california solar initiative (csi) research, development ... · research, development,...

California Solar Initiative (CSI) Research, Development,

Demonstration, and Deployment Program Grant # 2

Demonstration

Supported by the California Public Utilities Commission (CPUC), managed by Itron, Inc, and hosted at the

University of California, San Diego Campus

February 16, 2012

Agenda

• Opening Remarks (ITRON) • Grant Objectives (Nancy Miller) • About Viridity, About E3(Rick Schaal, Eric Cutter) • Introductions and Acknowledgments, Project Overview (Nancy Miller) • Policy Context for Managing Distributed Energy Resources (Eric Cutter) • UCSD Campus Resources Overview (Priya Sreedharan) • UCSD Baseline and Historical Operations (Priya Sreedharan) • Master Controller Overview (Chuck Richter) • VPower Demonstration (Chuck Richter) • UCSD Optimization Strategies (Chuck Richter) • Project Discoveries (Nancy Miller) • Next Steps • Questions

2/16/2011 2 CSI Grant #2 Mid-Project Demonstration

Working with the California Independent System Operator (CAISO), San Diego Gas and Electric (SDG&E) and our host site, the University of California San Diego (UCSD), one of the most advanced microgrids in the state, Viridity Energy and Energy+Environmental Economics (E3), have merged their respective expertise to:

• Analyze and demonstrate how load can serve as a flexible resource to integrate high penetration solar

• Demonstrate that the barriers to deployment of solar resources can be overcome through providing appropriate incentives for real-time management and optimization of distributed energy resources.

• Provide economic, reliability and market price benefits to California ratepayers that accompany increased deployment of solar resources

CSI Grant Overview


© 2011 Viridity Energy Inc. All Rights Reserved 4

© 2011 Viridity Energy Inc. All Rights Reserved

Key Company Facts

Founded in 2008 by well known industry executives

Unparalleled Solutions

Patent pending software platform optimizes and integrates energy assets such as controllable loads, cogeneration, renewables, and energy storage systems

Integrated energy management from Demand Response to Dynamic Load Management and Microgrids

Network Operating Center supports customer participation in wholesale power markets

Viridity Energy - Recognized Leader in Smart Grid Technology

Recipient of multiple industry awards and grants

Experienced Team: • Experienced utility and power market

executives • Technology innovators • Regulatory policy and affairs experts

Types of Clients

Utilities

Commercial

Data Centers Military

Industrial

Institutional

2/16/2011 CSI Grant #2 Mid-Project Demonstration 4

Intelligence Moving to The Edge of the Grid: Real-Time Generation + Dynamic load

Source: iTeres

WindFarm

Offices

Central Power Plant

Storage

Industrial Plant

Generators

Isolated Microgrids

HousesSolar Panels

Storage

Storage

Generators

Storage

The Grid is a Single Machine Global proliferation of distributed energy resources: • Solar • Electric vehicles • Distributed generation • Smart buildings

Power Grid will evolve from centralized command and control to distributed interconnected systems that are:

• Inter-coordinated • Self-scheduling • Self-healing

Viridity Energy enables the seamless integration of distributed resources, smart loads and microgrids as Virtual Power into real time power grid operations

5 © 2011 Viridity Energy Inc. All Rights Reserved

Major Trend #1:

Major Trend #2:

2/16/2011 CSI Grant #2 Mid-Project Demonstration

About E3

E3’s expertise in 8 key practice areas has placed us at the center of energy planning, policy and markets in California and the West.

6

Emerging Technology Strategy

Energy Markets and Financial Analysis

Transmission Planning and Pricing

Energy and Climate Policy

Cost of Service and Rate Design

International Electricity Policy and Planning

Energy Efficiency and Distributed Resources

Resource Planning and Procurement

Project Overview

• Expand the model of UCSD campus resources

• Propose strategies for the integration of high penetration PV solar systems

• Propose tariffs and incentives for Distributed Energy Resources (DER) technologies and Load Management Strategies

• Establish Baseline performance for UCSD DER operation under current rates and incentives

• Refine and test proposals with VPower™ in simulation and realtime mode

• Provide a robust cost-benefit analysis of the DER management strategies

• Provide an transparent analysis tool for public use as a part of the final report

