WEM ESOO Demand Forecast Tracking

The Sprint

Add WEM ESOO demand/supply forecasts as a Siren-web scenario option, and track actual demand against past ESOO vintages to test for systematic bias.

Background

Each year, AEMO publishes the Wholesale Electricity Market (WEM) Electricity Statement of Opportunities (ESOO) — a 10-year outlook of electricity consumption, peak demand, and supply adequacy for the South West Interconnected System (SWIS). The report models Low, Expected, and High demand growth scenarios and underpins key reliability settings such as the Reserve Capacity Targets.

Because each year’s ESOO is a fresh “vintage” — a snapshot of AEMO’s best estimate at that point in time — a running archive of past editions creates a natural test bed: how have previous forecasts held up against what actually happened on the SWIS?

Siren-web already ingests live AEMO SCADA data and tracks renewable energy progress through its RET Dashboard. This Sprint extends that foundation by folding ESOO forecasts into the platform as a selectable scenario input, and by building the comparison logic to check those forecasts against reality.

Why it matters

If ESOO forecasts are systematically over- or under-estimating demand, that has real consequences for how much confidence SEN, policymakers, and the public should place in AEMO\’s Reserve Capacity Targets and reliability outlook — and it strengthens (or tempers) the case for particular renewable and storage investment pathways in the SWIS.

What this Sprint will deliver

• A new scenario type in Siren-web’s Powermatch/Powerplot modules, letting users select a WEM ESOO vintage (e.g. 2023, 2024, 2025, 2026) and demand scenario (Low / Expected / High) as a modelled demand/supply input
• A data pipeline to extract and store the relevant forecast tables from each published ESOO (PDF or published data tables) into Siren-web’s data model
• A bias-tracking view that overlays actual SCADA-derived demand against each historical ESOO’s forecast for the same period, surfacing whether forecasts have tended to run high, low, or true over time
• Summary statistics (e.g. mean error, mean absolute error by year-ahead horizon) to quantify any systematic bias, viewable on the RET Dashboard or a new report page

Team structure

Like other SEN Sprints, this initiative runs as a small team:

• Sprint Leader — defines scope, manages the team membership, schedules activities, reports to the committee.
• Sprint Members — conduct research, contribute towards objectives, document conclusions, review others’ contributions

Knowledge and Skills that would help

• The SWIS, AEMO, Government agencies related to Electrical Energy Systems
• Research
• Extracting and Processing tabular data (the ESOO reports are published as PDFs and a supporting Excel workbook, with some data also in AEMO’s WA data portal)
• Statistical Analysis
• Technical Writing

---

Siren-web is SEN’s open modelling platform for the South West Interconnected System, tracking progress toward 100% renewables. This is one of several active Sprints — small, scoped initiatives run by a Leader and volunteer Members.

EV Uptake & Charging Load Modelling

The Sprint

Analyse CSIRO EV uptake projection datasets, build charging-time/location assumptions into SIREN, and compare against actual uptake as it accumulates.

Background

CSIRO publishes EV Uptake Projections — Low, Medium and High projections of electric and fuel-cell vehicle fleet numbers and electricity consumption from 2021 to 2050, down to postcode level. CSIRO is also commissioned to produce annual Electric Vehicle Projections reports for AEMO\’s Integrated System Plan, which set out detailed charging-profile assumptions (home, public, commercial charging shares; time-of-use responsiveness; weekday vs weekend shape) under scenarios such as Progressive Change, Step Change, and Green Energy Exports/Accelerated Transition.

EV charging load is one of the more consequential unknowns for the SWIS: unlike passive demand growth, when and where people charge can shift a large, flexible load onto (or off) the grid at exactly the hours that matter for renewable integration. Siren-web currently models generation, storage and demand for the SWIS but has no explicit EV charging layer. This Sprint would add one, grounded in CSIRO’s published assumptions rather than ad hoc guesses.

Why it matters

EV charging could add a large and fast-growing source of new demand to the SWIS over the next decade — but how much, and when during the day, depends heavily on which uptake and charging-behaviour assumptions turn out to be right. Building this into Siren-web lets SEN test how sensitive the renewable transition pathway is to EV charging patterns, and flags early if actual uptake is running ahead of or behind the assumptions underpinning AEMO’s own planning.

