GridCORTEX Live · Wholesale Markets & Trading  ·  ← All Demos
On this page What you are watching The business case Run this at your utility Where you see it and how you say yes

The Spread Synthetic Data · Simulation

In a nodal market the money lives in the differences, and the differences are physics. Forty-eight hours at BrightPlain Energy's trading desk: a planned transmission outage meets a forecast wind ramp, and the market-physics model calls it 36 hours early, the Elm–Wood River 230 kV constraint binds tomorrow evening, and CEDAR FLATS diverges $60/MWh from the hub. Watch the desk get positioned while the screen still shows $2 of separation: day-ahead offers reshaped before the crash is public knowledge, the wind hedged at DA prices, the gas unit offered into the spike, the battery scheduled across the spread, and the constraint-path FTR valued before the auction window closes. Every order placed by a human trader. The model just makes sure they see tomorrow first.

HOUR 0
MARKET PHYSICS SCAN
⏳ DECISION POINT: TIME SLOWED
SYNTHETIC DATA
Hub LMP CEDAR FLATS LMP Constraint binding Forecast cone Desk position

The desk saw tomorrow 36 hours early.

What a nodal trading desk looks like when the price forecast knows the physics
,
48-hour desk P&L
,
Peak spread called at
,
Wind revenue protected
The Market
WithoutWith GridCORTEXΔ
The Desk
WithoutWith GridCORTEXΔ
Illustrative simulation on synthetic data, BrightPlain Energy and all nodes, constraints, and prices are fictional; no real RTO, market participant, or market outcome is depicted. This is decision-support software, not trading advice, and nothing here is an offer or recommendation to transact; every order shown is placed by a human trader under the participant's own risk policies and market rules. See UC 11.1.
$2
Hub–node spread (live)
,
Constraint bind probability
$0
48-hr desk P&L
0%
Wind curtailed
Trading Desk Feed, physics · positions · every order placed by a human
Physics scan
The call
DA offers
FTR window
RT binds
Settle
The Validated Use Cases Behind This Scenario
UC 11.1
Nodal Price Forecasting Engine
GPU market physics, not statistics: the $60 divergence at CEDAR FLATS called 36 hours out, with the constraint and the hour attached.
UC 11.2
Bid & Offer Optimization Copilot
The reshaped day-ahead stack: wind locked at DA, the gas unit offered into the evening spike, the battery scheduled across the spread.
UC 11.3
Congestion & Curtailment Risk Analytics
The constraint that actually binds, outage plus wind ramp modeled together, curtailment exposure quantified per asset before it happens.
187 UCs
One Framework
The Spread is one of 187 validated use cases across 10 solution areas and 23 utility domains.
Inside the Demo
What you are watching, and what it proves

BrightPlain Energy is a fictional power trading firm in a "nodal" wholesale market. In a nodal market, every location on the grid gets its own electricity price. When a power line clogs, prices a few miles apart can split by huge amounts, and that split is where trading desks make or lose money. The simulation covers 48 hours at the desk, from Day 1 at 08:00 through final settlement at the end of Day 2. The desk manages three assets: a 180 megawatt wind farm at a local pricing point called CEDAR FLATS, GT-2, a 240 megawatt gas power plant at the hub (the region's main reference trading point), and a 100 megawatt battery. At the open, the hub and CEDAR FLATS prices sit just $2 apart. It looks like a normal Tuesday. Overnight, the forecasting model reads three things together: tomorrow's published notices of planned line maintenance, the wind forecast, and the map of how every line and substation connects. One corridor stands out. A major transmission line, Elm–Wood River, is scheduled for maintenance work, exactly when a 34 mph wind ramp will push extra wind power onto that part of the grid. The model says the remaining lines hit their safe carrying limit (traders call this "binding") on Day 2 from 17:00 to 21:00, at 78% probability and rising. When a line binds, cheap power gets trapped behind it. The wind farm's local price crashes toward $0 while the hub price spikes, a gap of $58 to $64 per megawatt-hour, the standard unit price of wholesale electricity. The model makes this call 36 hours before the market shows it.