Additional project information can be found at: http://www.calsolarresearch.ca.gov/Funded-Projects/solicitation2-viridity.html


http://www.calsolarresearch.ca.gov/Funded-Projects/solicitation2-viridity.html





POLICY CONTEXT FOR MANAGEMENT OF DISTRIBUTED ENERGY RESOURCES

CSI Grant #2 Mid-Project Demonstration 8 2/16/2011

California’s Renewable Energy Challenges


Managing RPS costs with “smart”, cost-effective interconnection and management of distributed energy resources (DER)

1) Increase Interconnection Potential for Distributed PV


Management of distributed energy resources can potentially interconnect PV at 3 X today’s levels

Graph shows “raw” PV solar potential based on minimum substation loading

0

2,000

4,000

6,000

8,000

10,000

12,000

14,000

16,000

18,000

2020

DG

Cap

acity

(MW

)

Ground > 10 MWGround < 10 MWCommercial RoofsResidential Roofs

Relative substation interconnection potential

“Raw” PV potential based on substation loading

2) Provide Low/No Carbon Flexible Resource


• There is an identified need for flexible resources to integrate renewables.

• BUT, there is excess capacity based on planning reserve margin

Can distributed energy resources displace fossil generation as an economic and flexible resource?

Fossil

Storage

3) Meet System Need for Flexibility


1 15 33 51 68 85 106 130 154 178 202 226 250 274 298 322 3460

100K

200K

300K

400K

Days in a Year

Selected Daily Revenue Benefits

RegulationLoad FollowingSpinning Reserves

Non-Spinning ReservesEnergy Price Arbitrage

Battery dispatched primarily to provide regulation, load following and spinning reserve

Energy dispatch limited to a few high priced hours in the summer

• 50 MW, 4 Hour Bulk Battery

• California 2020 RPS Compliant Scenario

• Modeled with PLEXOS and EPRI/E3 Energy Storage Valuation Tool

Energy storage is dispatched for flexibility… not energy. Shouldn’t we harness the flexibility in distributed resources?

Annual Revenues

System Capacity

Regulation

Load Following

Energy

Illustrative Flexible Load Potential in California


0

500

1,000

1,500

2,000

2,500

College Industrial

MW

Pumps Fans DrivesCoolingInterior LightingVentilationHeatingAir CompressorsOffice EquipmentMotorsMiscellaneous

CAISO LTPP Estimated Need for Flexible

Resources

LTPP: Long Term Procurement Planning

Distributed, flexible resources and loads can provide a significant portion of our anticipated need for flexible capacity

OVERVIEW OF UCSD RESOURCES


UCSD microgrid

With a daily population of over 45,000, UC San Diego is the size and complexity of a small city.

11 million sq. ft. of buildings, $250M/yr of building growth

UC San Diego operates a 42 MW microgrid

UC San Diego grid imports 2007

Self generate ~ 90% of annual demand •30 MW natural gas Cogen plant •2.8 MW of Fuel Cells in operation

•1.2 MW of Solar PV installed, additional 2 MW planned

•Twice the energy density of commercial buildings

Central plant resources

• The central plant is rich with dispatchable resources • Two 13 MW natural gas

generators • One 3 MW steam generator • Three Steam driven chillers

(~ 10,000 tons capacity) • Eight electric driven chillers

(~ 7800 tons capacity) • 3.8 million gal thermal storage

tank • Backup diesel generation

• >1 MW of solar PV • ~ 1.4 MW of DR-ready reducible

building load • Visibility at the building level

Chilled water tank at UCSD campus


Simple illustration of cogeneration

• Let’s review the concept of ‘cogeneration’ or ‘combined heat and power’

systems • Electricity generated and can be used onsite and/or exported to the grid • Heat is recovered used for various purposes

Source: EPA Combined Heat and Power Partnership website. http://www.epa.gov/chp/basic/index.html

Gas turbine/engine with heat recovery

UCSD’s cogeneration system

• The UCSD cogeneration system has the following

• Natural gas turbine generates electricity for onsite use & steam

• Steam is used to generate hot water, more electricity (steam turbine) and chilled water (steam-driven chillers)

• Cogeneration offsets imported electricity, boiler fuel use for generating hot water & chilled water, and electricity consumption required to generate chilled water

Steam chiller at UCSD campus

Cogeneration is a core component of UCSD’s microgrid

Flexibility of the UCSD resources

• UCSD resources can shift load, reduce load and move load between gas and electricity fuels • Natural gas generators produce electricity and steam;

once on, can go up/down ~ 6 MW • Tank can deliver significant portion of campus chilled water

needs and can be variably charged/discharged • Steam from generators can be used to generate hot water,

chilled water and additional electricity in varying amounts • Building load can be reduced ~ 1.4 MW for DR events


