.

Interruptible Demand Firm Net Temp at 12hr Lag 24hr Lag Contribution Response Contribution Growth System Peak Temp Temp Year to Peak (reduction in to Peak Notes Peak Firm Peak only, (%) (MW) MW) (MW) (deg C) (deg C) (MW) (deg C) - December...

AI summary The table presents data on interruptible demand and firm peak contributions to system peak in Nova Scotia from 2016 to 2018. It includes metrics such as net growth, temperature at peak, and temperature lags for each year, with notes indicating specific dates and conditions.

t . REDACTED (CONFIDENTIAL INFORMATION REMOVED) 2026 Load Forecast Report Appendix B Page 1 of 34 Appendix B – Forecast Model Details 2026 NS Power Load Forecast REDACTED (CONFIDENTIAL INFORMATION REMOVED) 2026 Load Forecast Report Appendi...

AI summary The residential average use SAE model is based on end uses such as heating, cooling, and other uses, incorporating variables like short-term utilization, efficiency trends, and saturation. The model uses elasticity factors and a base year of 2020 for calculations.

s its own elasticity, and the base line year is 2020. HeatIndex is defined as: HeatIndex = f(Heating Saturation, Efficiency, Shell Integrity, Square Footage) The cooling variable is defined as: XCool = CoolUse × CoolIndex REDACTED (CONFIDE...

AI summary The text defines key variables and indices used in load forecasting, including HeatIndex, CoolIndex, and XOther. These indices are calculated using factors like elasticity, baseline year 2020, and variables such as Heating Degree Days, household size, and electricity price. The formulas incorporate various factors affecting energy use and efficiency.

Factor REDACTED (CONFIDENTIAL INFORMATION REMOVED) 2026 Load Forecast Report Appendix B Page 16 of 34 General Service The General Service rate class model is estimated on a total monthly sales basis where total monthly billed sales is a fu...

AI summary The General Service rate class model estimates monthly billed sales based on heating, cooling, and other usage, incorporating factors like price elasticity, GDP, employment, HDD, CDD, and days in a month. Adjustments are made for specific events and post-pandemic usage shifts, with an ARMA process improving the model.

basis: OtherAvgMWm = OtherLoadm/ Daysm /24 The PkWindVarm is the average daily windspeed on monthly peak days. Variable Coefficient StdErr T-Stat P-Value mVarsNew.Heat_Var 1.605 0.097 16.529 0.00% mVarsNew.Cool_Var 0.944 0.185 5.111 0.00%...

AI summary The text provides a statistical model related to load forecasting, including variables such as heat and cooling demand, monthly other load factors, and peak wind speed. The model includes coefficients, standard errors, t-statistics, and p-values for each variable, indicating strong statistical significance.

INFORMATION REMOVED) 2026 Load Forecast Report Appendix B Page 33 of 34 Peak Model Statistics Model Statistics Iterations 1 Adjusted Observations 120 Deg. of Freedom for 105 Error R-Squared 0.975 Adjusted R-Squared 0.971 AIC 8.140584 BIC 8...

AI summary This section presents statistical metrics for a peak load forecasting model, including R-squared, adjusted R-squared, AIC, BIC, and other performance indicators, with some values marked as #NA. It also includes diagnostic statistics such as the Durbin-Watson and Ljung-Box tests, as well as measures of skewness and kurtosis.

Figure C4: Energy Forecast Accuracy NSR less mills Forecast Forecast Forecast Forecast Forecast Forecast Forecast Forecast Forecast Forecast Forecast for: for: for: for: for: for: for: for: for: for: Issued 2016 2017 2018 2019 2020 2021 20...

AI summary This table presents energy forecast accuracy data for NSR less mills from 2015 to 2025, comparing forecasted values with actuals. The data shows variations in forecast accuracy over time, with discrepancies between forecasts issued in different years and actual outcomes.

. From these annual forecast distributions, the various probabilistic forecasts can be obtained as well as sensitivity diagrams that show the relative impact of the variables in each year. In Figure D4 the probabilities of system peak fore...

AI summary The text discusses probabilistic forecasts and sensitivity diagrams for system peak demand, highlighting the impact of variables on peak demand forecasts. It explains the asymmetry in peak forecast distributions due to the use of the MAX function on monthly heating degree day data.