Four decision points follow, each staged for a human trader. Nothing is ever submitted automatically. First, the congestion call: the head trader approves a full review of every position against the predicted price split. The review shows the wind farm has 8.4 gigawatt-hours of energy at risk, energy that would be sold at crashed prices or shut off and wasted entirely. Second, the day-ahead offer reshape, before the 10:30 deadline. The day-ahead (DA) market is where tomorrow's power is sold today at locked prices; the live minute-by-minute market is called real-time (RT). The desk sells the wind farm's output day-ahead: it clears at $26.40, locking today's healthy price before tomorrow's real-time price crashes to $4. It also raises GT-2's offer so the gas plant sells into the predicted evening spike, and schedules the battery to charge at cheap CEDAR FLATS prices and sell back at expensive hub prices. Together this is worth an extra $188K versus doing nothing. The offers go in at 10:28, two minutes before the deadline. Third, the FTR window. A financial transmission right (FTR) is a contract that pays its owner when a specific transmission path gets congested, so it works like insurance on congestion. The same model shows this corridor will bind 9 of the next 12 months under the maintenance plan, making the contract worth about $96K more than the auction price suggests. The desk buys it 30 minutes before the auction closes.

Then the event arrives on schedule. At 17:10 on Day 2, Elm–Wood River binds: CEDAR FLATS trades at $6.80 and falling, the hub at $52 and climbing. By 19:00 the gap peaks at $62 per megawatt-hour, inside the predicted band. The fourth decision point fires mid-event: the maintenance crew keeps the line out 3 hours longer than planned. The model recalculates in minutes and recommends running the battery two hours longer and keeping GT-2 on through 23:00, worth another $128K. The line returns at 22:00. Ending state: the forecast was within $3 at the peak, only 1.1 gigawatt-hours of wind was wasted, and all of it had already been sold at day-ahead prices, and the settlement preview shows a $412K profit over the 48 hours. Every order was placed by a human, and every step was recorded in Relay, the built-in audit log, for the risk committee. Skip the approvals and the run ends the other way: a loss of $486K on an event that was in the public maintenance notices all along.

Without GridCORTEX

The maintenance notice sits in one browser tab. The wind forecast sits in another. The price forecast, built on historical averages, shows a normal Tuesday, because nothing like this exact combination is in its history. Nobody connects this maintenance job to this wind ramp, so the day-ahead market closes on autopilot: the wind sold as usual at CEDAR FLATS, the gas plant's offers unchanged. At 17:10 the line binds anyway. CEDAR FLATS crashes to $4. The wind farm gets paid the crashed live price, then is shut off entirely, wasting 8.4 gigawatt-hours of energy. GT-2, never positioned for the evening, watches the $60 hub spike from the sidelines. The 48-hour result: a $310K loss on the wind farm's crashed and wasted output, $106K of missed profit on GT-2, and $70K of battery mistiming, for a total loss of $486K, plus a missed insurance contract and a post-mortem that starts with "what happened?"

With GridCORTEX

The market physics engine (use case UC 11.1) runs on graphics processors, the chips that make heavy simulation fast, and solves demand, wind, maintenance notices, and the grid map together. So the $60 price split appears in the forecast 36 hours before it appears on any screen, with the exact line and the exact hours named. The offer copilot (UC 11.2) recommends the reshaped day-ahead sales (worth $188K), the congestion analytics (UC 11.3) price how much wind energy is at risk, the congestion-insurance contract is valued in time to buy it (worth $96K), and the live recalculation captures the 3-hour maintenance extension (worth $128K). A person stays in charge at every step: all four recommendations wait for the head trader's approval, they expire if ignored rather than executing themselves, and traders place every order under the desk's own risk rules. The result is a $412K profit, a $898K swing versus the unprepared desk.