UCSD online energy dashboard

Source: http://energy.ucsd.edu/campus/campus.php


http://energy.ucsd.edu/campus/campus.php



Renewables integration with VPower

• Hypothetical load following scenario using UCSD actual data for 6/7/2011 • Generators are ramped up and down within their acceptable

operating range throughout the day

0

5

10

15

20

25

30

35

40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

MW

Hour

ImportsGenerators' output with load followingCampus total consumptionNormal generators' output


BASELINING THE CAMPUS RESOURCES


Review of baselining effort

• Overall goals for baselining • Understanding thermal & electrical campus needs • Quantify efficiencies & understand operating strategies • Identify opportunities for operational improvement • Validate / calibrate VPower to ensure tool generates plausible

operational strategies

• Key questions • How do the campus needs vary by hour, season, month? • What are the overall system efficiencies (CHP, steam utilization)

and equipment-level efficiencies (chillers, generators, etc.)? • What are the regular modes of operation of each system? • How much capacity is available for changing operations?


UCSD data resources

• Baselining data obtained from multiple sources: • MSCADA system: 15-minute power data • Johnson Control Metasys System: thermal storage tank data • ‘BOP’ System: steam data for boilers, generators • ‘Efftrack’ chiller diagnostic system: chiller data • Daily central plant logs: gas usage • UCSD expert knowledge (i.e., John Dilliott, Energy Manager) • Solar data from Prof. Jan Kleissl

• Future: disparate data sources will be integrated into a campus-wide data historian


Analytical framework

Inputs Outputs

Nat gas

Diesel

Import kW

Solar PV

Boilers

NG gens

Backup gens

Steam

Steam

KW

St gen

HX’s

Stm chs

Elec chs

Aux eqp

KW

HW

CHW

Heating

Cooling

Non CUP HVAC

Lighting

Plug loads

Water heating

TES

Baseline data gives input / output relationships, equipment efficiency, capacity factors, seasonal and diurnal patterns of usage, historical costs.

Import + gen

Central Utility Plant End Use

Methodology steps

• Collect data • Electrical data: Imports, central plant generators, chillers, auxiliary, solar • Thermal data: Chillers, boilers (partial), campus hot water & chilled water

needs, thermal storage tank operation, generator fuel • Up to 3 years of data collected

• Assemble data • To understand system level operation: generate data sets with consistent

time stamps to understand inputs/ outputs across the energy flow diagram • Quantifying individual operating efficiencies and capacities: all individual

equipment data can be used

• Visualize and analyze data • Daily, monthly, yearly summaries of main loads • Calculate equipment level efficiencies • Calculate system level efficiency (overall inputs to outputs) • Forecast campus needs based on binning analysis


SNAPSHOTS FROM BASELINING EFFORT


Overview of campus needs: timeseries across entire data set

0

5,000

10,000

15,000

20,000

25,000

30,000

35,000

40,000

45,000

0

50,000

100,000

150,000

200,000

250,000

300,000

2008

-120

08-2

2008

-320

08-4

2008

-520

08-6

2008

-720

08-8

2008

-920

08-1

020

08-1

120

08-1

220

09-1

2009

-220

09-3

2009

-420

09-5

2009

-620

09-7

2009

-820

09-9

2009

-10

2009

-11

2009

-12

2010

-120

10-2

2010

-320

10-4

2010

-520

10-6

2010

-720

10-8

2010

-920

10-1

020

10-1

120

10-1

220

11-1

2011

-220

11-3

2011

-420

11-5

2011

-620

11-7

2011

-820

11-9

2011

-10

2011

-11

MW

h

MM

Btu ―

Gen 1 & 2 gas input in MMBtu Hot Water Demand in MMBtuChilled Water Demand in MMBtu Campus Total Consumption Net CUP in MWhChiller Consumption in MWh Output of Gen 1 & 2 in MWh

Seasonal dependence of CHW, HTW w/ base requirements; generation satisfies much of total campus needs


Overview of campus needs: comparison of monthly totals

0

5,000

10,000

15,000

20,000

25,000

0

50,000

100,000

150,000

200,000

250,000

2008-2 2009-2 2010-2 2011-2 2008-8 2009-8 2010-8 2011-8

MW

h - -

-

MM

Btu

Monthly Total Values Gen 1 & 2 Gas input in MMBtu Hot Water Demand in MMBtu Chilled Water Demand in MMBtu

