{
  "schemaVersion": 1,
  "id": "atlas-lesson-grid-interconnection-readiness",
  "title": "Grid interconnection and deliverable electrical capacity",
  "category": "learning",
  "format": "Atlas lesson",
  "description": "Connects facility demand to location-specific network, equipment and operating readiness.",
  "canonicalUrl": "https://semiaifoundry.com/learn/atlas/grid-interconnection-readiness/",
  "publishedDate": null,
  "revisionDate": "2026-10-07",
  "author": null,
  "topic": "Data-center physical systems",
  "series": null,
  "license": null,
  "rightsUrl": "https://semiaifoundry.com/ai-atlas/licenses.html",
  "version": "1.3.0",
  "status": "Published educational lesson",
  "evidenceScope": "PJM is one regional example, not global interconnection law. The source’s inconsistent winter headline is excluded; no unverified local queue position is claimed.",
  "contentCoverage": "published-lesson",
  "artifactUrl": "https://semiaifoundry.com/ai-atlas/lesson-data/grid-interconnection-readiness-e44533ceda8a91e6.json",
  "interactiveUrl": "https://semiaifoundry.com/ai-atlas/#grid-interconnection-readiness",
  "related": [],
  "supersedes": [],
  "supersededBy": [],
  "contentHash": "1b5fe69786916234325c609dd9a58efe59dbde14e05fe4ab3aff164cf940b9ab",
  "contentLength": 5172,
  "jsonUrl": "https://semiaifoundry.com/agents/publications/atlas-lesson-grid-interconnection-readiness.json",
  "markdownUrl": "https://semiaifoundry.com/agents/publications/atlas-lesson-grid-interconnection-readiness.md",
  "sections": [
    {
      "id": "section-1",
      "title": "Grid interconnection and deliverable electrical capacity",
      "start": 45,
      "end": 438
    },
    {
      "id": "section-2",
      "title": "The mechanism",
      "start": 438,
      "end": 1670
    },
    {
      "id": "section-3",
      "title": "Science and assumptions",
      "start": 1670,
      "end": 1698
    },
    {
      "id": "section-4",
      "title": "Backup duration",
      "start": 1698,
      "end": 2480
    },
    {
      "id": "section-5",
      "title": "Engineering example",
      "start": 2480,
      "end": 2696
    },
    {
      "id": "section-6",
      "title": "Distinctions and tradeoffs",
      "start": 2696,
      "end": 3125
    },
    {
      "id": "section-7",
      "title": "Specifications and interfaces",
      "start": 3125,
      "end": 3430
    },
    {
      "id": "section-8",
      "title": "Interfaces",
      "start": 3430,
      "end": 3635
    },
    {
      "id": "section-9",
      "title": "Check your understanding",
      "start": 3635,
      "end": 3664
    },
    {
      "id": "section-10",
      "title": "A battery can deliver 20 MWh of usable energy at the load-side boundary while supplying a constant 10 MW. What is its ideal duration?",
      "start": 3664,
      "end": 4043
    },
    {
      "id": "section-11",
      "title": "Evidence and scope",
      "start": 4043,
      "end": 4398
    },
    {
      "id": "section-12",
      "title": "Sources",
      "start": 4398,
      "end": 4819
    },
    {
      "id": "section-13",
      "title": "Continue exploring",
      "start": 4819,
      "end": 4842
    },
    {
      "id": "section-14",
      "title": "Foundations",
      "start": 4842,
      "end": 5044
    },
    {
      "id": "section-15",
      "title": "Related lessons",
      "start": 5044,
      "end": 5172
    }
  ],
  "citations": [
    {
      "title": "Explore in Atlas",
      "url": "https://semiaifoundry.com/ai-atlas/#grid-interconnection-readiness"
    },
    {
      "title": "PJM’s Updated 20-Year Forecast Continues To See Significant Long-Term Load Growth",
      "url": "https://insidelines.pjm.com/pjms-updated-20-year-forecast-continues-to-see-significant-long-term-load-growth/"
    },
    {
      "title": "Published lesson data",
      "url": "https://semiaifoundry.com/ai-atlas/lesson-data/grid-interconnection-readiness-e44533ceda8a91e6.json"
    },
    {
      "title": "Atlas releases and licensing",
      "url": "https://semiaifoundry.com/ai-atlas/licenses.html"
    },
    {
      "title": "Uninterruptible power supply and backup chain",
      "url": "https://semiaifoundry.com/learn/atlas/ups/"
    },
    {
      "title": "Power supply and power shelf",
      "url": "https://semiaifoundry.com/learn/atlas/power-supply/"
    },
    {
      "title": "Commissioning a usable AI cluster",
      "url": "https://semiaifoundry.com/learn/atlas/cluster-commissioning-gates/"
    }
  ],
  "recordHash": "209c73e3a30423a1808ee32a00c24d2185074cfed3fc6e25dd3607f7611439c4",
  "contentFormat": "text/markdown",