The key numbers (KPIs), side by side
KPIWithout GridCORTEXWith GridCORTEXDelta
The $60 spreadthe price gap between the hub and CEDAR FLATS, the number this whole story trades on (D1 and D2 are Day 1 and Day 2)discovered at 17:10 D2called at 08:00 D136 hours of warning
Constraint IDnaming the exact power line that will hit its safe limit, and the exact hourspost-mortem findingnamed, timed, 78→94%predicted from grid physics, not past averages
Wind at CEDAR FLATShow the wind farm's output was sold: at yesterday's locked price, or at the crashed live pricerode RT down, then curtailedsold at DA $26.40price risk closed a day early
Curtailmentwind energy shut off and wasted because it had nowhere to go on the clogged grid8.4 GWh lost1.1 GWh, all pre-soldalmost all the waste avoided
GT-2 and the spikewhether the gas plant was positioned to sell into the evening price spikeunderoffered, watched itcleared into 17–21+$106K captured
RT outage extensionthe crew kept the line out 3 hours longer than planned, a mid-event surprisea second surprisere-solved in minutesthe surprise paid the desk
48-hour desk P&Lthe desk's total profit and loss over the two days−$486K+$412K$898K swing
DA offer valueextra profit from reshaping what the desk sold into the day-ahead marketautopilot stack+$188K reshapeddone before the 10:30 deadline
FTR paththe congestion-insurance contract that pays when this specific line clogswindow missed+$96K at model marksvalued before the auction closed
RT adjustmentschanges made in the live market as the event unfoldednone+$128Kadded on top of the day-ahead win
Risk filethe record the risk committee reviews after the event"what happened?"Relay-traced, every stepthe committee sees why
Human controlwho actually submits the ordersn/aevery order trader-placeddecision support only
Live numbers (KPIs) on the dashboard
Hub–node spread (live)The live gap between the hub price and the CEDAR FLATS price, in dollars per megawatt-hour. Around $2 is a normal quiet market; $62 at the peak is the event the desk spent two days positioning for.
Constraint bind probabilityThe model's odds that the Elm–Wood River corridor hits its limit in the predicted window. Low and falling would mean a quiet evening; here it climbs from 78% to 94%, meaning the price split is coming.
48-hr desk P&LThe desk's running profit and loss across the event. Climbing toward a $412K profit means the approvals are working; falling toward a $486K loss is what an unprepared desk looks like.
Wind curtailedHow much of the wind farm's 8.4 gigawatt-hours of at-risk energy is actually wasted. 1.1 GWh, all sold in advance, is the good reading; 8.4 GWh unsold and shut off is the bad one.

The Business Case: Safety, Hours, and Cost

A utility does not buy a demo. It buys a safety exposure that goes away and a cost that goes down. Below is that case for every use case behind The Spread, written the way a plant manager, a safety lead, and a CFO each need to read it. Every hour and every dollar is a formula you run with your own rates and volumes. There are no vendor benchmarks in here and no invented percentages. If a number is not yours, it is not a number.
UC 11.1 Nodal Price Forecasting Engine

What happens today, without this

Every morning a day ahead desk analyst builds tomorrow's price view by hand. They pull the market operator's posted locational marginal prices, called LMPs, for the specific grid locations where your units and contracts settle, paste them into a workbook next to gas forwards and the load forecast, and shape the hours off last year's similar days. The same analyst does a lighter version of that work again before each real time window. When the person who owns the workbook is out, the desk positions off the market operator's own forecast and a gut call.

What it replaces or shrinks

  • The morning workbook build of a nodal price curve from posted history, gas forwards, and the load forecast
  • Hand reshaping of hourly price curves off similar day lookups
  • The lighter hour ahead refresh an analyst redoes before each real time window, which shrinks rather than disappears
  • Chasing node level data across market operator portals and pasting it into the desk workbook
  • The forecast versus settled price reconciliation that nobody currently has time to run
  • One off requests to the analyst for a price view at a node the desk does not normally cover

Why it is safer

This use case has no direct safety benefit and we are not going to invent one. The only honest link is indirect: a desk that is positioned correctly buys less energy in a scramble, and fewer scrambles means fewer unplanned unit starts and fewer off hours callouts for plant and switching staff.

Counted in units you already track:

  • Night driving hours for off hours callouts to start units the desk did not plan to run
  • Road miles driven by on call plant and field staff responding to unplanned starts
  • Switching operations performed under time pressure to support an emergency purchase or an unplanned dispatch

Man-hours it gives back

Analyst hours come back to the trading desk, and the desk's morning moves from building a number to arguing about it.