Campus Total Consumption in MWh Output of Gen 1 & 2 in MWh Chiller Consumption in MWh

* Chiller data only available in latter part of 2011

Significant onsite generation across months/years; Little change over 3 years; significant baseload CHW, HTW

February August 2/16/2011 CSI Grant #2 Mid-Project Demonstration 29

Flexibility in providing campus hot water and chilled water

Campus needs can be met amply through existing resources and there is flexibility to meet the need through different combinations

Steam

Elec

Steam & Elec

Steam & TES

0 500 1000 1500 2000 2500 3000 3500 40005

25

45

65

85

105

125

145

165

185

205

225

Hours

MM

Btu/

hr

Histogram of CHW Demand

Nameplate Capacity Values

Boiler 2 Boiler 3

Boiler 4

Calculated Maximum

Potential Cogen Output

All Boilers

0 500 1000 1500 2000 2500 3000 3500 4000 4505

25

45

65

85

105

125

145

165

185

205

225

Hours

Histogram of HTW Demand

Nameplate Capacity Values


Overall system efficiency

0

50%

100%

0

2000

4000

6000

8000

10000

12000

Ove

rall

Effic

ienc

y

MM

Btu

Gen 1&2 elec out in MMbtu STG MMbtu out CHW from Steam ChillerHTW from Gen's GEN GAS USE Overall CHP Efficiency useful out/ in

0

5

10

15

20

25

30

35

0

50

100

150

200

250

300

350

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

MW

- -

-

MM

Btu/

hr ―

Hour

Generator fuel input in MMBtu/hr Hot Water Demand in MMBtu/hr

Chilled Water Demand in MMBtu/hr Campus Total Consumption Net Chillers in MW

Chiller Consumption in MW Campus Plant Generation in MW

Solar in MW Imports in MW

Actual operations: June 7, 2011

Load shifting from thermal storage lowers electrical chiller consumption

Thermal and electrical needs vary over the day but contain significant baseload components (high load factor).

Solar PV output ~ 1 MW 2/16/2011 CSI Grant #2 Mid-Project Demonstration 32

Binning analysis to forecast loads

• Analyzed chilled water, hot water, campus power requirements over 3 years

• Preliminary insights: • Significant baseload across all hours • Seasonal dependence exists but relatively tempered • Weekday vs. weekend operation variability small • Variability across years exists but not growth related


Snapshot: variation across months

05

1015202530354045

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24Cam

pus T

otal

Con

sum

ptio

n (M

W)

Hour

February: Weekdays

90th PercentileMedian10th Percentile

0

20

40

60

80

100

120

140

160

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

Chill

ed W

ater

Loa

d (M

MBt

u/hr

)

Hour

February: Weekdays


0

20

40

60

80

100

120

140

160

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

Chill

ed W

ater

Loa

d (M

MBt

u/hr

)

Hour

September: Weekdays 90th PercentileMedian10th Percentile


Overall observations

• Majority of operating rules and efficiencies have been validated

• Loads have significant baseload component; variability due to hour, season, year

• Flexibility in resources to meet loads confirmed


PRELIMINARY BUSINESS ANALYSIS


Preliminary business analysis

• How might a microgrid maximize its revenues and reduce costs? • Should it run cogen to maximum capacity? • How should storage be utilized? • How should resources be scheduled including loading order of

heat recovery systems?


Scenarios analyzed

• What is the value of controllable UCSD assets in support of high penetration PV and lowering campus costs?

• Spreadsheet model developed to analyze the following scenarios using real UCSD load data • Scenario 1: full import • Scenario 2: full import + thermal storage tank • Scenario 3: cogen (natural gas generators, steam chillers, hot

water heat exchanger, steam generator) • Conducted sensitivities on 1 vs. 2 generator, steam gen

• Scenario 4: cogen + thermal storage tank • Conducted sensitivities on steam gen


Methodology

• Spreadsheet model developed to analyze the cases using real UCSD load data from June-Oct 2011 • Consistent with campus thermal and elec needs • Loading order consistent with UCSD operations • Efficiencies & capacities from baseline effort used

Results of preliminary business analysis:

