B-4: Reducing Peak Demand through Distributed Grid Edge Control

Transcription

1 B-4: Reducing Peak Demand through Distributed Grid Edge Control Shane Smith UDWI Bloomfield, USA Damien Tholomier Varentec Santa Clara, USA

2 Introduction Utilities District of Western Indiana REMC (UDWI) A Touchstone Energy Cooperative, which is a member of the Hoosier Energy Power Network and the 4 th -largest electric co-op in southern Indiana, serving over 19,000 meters in Clay, Daviess, Greene, Knox, Lawrence, Martin, Monroe, Owen, Putnam, Sullivan, and Vigo counties for 80 years. UDWI REMC is part-owner of Hoosier Energy, a Generation and Transmission (G&T) cooperative providing wholesale electric power and services to 18 member distribution cooperatives in central and southern Indiana and southeastern Illinois. Based in Bloomington, Indiana, Hoosier Energy operates coal, natural gas and renewable energy power plants and delivers power through a 1,500-mile transmission network.

3 Problem Statement In 2010, Hoosier Energy Implemented Seasonal Rates Rate was designed to support / promote DSM Programs Included seasonal demand ratchets off peak months were charged at the average of the preceding three on-peak months With 94% residential meter and a 61% Load Factor, increases in demand charges affect us more. In 2011, Hoosier Energy increased demand component 41% of 2016 s purchased power costs were demand charges Announced intent to increase demand component in future rates Up to 50% increase in demand costs Hoosier can only bill for production demand if they call for a DSM event.

4 Traditional CVR Never Considered Existing System could not support 51% of system still copperweld 28% 6A 21% 8A 15% 4ACSR Only 27% Standard conductors (336.4MCM, 1/0ACSR, 2ACSR) Construction Work Plans only a priority for last 10 years Line Loss is our CVR AMI Data confirms Sampling of AMI data showed frequent ANSI-A Violations Older AMI meters with some limitations (± 5% voltage accuracy) DSM programs significantly underperforming Net Loss

5 VVO / CVR Project Objectives Optimize the Primary and Secondary VVC Equipment to maximize VVO/CVR Demonstrate ability to regulate service voltage within ANSI-A band by reducing voltage fluctuations caused by dynamic loads Reduce voltage by a minimum of 3.7% during Peak Time (Peak Demand Reduction) Reduce Technical Losses Measured Field Performance during Peak Time Voltage Fluctuation reduced by 3.3% Incremental Voltage Margin by 2.2% Maximum Voltage Reduction of 4.2% (from 126V to 120.5V) Reduction of the Line Losses by 1.5%

6 12.47kV Switz City s/s 5.23 MW Peak Load, PF = 0.97 Three Voltage Control Zones (VCZ) with residential load One LTC (SP=126 ±1 V) with two Feeders: (1.28 MW) (3.95 MW) Two LVRs: Three-phase LVR (SP=126 ± 1 V) Single-Phase LVR (SP=126 ± 1 V) Six Capacitor Banks Two Fixed, 600 kvars: in service (CAP868, CAP872) Two Switched, 600 kvars: out of service Feeder length 11 miles 558 distribution transformers, 692 customers Project Scope ENGO unit deployment Initial Deployment - 43 ENGO units Zone 1 (LTC, 60% load) with 34 units (28 on A, 1 on B and 3 on C) Zone 2 (LVR1, 34% load) with 5 units (3 on B and 2 on C) Zone 3 (LVR2, 6% load) with 4 units (4 on A) After Relocation of LVR2 (Toms) - 43 ENGO units Zone 1 (LTC, 57% load) with 16 units (11 on A, 3 on B and 2 on C) Zone 2 (LVR1, 35% load) with 8 units (4 on B and 4 on C) Zone 3 (LVR2, 8% load) with 19 units (19 on A) After the Optimization of Primary and Secondary VVC equipment - 23 ENGO units Zone 1 (LTC, 57% load) with 12 units (10 on A and 2 on C) Zone 2 (LVR1, 35% load) with 8 units (4 on B and 4 on C) Zone 3 (LVR2, 8% load) with 3 units (3 on A) Voltage Control Zone

