Inside CAN Networks for Smarter BESS Co‑location

Most co-located solar-plus-storage projects don’t fail because of batteries or inverters. They fail because those devices can’t talk to each other reliably.

As utility-scale renewables race ahead in late 2025, the quiet bottleneck isn’t cell chemistry or panel efficiency. It’s communication. Wind, solar, EV chargers, and multi‑megawatt battery energy storage systems (BESS) are being stitched together with networks that were originally designed for cars and industrial machines. When those Controller Area Network (CAN) buses start dropping messages or clashing on IDs, your “green technology” quickly turns into an expensive, underperforming asset.

This post unpacks the core ideas behind the recent “Inside CAN Networks” webinar with HMS Networks and adds a layer the webinar only hinted at: how these communication decisions directly impact project bankability, lifetime emissions, and the real ROI of green technology.

Why communication is now a critical piece of green technology

The core point is simple: co‑located renewables and BESS only deliver their promise if the communication layer is rock solid.

Co‑location is no longer a niche idea. We’re seeing:

• Solar + BESS projects in the 10–500MWh range as standard
• Wind farms retrofitted with batteries to smooth output
• Hybrid plants that mix PV, wind, BESS, and sometimes EV charging

On paper, these plants do everything sustainability teams want:

• Store excess clean generation instead of curtailing it
• Shift energy into evening peaks when grids are dirtiest
• Provide fast frequency response and other grid services

But here’s the thing about hybrid plants: they’re complex cyber‑physical systems. You’ve got:

• Battery racks and modules from one vendor
• BMS, inverters, and PCS from others
• Site controllers, SCADA, and utility interfaces on top

Most of these building blocks speak via CAN networks somewhere in the stack. If that CAN layer is noisy, misconfigured, or insecure, you get:

• Wrong data going into your EMS/AI optimizer
• Slower control loops and missed market signals
• Nuisance trips, fault storms, and truck rolls

That’s not a minor annoyance. It directly erodes revenue, increases lifetime emissions (because the system underperforms), and destroys the business case for green technology.

Inside CAN networks: what actually goes wrong?

At a high level, CAN works well: it’s robust, simple, and widely used. But once you scale from a vehicle to a 50MW/100MWh site with thousands of nodes, the edge cases show up fast.

1. Data integrity in large BESS systems (10MWh and up)

The bigger the BESS, the more ways data can be corrupted or delayed.

Common issues include:

• Electrical noise from high‑power switching and long cable runs causing bit errors
• Improper bus termination leading to reflections and intermittent faults
• Overloaded buses where too many nodes or too much traffic push CAN toward its limits
• Latency spikes that break timing assumptions in the BMS and controllers

This matters because your energy management system (EMS) and any AI‑based optimization rely on clean, timely data:

• If temperature readings arrive late or drop out, the BMS may derate the system “just to be safe.”
• If SOC values from racks are inconsistent, dispatch algorithms can’t trust them and will operate more conservatively.
• If alarms flicker on and off due to bus errors, you get alarm fatigue and missed real issues.

Practical strategies that work:

• Design CAN segments with clear node limits and cable length budgets instead of “it’ll probably be fine.”
• Use industrial‑grade repeaters, gateways, and isolators in high‑noise zones, especially near inverters and switchgear.
• Standardize on known‑good termination and shielding practices across all vendors on site.
• Implement monitoring on the CAN layer itself (error counters, bus load, retransmissions) and feed that into your asset monitoring stack.

If you’re serious about digital twins or AI‑driven optimization of your green assets, you need this foundation. Garbage in, garbage optimizations out.

2. Secure remote access: cutting TCO without risking the plant

One of the strongest themes from the webinar is that secure VPN‑based remote access is non‑negotiable for modern BESS and hybrid plants.

Why? Because sending engineers to site for every firmware tweak, diagnostic check, or configuration change is slow, expensive, and emissions‑heavy. For fleets spread across regions, it’s also impossible to scale.

When done well, remote access over hardened VPN:

• Slashes truck rolls, cutting O&M costs and associated CO₂
• Enables faster incident response when a BMS or inverter misbehaves
• Makes OTA (over‑the‑air) updates realistic for dozens or hundreds of sites
• Supports centralized expertise – your best engineers can triage issues everywhere

But “remote access” can mean anything from a password‑protected web portal to a security nightmare.

What good looks like:

• VPN tunneling via hardened industrial gateways
• Role‑based access control with MFA
• Session logging for compliance and traceability
• Network segmentation: access only the devices needed, not the whole plant

From a green technology perspective, this isn’t just IT hygiene. Remote access is a key enabler for:

• Rolling out new optimization algorithms across a fleet
• Fine‑tuning charge/discharge strategies as markets change
• Updating cybersecurity controls without site visits

In short, secure connectivity is part of the sustainability story because it keeps assets efficient over their 15‑20 year lifetimes.