Estimated energy costs between June 1 and October 31 2011 for the idealized scenarios $Millions Electrical Gas Total SavingsFull import 6.8$ 1.3$ 8.1$ 0%Full import and thermal storage 6.7$ 1.3$ 8.0$ 1%Cogeneration 1.0$ 5.3$ 6.3$ 22%Cogeneration & thermal storage 0.8$ 5.3$ 6.1$ 25%


Breakdown by month and energy/demand charges

• Demand charges significant component of elec charges in both summer/ winter

• More significant cost component in the cogeneration cases

• Savings from load shifting weighted towards energy savings

$-

$0.2

$0.4

$0.6

$0.8

$1.0

$1.2

$1.4

$1.6

Total electric Demand Gas Total electric Demand Gas

July October

Cost

s ($M

illio

ns)

Full import

Full import and thermal storage

Cogeneration

Cogeneration and thermal storage


Insights and opportunities for further optimization

• Preliminary scenario analysis suggests general operating strategy saves campus $

• Baseline and preliminary analysis suggest opportunities for enhancement • Tank operation, staging of equipment

• Limitations • Analysis performs limited number of sensitivities that assumes

an operating scenario (rather than solving for it) • Assumes perfect foresight of campus needs • To manage the complexity of the problem space, a realtime

optimization tool is beneficial --- > VPower


MASTER CONTROLLER


Microgrid controller

• Making load elastic/controllable/price-responsive is facilitated through decentralized ‘microgrid’ control.

• Improving microgrid control: • can help with transmission congestion constraints • can improve load participation in DR programs or energy markets (in some

cases through load control which looks a virtual generator to the wholesale market)

• can improve grid utilization through the integration of distributed energy generation, including intermittent renewables

• Typical microgrid control: • includes real-time monitoring and control of resources (SCADA) • advanced analytics to perform load flow analysis and optimization • interaction with the electric power markets


Distributed resources need to be combined and optimized The resources appear to the system operator as a “virtual generator” that is integrated into the regional dispatch

Microgrids include a “central nervous system” that directs the operations of distributed resources

Microgrid Master Controller

Distributed Generation

Controllable Loads

Distributed Storage

The Microgrid Master Controller regulates the power distribution infrastructure within the microgrid footprint

Distributed Resources 2/16/2011 44 CSI Grant #2 Mid-Project Demonstration

UCSD Master Controller


Data Acquisition, Storage and

Control (PI ™)

Master Controller

Resource Optimization (VPower™ )

Power Analytics (Paladin ™)

CA ISO Realtime and Day Ahead Pricing Info

Distributed Resources

Weather, including

cloud cover temperature,

etc

Setpoints can include things like generation MWs, HVAC controls for buildings, battery charging and discharging levels, chiller controls, etc.

Facility Manager

Master Controller: VPower Optimization

• VPower resource models include • Electric (including renewables like solar) and steam generators • Chillers and heat exchangers • Electric and thermal storage devices • Fixed and interruptible load

• VPower study cases are prepared with • Resource parameters and efficiency rates • Period-by-period operator-entered over-rides (if desired) • Electrical and thermal (hot/chilled water) requirements • RT or DA electric energy price data from CAISO • Weather forecasts

• Economically optimized cases schedule the resource output (MMBTU and MW) to minimize costs subject or maximize revenues while meeting thermal and energy requirements.

• Each day, multiple cases can be prepared to investigate the impact of various inputs, scenarios, etc.


Master Controller: VPower Optimization

• The Master Controller handles the data exchange between VPower and Paladin.

• It passes VPower’s economically optimized schedules to Paladin where they are checked for reliability.

• Paladin models the electric network in detail, and determines if the feeder and equipment energy flows are within their appropriate ratings.

• Paladin notifies VPower if any schedule modifications are needed for microgrid reliability, including optimization constraints to be included.


Steam Header GT1 COGEN-Elec

GT1 COGEN-Steam

GT2 COGEN-Elec

GT2 COGEN-Steam

THERMAL/FUEL

Gas Turbine 1

Gas Turbine 2

THERMAL (exhaust heat)

THERMAL/FUEL THERMAL (exhaust heat)

Steam Turbine ST1

Steam Turbine

Boiler

Boiler

THERMAL/FUEL

THERMAL (STEAM)

