How can buildings coordinate solar panels, batteries and EV charging?

Solar panels, a battery and a row of electric-vehicle charge points can work together to cut peak demand, use more of the building's own generation or keep essential loads running. They cannot do all three at once, because those goals compete for the same battery capacity and the same charging hours. Coordination therefore starts with an agreed energy-management strategy: what the building is trying to achieve, in what order of priority and within which limits. Only then does it make sense to choose the software that carries the strategy out.

Coordination inside a single building is the subject here. Sharing energy across several buildings or a district raises further questions of shared connections and governance.

A directive that puts all three on site

In the EU, the recast Energy Performance of Buildings Directive requires new buildings to be designed to optimise their solar potential and obliges member states to ensure suitable solar installations on many new and some existing non-residential buildings, where technically and economically feasible. It also requires the recharging points it mandates to be capable of smart and, where appropriate, bidirectional recharging. More buildings will therefore have all three assets, and with them the problem of making them cooperate.

Rank the objectives before anything else

The US Federal Energy Management Program's guide to grid-interactive efficient buildings lists the benefits owners pursue: avoiding demand charges and peak-period prices, reducing bills or earning payments from utilities, resilience to grid outages, decarbonisation and better energy management. It describes four demand-management strategies: efficiency, load shedding, load shifting and modulation. The owner's job is to pick the objectives that matter for this building and put them in order. A battery held full for resilience cannot also be emptied every afternoon to shave peaks, and the software needs to be told which matters more.

The limits the strategy must respect

Optimisation works inside constraints, and those need recording before anyone discusses algorithms:

• the grid connection capacity and the terms of the connection agreement, including any export limit;
• the capacity of the building's own electrical distribution;
• the charging commitments made to vehicle users: who needs how much charge, and by when;
• equipment limits, such as the battery's state-of-charge range, charge and discharge rates and inverter ratings;
• any battery reserve kept back for resilience.

Ordinary grid-connected operation and backup supply are different things. A battery and a solar array do not automatically keep a building running in a power cut. Operating safely without the grid requires appropriate electrical design, protection and approvals by qualified electrical engineers.

Who decides and who enforces

A workable design separates the system that plans from the devices that protect themselves:

The interfaces between these layers should be documented, along with what happens when communication fails; normally each device falls back to a safe, standalone mode. The economics deserve the same rigour. Costs should be evaluated against the actual tariff and operating profile, including battery degradation, round-trip losses and users' charging needs, because not every shifted kilowatt-hour is worth the same.

Policy is the owner's job, not the software's

The software can enforce a policy but cannot set one. Someone has to decide whether visitors' vehicles are charged on the same terms as staff, whether any vehicles take priority, how much battery capacity is held back for resilience and whose budget pays for charging. These are questions to settle with the people affected before the system is configured, and to revisit once there is operating data to judge them by.

Questions for bidders

• Which system holds the priority order, and how is it changed when tariffs, the connection agreement or charging policy change?
• How are drivers' charging needs captured, and what happens when they cannot all be met?
• What happens on loss of communication between the supervisory system and each device, and how has that been tested?
• How is the import limit enforced if the supervisory system fails?
• What does the facilities manager see, and what can they override?
• What data is retained to show afterwards that the strategy worked as intended?

Those answers should then be proved on site. Witness tests should include an attempt to exceed the import limit, a communications failure, a manual override and a day on which the charging requirement cannot be met, and each should behave as the control philosophy says it will.