Geo-spatial Electrification Planning for Myanmar Sustainable Engineering Lab Director: Vijay Modi Department of Mechanical Engineering (SEAS) and Earth.

Post on 29-Mar-2015

214 Views

Category:

Documents

1 Downloads

Preview:

Click to see full reader

Transcript

Geo-spatial Electrification Planning

for Myanmar

Sustainable Engineering LabDirector: Vijay Modi

Department of Mechanical Engineering (SEAS) and

Earth Institute, Columbia UniversityNay Pyi Taw, Myanmar, November 22, 2013

1

Outline

• Rationale • Our Approach: analysis platform for decision making:

where what technology, prioritization, near-term decisions guided by long-term, least-cost

• Not engineering designs, but rapid planning • Examples: village, region and national scales • Combine International Experience + Local Expertise • Our Project Plan and Starting Work in Myanmar

Rationale

• Organize information in a systematic digital form

• Useful for – Quickly budgeting– Responding to internal and donor needs– Transmission upgrades/generation– Communication bet township/state/national

• Co-ordination between grid and off-grid projects

Benefits of National Geospatial Planning

• Accelerate national access at scale:– access and service standards can be applied consistently– important for remote, under-served populations

• Improve allocation of investments:– ensure on-grid generation benefits from economies of

scale– efficient targeting of off-grid systems for smaller

communities– enable rapid design and bulk procurement in roll-out

• Provide a coordinated investment framework:– help donors & government prioritize according to local

development goals– reduce risk for private sector investors and entrepreneurs

• NOTE: The difference between design vs. planning

4

Approach (1)

• Acquire detailed geo-spatial, cost and technical information:– Demand points (settlements)– Electricity Infrastructure (MV distribution

lines)– Demand level– Growth rates (population, economic)– Cost Factors (grid and off-grid, initial &

recurring)

5(with off-grid)

Approach (2)• Project demand and cost for a defined time-

horizon:– Apply growth rates, wealth / income mapping

• Algorithmic computation of least-cost electricity system:– Grid extension– Mini- or Micro-grids (renewable, hybrid)– Household Systems (solar)

• Generate phased Roll-Out plan for grid and distributed systems

6(with off-grid)

NetworkPlanner:An Algorithmic Planning System

NetworkPlannerHow it Works

• A free, web-based tool (no license fees) Designed by our lab at the Earth Institute

• Accepts essential local inputs of geospatial demand points, costs for electricity technologies, growth rates and other key parameters.

• Algorithmically plans least-cost electricity systems in these steps:

1. Computes the lifetime costs (initial and recurring) for meeting projected electricity demand at every point for all electricity options

2. Chooses the lowest-cost electricity technology for each point

3. Creates detailed map outputs for the least-cost grid and all standalone systems.

4. Creates detailed tabular outputs describing investment needs, recurring costs, additional generation required, and many other planning outputs.

Kenya / EI Electricity Planning Project:• Results highlighted

the need to electrify western, under-served areas

• Region specific generation req. and transmission upgrades

• Ultimately led to major loan 1.3B from World Bank 9

Kenya – The First Major Use of Geospatial Planning

Example: Electrification Planning at the Regional

scale

Flores, Indonesia: 6,300 settlements

in Eastern Indonesia10

Inputs that went into access planning

• Location of each settlements • Population of each settlement • Where is the existing medium-voltage network?

• Unit costs of utility infrastructure, off-grid• New customer demand

     Initial Costs  Recurring Costs

Grid Extension

$30/m MV line (with poles)$600 Total HH connection Costs

(incl: service drop/LV line)$200/ kVA Transformer (≥15 kVA)

$0.35 / kWh "bus bar" costAnnual O&M:

1% of line costs3% of transformer costs

 

Village Solar  $1.00/W panels (5 peak sun hours / day)$267/kWh batteries, 5 kWh/kW$100/HH for LV wire to home$1/W BOS (electronics)

$267/kWh battery replacement (every 3.3 years)Annual O&M:

1% of panel cost 

Off-Grid / HH Solar

$2.15/W panels (5 peak sun hours/day)$267/kWh batteries5kWh/kW battery capacity

$267/kWh battery replacement (every 2.5 years)Annual O&M:

2% of panel cost

13

Essential Local Contribution: Technical and Cost Parameters

Example of Grid-Rollout over 20 Years(Flores Island, Eastern Indonesia)

15

16

17

18

19

A complete electricity plan specifies locations with grid access, mini-grids, and off-grid (solar

home systems)

0-1011-20

21-50

51-100

101-250

251-500

501-1000>1000

0

20,000

40,000

60,000

80,000

100,000

120,000

140,000

Household SolarVillage Solar Mini-GridProposed GridGrid Connected

Bins: Number of Households per Settlement

Total Number of Households 

by Tech Type

Proposed Grid: 166,000 HHsVillage Solar Mini-Grid:

84,000 HHs

21

Model results provide quantitative outputs for locations served by all system types, with

capacities and costs.

Investments (250,000 new HH with elec)

Pre-Existing Grid:

Pre-existing household grid connections (2010 Census) 170,130

Proposed New Grid Conn 165,000Total Initial cost for grid network (MV+LV) $165 Million

Village Solar Mini-Grids Conn 84,000Total Initial cost for all Mini-Grid systems $94 Million

Household Solar (model output) 3000$5M

Generation Costs for Grid Connected HHCoal + Solar + Geothermal + Wind

22

Grid: MetricsPre-Existing Grid: Units

Pre-Existing MV line length kilometers 1,505

Pre-existing connections (2010 Census) Households 170,130

Existing MV Line Length per HH meters 8.8

Scenario Results Units 480 kWh/yr 240 kWhProposed MV line length kilometers 1,934 94Proposed new grid connections Households 166,141 19,400Proposed MV line per new HH meters 11.6 4.8Proposed Total New Capacity MW 21.8 1.3Proposed New Capacity per HH kW 0.130 0.065

Scenario Results 480 kWh/yr 240 kWh/yr

 Grid Costs Totalper new HH conn. Total

per new HH conn.

Total Initial cost for grid network (MV+LV) $163,479,151 $984 $14,337,188 $736 for MV grid network $53,638,088 $323 $1,904,402 $98 for LV, Svc Drop, HH wiring $109,841,063 $661 $12,432,786 $638Total levelized cost for Grid power $0.63 / kWh $0.72 / kWh

23

Solar Mini-Grid: Metrics

Solar Mini-Grid: Costs

Scenario Results 480 kWh/yr 240 kWh/yr UnitProposed Total Capacity of Mini-Grids 35 49 MWProposed new mini-grid connections 84,430 215,840 HouseholdsProposed Capacity per HH 416 225 W/HH

Scenario Results 480 kWh/yr 240 kWh/yr

Totalper new HH conn. Total

per new HH conn.

Total Initial cost for system $94,032,500 $1,114 $149,769,300 $694System discounted cost $366,324,716 $4,339 $509,826,782 $2,362Total demand met by mini-grid (in kWh) 502,893,767 5,956 642,808,188 2,978 Total levelized cost per kWH for Grid power $0.76 /kWh $0.83 /kWh

10% of the kWh from diesel backup for solar mini-grids brings reliability from 80% to 90%

24

Solar Home Systems: Metrics and Costs

Scenario Results 480 kWh/yr 240 kWh/yrProposed Total watts of SHS 946 kW 3,197 kWProposed new SHS connections 2,649 Households 17,907 HouseholdsProposed Size for each SHS 356 W/HH 179 W/HH

480 kWh/yr 240 kWh/yr

 Scenario Results Totalper new HH conn. Total

per new HH conn.

Total Initial cost for system $5,321,274 $2,009 $17,992,857 $1,005System discounted cost $18,141,961 $6,849 $61,343,525 $3,426Total demand met by Grid (in kWh) 15,778,344 5,956 642,808,188 35,897Total levelized cost for SHS power $1.16 /kWh $1.16 /kWh

Solar Home Systems Solar: Costs

25

Marginal costs of grid connections can increase with roll-out

Model ResultsBackground GIS PlatformPopulation Modeling Solar MicroGrids Talking Points

26

Mini-grid

Mini-Grid Rollout: An example prioritizing higher demand villages

Grid?