WC-1 Steam Chiller

WC-2 Steam Chiller

WC-8 Steam Chiller

TES Stratefied Cold W

ater Tank

Chiller 9

Chiller 6

Chiller 3

Chiller 7

Chiller 5

Chiller 2

Chiller 1 ELECTRIC

Spent Cold Water

STEAM HOT WATER

CAMPUS WATER SUPPLY

RETURN

STEAM

STEAM

STEAM

STEAM

GEN DIESEL 4kV 1

THERMAL FUEL

GEN DIESEL 4kV 2

GEN DIESEL 4kV 3

GEN DIESEL 1

GEN DIESEL 2

Fuel Cell 1

EV Rapid Charge Station - Public

EV Slow Charge Station -Public

EV Rapid Charge Station - UCSD

EV Slow Charge Station - UCSD

EV Rapid Charge Station - Buses

EV Slow Charge Station - Buses

PV-1 XFMR-336

SUN/CLOUD COVER

PV-2 Campus Services

PV-3 Price Center

PV-4 Gilman Parking

PV-5 Powell Lab

PV-6 Hopkins Parking

PV-7 School of Management

PV-1MW Energy Storage 7.6MWH

Integrated PV Storage-Battery

Integrated PV Storage - Solar

Comfort Index Johnson Control

(JCI Unitary)

UCSD Hot

Water System

SUN/CLOUD COVER

Trade Street Warehouse PV

Supply Contract •MW •Price

Market Forecasts •Energy Price •Regulation Price •Spinning Price •G&T_Rates

Forecasts •Load MWs •Hot Water •Cold Water

OFF-CAMPUS UCSD MAIN CAMPUS

THERMAL/FUEL

6KGPM Max

Hot Water

Strategy

UCSD Cold

Water System

Campus Hot water requirements per period

HX

Steam Header 1

Steam Header 2

Fixed discharge schedule. This can be over-ridden by data from Paladin.

Chilled water requirements per period, Given TES discharge


VPower Model of UCSD Resources

Dispatchable Virtual Generation

Detailed microgrid model

Inelastic Consumption

Controllable Consumption

Generation resources

High level microgrid model

External perspective (Energy market or utility)

~ Dispatchable virtual generator


VPOWER DEMONSTRATION


VPower Live Demo

2/16/2011 CSI Grant #2 Mid-

Project Demonstration

51

UCSD OPTIMIZATION STRATEGIES


Optimization Strategies and Next Steps

• Short term: tuning VPower models • Detailed comparison between baseline data and VPower results

• Cases: supply from grid; grid+elec from GTs; grid+GT including steam; add-in TES tank, etc. for peak days, typical days, etc.

• Mid term: When tuning is complete, use VPower to assess UCSD microgrid impact and opportunities

• Post analysis of operations (what would have been the best schedule?) • Analyzing operational rules or control actions

• Best periods for interrupting load? • Best time to charge thermal electrical storage tank • Suggested Electric Vehicle charging schedules • When to shift electric and thermal usage to reduce costs, or maximize revenues

• Determination of the value of supply contracts • Analyzing tariff changes or potential DR agreements/programs

• Mid to Longer Term: Further involve UCSD operations staff using the Master Controller • Review predicted results with actual results • Monitor data collected in central repository and verify efficiencies • Support energy market opportunities and benefits

2/16/2011 53 CSI Grant #2 Mid-

Project Demonstration

PROJECT DISCOVERIES


Project Discoveries


Data Gathering and Preparation Takes Time

Real-time Controls

Good working relationship with the end user is important

Model Complexity

Security – Physical and Cyber

Interoperability between systems

Demand Charges

NEXT STEPS


Firm

Distributed Firm & Flexible Resources


Flex

Smart Grid

< 5 min 20 min 1 hr 1 day 1 year

Distribution

System

Customer • Improve power quality • Reduce trips/faults

• Reduce frequency and duration of outages

• Reduce load & generation forecast error

• Provide • Fast ramp (MW/min) • Load following • Load for overgeneration

• Prevent Islanding • Prevent reverse flow • Reduce Volt/VAR fluctuation • Manage uncertain interactions

• Decrease O&M

• Provide frequency regulation

• Provide firm/predictable resource

• Increased equipment life

• Bill reduction • Risk reduction (financial) • Demonstrate physical

assurance • Reduce interconnection cost

• Provide firm/predictable resource for:

• system peak load (RA/ NQC)

• N-1 contingencies (NERC)

