WalMarts View on Demand Response Program Design

1
Wal-Marts View on Demand Response Program Design

• Anoush Farhangi
• Angela Beehler

2
Metering Concerns

• A cornerstone of deregulation is advanced
  metering. Without measuring consumption in finer
  intervals it is not possible to verify
  participation in demand response programs.
• The cost of advanced metering should be lowered
  by removing unreasonable regulation, and
  promoting competition and innovations.
• Customers should be allowed to install their own
  advanced meters given that they are in compliance
  with the standards set by the regulatory
  authority.
• Customers should receive full credit for meter
  ownership.
• Customers or their authorized representative
  should have full, frequent and easy access to
  their own meter data.
• The data should be made available to customers in
  a standard format set by the regulatory
  authority.
• The customer should not bear any cost for access
  to their own meter data.

3
Identify Target Market

• The highest priority of the Commission should be
  the development of rules, regulations and
  incentive plans that would encourage Demand
  Response among all customers.
• Rules should not be developed for the benefit of
  a handful of market participants.
• Large industrial customers that are the sole
  beneficiaries of the current regulations could
  engage in one on one contracts with Ercot.

4
Terms and Conditions

• Customers or their authorized representative
  should be able to directly participate in demand
  response programs without having to go through a
  third party aggregator and without having to pay
  any participation fees.
• Paperwork, and program terms should be kept
  simple and standard across TDSPs and systems.
• A customer should be able to aggregate its entire
  load from multiple sites in one system or zone.
• Minimum per site requirement is discriminatory
  and prevents the full participation of commercial
  customers.

5
Market Reforms

• Market data should be available to all customers
  in a timely basis.
• Market should be transparent. After the fact
  changes to market data should be kept to a
  minimum.
• Price signals should be accurate and allocated to
  specific intervals in which the cost occurs.
• Balancing accounts, uplift charges, etc. that
  apply to load ratio shares not to hourly prices
  distort market signals and should be minimized.
• Customers must be made aware of severe market
  conditions well ahead of time to prepare for full
  load participation.

6
Incentives Reform

• There should be a parity between demand and
  supply alternatives.
• Reliability pay should be proportional to the
  response time, not to size.
• A customer able to reduce one KW in 10 minutes
  should get paid more than a customer able to
  reduce 1 KW in 30 minutes.
• A customer able to reduce load by 10,000 KW in 10
  minutes should get the SAME rate per KW as the
  customer that can reduce 1 KW in 10 minutes.
• Demand Resources can be used to reduce load,
  shift load, provide reliability support or to
  lower system ramp rate.
• Demand resource deployments can result in
  shifting load from high heat rate periods to low
  heat rate periods, or in conservation.
• In addition to providing reliability benefits,
  demand resources provide environmental attributes
  as well.
• The Commission should consider providing to the
  customers environmental incentives for deployment
  of demand resources such as RECS or White Tags.

7
Base Line Development

• Calculation of baseline energy should be based on
  the conditions at the time of interruptions and
  not based on historical data.

8
Experiments
9
Connecticut Experiment 8/1/2006 and 8/2/2006
The Experiment resulted in around 3 MW of load
reduction in 35 facilities in CT in two critical
periods in the Northeast.
10
Light Dimming Experiment in Alice Texas
Lights were dimmed to take advantage of natural
daylight Spikes in lighting load as cloud cover
reduced natural daylight Direct effect
Significant reduction in lighting load around
70KW Indirect effect Reduction in Air
Conditioning load around 40 KW Currently, over
100 stores in Ercot are equipped with the dimming
technology.
11
Pre Cooling Experiment, Tyler Texas

• The store was pre cooled in the early morning
  hours.
• Set point was returned to normal at 7 am.
• 250 kwh increase in AC, 120 KWH reduction in
  Refrigeration, a net of 130 KWH gain.
• Can be used to lower system ramp rate.

12
Texas 4/17 and 4/18 load
Significant reduction on 4/18 after AC set point
was raised by 3 degrees
13
Baseline estimated from historical data is not an
accurate measure of savings