HOURS AVOIDED PER YEAR = trading days per year x analyst hours per day spent building the day ahead price view, plus real time windows per day x trading days x minutes per hour ahead refresh, plus one off node requests per month x 12 x hours per request, minus the review time an analyst still spends checking the forecast against their own market read before the desk positions.

The numbers we need from you to run that formula:

  • Trading days per year and analyst hours per day spent building the day ahead view
  • Hour ahead refreshes per day and minutes each one takes
  • One off node level price requests per month and hours per request
  • Loaded hourly rate for a desk analyst and for a scheduler
  • The nodes you settle at and how many of them the desk currently covers by hand

Where the dollars come from

Cost driverHow it is calculated, from a rate you supply
Analyst laboranalyst hours avoided x your loaded desk analyst rate
Imbalance exposureyour megawatt hours settled in real time against a day ahead position x your own average absolute difference between the two prices x the share of that gap you believe a better forecast closes, a share you set, not us
Basis and hedge sizingyour hedged notional volume x your own realized basis between hub and node x the share of basis error you attribute to a price view built from history rather than from grid conditions
Vendor consolidationthe annual subscription fee for any third party price forecast this replaces or lets you downgrade
Missed positionyour desk's own margin per megawatt hour x the volume you did not commit because the price view was not ready before the market deadline

Reliability and maintenance

Reliability
This does not move SAIDI or SAIFI and we will not pretend it does. It moves financial risk: the exposure that appears when your position and the settled nodal price disagree, and the pressure a badly positioned desk puts on operations to start units late. Measure it as the variance of your real time settlement against your day ahead schedule, which your back office already reports.
Maintenance
There is no physical maintenance effect here. The nearest equivalent is model upkeep: forecast error is scored against settled prices every single day, so a model that starts to drift shows up as a trend in the daily score rather than as a bad quarter someone notices at the close.

What else it moves

ComplianceEvery position has a dated, reproducible price basis attached to it, which is what a market monitor inquiry or an internal risk audit asks you to produce.
WorkforceYour analysts stop being spreadsheet mechanics and start being market readers, which is the part of the job that keeps a good analyst from leaving for a hedge fund.
CustomerFor the load serving side of the house, a lower cost of supply flows to customers through your fuel and purchased power adjustment, which regulators will ask you to show.

What it costs you, stated honestly

You pay for the GridCORTEX forecasting service, for the market data feeds it needs, which you may already license, for the integration into your energy trading and risk management system, called an ETRM, and for your own analyst time to run the forecast in shadow mode next to the current process for a full season before you trust it. The shadow season is the real cost and it is not small.

How to build the payback case

Payback is usually carried by analyst hours and by the imbalance number, because those are the two you can audit from your own settlement statements. Treat trading margin improvement as upside, because it is the number your risk committee will trust least.

This is a planning model driven by your rates, your volumes, and your own settlement history, not a vendor claim. Re-run it with your actuals after the shadow season before you size the subscription.
UC 11.2 Bid and Offer Optimization Copilot

What happens today, without this

Before every day ahead market deadline, a senior trader and a scheduler rebuild the offer stack for each generating unit by hand. They take the current gas price, the unit's heat rate curve, its start and no load costs, whatever derate the plant emailed over that afternoon, and the must run and emissions constraints that live largely in one person's head, and they type an offer curve per unit into the market interface. Storage is bid on a rule of thumb against a price curve someone drew that morning. When the deadline is close, yesterday's offers get copied forward.

What it replaces or shrinks

  • The manual rebuild of each unit's offer curve from heat rate, fuel price, start cost, and no load cost
  • Copying yesterday's offer stack forward when the market deadline is closing
  • Chasing the plant by phone and email for current derates and outage status before the deadline
  • Hand calculation of a battery charge and discharge schedule against a price curve
  • Re-keying the same offer data into the ETRM and the market interface
  • The monthly look back where someone tries to work out which units were mis-offered and why

Why it is safer

There is no direct safety benefit in offer construction and claiming one would be dishonest. The indirect mechanism is real but modest: offers that reflect the plant's actual condition produce fewer out of merit and emergency starts, and emergency starts are what generate short notice switching and after hours callouts.