Focus on flexibility and capacity value of distributed resources

Address Tariff, Market and Regulatory Barriers


Tariff Barriers

Market Barriers

Regulatory Barriers

Low resolution metering and billing determinants

Rate complexity and uncertainty

Demand and stand-by charges

Complexity of settlement process

Infrastructure requirements

Limited capacity payment options

Net energy metering

Interconnection process Customer

Utility

CPUC

ISO Baseline calculation

Dual participation / double payment

Limited incentives of dynamic pricing

Limited access to wholesale markets

Limited capacity payment options

Limited payments for dispatchability

Address some (but not all) barriers to interconnection of distributed resoruces

Innovative Business Models

• Two-part rates • Demand Subscription • Firm Service Level

• Scheduled Load

• Provide load schedule • Allow re-schedule by

utility/ISO

• Incentives for flexibility/dispatchability

• Load Participation

2/10/2011 59 ©2011 Viridity Energy Inc. Proprietary and Confidential All Rights Reserved

• Allows greater flexibility in designing incentives and recovering fixed revenue requirement

• Provide generation/load

balancing

• Emphasis on high value benefits that are not currently captured.

Incentives Reason

Dynamic Pricing and Load Participation are not sufficient for renewable integration or grid support

Cost-Benefit Analysis Tool


Web Based Demonstration Model

http://www.ethree.com/public_projects/energy_storage_cost_effectiveness_

evaluation.php

http://goo.gl/kYNpv

(Will be available February 17, 2012)

http://www.ethree.com/public_projects/energy_storage_cost_effectiveness_evaluation.php




http://goo.gl/kYNpv

http://goo.gl/kYNpv

Daily Revenue by Benefit/Market


Present Value Cost/Benefit Comparision


QUESTIONS


THANK YOU FOR PARTICIPATING


APPENDIX SLIDES: BASELINING AND PRELIMINARY ANALYSIS


Resources VPower will optimize

Inputs

Outputs

Nat gas

Diesel

Import kW

Solar PV

Boilers

NG gens

Backup gens

Steam

Steam

KW

St gen

HX’s

Stm chs

Elec chs

Aux eqp

KW

HW

CHW

Heating

Cooling

Non CUP HVAC

Lighting

Plug loads

Water heating

TES Import + gen

Central Utility Plant End Use Denotes resources for which schedules will be optimized by VPower

Note: VPower also models JCI interruptible load & hot water strategy.

Services that VPower will treat as constant: CUP CHW, CUP HW, kW net CUP electric chillers 2/16/2011 CSI Grant #2 Mid-Project Demonstration 66

Data collected and analysis completed

• Daily data • Greensheet (dating back 2005) • Imports, hours of op, CUP gen,

gen gas use, some chiller kW

• Hourly data • Chilled water & hot water to

campus (dating back 2007)

• 15 minute interval data • MSCADA: Imports, CUP gen • JCI system: TES • BOP: steam data for boilers,

generators

• Solar data • 15 minute inverter & meter

data dating back 2009

• Historical price info • Gen & delivery prices dating

back 2009 (CAISO day-ahead, NGI; ALTOU, EG and GTNC tariffs)

• Efftrack • Chiller data obtained • Nameplate capacity and

efficiency info • Historical data dating back to

June 2011


Snapshot: hourly/15 min analysis

-30

-25

-20

-15

-10

-5

0

5

10

0

5

10

15

20

25

30

35

40

5/22

/201

1

5/23

/201

1

5/24

/201

1

5/25

/201

1

5/26

/201

1

5/27

/201

1

5/28

/201

1

5/29

/201

1

5/30

/201

1

5/31

/201

1

6/1/

2011

6/2/

2011

6/3/

2011

6/4/

2011

MW

h

CUP Generation ImportsCampus Total Consumption Solar Production (right axis)CUP Plant Consumption


Actual operations chilled water system: June 7, 2011

-4000

-2000

0

2000

4000

6000

8000

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

Tons

Hour

TES Discharge Electric Chiller OutputSteam Chiller output TES ChargeCampus chilled water supply Campus chilled water demand


Individual equipment efficiencies


Equipment level metrics (3)

• Thermal storage tank was analyzed: charge, discharge times, rates, losses generated

Quantity 10th % Median 90th % Daily discharge - ton-hr 6,401 16,250 24,504 Charge start time (hr starting) 21 23 1 Charge duration (hr) 8 10 13

Discharge start time (hr starting) 8 9 12 Discharge duration (hr) 10 14 16 Flow rate (gpm) 663 2,491 4,958 Delta T (deg F) 11 13 14 Overall tank efficiency over measurement period Charge ton-hr 6,751,512 Discharge ton-hr 6,454,207 Losses 4.4%