REDACTED (CONFIDENTIAL INFORMATION REMOVED) REDACTED 2026 Load Forecast Report Appendix D Page 6 of 8 Figure D5: Relative Sensitivity of Peak In terms of the sensitivity of the energy sales forecast to the various input variables, Figures...

AI summary The document discusses the sensitivity of energy sales forecasts to various input variables, highlighting that weather has the strongest near-term impact while economics becomes equally important in the long term. Demand-side management (DSM) has the largest impact on both energy and peak demand, with solar, EVs, and hydrogen facilities also having significant effects.

. REDACTED (CONFIDENTIAL INFORMATION REMOVED) 2026 Load Forecast Report Appendix E Page 7 of 21 EV Load • 2025 sales higher than forecast in spite of the cancellation of federal and provincial rebates (2,400 vs forecast of 1,700). • 75% ZE...

AI summary The 2026 Load Forecast Report highlights higher-than-expected EV sales in 2025 despite rebate cancellations, a lower ZEV target by 2035 compared to the 2025 forecast, and slower-than-expected uptake of behind-the-meter solar installations in 2025 due to the end of certain rebate programs.

D (GWh) (GWh) +477 2035 11,365 11,842 (4.2%) Growth -0.2% / y 0.4% / y (10y avg) 16 . REDACTED (CONFIDENTIAL INFORMATION REMOVED) 2026 Load Forecast Report Appendix E Page 17 of 21 Forecast Comparison – Energy • 2026 Net System Requirement...

AI summary The 2026 Load Forecast Report indicates that the Net System Requirement (NSR) is expected to be higher than the 2025 Load Forecast due to changes in the timing of RTR migration, decreased solar generation, and increased EV load. The greatest variance between the 2025 forecast and actual NSR is attributed to large customer load, primarily from one customer.

Muni Forecasts Apr-09 to Mar-10 Domestic Commercial Industrial Losses Total 74.6 97.2 20.6 7.7 200.10 GWh 37.3% 48.6% 10.3% 3.8% 100% old1 37.3% 48.6% 10.3% 3.8% 100% old2 38.5% 47.5% 8.0% 6.0% 100% TOTAL Total NS SECTOR TOTALS: MUNICIPAL...

AI summary The text presents electricity consumption forecasts for the period April 2009 to March 2010, including domestic, commercial, industrial, and losses data. It provides percentages and total consumption in gigawatt-hours (GWh) for different sectors and includes historical comparisons.

Incremental ENERGY Total load Load Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Total 2016 2017 0 2018 0 2019 0 2020 0 100.0 2021 0 2022 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 2023 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0...

AI summary The table presents incremental energy load data across various years, showing total load and monthly breakdowns from 2016 to 2028. The data indicates a significant increase in load starting from 2024, with notable figures in 2026 and 2027, and a large total value of 2475.0 in 2026.

Cumulative Incremental PEAK Peak Peak Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Total 2016 2017 2018 2019 2020 2021 Check 2022-2024 2022 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 2023 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0...

AI summary The text presents a table showing projected peak energy demand from 2016 to 2033, with incremental and cumulative values for each year. The data indicates a steady increase in demand, particularly from 2024 onwards, with significant growth expected by 2033.

Incremental ENERGY Total load Load Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Total 2016 2017 0 2018 0 2019 0 2020 0 2021 0 2022 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 2023 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0....

AI summary The document presents a table showing incremental energy load data from 2016 to 2029, with specific values for each month and the total annual load. The data shows a gradual increase in energy load, with a significant jump in 2024 and a decline in 2025 followed by a steady increase from 2026 onwards.

Incremental ENERGY Total load Load Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Total 2016 2017 0 2018 0 2019 0 2020 0 2021 0 2022 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 2023 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0....

AI summary The document presents a table showing incremental energy load data from 2016 to 2028, with the majority of years showing zero load except for 2025 and onwards, where small increases are noted starting in 2025 and continuing through 2028.

Incremental ENERGY Total load Load Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Total 2016 2017 0 2018 0 2019 0 2020 0 2021 0 2022 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 2023 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0....

AI summary The document presents a table showing incremental energy load data for various years from 2016 to 2028, with load values for each month and the total annual load. The data shows minimal changes in load from 2022 to 2024, with a slight increase beginning in 2025 and continuing through 2028.

Incremental ENERGY Total load Load Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Total 2016 2017 0 2018 0 2019 0 2020 0 2021 0 2022 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 2023 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0....

AI summary The table presents incremental energy load data for various years, showing total load and monthly breakdowns. The data starts from 2016 up to 2029, with significant values beginning in 2025 and increasing slightly each year.