Counted in units you already track:

  • Switching operations performed under time pressure for out of merit or emergency unit starts
  • Night driving hours for schedulers and plant staff called in around a late or missed offer deadline
  • Road miles driven for unplanned unit starts that a correctly priced offer would have avoided

Man-hours it gives back

Trader and scheduler hours come back at every market deadline, and the plant stops fielding a phone call about derates before each one.

HOURS AVOIDED PER YEAR = market deadlines per year x generation and storage assets offered per deadline x minutes per asset to build and enter an offer, plus offer performance reviews per year x hours per review, plus derate confirmation calls per deadline x minutes per call, minus the review and edit time a trader still spends on each drafted offer before submitting it.

The numbers we need from you to run that formula:

  • Generation and storage assets offered per deadline and minutes spent per asset today
  • Market deadlines per year, counting day ahead plus any intraday or reoffer windows
  • Hours spent per month on offer performance review and settlement look back
  • Loaded hourly rate for a senior trader, a scheduler, and a settlement analyst
  • Start cost, no load cost, and current heat rate curve for each unit, which the copilot needs anyway

Where the dollars come from

Cost driverHow it is calculated, from a rate you supply
Desk and scheduling labortrader and scheduler hours avoided x your loaded rates
Offer qualityyour dispatched megawatt hours x your own margin per megawatt hour x the share of margin you attribute to offers that carry today's fuel price, derates, and constraints instead of yesterday's, a share you set from the shadow run
Start cost recoverystarts per year x your own start and no load cost per unit x the share of starts where the offer did not recover them because the cost data was stale
Storage cycle valueyour battery's cycles per year x usable megawatt hours per cycle x the difference between the spread you realized and the spread the shadow run captured over the same period
Settlement and make whole exposureyour make whole and uplift payments per year x the share your settlements group can trace to offer construction errors

Reliability and maintenance

Reliability
This touches commitment quality more than customer reliability. Units offered at levels they can actually deliver get committed at levels they can actually deliver, which means fewer real time deviations, fewer forced derate notifications to the market operator, and fewer performance penalties. It does not change SAIDI or SAIFI.
Maintenance
Plant condition finally reaches the market position on the same day it is known, so a derate is priced into the offer instead of discovered at dispatch. Over time the record of offered versus achievable output tells the plant which units are quietly degrading, which is a maintenance signal you do not have today.

What else it moves

ComplianceEach offer carries its cost basis, which is exactly what a market monitor asks for when it reviews cost based offers or a mitigation question comes up.
WorkforceThe deadline crunch stops being the defining feature of the job, which matters on a desk where one person's absence changes the quality of every offer that day.
CustomerFor a load serving utility, better offer economics on the owned fleet reduce net supply cost, which flows through to rates.

What it costs you, stated honestly

You pay for the copilot, for the integration into your ETRM and your unit cost data, and for the plant data plumbing that keeps derates and outage status current, which is often the hardest part. Add your own trader time to run the copilot in shadow mode against submitted offers for at least one full season, because that shadow record is the only evidence your risk committee will accept.

How to build the payback case

Payback is dominated by the shadow run result on offer margin, plus desk labor. Build the case on the shadow comparison, since it is measured against your own submitted offers and settled outcomes rather than against anyone's benchmark.

These are planning models built from your unit costs, your deadlines, and your own settlement data, not vendor claims. Re-run them with the shadow season actuals before you commit to a term.
UC 11.3 Congestion and Curtailment Risk Analytics

What happens today, without this

A congestion desk analyst starts each day reading the market operator's constraint postings, shadow prices, and transmission outage schedule, and keeps a personal spreadsheet of which constraints have bound recently. They phone transmission planning to find out what outages are coming on the paths that matter, and they email curtailment warnings to renewable asset managers one at a time. When a congestion loss shows up in settlement, the same analyst writes the explanation for the risk committee weeks after the fact.