Snapshot: solar PV analysis


Summary of UCSD solar PV output

Max. Solar Meter Output (kW)

2009 2010 2011Birch Aquarium BIRC_1608 9/8/2009 49.2 95% 46.7 44.6 47.0 45.1 Campus Services Complex - Buildings A to E & Trade Shops CSC_1604 9/17/2009 258.4 97% 249.7 249.7 258.9 261.8 Campus Services Complex - Fleet Services Building FLSV_1361 Pre-2009 28.7 96% 27.5 24.4 24.4 24.4 East Campus Central Util ities Plant ECUP_1287 Pre-2009 28.7 96% 27.4 24.5 24.4 24.5 Engineering Building Unit 2 EBU_1362 and EBU_1363 Pre-2009 80.4 95% 76.5 66.6 66.6 66.5 Gilman Parking Structure GILM_1364 Pre-2009 195.0 99% 192.9 138.1 137.8 135.3 Hopkins Parking Structure* HOPK_1365 Pre-2009 338.0 99% 334.3 370.8 373.8 370.8 Price Centers, Buildings 1 to 4* PRIC_1409 Pre-2009 206.6 96% 199.3 199.4 200.8 219.4

Total 1,185.0 1,154.3 855.6 1,075.6 1,071.9 *Solar Meter Output from July 1, 2010 to June 30, 2011 was only provided as a daily average (kWh).

Rated Power (kW AC)

Total Rated Power (kW DC)

First Day of OutputIDSolar Meter Location

Inverter Efficiency

Annual Energy (kWh) Operating Hours Capacity Factor (%)

2009 2010 2011 2009 2010 2011 2009 2010 201116,471 68,389 65,890 2,751 8,756 8,754 12.8% 16.7% 16.1%

87,132 399,739 415,048 2,928 8,760 8,760 11.9% 18.3% 19.0%

50,805 47,928 45,293 8,759 8,759 8,008 21.1% 19.9% 20.6%44,458 47,745 50,470 8,172 8,759 8,759 19.8% 19.9% 21.0%

127,833 122,500 116,428 8,760 8,760 8,760 19.1% 18.3% 17.4%220,738 206,717 202,387 8,759 8,510 7,854 13.1% 12.6% 13.4%536,719 504,082 524,575 8,759 8,759 8,753 18.3% 17.2% 17.9%281,352 292,998 338,367 8,760 8,760 8,760 16.1% 16.8% 19.4%

1,365,409 1,690,041 1,742,042 8,760 8,760 8,760 13.5% 16.7% 17.2%

Birch AquariumCampus Services Complex - Buildings A to E & Trade ShopsCampus Services Complex - Fleet Services BuildingEast Campus Central Util ities PlantEngineering Building Unit 2Gilman Parking StructureHopkins Parking Structure*Price Centers, Buildings 1 to 4*

Total

Solar Meter Location


Day type influence

05

1015202530354045

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24Cam

pus T

otal

Con

sum

ptio

n (M

W)

Hour

September: Weekdays


05

1015202530354045

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24Cam

pus T

otal

Con

sum

ptio

n (M

W)

Hour

September: Weekdays with Weekends


0

20

40

60

80

100

120

140

160

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

Chill

ed W

ater

Loa

d (M

MBt

u/hr

)

Hour

September: Weekdays


0

20

40

60

80

100

120

140

160

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

Chill

ed W

ater

Loa

d (M

MBt

u/hr

)

Hour

September: Weekdays with Weekends



Year to year differences of campus power consumption: January look


Year to year differences of campus chilled water needs: February look

0.7

0.9

1.1

%

%

%

%

%

0

20

40

60

80

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

Chill

ed W

ater

Loa

d (M

MBt

u/hr

))

Hour

February: Weekdays

2008 - Median 2009 - Median 2010 - Median 2011 - Median


Project Team Contact Information

Name E-mail Phone

Eric Cutter [email protected] (415) 391-5100

Laura Manz [email protected] (858) 354-8333

Nancy Miller [email protected] (206) 718-0490

Snuller Price [email protected] (415) 391-5100

Chuck Richter [email protected] (425) 765-4349

Rick Schaal [email protected] (484) 534-3819

Priya Sreedharan [email protected] (415) 391-5100


mailto:[email protected]







california solar initiative (csi) research, development ... · research, development,...

Documents