Incremental ENERGY Total load Load Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Total 2016 2017 0 2018 0 2019 0 2020 0 2021 0 2022 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 2023 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0....

AI summary The text presents a table showing incremental energy load data from 2016 to 2029, with significant increases noted starting in 2025. The data shows a pattern of minimal load in earlier years and a gradual rise beginning in 2025, with consistent values for most months after that.

2952.5 2339.8 5,000 Historic Sales Previous Forecast Current Forecast 2021 10902.2 1.7% 10902.2 201620172018201920202021202220232024202520262027202820292030203120322033203420352036 2973.5 2480.3 2022 11133.9 2.1% 11133.9 3088.3 2480.2 Hist...

AI summary The text presents numerical data including sales figures, percentages, and forecasts spanning multiple years, with references to terms such as 'Historic NSR' and 'Current Forecast'. These figures appear to be related to energy sales and demand projections over time.

LG LI LG LI Current Forecast Current ForecasPrevious Forecast Previous Forecast 2010 416.1 3162.7 2011 414.9 2769.8 2012 410.8 1417.3 2013 404.1 1863.6 2014 393.5 1799.0 Other Industrial Sales Large General Sales 2015 414.5 1723.2 420 2016...

AI summary The text presents historical data on Large General Sales and Other Industrial Sales from 2010 to 2021, showing trends over time with numerical values for each year. It includes a visual representation with a graph indicating changes in sales volume.

Codes and Residential Commercial Residential Commercia Total Residential Residential Commercial Commercial Industrial Industrial LED Standards Total Total Loss Loss Residential Cummulativ Commercial l Industrial Industrial Incrementa Total...

AI summary The text presents a table with data on residential and commercial energy usage, including incremental and cumulative values, percentages, and loss metrics over several years, starting from 2008 to 2019. The data appears to track energy efficiency measures, such as LED standards, and their impact on energy consumption and losses.

Energy Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Total 2026 1265.4 1164.3 1149.6 960.9 844.4 774.3 858.8 835.1 755.3 825.9 958.6 1182.0 11574.6 2027 1250.1 1149.7 1130.7 945.8 834.1 765.0 847.0 828.2 747.2 813.7 942.6 1170.3 11424.5...

AI summary The document presents a table with monthly and annual energy consumption data from 2026 to 2036, showing a consistent pattern of energy usage across years. The data is part of a load forecast report, which is redacted and confidential.

1 Request IR-1: 2 3 Reference: Exhibit N-1, Section 4.5.2. 4 5 (a) Please provide a narrative description of each of the "potential hybrid programs 6 which have been modelled by NS Power." (p. 37) 7 8 (b) Please provide the modelling refer...

AI summary The document contains a request (IR-1) directed to NS Power, asking for detailed information on potential hybrid programs, modeling data, documents from a hybrid heating working group, details on heat pump systems, explanations for setting heat use to zero, and clarification on peak mitigation forecasts and energy reduction figures.

(c) Please refer to CA IR-1 part (f). Hybrid event trigger scenario Event hours CPP events 72 Temperature < -10o C 225 Top 88 NSR hours 88 1 (d) 2 (i) Based on a residential average use of 10,367 kWh for 2036 (from 2026 Load 3 Forecast Rep...

AI summary The text discusses hybrid event trigger scenarios and their associated event hours, referencing load forecasts and energy usage data. It also includes a request and response related to hybrid heating scenarios in the 2026 Load Forecast and their alignment with the Net Zero Atlantic study.

NON-CONFIDENTIAL 1 Request IR-1: 2 3 Figure 3: Historic and Forecast Net System Requirement and System Peak. 4 5 (a) The Net System Requirement (NSR) forecast percentage change in 2027 is negative. 6 Aside from Renewable to Retail sales, w...

AI summary The document includes a request (IR-1) asking Nova Scotia Power (NS Power) to explain the factors leading to a decline in Net System Requirement (NSR) in 2027, the reasons for the projected NSR increase in 2031 and 2033, and why the system peak forecast in 2025 underestimated the peak for 2028 to 2032. A response begins addressing the first part of the request.

1 Request IR-1: 2 3 Refer to M12861, Exhibit-N1-Nova Scotia Power Inc. 2026 Load Forecast Report (filed May 4 15, 2026) (the "Report"), Figure 37, and respond to the following: 5 6 (a) Figure 37 shows cumulative industrial electrification...