3. Eliminating CAN ID conflicts, especially with second‑life batteries

As the industry pushes for more circular, low‑carbon solutions, second‑life batteries are showing up in commercial and utility‑scale BESS. That’s great for sustainability—and brutal for CAN configuration.

Why it hurts: many devices, especially legacy modules and BMS units, ship with fixed or overlapping CAN identifiers. In a car that’s fine. In a multi‑rack, multi‑vendor BESS, it’s a mess.

Symptoms:

• Two devices respond to the same ID
• Messages from one rack overwrite another’s
• The site controller “sees” phantom devices or garbled values

You can’t optimize or even safely operate a hybrid plant if you’re not sure which rack a voltage reading came from.

How teams are solving this:

• Using CAN gateways that can remap IDs on the fly, creating a clean logical namespace
• Establishing a site‑wide CAN ID plan early in design, not during commissioning
• Standardizing interface profiles when procuring second‑life modules (“here’s how you must talk on the bus if you want into our ecosystem”)

My view: if you’re planning to use second‑life batteries but haven’t budgeted for robust CAN management, you’re underestimating both cost and complexity.

Where AI and advanced software fit into CAN‑based renewables

AI shows up twice in this story: above the CAN layer (for optimization) and around it (for diagnostics and reliability).

AI on top of clean CAN data

Once your CAN network is delivering reliable, time‑synchronized data, AI becomes genuinely useful:

• Forecast‑driven dispatch: algorithms combine weather, price, and grid signals with real‑time battery health data.
• Degradation‑aware control: EMS systems adjust setpoints to extend life based on actual cell behavior.
• Co‑optimization of multiple assets (PV, wind, BESS, EV chargers) to hit both financial and carbon targets.

No amount of clever modeling compensates for bad or missing data at the device level. Stable CAN networks are what make AI‑powered green technology more than a slide in a pitch deck.

AI for communication health

A growing number of operators are also using analytics to monitor the communication fabric itself:

• Detecting abnormal error rates or bus load patterns before failures
• Flagging nodes that regularly cause retransmissions
• Correlating CAN issues with environmental data (temperature, humidity, switching events)

Treat your communication network as an asset, not an afterthought. If you want true “smart” grids and smart plants, you start by making the nervous system observable and predictable.

Designing CAN networks for future‑ready hybrid plants

If you’re planning or upgrading co‑located renewables and BESS, here’s a practical blueprint that reflects the webinar insights plus what I’ve seen on real projects.

1. Start early: include CAN and connectivity in concept design

Too many projects leave communication details to late‑stage EPC or even commissioning. That’s how you end up with:

• Vendor‑specific islands that don’t talk cleanly to each other
• Ad‑hoc gateways and protocol bridges that are impossible to maintain
• Hidden constraints that block later software upgrades

From day one, define:

• Which networks you’ll use where (CAN, Modbus, Ethernet, etc.)
• How device counts and distances map to bus segments
• Naming and ID conventions for CAN across vendors

2. Treat CAN gateways as strategic, not just “adapters”

Modern CAN gateways, like those marketed by HMS Networks, aren’t just plug adapters. They’re:

• Protocol translators (CAN to Ethernet/IP, OPC UA, MQTT, etc.)
• Security enforcers (firewalls, VPN endpoints)
• Abstraction layers that shield your software stack from vendor quirks

A well‑chosen gateway strategy makes it easier to:

• Swap out hardware vendors in future without rewriting software
• Integrate new assets (like EV chargers or thermal storage) into existing plants
• Standardize fleet‑wide monitoring and control architectures

3. Build for remote operations from day one

If your design still assumes “we’ll just go onsite when something breaks,” you’re designing for 2010, not 2030.

Remote‑friendly design means:

• Embedded secure access points at each plant
• Clear separation between OT, IT, and external access zones
• Ability to update firmware and configurations via VPN without manual fieldwork

This approach doesn’t just cut O&M costs. It also supports greener operations by minimizing travel and enabling more precise, real‑time control that squeezes more clean energy value out of every kilowatt‑hour stored.

Why this matters for the future of green technology

If you care about green technology as more than marketing, you have to care about networks like CAN.

Robust communication in co‑located renewables and BESS:

• Increases real, delivered clean energy by reducing curtailment and downtime
• Improves asset life through smarter, data‑driven operation
• Enables AI and advanced analytics to actually function as promised
• Supports circularity by making second‑life batteries practical at scale

Most companies focus their innovation stories on shiny hardware or AI algorithms. The reality is more grounded: the plants that consistently hit their performance and sustainability targets are the ones where the “boring” communication layer was designed with care.

If your team is wrestling with CAN bus errors, remote access headaches, or ID conflicts in co‑located projects, that’s not just an engineering nuisance. It’s a strategic problem that directly touches revenue, risk, and your environmental impact.

The better way is clear: treat communication as a first‑class design concern. Get your CAN networks stable, observable, and secure. Then let your optimization software and AI do what they’re good at—turning well‑behaved data into cleaner, more profitable energy.