Source-linked AI summary
EnergyNet in Practice: Long Grid, Short Grids and Mobility-Based Energy Peering
Jonas Birgersson, Max Collins, Marc A. Weiss, Jimmy Chen, Daniel Kammen, Tomas Kåberger, Mark Z. Jacobson, Newsha K. Ajami, Franklin Carrero-Martínez, Michael Menser, Henrik Ny, Lou Riordan, Jill G. Ferguson
TL;DR
The paper asks how much upstream network investment remains necessary when local resources and mobile energy support are coordinated, and proposes Short Grids as the deployment unit. It evaluates the architecture in modelled cases, where reserve levels shape critical-service outcomes; field measurement remains future work.
Problem
The paper asks which network investments remain necessary when local generation, storage, priority-based operation and mobile support are designed to work together.
Source: §1 · p. 1Method
The paper develops a Short Grid deployment model and evaluates coordinated operation in modelled cases at neighbourhood, paired-site and district scales.
Source: §9 · p. 10Results
About 30% reserve enables full critical service in the deterministic case; at that reserve, summer service was full in 1,000 trials and winter median sustained critical fraction was about 92%.
Source: §5.1 · p. 6Takeaways & Limitations
The paper’s design rule is to build coordinated local capability first and reinforce the Long Grid only for what remains.
Source: §9 · p. 10The Lund configuration is under construction, and field data will test the simulated results against measurement.
Source: §5 · p. 6, §9 · p. 10
Abstract
from arXiv · showhide
Electrification is planned as a network problem: more demand, more upstream capacity, local resources waiting for reinforcement. We argue that a material part of that bill, in money and time, follows from architecture, not electrification itself, and propose the Short Grid as the deployment unit: a locally controlled building, block or campus whose demand, generation, storage and local market operate behind Energy Routers enforcing an approved connection envelope to the Long Grid. Because the envelope governs exchange, not installed capacity, local resources can grow and many sites can build in parallel at the pace of PV and EV installation, not grid reinforcement. PV and EV batteries dominate today, but the architecture is general: any generation or storage, fixed or mobile, fits behind the same boundary and interfaces. EnergyNet Operators coordinate exchange among Short Grids and with the Long Grid; protection stays local; Vehicle-to-EnergyNet (V2EN) and Mobility-Based Energy Peering (MEP) make mobility-financed batteries accessible within and between them; priority-based operation protects critical services in shortage. We test it in models, not yet the field: a ten-building neighbourhood in Lund, Sweden; a bus-linked microgrid pair at Stanford and Half Moon Bay; and a proposed 20-site network in Rockaway, New York. Households and services are the largest demand category in Sweden, Denmark, Norway, California and New York (484 TWh/yr); halving their grid imports would be about 242 TWh. Europe's network investment to 2040 is estimated above EUR 1.2 trillion; modelling shows it depends on system design. Negagrid value, following the negawatt, counts network investment avoided, downsized or deferred, net of local substitute costs. "Don't pay twice": build coordinated local capability first; reinforce the Long Grid for what remains.
1. From network architecture to repeatable implementation
The paper develops EnergyNet’s architecture into a repeatable deployment model for locally controlled Short Grids that coordinate resources behind controlled boundaries. It links this model to vehicle-based energy services, evaluates its potential in cases, and frames investment around network costs avoided, downsized or deferred.
1. From network architecture to repeatable implementation
- The paper asks which network investments remain necessary when local generation, storage, priority-based operation and mobile support work together.Its “don’t pay twice” proposition compares additional network and dedicated-storage investment with the full cost of local capability.Source: §1 · p. 1
- The paper translates an architecture of Energy Routers, local and wide-area networks, the Energy Protocol and operator-scale management into independently deployable Short Grids.Its proposed domains cooperate to support electrification, critical services and reduced infrastructure requirements.Source: §1 · p. 1, §1.1 · p. 1
1.1 Relationship to adjacent work
- The paper’s related-work framing connects independently controlled, asynchronously interconnected subgrids with a deployment and evaluation model spanning local domains, operators, vehicle interfaces and regional adoption.It uses the companion framework for computational examples and a practitioner account for deployment rationale.Source: §1.1 · p. 1
1.2 Three contributions
- Short Grids provide a repeatable deployment domain for autonomous local resources behind an enforceable boundary, coordinated through interoperable interfaces.The paper presents V2EN and MEP as a service model for mobility-financed batteries, and assesses the approach through practical-operation cases.Source: §1.2 · p. 2
1.3 The argument in brief
- The argument links buildings’ electricity demand and growth to local generation, storage and charging, then to modelled cases, scale estimates and negagrid accounting.Its proposed design rule is to build coordinated local capability first and reinforce the Long Grid for what remains.Source: §1.3 · p. 2
1.4 Terms used in this paper
- The paper defines Short Grid as a locally controlled electrical domain and negagrid value as network investment avoided, downsized or deferred, net of local substitution costs.Its terminology also distinguishes the existing Long Grid from EnergyNet’s wide-area coordination network.Source: §1.4 · p. 2
2. The Short Grid as a deployable electrical domain
The Short Grid is a locally coordinated electrical domain that manages resources behind a controlled Long Grid boundary. The design combines local autonomy, priority-based service and parallel deployment while respecting connection and equipment constraints.