AI summary The text requests Nova Scotia Power Inc. to explain the consistency of its 2026 Load Forecast with net-zero goals, the inclusion of federal strategies, and the basis for electrification projections. The response indicates that electrification involves heating and cooling conversions but does not assume industrial process electrification, and no cost-benefit analysis was conducted for industrial electrification.

REDACTED 2026 Load Forecast Report Synapse IR-1 Attachment 1 Page 1 of 196 REDACTED (CONFIDENTIAL INFORMATION REMOVED) Average of large customer co 2018 5 232.9 0.0 2018 6 143.8 3.8 2018 7 15.0 53.2 2018 8 0.1 78.7 2018 9 61.2 21.1 2018 10...

AI summary The document presents a redacted load forecast report containing historical data on average large customer consumption and heating/cooling degree days (HDD18 and CDD18) from 2018 to 2023. The data appears to be used for forecasting and analysis purposes, though key details are redacted.

REDACTED 2026 Load Forecast Report Synapse IR-1 Attachment 1 Page 12 of 196 REDACTED (CONFIDENTIAL INFORMATION REMOVED) Year Cust Count Change in Res Customers Population Change in Population Housing Completions Starts Household Size Year...

AI summary The document presents a table showing historical data on customer count, population changes, housing completions, and household size from 2004 to 2007. The data is part of a 2026 Load Forecast Report and is used for analyzing trends in residential energy consumption.

REDACTED 2026 Load Forecast Report Synapse IR-1 Attachment 1 Page 20 of 196 REDACTED (CONFIDENTIAL INFORMATION REMOVED) CPI HouseholdIncome Household Income (mil $2002) % change SFCompletions MUCompletions HousingCompletions % change GDP M...

AI summary The document presents economic and housing data from 2034 to 2036, including household income, GDP, employment, and housing completions. It also includes forecasts for peak reduction and energy reduction from 2028 to 2031. The data is presented in tables with various metrics and percentages.

REDACTED 2026 Load Forecast Report Synapse IR-1 Attachment 1 Page 34 of 196 REDACTED (CONFIDENTIAL INFORMATION REMOVED) 2023 2024 2025 billed 2025 accrued 2026 Year HP Heat (GWh) HP Cool (GWh) Baseboard Heat (GWh) Water Heat (GWh) 2026 115...

AI summary The document presents a load forecast report for the years 2026 to 2036, including data on heat and cooling demand, water heating, and baseboard heating in gigawatt-hours. It also includes temperature lag averages and peak demand models.

SUMMARY OUTPUT Regression Statistics Multiple R 0.81801092 R Square 0.66914186 Adjusted R Square 0.66905068 Standard Error 143.091653 Observations 14520 df SS MS F Significance F Regression 4 601065795.8 150266449 7338.94 0 Residual 14515...

AI summary The document presents statistical regression analysis with a focus on load forecasting, including coefficients for variables like wind, 12-hour average temperature, and weekdays. It includes forecasted and actual peak loads for 2025 and 2026, as well as temperature and load data for specific dates in 2022.

REDACTED 2026 Load Forecast Report Synapse IR-1 Attachment 1 Page 189 of 196 REDACTED (CONFIDENTIAL INFORMATION REMOVED) Item 2026 NSR Energy (GWh) 2026 Firm Peak (MW) 2036 NSR Energy (GWh) 2036 Firm Peak (MW) 2022 155 - 2,061 2,216 12.6 -...

AI summary The document presents a load forecast report with energy consumption and peak demand data for various years, highlighting trends in NSR energy and firm peak demand from 2022 to 2036. Forecasted values and historical data are compared, indicating changes in energy usage patterns over time.

REDACTED 2026 Load Forecast Report Synapse IR-1 Attachment 1 Page 192 of 196 REDACTED (CONFIDENTIAL INFORMATION REMOVED) Forecast Forecast Forecast Forecast Forecast Forecast Forecast Forecast Forecast Forecast Forecast † Temperature at Pe...

AI summary The text presents a table from a 2026 Load Forecast Report, showing forecast data related to temperature, HDD, CDD, and economics. It includes contribution to variance and rank correlation values for different assumptions.

2 (b) Binary variables and a first-order moving average variable that were included in the 2025 3 peak model were removed from the 2026 peak model. These were introduced to improve 4 model fit in 2025 but did not improve the model fit this...