What it replaces or shrinks

  • The morning read of constraint, shadow price, and outage postings into a hand kept watchlist
  • The personal spreadsheet of historically binding constraints maintained from memory
  • Phone calls to transmission planning to find out which outages are scheduled on the paths you are exposed to
  • Manual mapping of open positions to the specific constraints that could move them
  • Curtailment warnings emailed to renewable asset managers one asset at a time
  • The after the fact congestion loss explanation written for the risk committee, which shrinks because the exposure was already flagged

Why it is safer

There is no direct field safety benefit from congestion analytics and we will not manufacture one. The honest indirect link is that a constraint that surprises the system is relieved with short notice reconfiguration, and short notice switching is where operator error risk lives.

Counted in units you already track:

  • Switching operations executed on short notice to relieve a constraint nobody anticipated
  • Energized area entries associated with unplanned reconfiguration work
  • Night driving hours for field staff supporting an emergency transmission reconfiguration

Man-hours it gives back

Analyst hours come back to the congestion desk, and transmission planning gets the same watchlist without anyone having to call them for it.

HOURS AVOIDED PER YEAR = trading days per year x analyst hours per day building the constraint watchlist, plus congestion loss post mortems per year x hours per post mortem, plus curtailment notifications per year x minutes per notification, plus planning inquiries per month x 12 x hours per inquiry answered, minus the analyst time still spent confirming each flagged constraint against the market operator's own outage schedule.

The numbers we need from you to run that formula:

  • Analyst hours per day currently spent building the watchlist and reading postings
  • Congestion loss post mortems per year and hours spent on each
  • Curtailment notifications sent per year and minutes per notification
  • Loaded hourly rate for a congestion analyst, a renewable asset manager, and a planning engineer
  • The nodes, paths, and constraints the desk tracks today and how many it would like to track

Where the dollars come from

Cost driverHow it is calculated, from a rate you supply
Congestion exposureyour megawatt hours exposed to basis between source and sink x your own realized congestion component x the share of that cost you judge was foreseeable a day ahead, a share you set, not us
Hedge qualityyour financial transmission rights portfolio notional, called FTRs, x your own auction to settlement spread x the share you attribute to better constraint selection
Curtailmentforecast curtailed megawatt hours per renewable asset x your own power purchase agreement or merchant price per megawatt hour, counting only the curtailment you could have rescheduled or repositioned around
Analyst and asset manager laborhours avoided x your loaded rates
Duplicate study effortconstraint studies that planning no longer repeats per year x hours per study x your loaded planning engineer rate

Reliability and maintenance

Reliability
This is transmission risk visibility, not customer reliability, and it should be sold as such. What it changes is when you learn that an element will bind, which gives planning the option to move a maintenance outage rather than accept the congestion cost. Any SAIDI effect is second order and depends on how often local constraints drive your interruptions, which your outage cause coding already tells you.
Maintenance
Transmission maintenance outages can be scheduled off the days a constraint would bind, which turns a congestion cost into a scheduling decision. Over a season, the record of which elements bind repeatedly is the evidence planning needs to justify an upgrade or a reconductor.

What else it moves

CompliancePositions and hedges carry a documented, dated congestion rationale, which is what the risk committee and any market monitor inquiry ask for.
EnvironmentCurtailment that gets rescheduled around is zero carbon energy that reaches the grid instead of being spilled.
WorkforceThe desk keeps its constraint knowledge in a model rather than in one analyst's spreadsheet, which matters the week that analyst is on leave.

What it costs you, stated honestly

You pay for the analytics service, for the network model and market data it consumes, for integration into your ETRM and into planning's study queue, and for analyst time to score predicted binding against actual binding for a season before the desk positions on it. Getting a current, accurate network model into the service is usually the integration item people underestimate.

How to build the payback case

Payback is normally led by curtailment you avoid and analyst hours, both of which you can audit. Congestion exposure avoided is the larger number and the softer one, so put it second in the business case, not first.