2. The Short Grid as a deployable electrical domain
- A Short Grid coordinates loads, generation, storage and controllable connections behind an Energy Router boundary, with local resources represented through a common Energy Protocol abstraction.The architecture is technology-agnostic, accommodating stationary or mobile generation and storage while the connection envelope governs external exchange rather than installed local capacity.Source: §2 · p. 2, §2.1 · p. 3
2.1 The envelope governs exchange, not installed capacity
- The approved envelope governs import, export, ramp-rate, power-quality and protection requirements, while local cable and converter ratings and shared-feeder limits still apply.Installed resource capacity is not itself the external exchange limit.Source: §2.1 · p. 3
2.2 Autonomy, operators and the local market
- EnergyNet Operators coordinate resources and exchanges, while owners retain asset-policy authority and local controllers enforce protection, reserves and service priorities.Local control remains enforceable if wide-area communications or an operator platform fails; local exchange is subject to ownership, permission, priority and compensation constraints.Source: §2.2 · p. 3, Figure 1 · p. 3
2.3 Priorities preserve essential service when full service is impossible
- When resources cannot meet all demand, local priority-based operation maintains critical demand whenever physically feasible and reduces or defers flexible demand.Priorities require independently controllable loads and can preserve stored energy for critical services; V2EN and MEP can extend the service period.Source: §2.3 · p. 3
2.4 The office LAN precedent: build locally, interconnect at the boundary
- The office-LAN analogy separates local network construction and traffic from upstream connectivity, while recognizing that remote and cloud-dependent services still require external access.The paper applies this boundary principle to local generation, storage and charging; local service during upstream disruption depends on available local equipment, power and services.Source: §2.4 · p. 4
2.5 Deployment at the pace of local installation
- Within an approved envelope, local-resource deployment can proceed with equipment installation rather than await upstream reinforcement, allowing sites to build in parallel.The figure contrasts serial upstream-first commissioning with repeatable local work; engineering and approval still apply, and arrow lengths are not time estimates.Source: §2.5 · p. 4, Figure 2 · p. 4
3. V2EN and Mobility-Based Energy Peering
V2EN and MEP make compatible vehicle batteries available as local and inter-domain energy resources while retaining mobility as their primary service. The proposed service abstraction coordinates exchange under stated operating conditions, and the paper distinguishes accessible capacity from merely installed capacity.
3.1 One service abstraction, different physical devices
- The proposed V2EN service profile lets a neighbouring router and vehicle express capabilities and requirements through a common resource abstraction, with transfer beginning when terms overlap and receiving intake is positive.Energy Protocol provides some functions while complementary open standards support others; Table 1 combines execution behaviour with service-level responsibilities, not all of which are native protocol fields.Source: §3.1 · p. 4, Table 1 · p. 5
3.2 Storage that is accessible, not merely installed
- Vehicle-battery value depends on connection, location, power, owner consent and energy above protected reserves, not gross fleet capacity.In Swedish measurements of 334 EVs, 70% of charging energy was supplied at home and the most-frequent parking duration was 14 hours; parking alone does not establish connection or permission to discharge.Source: §3.2 · p. 5
3.3 MEP adds a time-dependent support path
- MEP transfers stored energy between Short Grids through a delayed, capacity-limited vehicle journey rather than a continuous electrical line.The paper distinguishes opportunistic transfers on unchanged journeys from scheduled support visits that preserve ordinary vehicle duty while adding travel; both can replenish local reserves.Source: §3.3 · p. 5
4. Evaluation: deployment, operation and maintained service
The evaluation separates operating value from deployment value across three modeled settings, with results drawn from simulations and a scenario summary rather than field data. It reports lower imports and peak demand in Lund, mobile support between Stanford and Half Moon Bay, and emergency supply in Rockaway.