What this Sprint will deliver

• A data pipeline to ingest CSIRO’s postcode-level EV uptake projections (Low/Medium/High) and derive SWIS-wide fleet numbers over time
• A tracking view comparing actual EV registrations/uptake in WA (as data becomes available) against the CSIRO projection curves, to see which trajectory the real world is tracking toward
• EV Charging as a demand factor in Siren-web as an input to demand forecasting
• EV V2G – as a source of supply.

Team structure

Like other SEN Sprints, this initiative runs as a small team:

• Sprint Leader — defines scope, manages the team membership, schedules activities, reports to the committee.
• Sprint Members — conduct research, contribute towards objectives, document conclusions, review others\’ contributions

Knowledge and Skills that would help

• The SWIS, AEMO, Government agencies related to Electrical Energy Systems.
• Research into transport electrification.
• Extracting and Processing tabular/geospatial datasets (CSIRO\’s data is postcode-level).
• Statistical Analysis.
• Technical Writing.

---

Siren-web is SEN’s Python-based open-source modelling platform for the South West Interconnected System. 

End-Use Electricity Demand Decomposition

The Sprint

Determine a breakdown of end-use electricity demand into different factors.

Background

SWIS electricity demand isn’t one thing growing at one rate — it’s the sum of very different end uses, each on its own trajectory:

• Residential base load (heating, cooling, appliances)
• Rooftop solar and household batteries, which are reducing grid-supplied demand during daylight hours
• Electric vehicle charging, which is only just beginning to scale up (see the companion EV Uptake Sprint)
• Large Industrial Loads (LILs) — mining, minerals processing, and other big single-site consumers, which AEMO tracks individually
• Emerging loads such as data centres, which the Data Centre Energy Demand report prepared for AEMO shows can materially exceed baseline forecasts if adoption runs hot
• Commercial and business electrification more broadly

AEMO’s own Electricity Demand Forecasting Methodology reflects this: it forecasts customer connections, technology adoption, and consumption separately for different segments (including individually surveyed LILs), rather than applying one aggregate growth rate. Lumping all demand together and applying a single growth rate to it hides the fact that some components are shrinking (grid demand offset by rooftop solar), some are just starting to ramp (EVs, data centres), and some are lumpy and site-specific (LILs).

Siren-web currently models total SWIS demand as an aggregate. This Sprint is the foundational data work needed before demand can be forecast end-use by end-use, each with its own growth assumption, rather than as one blended trend line.

Why it matters

Every other demand-forecasting Sprint depends on this breakdown existing. Without it, EV charging growth, rooftop solar offset, and industrial load changes all get blended into one number, masking the divergent trends SEN needs to understand and communicate; and making it much harder to spot when actual demand starts departing from any single assumed growth path.

What this Sprint will deliver

• A defined set of end-use demand factors for the SWIS (e.g. residential base load, rooftop PV/battery offset, EV charging, LILs/industrial, data centres, commercial), informed by AEMO’s forecasting methodology and available WA-specific data
• A decomposition of historical SCADA/consumption data into these categories as far as the data allows, quantifying each category’s current share of total demand
• Documented growth assumptions per category, sourced from AEMO, CSIRO, ABS and other public data, so each end use can be projected on its own trend rather than a single blended rate

Team structure

Like other SEN Sprints, this initiative runs as a small team:

• Sprint Leader — defines scope, manages the team membership, schedules activities, reports to the committee.
• Sprint Members — conduct research, data entry and demand factor modelling in Siren-web, document conclusions, review others\’ contributions

Knowledge and Skills that would help

• Research.
• Familiarity with, or willingness to dig into, AEMO/CSIRO/ABS public data sources.
• Statistical Analysis
• Technical Writing

---

Siren-web is SEN’s Python-based open-source modelling platform for the South West Interconnected System.

Renewable buildout and generation share

Sprint

To track the forward pipeline of proposed facilities and calculate implied renewable share (currently supported, though proposed-facility data is incomplete and contestable), plus a separate question of whether a “missing money” problem, possibly tied to the federal CIS scheme, is constraining new entry.