This is a planning model built from your own exposure, your realized congestion, and your rates, not a vendor claim. Re-run it against a full season of scored predictions before you commit.
Each of these opens in full on the use case page, alongside the integration plan, the data ask, the path to production, and the operator console. Open the use case library.
For Your Architects and Data Owners
Run this at your utility

What is this, exactly? It is AI software: intelligent agents and models built and delivered by SoftServe, running on NVIDIA accelerated computing. It is not a hardware appliance and it does not replace the systems you run today. It deploys in your own cloud or on your premises, connects read-only to your existing systems, and recommends; your people approve every action, starting in shadow mode until it earns trust.

A forecast feed for the trading and scheduling desk: hour-ahead and day-ahead price forecasts at the specific grid locations, called nodes, where your assets settle, with confidence ranges the desk can position against. The demo above uses synthetic data; everything below describes what the real deployment needs from your organization.

Systems it connects to

Your systemTypical productsHow we connect
Market and grid operator interfacesPJM, MISO, ERCOT, CAISO portals; OASISread-only API
Energy trading and risk (ETRM)ION Allegro, openlink Endurdatabase replica refreshed nightly
Weather and environmentNational Weather Service feeds, commercial forecast servicesread-only API
Plant control (DCS) for generationplant historian, Emerson Ovation, GE Mark VIehistorian mirror (one-way feed)
Planning and study toolsPSS/E, PowerWorld, TARAscheduled file export (CSV or CIM XML)

Data it needs from you

How it runs on your systems

Runs in your own cloud account on GPU instances; most inputs are public market data, so setup is light. Connections to your trading systems are read-only through your existing data zone; the system takes no positions and executes nothing, running first as a backtest beside the desk's current model.

Path to production

Weeks 1-3
Connect grid operator feeds and the forecast archive; public data makes this the fastest setup phase.
Weeks 4-10
Backtest 90 days of nodal forecasts at five representative nodes against the current model.
Weeks 11-12
Review accuracy improvement by node and hour type and make the go or no-go call.
Months 4-5
Security review, monitoring, and delivery of the live feed into the desk's tools, with training.
Month 5 onward
The desk starts each day with the nodal forecast in its workflow, growing coverage to the full node set.

What we need from your team

Full integration, data, and timeline detail for each use case in this scenario: UC 11.1 · UC 11.2 · UC 11.3
For Your Operators and Dispatchers
Where you will see it and how you say yes

The Approve button you just clicked in the demo above is the real workflow. This is what it looks like on the screen of the day-ahead trading desk analyst in the GridCORTEX console:

GridCORTEX ConsoleSigned in: the day-ahead trading desk analyst
Notifications
Nodal forecast: HUB_WEST HE17-19 tomorrow at $310-$425/MWh, 3.2x seasonal norm; driver is a wind lull plus heat
Daily model refresh complete; all connected feeds healthy
Recommendation
Accept tomorrow's nodal price forecast for the desk
  • 28 percent more accurate than the current model over 90 days
  • Spike driven by a modeled 2.1 GW wind drop against a 104 F peak
  • Confidence range at HUB_WEST is $310 to $425
✓ Accept forecastModifyDecline
After you approve: The forecast lands in the energy trading and risk (ETRM) system as the pricing reference for traders' own position decisions, and an audit entry records who approved it and why.
Computed from data as of 17:42:10 local; every card shows the timestamp of the data behind it.

What happens when you hit approve

Accepting writes the forecast into your ETRM as an advisory pricing reference through its API; no positions are touched. Traders make and enter their own position decisions in the ETRM; GridCORTEX never trades.

How you tell it what it cannot see

Fully automatic from connected market, weather, and grid feeds; there is nothing to enter.

Live data, not stale data

Market data refreshes every 5-minute interval and weather hourly; each curve shows the as-of timestamp of the run that produced it.

Where it lives day to day

Forecast curves live in the GridCORTEX console and the ETRM desk view; projected spikes beyond set bounds send a Teams push. The console runs in a browser beside your existing screens on day one; embedding into your own systems is a roadmap step once the read-only phase has earned trust. Approve, Modify, and Decline are all captured in an audit trail your compliance team can pull, and GridCORTEX never blocks or overrides anything in the systems you run today.

The Gap: Why Your Existing Systems Don't Already Do This

The fair question from any head of trading: "We have market data terminals, a price forecast vendor, and traders who know this footprint cold, what's new here?" Here's the honest answer.

