The ESKOM Pilot Testing and Findings of the Residential Time-of-Use Tariff (Homeflex) Project by Vashna Singh and Marcus Dekenah 17 October 2006.

The ESKOM Pilot The ESKOM Pilot

Testing and Findings of the Testing and Findings of the

Residential Time-of-Use Residential Time-of-Use

Tariff (Homeflex) ProjectTariff (Homeflex) Project

by Vashna Singh and Marcus Dekenah

17 October 2006

Contents

• Objective of Homeflex• Background• Market Drivers• Pilot Objectives• Pilot Testing• Pilot Results• Conclusions• Questions

The Objective

• The objective of the Homeflex Pilot Project is to develop and introduce a residential time-of-use tariff that will penetrate the market and provide incentives and benefits to customers which will ultimately result in the optimization of the country’s peak demand curve profile.

Background • Eskom identified the need for Residential

Time-of Use tariff.• A tariff (Homeflex) was developed,

various pilots were run - less than 5% load shift.

• Various other load management & efficiency strategies were studied during this period - not integrated with the tariff.

• Changing environment - WEPS, ISEP, ESI, EDI - tariff become outdated.

• A revised Business Case was proposed and accepted by Distribution Group Sales - recommended piloting of tariff with load management technologies.

Market Drivers

• The continued increase in Eskom’s peak demand and requirements for increased capacity decisions that need to be made.

• A requirement for increased sales in off-peak times.

• The Energy Policy White Paper and DME’s Electricity Pricing Policy stipulations for time-of-use tariffs, cost reflective tariffs, differentiated capacity charges and sophisticated tariffs for the upper market.

• Increased customer needs for flexibility and lower costs - increase in customer service due to reduced bills and more value adding options.

Market Drivers

• Load shifting in an all electric market reduces the Distributor’s purchase costs and increase profitability.

• Better alignment to WEPS - increased incentive to shift load, resulting in improved net contribution for the Distributor.

• A target market exists for medium to high residential consumers who have ability to shift load.

• Major market for Homeflex lies in municipal areas (still benefit Eskom Generation).

Pilot objectives

• Primary aim: to determine whether a TOU pricing signal would encourage customers to shift their electricity consumption from peak to off peak periods.

• Other aspects tested: Customer’s acceptance of and

response to the tariff and load management technologies.

The Distributor’s support structures required for the implementation of Homeflex e.g. Metering, Billing, Contact Centre, Field Services.

Pilot Testing

• Pilots funded by Eskom’s DSM.• Revised pilots were launched at eThekwini

Electricity and Sandton, Tableview was a new site.

• Three pilots were run - Tableview, Sandton and Durban Metro.

• Each Pilot sample was made up of – 50 Test customer on 2 Part Homeflex tariff,– 50 Test customers on 3 Part Homeflex

tariff,– 50 Control customers (on Homepower)

• An ABB Load Research installed at each together with current domestic meter.

Pilot Testing

• Durban Metro pilot (customer controlled load management - timers used), pilot is closed.

• Tableview Pilot (utility controlled load management - Flexicon system used), pilot is closed.

• The Sandton pilot is still in operation to test other metering options.

Pilot Results

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

0 250 500 750 1000 1250 1500 1750 2000 2250

Consumption [kWh/mth]

Gey

ser

Ow

ners

hip

[pu]

Source: NRS/TSI LR projects 1994-2003

Who has a geyser?

Customers over 500kwh/mth

Who is the target?

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0 500 1000 1500 2000 2500 3000 3500 4000 4500

Consumption [kWh/mth]

Cu

m p

rob

ab

ilit

y o

f c

on

su

mp

tio

n

Analysis of C1,C2,C3 sales Y2002

Load shift: Sandton

0

1

2

3

4

5

6

7

0 2 4 6 8 10 12 14 16 18 20 22

Time of day (hour)

Ave

rag

e lo

ad (

kW)

CON

TOU2

TOU3

Avg. Winter weekday, 2002

Mid-range customers only

Load shift: Tableview

0

0.5

1

1.5

2

2.5

3

3.5

0 2 4 6 8 10 12 14 16 18 20 22

Time of day (hour)

Ave

rag

e lo

ad (

kW)

CON

TOU2

TOU3

Avg. Winter weekday, 2002

Mid-range customers only

Modeled response to TOU

0

1

2

3

4

5

6

7

0 2 4 6 8 10 12 14 16 18 20 22

Time of day (hour)

Avg

erag

e w

inte

r w

eekd

ay l

oad

(kW

)

STOU

SCON

TTOU

TCON

Avg. Winter weekday

Summary of load shift*

AM07h00-10h00

PM18h00-20h00

Sandton 0.9 kW/hh(-17%)

0.5 kW/hh(-7%)

Tableview 0.7 kW/hh(-30%)

0.9 kW/hh(-29%)

Eskom target market

0.76 kW/hh(-23%)-86MW

0.72kW/hh(-18%)-82MW

*Avg. high season weekday per model. Highest mean saving per time-interval in PSO peak-slot shown.

Summary (Contd.)

• TOU2/TOU3 does not affect consumption.

• Profile of TOU2/TOU3 the same.• Profile of TOU2/3 different from

Homepower control. Geyser control is profile modifier.

• Difference between local/remote load control.

• Elasticity not modelled.

Outputs

• Statistical conclusions based on 200 consumers.

• Annual hourly profile model with household (C, TOU2/3) and network (WEPS) sales and load.

• TOU Tariffs can be tested on model.• Profile model can be applied to

each Accops.

Implementation Strategies

• Pilot testing proved that Eskom’s Distribution support structures can adequately handle Homeflex.

• Lessons from pilot test show that the metering technologies and integration with billing system needs further refinement in order to develop a more feasible and cost effective package for roll out.

• A request for a metering, load management and data management solution was sent out to the market.

• The Request for Proposal is currently being evaluated.• Chosen technologies will be tested for suitability,

implementation readiness and customer acceptance in a Proof of Concept Phase later this year.

Conclusion

• Homeflex may be usefully practiced on consumers using more than 500kWh/month.

• Response to a 2 part or 3 part tariff is not distinguishable, but was distinguishable from the control groups.

• The introduction of the tariff did not change the levels of consumption.

• The tariff may be the “glue” that keeps the load shedding devices in place, operating normally and untampered.

Conclusion

• To make load management strategy most successful, the customer must see direct benefits. A time-of-use tariff is an extremely effective strategy which provides immediate incentives for customers to do load shifting, hence reaping direct benefits.

Questions?