  "content": "Data-center physical systems · Atlas v1.3.0\n\n# Grid interconnection and deliverable electrical capacity\n\nConnects facility demand to location-specific network, equipment and operating readiness.\n\n[Explore in Atlas](https://semiaifoundry.com/ai-atlas/#grid-interconnection-readiness)Manufacturing supply to commissioned hall; time-dependent model\n\nLearning objectiveSeparate proposed demand, local electrical service and backup duration.\n\n## The mechanism\n\nA site’s proposed demand must fit a real electricity network at a real location. Interconnection planning examines service arrangements and required infrastructure rather than treating a national generation total as locally available power. Substations, transformers, transmission and distribution work have their own schedules, while load forecasts assess how much proposed demand is likely to materialize. PJM’s 2026 methodology discusses screening large-load requests and adjusting forecasts. A signed capacity agreement, completed electrical equipment and permission to operate are different states. Backup energy and demand response can support operation but do not automatically replace a continuous grid connection.\n\n1.  Define the facility load boundary and the expected time-dependent demand.\n2.  Identify the required grid connection and screen the project against duplicate or speculative requests.\n3.  Track studies, service commitments and required network or substation work separately.\n4.  Verify equipment installation and operational acceptance before assigning deliverable capacity.\n5.  Model backup duration and demand-response constraints without treating them as unlimited continuous supply.\n\n## Science and assumptions\n\n### Backup duration\n\n$$\nt=\\frac{E_{\\mathrm{usable}}}{P_{\\mathrm{load}}}\n$$\n\nBackup duration is already-usable energy divided by constant load power.\n\nThis load-side energy convention differs from a stored-energy input that still needs conversion efficiency.\n\nSymbols and units\n\nt\n\nDuration supported by already-usable energy\n\nh\n\nE\\_usable\n\nEnergy usable at the declared load-side boundary\n\nMWh\n\nP\\_load\n\nConstant supported real load power\n\nMW\n\nEusable is energy deliverable at the load-side boundary in MWh, Pload is constant output power in MW, and t is hours.\n\nAssumptions Positive constant load; usable energy already accounts for conversion losses, reserves and the stated battery condition. Changing load, temperature or degradation requires a revised usable-energy estimate.\n\n## Engineering example\n\nSynthetic commissioning plan has racks arriving in March and a required transformer accepted in June. The electrical readiness gate can delay operation despite the earlier hardware delivery.\n\n## Distinctions and tradeoffs\n\nAn announced megawatt connection or generation project is not proof that a specific hall can draw that power today.\n\nA site with favorable land or fiber may need longer electrical upgrades; on-site support equipment adds cost and operational dependencies.\n\nAccelerators cannot deliver sustained workload capacity until electrical service and the corresponding facility systems are actually ready.\n\n## Specifications and interfaces\n\nPower versus energy\n\nMW capacity and MWh duration\n\nA backup device’s discharge rating does not establish how long it can support the site.\n\nGeographic boundary\n\nSpecific connection and service territory\n\nRegional generation totals cannot be assigned freely to a campus.\n\n### Interfaces\n\nFacility load request Input\n\nDemand profile, location and proposed operating date\n\nElectrical service readiness Output\n\nCompleted connection with defined capacity and operating conditions\n\n## Check your understanding\n\n### A battery can deliver 20 MWh of usable energy at the load-side boundary while supplying a constant 10 MW. What is its ideal duration?\n\n1.  Two hours\n2.  Twenty hours\n3.  An unlimited period\n\nAnswer and reasoning\n\nTwo hours\n\n20 MWh divided by 10 MW gives two hours. Conversion losses and reserved energy have already been accounted for in the stated load-side usable energy.\n\n## Evidence and scope\n\nPJM is one regional example, not global interconnection law. The source’s inconsistent winter headline is excluded; no unverified local queue position is claimed.\n\nPropagate a stated facility power budget, PUE, readiness and utilization to illustrative work; detailed conversion, battery, cooling and grid models are not included.\n\n## Sources\n\n-   [PJM’s Updated 20-Year Forecast Continues To See Significant Long-Term Load Growth](https://insidelines.pjm.com/pjms-updated-20-year-forecast-continues-to-see-significant-long-term-load-growth/)\n\n[Published lesson data](https://semiaifoundry.com/ai-atlas/lesson-data/grid-interconnection-readiness-e44533ceda8a91e6.json) · [Atlas releases and licensing](https://semiaifoundry.com/ai-atlas/licenses.html)\n\n## Continue exploring\n\n### Foundations\n\n-   [Uninterruptible power supply and backup chain](https://semiaifoundry.com/learn/atlas/ups/)\n-   [Power supply and power shelf](https://semiaifoundry.com/learn/atlas/power-supply/)\n\n### Related lessons\n\n-   [Commissioning a usable AI cluster](https://semiaifoundry.com/learn/atlas/cluster-commissioning-gates/)"
}