Federated batteries for Hawaiʻi: capacity at the customer meter.

Act 266 of 2025 requires the Public Utilities Commission to establish a goal for the installation, by 2030, of new customer-sited distributed energy resources. It also directs the Commission to establish grid-services, microgrid, and community-renewable tariffs with storage riders, aggregation provisions, and device-level measurement and verification for storage.

In April 2026, the Commission opened Docket No. 2026-0084 to design a Virtual Power Plant grid-services program across Hawaiian Electric’s service territories. Its initial guidance calls for remote dispatch, performance-based compensation, and expanded access for low- and moderate-income households. That makes the central design question practical: what must a customer receive in return for making a device available to the grid?

The policy should use the term federated fleet rather than a single, centralized plant. The resources remain owned or hosted at homes and businesses, but they can respond in a coordinated way when a tariff, contract, and communications system make that service reliable. The public value is measured performance at a known time and location, not the number of devices enrolled.

The U.S. Department of Energy defines virtual power plants as aggregations of distributed energy resources—such as solar-plus-storage, controllable appliances, electric vehicles, and flexible loads—that can provide utility-scale and utility-grade services. In Hawaiʻi, those services can include evening capacity, ramp management, contingency response, and local load reduction when a feeder is constrained.

A federated battery-and-thermal fleet can reduce the need for selected centralized investments when it can deliver a contracted service with comparable confidence. It is not a substitute for all transmission, distribution protection, black-start capability, long-duration energy, or 24-hour firm supply. Planning must test each need separately rather than treating all kilowatts as interchangeable.

Thermal equipment expands the menu of flexible resources. A heat-pump water heater, ice storage, efficient cooling system, or refrigerated case can shift energy use when service temperatures remain within agreed limits. The control strategy must preserve customer comfort, food safety, and equipment warranties before it creates a grid dispatch option.

Hawaiʻi law directs the Commission to establish compensation values for the resiliency, capacity, and ancillary-service value of participating solar-plus-storage systems. That instruction is important because it moves the discussion beyond a single export price. It does not remove the need for clear contracts, independent measurement, and customer protections.

A credible VPP tariff should specify the minimum state of charge reserved for an outage, permitted dispatch windows, response time, notification rules, opt-out rights, payment calculation, equipment communications, warranty boundaries, and the consequences of a failed dispatch. Health, safety, storm, and public-safety-power-shutoff conditions require explicit customer overrides.

PMF recommends an equity adder for income-qualified households, no retail sale of device or usage data, open interoperability requirements, minimum cybersecurity controls, and public annual reporting on bill effects, dispatch performance, overrides, and feeder-level benefits. These are policy recommendations, not requirements currently set by the Commission.

Hawaiian Electric’s July 2026 integrated-grid-planning solicitation seeks renewable energy, grid-forming resources, and firm capacity through a Commission-supervised process. The utility has said that firm-generation options should be evaluated in a transparent forum rather than presumed through a sole-source arrangement.

The same discipline should apply to federated capacity. A procurement should state the required location, availability window, duration, response, telemetry, outage reserve, and customer-protection terms, then allow utility-scale storage, customer batteries, controllable thermal load, efficiency, and targeted network upgrades to compete on measured total value.

This is a planning standard, not an argument that one resource class always wins. A decentralized portfolio may be faster to deploy, may reach customers who need bill relief, and may avoid some grid constraints. It may also have higher coordination costs or be unable to meet a particular system need. The evidence should decide the mix.