AI summary In the 2026 peak model, binary variables and a first-order moving average variable that were included in the 2025 peak model were removed. These variables were introduced to improve model fit in 2025 but did not contribute to model improvement in 2026.

7 (c) The analysis outlined in the Report, and subsequent model tests, provided evidence to 8 suggest that the magnitude of the January 2026 peak was at least in part due to the 9 prolonged period of cold, as opposed to just the temperatur...

AI summary The analysis in the report indicates that the January 2026 peak demand was influenced by a prolonged cold period, not just the 12-hour temperature leading up to the peak. Electrification is expected to increase future peak demand, with its effects incorporated into the peak model through heat use, cool use, and other use variables, as well as growth programs.

6 1 As can be seen, models perform very well for Residential class, but not for General Demand, even in class-tailored form. Restricting the training period to more recent years lead to minor model improvements, but using winter months onl...

AI summary The models perform well for residential class but poorly for General Demand, even with tailored training. Restricting training to recent years slightly improved models, but using only winter months did not. Poor performance of non-residential class models limited their use in estimating coincident peak demand.

2026 Load Forecast Report (NSEB M12861) NSPI Responses to Synapse Energy Economics Inc. Information Requests 1 Request IR-5: 5 the combined intensity for non-hybrid and hybrid heat pumps. The calculation for 6 non-hybrid heat pumps is as f...

AI summary The document details NSPI's methodology for calculating the combined intensity for non-hybrid and hybrid heat pumps, using 2020 as the base year for non-hybrid and 2028 for hybrid heat pumps. It also explains the relationship between two figures in the 2026 Load Forecast Report and references a prior study on the load and sales impact of a heat pump on-bill financing program.

NON-CONFIDENTIAL 1 and collect heat pump and total house hourly load data, and measure heat-pump 25 (i) Refer to Attachment 1 of the Report, "calibration" tab, for the input heat pump 26 heating intensity per household ("HPHeat"). 27 28 (i...

AI summary The text discusses the methodology for collecting and analyzing heat pump and total house hourly load data, including calibration of heating and cooling intensities, and the use of efficiency data in forecasting models. It references attachments and figures within the report for detailed input data.

NON-CONFIDENTIAL 1 to heating) is then compared to the modeled increase in energy consumption from 2 E3 over the same time period, and the difference between the two energy increases 3 is found in Figure 26. Please refer to Attachment 1 fo...

AI summary The document discusses the comparison between E3's modeled increase in energy consumption and the modeled increase from NS Power, highlighting the difference in energy increases and the adjustment calculations for hybrid peak demand impact. Attachments provide supporting calculations and load shapes.

Q: What are your findings? A: NS Power has embedded a number of unexplained and unsupported assumptions in its residential load forecast. In addition to the hard-coded factors I mentioned above, there are unexplained factors in several oth...

AI summary The reviewer found that NS Power's residential load forecast contains unexplained and unsupported assumptions, with unclear sources for baseline data and outdated Itron-supplied EIA data. These issues were identified in multiple worksheets, and similar gaps may exist in other workbooks.

Q: Why does this matter? A: NS Power's hybrid heating forecast depends heavily on the data included in this workbook. Yet it is impossible to clearly understand, much less validate, NS Power's forecast. As shown above, NS Power's views on...

AI summary NS Power's hybrid heating forecast is critical for investment plans and program designs, but the forecast lacks clarity and validation opportunities due to inconsistent data and unexplained historical basis from 2023 to 2026.

Q: Did NS Power implement Synapse's recommendation to validate heating intensity assumptions? A: No. In its decision on the 2025 Load Forecast Report, the Board endorsed NS Power's acceptance of Synapse's recommendations. Synapse recommend...

AI summary NS Power did not implement Synapse's recommendation to validate heating intensity assumptions with AMI or empirical data, despite the Board's endorsement of the recommendation. NS Power claims the assumptions are based on a 2022 Itron study, but the load forecast does not clearly reference this data. NS Power is currently working on techniques to use AMI data for evaluating heating loads.

Q: Is NS Power's residential hybrid heating forecast reasonable? A: No. The energy-to-demand reduction ratio is far too high. NS Power forecasts that the maximum demand reduction per customer is 2.76 kW during the peak hour and, depending...

AI summary The residential hybrid heating forecast by NS Power is deemed unreasonable due to an overly high energy-to-demand reduction ratio. The forecast assumes a maximum demand reduction of 2.76 kW per customer and significant annual energy savings, but it is criticized for misinterpreting data from a 2022 Itron study and not accurately representing hybrid heating scenarios.