4. Evaluation: deployment, operation and maintained service
- The evaluation compares dispatch with fixed demand, weather, mobility requirements and installed resources, and deployment with the same connection, commissioning date and required services but potentially different deployable portfolios.Source: §4 · p. 5
- The optimization is a feasibility benchmark, not a reproduction of distributed Energy Protocol execution or converter dynamics; the cited calculations were not independently rerun here.Source: §4 · p. 5
- Lund shows about 60% lower annual imports and 50% lower peak import.Source: §4 · p. 6
- The Stanford Farm–Half Moon Bay case delivered about 21 kWh over three bus visits and replenished the receiving reserve.The bus minimum was 157 kWh above the 70 kWh mobility reserve.Source: §4 · p. 6
- In Rockaway, stationary supply was 49.1 MWh of a 129.4 MWh critical requirement; the 86.4 MWh emergency target was met with 700 EVs and 30 buses.Source: §4 · p. 6
- The five-system 50% import-reduction illustration is conditional arithmetic, not a forecast.Source: §4 · p. 6
5. Lund: deployable capacity, grid relief and fault tolerance
The Lund model tests whether a ten-building neighbourhood can add local resources within existing connections and localize service shortfalls when links fail. Its annual import and peak reductions are project estimates, while field measurements remain forthcoming.
5. Lund: deployable capacity, grid relief and fault tolerance
- The Lund configuration was under construction, and measured results were to be reported separately.Physical disturbance containment still requires boundary and protection validation.Source: §5 · p. 6, §5.1 · p. 6
5.1 Configuration and results already available
- The model represents ten buildings and 278 homes with 650 MWh annual demand, 500 kWp PV, and 1 MWh distributed battery storage.It uses a ten-router closed ring and two Long Grid interfaces.Source: §5.1 · p. 6
- All demand was served when one grid interface or one ring link was lost for 18 hours.The router remained intact in the grid-interface case.Source: §5.1 · p. 6
- 65% of island critical load was sustained with 10% protected reserve after two links failed and R4 and R5 were isolated for 18 hours.The other eight routers were fully served.Source: §5.1 · p. 6
- About 30% reserve enabled full critical service in the deterministic case; at that reserve, all 1,000 summer trials had full service and winter median sustained critical fraction was about 92%.The winter figure is critical load maintained during the imposed partition, not annual reliability.Source: §5.1 · p. 6
5.2 Annual imports and peak load: project estimates
- The project estimates approximately 60% lower imports and about 50% lower peak demand, extrapolated from seasonal dispatch and comparable Swedish district studies.These estimates are distinct from the representative 72-hour simulations; the full-year comparison and supporting peak series are forthcoming.Source: §5.2 · p. 6
5.3 From pilot to programme: deployment sequence and regulatory basis
- Deployment proceeds by settling the existing connection agreement, commissioning routers and local resources, onboarding an operator, then adding buildings in parallel within the envelope.The evidence sequence is simulation first, field comparison against Section 4 metrics next, and scaling after verification.Source: §5.3 · p. 6
- Swedish exemptions permit qualifying internal networks, but where energy-sharing rules are not transposed, transposition remains the binding constraint.The connection to the concession grid remains governed by the connection agreement.Source: §5.3 · p. 7
6. Stanford Farm–Half Moon Bay: a mobile support path
The model tests whether a campus bus can support a remote site before its reserve is exhausted while completing its ordinary duty. In the reported low-PV case, bus visits replenish the receiving site's battery, but the model retains depot charging.
6.1 What the reported case establishes
- The model links two separately located energy systems without a direct electrical line, with local reserves carrying the receiving site until the bus arrives.Half Moon Bay is already off-grid in the model, and depot charging remains available; the case does not establish service after simultaneous loss of all charging supplies.Source: §6 · p. 7, §6.1 · p. 7
6. Stanford Farm–Half Moon Bay: a mobile support path
- About 21 kWh was delivered over three visits while the bus retained at least about 157 kWh, above its 70 kWh mobility reserve.Without support, the receiving battery approached depletion; with support, its reserve was replenished.Source: §6 · p. 7
- The receiving site had PV output at 20% of the clear-day profile, and forecasts triggered a preventive visit after the bus completed its normal campus duty.The bus returned to the depot and recharged overnight.Source: §6 · p. 7
7. SAER in Rockaway: shared resources and mobile replenishment
SAER proposes an interconnected Rockaway network that shares local energy, prioritises essential services and replenishes reserves with mobile storage. In a 48-hour winter outage model, the mobile configuration meets the reduced emergency-service energy target after stationary supply alone falls short.