7 Conventional VVO/CVR Conventional methods of achieving peak demand reduction for rural cooperatives in the range of 1%-2.5% control during heavy loaded condition without violating customer-side ANSI- A voltage limits Measured minimum weighted Voltage Margin at Switz City s/s during Peak Time 1.08% No margin in Zone 1 (57% load) Min 2.96% in Zone 2 (35% load) Min 5.08% in Zone 3 (8% Load) Voltage Margin as pct of MW Switz City PQ Data 5.3MW Peak Load by end of summer time and comprises residential and irrigation loads as well as petroleum pumping station and storage, a coal loading facility, an asphalt plant, turkey farms, grain drying, and a large grain processing facility

8 Voltage Fluctuation Limits Conventional VVO/CVR

9 Grid Edge VVC Solution

10 Secondary Voltage Profile w/o and with ENGOs

11 Voltage Margin as function of Load (per zone)

12 Feeder Voltage Flattening (15 min Data) ENGO ON ENGO OFF Max 126.3V Max 126.8V Zone 1 57% of Load Zone 1 Min 116.6V Max kvar Min 113.7V Voltage Margin increased from -0.3% to 2.2% Voltage fluctuation reduced from 10.9% to 8.1% Zone 2 35% of Load Zone 2 Max 126.2V Min 122.2V Max 74.2 kvar Max 129.9V Min 120.4V Voltage Margin increased from 5.3% to 6.3% Voltage fluctuation reduced from 7.9% to 3.3% Zone 3 8% of Load Voltage Margin increased from 6.2% to 7.4% Max 126.2V Max 125.6V Voltage fluctuation reduced from 3.5% to 2.8% All zones Zone 3 Min 122.8V Max kvar Min 121.4V Weighted Voltage Margin increased from 2.0% to 4.2% Voltage fluctuation reduced from 9.3% to 6.0%

13 Primary and Secondary Optimization Initial Deployment (43 ENGOs) Final Deployment (23 ENGOs and LVR2 Relocation

14 CVR Factor Theory % % : The theory of Conservation Voltage Reduction (CVR) resides in the formula for electric power, = The goal for CVR is to reduce demand or energy consumption by reducing voltage, however, the fundamental question to be answered is, How much power or Energy will be reduced if % voltage is reduced in a given system?. Therefore, a metric called -factor is defined. = % % = % %

15 CVR Factor ZIP configuration (constant impedance, Z; constant current, I; and constant power, P elements). CVR factor decreases when the voltage dependence of the load changes from a constant CVR to a constant power type. CVR factor will depend on the time of the day/month of the year/mix of load that voltage regulation is enabled. Example of CVR factor for Power HVAC: o o Cooling mode: CVR varies from 0.57 at 82 ⁰(F), 0.41 at 95⁰(F) and 0.22 at 115⁰(F) Heating mode: i.e. CVR = 0.67 at 47⁰(F) Incandescent bulb: CVR = CFL bulbs: CVR = LED bulbs: CVR = Commercial lights with magnetic ballasts: CVR=1.204 Cooking range Top: CVR = Source: 1) EPRI

16 CVR Tests performed by Various Utilities Typical Value: Typical Value:

17 6 CVR Weekly Events Peak Demand Reduction: CVR Event # 1, 2, 3 and 4: 122V (LTC, LVRs) = 3.33% CVR CVR Event # 5: 121 V (LTC, LVRs) = 4.17% CVR CVR Event # 6: 120.5V (LTC., LVR2), 125 (LVR1) = 3.29% CVR Optimized Set-Point: 120.5V (LTC, LVR2) and 121V (LVR1) = 4.44% CVR

18 Example of CVR Event (Coop in Indiana) Peak Shaving: 229kW CVR factor = 1.2 MW Measured MW with no CVR 4% to 6% overall Peak Demand (MW) Reduction (combined Primary and Secondary voltage control) achieved by operating the voltage in the 114V to 120V range at the customer s meter. GEMS+ENGO solution delivers a Benefit-Cost Ratio (BCR) > 3.0 Substation Voltage 4 to 5 year payback period Fully Deployed and Functioning on an entire substation in a matter of weeks

19 Conclusion Volt-VAR Control (VVC) is at the heart of grid operations Conventional VVC with LVRs, and VARs from generators and MV cap banks have been sufficient in the past in creating a minimum floor for primary voltage Cannot meet dynamic and precise grid-edge control needed with multi-objective VVO/CVR, dynamic voltage mitigation and renewable integration that utilities are moving towards With more granular AMI information, gaps are becoming visible Utilities have few tools to manage Deployment in Progress at 3 new substations at UDWI SCADA upgrade, Automation of LTC/LVR controls, Optimization between Primary and Secondary VVC equipment Deployment of Secondary VAR controllers (ENGO devices) at strategic locations 19