Q: What is your recommendation to the Board? A: The Board should immediately reject NS Power's flawed residential hybrid heating forecast as unsupported by facts or sound reasoning. Normally, it would be reasonable for the Board to direct...

AI summary The respondent recommends that the Board reject NS Power's residential hybrid heating forecast due to its lack of factual support and sound reasoning. They suggest using NS Power's proposed hybrid heat pump adoption rate and energy savings data as a temporary basis for a more accurate forecast, while urging the use of AMI data for a comprehensive review by 2027.

SUMMARY OF PROFESSIONAL EXPERIENCE - 2023– Present Vice President, Grid Strategies, LLC . Provides research, technical assistance, and expert testimony on electric- and gas-utility planning, economics, and regulation. Reviews electric util...

AI summary The individual has extensive experience in utility planning, regulation, and energy policy, spanning roles in research, regulatory policy, and advocacy. They have provided expert testimony, designed energy efficiency and electrification programs, and worked on renewable energy and market data initiatives.

tion of energy efficiency in Georgia Power's 2013 integrated resource plan, including cost effectiveness, rate and bill impacts, and lost revenues, economics of fuel switching and renewable resources. South Carolina PSC Docket No. 2013-392...

AI summary The text discusses energy efficiency in Georgia Power's 2013 integrated resource plan, including cost effectiveness, rate and bill impacts, and the economics of fuel switching and renewable resources. It also references a South Carolina PSC docket involving testimony on capacity, energy efficiency, renewable energy alternatives, and the use of solar power.

The historical trend for firm peak demand shows a general increase, as shown in Figure 2 below. We plotted the actual rather than the weather-normalized historical peaks to show the year-to- year variations. The 2026 firm peak demand incre...

AI summary The historical trend for firm peak demand shows a general increase, with a 244 MW (10.4%) rise by 2026, primarily driven by increased electrification. The modeled values from the SAE modeling and subsequent adjustments are detailed in Figures 65 of the Report.

1.2. Sector Overview The following table shows the forecast energy use by sector. Overall, industrial load is the only sector with an energy decrease. The remaining sectors have increases between 2026 and 2036. The pattern seen in [Table 3...

AI summary The document provides an overview of energy use forecasts by sector, noting that industrial load is the only sector with a projected energy decrease, while other sectors are expected to see increases between 2026 and 2036. This differs from the 2025 forecast, where municipal and other load was the only sector with an increase.

2. BOARD DIRECTIVES AND PRIOR SYNAPSE RECOMMENDATIONS In its Decision concerning NS Power's 2025 forecast, in Matter 12349, dated December 19, 2025, the Board issued several directives to NS Power for its 2026 forecast. In this Decision, t...

AI summary The Board issued directives to NS Power regarding its 2026 forecast, including implementing prior recommendations and monitoring various factors such as housing completions, forecast variances, and government policies on electrification. The Board also encouraged continued monitoring of battery storage price trends and noted some intervenor suggestions for future planning.

3. ENERGY FORECAST NS Power forecasts load using different methodologies for different rate classes, reflecting the distinct drivers and data available for each. For the residential and commercial classes, NS Power uses Statistically Adjus...

AI summary NS Power uses different forecasting methods for various rate classes. Residential and commercial classes use SAE models, which combine end-use and econometric methods. Industrial classes use econometric models and customer surveys, while the municipal class is treated as an accounting matter due to third-party energy supply.

3.3. Commercial Sector The commercial class represents about 27 percent of total load. Like the residential class, it is forecast using an SAE model, but the commercial model produces separate forecasts for its two rate classes, Small Gene...

AI summary The commercial class accounts for 27% of total load and is forecasted using an SAE model with separate forecasts for Small General and General Demand rate classes. Forecasts are based on end-use intensity projections, GDP, employment, real price, and monthly degree days.

Efficiency assumptions and heating intensity methodology The 2026 Load Forecast does not clearly identify which efficiency metric underlies NS Power's heat pump assumptions. Based on our review of the forecast workpapers, we assume that NS...

AI summary The 2026 Load Forecast does not clearly specify the efficiency metric used by NS Power for heat pump assumptions, leading to potential inaccuracies. The analysis suggests that NS Power may be using HSPF instead of HSPF2 and relies on a calibration process involving historical data and the Itron study for heating intensity values.