7. SAER in Rockaway: shared resources and mobile replenishment
- The proposed Rockaway network links 20 Short Grids through 83 Energy Routers and 28 inter-site connections, allowing sites to share energy rather than remain isolated.The internal symbols show differing site configurations; the links represent inter-site connections.Source: §7 · p. 7, Figure 3 · p. 8
7.1 Share, prioritise and reinforce
- SAER combines stationary sharing, priority-based demand reduction and mobile energy replenishment to support services during a prolonged regional outage.Fixed links distribute available energy, priority operation protects essential services, and MEP brings energy across the affected-area boundary.Source: §7.1 · p. 7, Figure 4 · p. 8
7.2 A 48-hour winter outage
- With 700 EVs and 30 buses making one recharge rotation, the model reports 86.4 MWh supplied against the 86.4 MWh emergency-service target.Stationary supply alone provides 49.0 MWh, or 56.7% of that reduced target; the original critical requirement is 129.4 MWh.Source: §7.2 · p. 7, Figure 5 · p. 8
8. The built environment: global scale, electrification and “don’t pay twice”
The section frames buildings as a large and growing electrification base, then considers how coordinated local capability could reduce grid dependence and network investment. Its scale examples are conditional, and negagrid value is assessed net of local substitute costs.
8.1 The largest demand category, everywhere it is measured
- Households and services are the largest electricity-demand category in all five reference systems, ranging from 51% in Norway to 87% in New York.Buildings account for more than half of global electricity consumption and are projected to represent 49% of additional global demand between 2025 and 2030.Source: §8.1 · p. 9
8.2 A conditional illustration of grid relief
- 121.2 TWh/year corresponds to halving half of the five-system demand proxy; halving the entire proxy corresponds to 242.3 TWh/year.These are conditional scale illustrations, not regional forecasts; achievable reductions depend on eligible sites, local generation, demand profiles, storage and adoption.Source: §8.2 · p. 9
8.3 A decarbonisation opportunity comparable in scale to major industries
- Building operations emit 9.9 GtCO₂ including supplied electricity and heat, while direct emissions are approximately 2.8 GtCO₂; grid-import reductions are not equivalent to emissions reductions.The emissions outcome requires separate assessment of displaced fuels, electricity emissions intensity and timing, conversion losses and added equipment impacts.Source: §8.3 · p. 9
8.4 Turning local resources into deployable capability
- Short Grids combine demand, PV, stationary storage and compatible vehicle batteries behind a controlled local boundary, potentially enabling deployment within existing connection envelopes.Where local coordination resolves the binding constraint, deployment can follow local equipment installation rather than upstream reinforcement; usable capacity may also complement industrial decarbonisation.Source: §8.4 · p. 10
8.5 From negawatts to negagrid value: part of the network bill is a design choice
- Net negagrid value is the present-value difference between network-investment pathways with and without Short Grids, less incremental local costs for substitute service.The European Commission estimates €730 billion for distribution and €477 billion for transmission development by 2040; other modelling identifies potential reductions or avoided upgrades, not EnergyNet-specific savings.Source: §8.5 · p. 10
9. Conclusion
The conclusion argues that coordinated local capability can do some of the work otherwise assigned to network expansion, with the investment case measured against equivalent services and local costs. The cases remain modelled outcomes; field verification is still pending.
9. Conclusion
- Buildings account for more than half of global electricity use, while conventional European distribution and transmission development is estimated to exceed €1.2 trillion by 2040.The five reference systems together represent about 484 TWh annually in household and service demand; halving Long Grid imports across that base is an illustration, not a forecast.Source: §9 · p. 10
- Short Grids deploy in parallel within existing connection envelopes, while operators coordinate exchange, mobility-based peering and priority-based operation across domains.Negagrid value counts network investment avoided, downsized or deferred, net of local costs required to provide substitute service.Source: §9 · p. 10
Data and code availability
- The reported case outcomes are simulated, not field data; Lund baseline and post-commissioning measurements of Long Grid imports and coincident peak are still to be reported.Scaling follows verification, and the data and code for modelled cases are held by their authors pending public release arrangements.Source: §9 · p. 10, Data and code availability · p. 11
Use of AI-assisted tools
- AI tools assisted source checking, editing, structural review and conceptual artwork, while the human authors retain responsibility for technical accuracy and final approval.Source: Use of AI-assisted tools · p. 11