What you own keeps doing its job

  • Market data platforms (Yes Energy/PCI-class), remain the eyes on the market; the physics engine consumes their feeds.
  • The ETRM, remains the book of record for every position and settlement.
  • Your risk policy, every limit, every approval chain, untouched. The model proposes inside the box your risk committee drew.
  • Your traders, every order is theirs. The AI forecasts and ranks; humans transact.

The gap GridCORTEX fills, above them, not instead of them

  • Statistical forecasts average away the exact hours that pay. Regression on history can't see a constraint that binds because THIS outage met THIS wind ramp. GPU-accelerated market physics (PhysicsNeMo-class) solves the network, so the $60 divergence shows up in the forecast before it shows up on the screen.
  • Congestion analytics and bidding live in different tools. The constraint call, the curtailment exposure, and the reshaped offer stack came out of one model run, the desk acted on all three before the DA deadline, not after the post-mortem.
  • The FTR auction rewards whoever models the paths best. Constraint-path valuation from the same physics engine, delivered inside the auction window with confidence bands, the desk decides, with the number in front of them.
  • Real-time surprises un-earn the day-ahead win. When the outage extension bound the constraint harder than DA cleared, the battery and the gas unit were re-ranked in minutes, the desk captured the extension instead of donating it.
  • The post-mortem finally has a receipt. Every forecast, every recommendation, and the trader's decision on it, Relay-traced. The risk committee sees why the desk was positioned, not just that it was.
Accent, don't replace: GridCORTEX reads your market feeds, outage cards, weather, and asset stack · solves the network physics, calls the spread with the constraint and hour attached, reshapes the offers, and values the paths · and your traders place every order under your risk policy. In a nodal market, seeing tomorrow first is the entire edge.
Under the Hood: What GridCORTEX Took Into Account in This Scenario

When someone asks "what did it actually calculate?", this is the list. In the simulation these factors drive the storyline; in a pilot they are computed from your market footprint and asset stack.

⚡ Market Physics

  • Network-solved nodal price forecasts: load, wind/solar, outage cards, and the transmission topology solved together, not regressed separately
  • Constraint-binding probability by hour, with the specific limiting element named (Elm–Wood River 230 kV, 17:00–21:00, 78%→94%)
  • Forecast cones that narrow as the physics converges, the desk sees confidence, not just a number

📊 Position & Offers

  • Asset-by-asset exposure to the called spread: the wind farm's basis risk, the gas unit's spike capture, the battery's arbitrage window
  • Offer-stack recommendations solved against unit economics, ramp limits, and the DA/RT two-settlement structure
  • Everything staged for trader review before the DA deadline, recommendations expire, they never auto-submit

🛤 Paths & Hedges

  • FTR/CRR path valuation from the same constraint model, delivered inside the auction window with confidence bands
  • Curtailment exposure per renewable asset under the binding scenario, the GWh at risk, priced
  • Portfolio view: how the DA position, the FTR, and the RT flexibility hedge each other across the event

🧾 The Record

  • Every forecast, recommendation, and human decision Relay-traced for the risk committee and compliance
  • Backtest-first deployment: the engine proves itself against 90 days of your settled history before a single live recommendation
  • Hard boundary: decision support only, no autonomous trading, ever

Presenter's one-liner: "A planned outage met a wind ramp, and the physics model called the $60 spread at CEDAR FLATS 36 hours before the market printed it. The desk locked the wind at day-ahead, offered the gas unit into the spike, ran the battery across the spread, and valued the constraint path before the FTR window closed. When real-time bound even harder, they captured the extension too. Every order human-placed. The model's only job was making sure they saw tomorrow first, and the P&L swing was $898K in 48 hours."

GridCORTEX Live Scenario Demo · Synthetic data throughout; all nodes, prices, and outcomes are fictional; no real RTO or participant is depicted · Decision support only, not trading advice; traders place every order · SoftServe + NVIDIA · Created by Ronnie Mauldin, NVIDIA Solutions Director, Power & Utilities, SoftServe · JUL 2026