6. RECOMMENDATIONS - 1. NS Power should continue to monitor the impact of trade policy and consider explicitly incorporating tariff impacts into its future forecast if they are expected to have a material impact on load growth. - 2. Concer...

AI summary The recommendations focus on improving NS Power's forecasting methods for load growth, hybrid heating participation, heat pump efficiency metrics, EV charging, solar installations, and DSM savings accumulation. Emphasis is placed on using more accurate modeling approaches, incorporating updated data, and clarifying assumptions to enhance forecast reliability.

notwithstanding the documented change made in 2024 to account for a large unallocated variance, which was determined to be driven by heating loads. The heat pump heating intensity, saturation, and an illustrative total forecast load are in...

AI summary NS Power acknowledges that variance in heat pump heating intensity and load forecasts is due to annual model calibration, incorporating updated billing data, saturation, and efficiency data, as well as changes in heat pump installation forecasts. This calibration helps optimize model performance.

Request IR-1: - Figure 3: Historic and Forecast Net System Requirement and System Peak. - a) The Net System Requirement (NSR) forecast percentage change in 2027 is negative. Aside from Renewable to Retail sales, what are the factors that a...

AI summary The document contains several requests for clarification regarding forecasts related to Net System Requirement (NSR) and system peak, as well as employment data sources and discrepancies in manufacturing GDP figures. Questions focus on the factors influencing NSR trends, the accuracy of system peak forecasts, and the rationale for using broader age groups in employment data analysis.

Request IR-8: Residential Hybrid Heating (Section 4.5.2, p. 37-42) - a. Refer to the following statements in the 2026 Load Forecast Report: "The 2026 residential SAE regression model has been updated to include peak and energy reductions r...

AI summary Request IR-8 focuses on residential hybrid heating analysis by NS Power, including data on load shapes, participation rates, modeling approaches, and documentation related to the 2026 Load Forecast Report. It asks for detailed explanations and supporting evidence regarding the impact of hybrid heating programs on energy and peak demand.

1 Request IR-09: 2 Reference: NS Power 2026 Load Forecast Report, page 42, lines 3 – 17 3 4 (a) What is the capacity reduction associated with commercial hybrid heating? 5 (b) If no capacity reduction has been modelled, does this imply the...

AI summary The text includes requests for information regarding the capacity reduction associated with commercial hybrid heating programs, the assumptions and methodologies used in the NS Power 2026 Load Forecast Report, and the potential risks and validation of post-model adjustments related to hybrid heating impacts.

1 Request IR-1: 2 3 Reference: Exhibit N-1, Section 4.5.2. 4 5 (a) Please provide a narrative description of each of the "potential hybrid programs which 6 have been modelled by NS Power." (p. 37) 7 8 (b) Please provide the modelling refer...

AI summary Request IR-1 seeks detailed information from NS Power on hybrid heating programs, including modeling data, heat pump system types, energy reduction assumptions, and alignment with provincial electrification policies. It questions the methodology behind energy use forecasts, backup heating system usage, and the validity of hybrid program outcomes.

1 M12861 2 3 NOVA SCOTIA ENERGY BOARD 4 5 IN THE MATTER OF: The Public Utilities Act, R.S.N.S. 1989, c.380, as amended 6 7 - and - 8 9 10 IN THE MATTER OF: NOVA SCOTIA POWER INCORPORATED's 2026 Load Forecast Report 11 12 13 14 INFORMATION...

AI summary The Nova Scotia Energy Board has issued an information request (IR-1) to Nova Scotia Power Inc. (NS Power) regarding its 2026 Load Forecast Report. The request was made by Solar Nova Scotia and is due by July 7th, 2026. The report is part of a proceeding under the Public Utilities Act.

July 21, 2026 VIA EMAIL Ms. Crystal Henwood Clerk of the Board Nova Scotia Energy Board 1601 Lower Water Street, 3rd Floor Halifax NS B3J 3S3 Dear Ms. Henwood: Re: M12861 - Nova Scotia J>ower Inc. (NSPI) - 2026 Load Forecast Report - REDAC...

AI summary The Small Business Advocate (SBA) reviewed NSPI's 2026 Load Forecast Report and acknowledges its thoroughness but raises concerns about the housing completions adjustment. The SBA notes that NSPI did not explicitly account for market conditions like development delays and financing constraints, despite incorporating a flat 10% annual decline in Signal49's forecast.