Solar home System

Mini-grid 

High

Vol

tage

Planning at Township Level 

Algorithm Results Technology Options

SolarMiniGrid

Phase 1 < 20

Phase 2 20 – 25

Phase 3 > 25

Investment in MV line per

connection (meters shown)

Grid Rollout: An example prioritizing most cost-effective branches

Incremental Infrastructure

• one could start local and where/when demand grows and grid comes closer connect to the grid

• Keeps initial investments small and modular• Does not strain utility immediately • Allows demand grows and entrepreneurship

to emerge organically

Our Project Plan

• The Earth Institute’s approach to Electrification Planning in Myanmar will include three phases with an in-country workshop roughly marking each phase, as outlined below.

Our Project Plan

1. Data Gathering– Milestone 1: Inception Workshop– Deliverable 1: Inception Report

– Milestone 2: Formation of a GIS dataset with acceptable resolution of population and MV network representation

– Deliverable 2: Interim Report

Our Project Plan

2. Demand Analysis & Quantify Supply Needed– Milestone 3: Workshop on Initial Results– Deliverable 3: Draft Final Geospatial Rollout Plan

3. Final Scaled-Analysis and Knowledge Transfer– Milestone 4: Final Results Workshop– Milestone 5: GOM feedback on Draft Report– Deliverable 4: Final Geospatial Rollout Plan &

accompanying datasets developed

Starting Work For Myanmar: Data Sets Needed for Electricity Modeling

• Settlement Data from Village or Village Tracts– Locations– Populations

• Existing Grid Distribution Data– map or digital file for medium-voltage lines

• Other drivers– Supply options, current/future– Demand/hh, population and demand growth, – unit costs of options

Village locations within Village Tracts

Village Tract data may be supplemented in some cases

Electrification Planning benefits from village level

data.

Settlement Data

• From MOLFRD, DRD• Village location and population, 2001 • In absence of more recent data, an excellent

starting point • Project to future• Can we updated as new figures become available

EXAMPLE:  State: Magway   Township:  Pakoku

Existing distribution grid

• Transmission Lines – Available through MEPE, up to 66 kV

• Medium Voltage lines, ESE, YESB – 33 kV and 11 kV lines – Currently not available in digital geo-spatial form– ESE is compiling paper/scan images from all

states/regions, estimated avai: 1 month • Smaller off-grid systems

– Expect to report later

Other drivers

• Large generation Sources– Hydropower, HGPE (generation), DHP (planning)

• Cost of Grid Generation in the future– Some uncertainty due to variable international

fuel supply, scale of demand growth– JICA study– Hydro potentials, thru MOPE

• Hydro <5MW, from ESE • Unit costs, demand

Original:jpeg with hand-drawn

MV lines (red)

GIS product:shapefile with digitized

MV lines (blue)

Training and Capacity Building : Example: Data collection with smartphones

44

Training and detailed work with local electricity

technicians and managers resulted in completion of a

detailed local medium-voltage distribution grid

map for a region in Eastern Indonesia

West Timor (IDN), MV grid lines following EI training & mapping

West Timor (IDN) MV grid initial estimate

Training & mapping of MV grid lines (IDN)

Training & mapping of MV grid lines (IDN)

Approach to data

• Settlement Data valuable for electricity, gas, roads, water, and other services

• Tools to gather and “maintain” data. • How to rapidly capture planning grade data for

existing lines • Work closely with Government/Utility to

embed tools/processes in their systems

Working with Ministries, Utilities, Enterprises 

• We would like to begin here in Nay Pyi Taw, possibly following up in other regions or offices. We ask your permission and support

• Our plan and approach is to work side-by-side with local experts and practitioners to ensure the relevance, completeness and accuracy of all data and outputs.

• Workshop tomorrow

top related