Background

Part 1 — the pipeline. AEMO’s own reporting shows a large pipeline of generation and storage sitting behind the SWIS’s committed projects: the 2025 WEM ESOO pointed to over 1,600 MW of committed/probable projects due online by 2027-28, plus a further ~2,300 MW submitted at the Expressions of Interest stage of the Reserve Capacity Cycle — on top of roughly 1,700 MW of ageing coal and gas expected to retire from 2027 to 2032. Siren-web already has a way of tracking proposed facilities and turning that into an implied future renewable share, but the underlying data is patchy: EOI-stage projects are self-reported, non-binding, and change often, so any renewable-share number built on them needs to be treated as indicative rather than a hard forecast — and clearly flagged as such.

Part 2 — is new entry being held back? Separately, there’s a live policy question of whether the SWIS is seeing a “missing money” problem — the well-established phenomenon where energy-only wholesale market revenue alone doesn’t cover the cost of new generation/storage, discouraging investment even where physical need is clear. The federal government’s Capacity Investment Scheme (CIS) was designed specifically to underwrite renewable generation and storage projects against this risk — including WA-specific tenders through the CIS WEM Design, which has already run several tender rounds contracting gigawatts of new WA capacity. The open question for this Sprint: is the scale and timing of CIS support keeping pace with what the pipeline data suggests is needed, or is there a gap — a “missing money” shortfall — that’s slowing projects down before they reach financial close?

Why it matters

The gap between “proposed” and “built” is where a lot of uncertainty about WA’s renewable transition lives. A single implied renewable-share number, if presented without its caveats, risks either over- or under-stating progress. And if new entry genuinely is being constrained by missing money — even with the CIS in place — that\’s an important, testable finding for SEN’s advocacy: it would mean the underwriting scheme’s current scale, timing, or design isn\’t fully solving the problem it was created for.

What this Sprint will deliver

• Clean and verified pipeline dataset in Siren-web — committed, probable, and EOI-stage proposed facilities, with probabilities of commissioning.
• SWIS Planned Transmission – Extracted from publicly available sources.
• An implied renewable share calculation built from that pipeline, shown as a scenarios based on probabilities.
• A tracked view of CIS activity in WA — tender rounds, capacity contracted (generation vs dispatchable), and timing — set against the pipeline and retirement schedule
• An explainer and indicator addressing the missing money question: is contracted/committed capacity (including CIS-backed projects) tracking ahead of, in line with, or behind what the retirement schedule and demand growth implies is needed
• Documentation of assumptions and data limitations, so the “implied renewable share” figure can be quoted publicly with appropriate caveats

Team structure

Like other SEN Sprints, this initiative runs as a small team:

• Sprint Leader — defines scope, manages the team membership, schedules activities, reports to the committee.
• Sprint Members — conduct research, data entry and demand factor modelling in Siren-web, document conclusions, review others’ contributions

Knowledge and Skills that would help

• Reseach.
• Comfort reading AEMO/CIS public disclosures and tender outcomes
• The ability to extract data from publicly available sources and technical publications
• Some grounding in electricity market economics would be very useful but not essential.
• Technical Writing

---

Siren-web is SEN’s python based open source modelling platform for the South West Interconnected System.

Global Leaders to Convene in Western Australia for ‘Global Citizen NOW: Perth’ Ahead of COP31 to Help Expand Clean Energy to 4 Million Homes

Sprint

Support for SEN’s participation at the event – man the stall – Perth Convention Centre 18–20 October 2026.

Background

refer GC Now presser 

What this Sprint will deliver

• Raising SEN’s brand as WA leading clean energy advocate
• Build our network

Team structure

Like other SEN Sprints, this initiative runs as a small team:

• Sprint Leader — defines scope, manages the team membership, schedules activities, reports to the committee.
• Sprint Members — attends the event, mans the stall

Knowledge and Skills that would help

• Public facing engagement – Sociable, for engineers this means friendly, outgoing, and liking the company of other people.
• Knowledge of SEN’s activities (advocacy, modelling, public education)
• Reliable