An on-premises operational historian and fault-pattern analysis platform for solar and battery storage facilities. It monitors plant data, learns the signature of failures that have already happened, and watches for those signatures returning — giving a lean operations team the reach of a much larger one.
Generation portfolios are getting larger, more distributed, and more heterogeneous — solar and storage across multiple sites, states, and equipment vendors — while operating teams stay lean. You cannot staff every site, and the assets are too dispersed to run by watching screens.
The result is that most operational knowledge is held by individuals rather than systems. When a fault occurs, the diagnosis depends on whether someone on shift happens to remember the last time it happened. When staff turn over, or when a portfolio is acquired or carved out, that memory leaves with them — and the fleet starts relearning the same failures from scratch.
Predictive maintenance and retroactive analysis are the same capability pointed in two directions of time. Retroactive analysis learns a fault's signature from what already broke. Predictive maintenance watches for that signature before it breaks again. One engine, one data model, one pattern library — looking backward to learn, forward to warn.
The capture layer is deliberately the foundation. It is the piece most operators are missing, it is the layer that determines what questions can ever be asked of the data, and it is the layer that keeps the entire system outside of any control path.
Enerdyne commissions the equipment this product monitors — power conversion systems, battery racks and management units, protection relays, plant controllers, revenue meters, and the networks connecting them. The fault library is derived from field experience, not from a generic anomaly-detection template. Representative fault classes the analysis layer is built to recognize:
| Fault class | Why humans miss it |
|---|---|
| Protective device does not operate | An energy path finds unprotected auxiliary or control wiring instead of the fused power path. Equipment is damaged with no trip, no alarm, and no breaker to reset — the only evidence is a thermal rise and a power supply changing state. |
| Transfer and reconnection transients | Open-transition switching reconnects out of phase across rotating loads that are still spinning and regenerating. The event is brief, occurs during commissioning or maintenance, and leaves damage that surfaces later. |
| Component rating misapplication | Hardware rated for a different voltage or grounding scheme is installed and appears to work. The signature is a steady-state current draw inconsistent with the nameplate load. |
| Stored-energy discharge on emergency stop | An emergency stop or breaker opening induces a discharge into protection devices, requiring extensive teardown to restore. Recurs whenever the interlock sequence is wrong. |
| Cell and module imbalance drift | State-of-charge divergence accumulates slowly until racks can no longer balance. Visible far in advance in trend data; invisible in any single reading. |
| Silent instrument loss | A sensor stops reporting. Downstream calculations continue to produce plausible numbers computed from data that no longer exists. |
| Integration and polling blind spots | A monitoring path is configured but has never successfully carried traffic. The absence of alarms is read as the absence of problems. |
Every fault class above shares one property: the plant does not announce it. There is no alarm to acknowledge and no trip to investigate. They are found by comparing behaviour against what that machine, or a sister machine elsewhere in the fleet, did before — which is exactly what a retained operating history makes possible and a live dashboard does not.
Capability is delivered in stages, each of which produces value on its own and earns the trust required for the next. Trust is built by proving the engine against outcomes that are already known before anyone is asked to rely on a forward-looking claim.
| Phase | Capability | Status |
|---|---|---|
| 1 — Capture | Unidirectional read-only acquisition from site systems into a central operating history. Establishes the data foundation and the retention baseline. | Available |
| 2 — Retroactive proof | Analysis run against the customer's own historical data, identifying faults that already occurred and the precursors visible beforehand. Validates the engine against known outcomes. | Available |
| 3 — Forward monitoring | Continuous monitoring of live data for recurrence of established fault signatures across the fleet. | Roadmap |
| 4 — Advisory operator | Operator-facing guidance: what to inspect, why, and the evidence supporting it. Human decides and acts. | Roadmap |
| 5 — Supervised response | Closed-loop action under human supervision. Not offered, and not proposed until phases 1–4 have an operating record at the customer's own sites. | Not offered |
Enerdyne claims that this system detects degradation early and prioritizes what to inspect. Enerdyne does not claim to predict the date on which a component will fail. The first claim can be demonstrated and defended; the second cannot, and vendors who make it lose credibility the first time it is checked.
The following describes the capabilities and constraints of the system as designed, provided to support a customer's own security review. Enerdyne LLC does not assert that this product is certified or compliant under any standard; applicability of any regulatory framework is determined by the registered entity operating the facility.
System integration, software authorship, and configuration: United States — Enerdyne LLC, a Montana limited liability company.
Hardware: assembled from commercial off-the-shelf components. Manufacturer, model, and country of origin for each component are identified individually on the as-built bill of materials supplied with the delivered system. Enerdyne does not represent the country of origin of components it does not manufacture, and supplies component manufacturers' documentation unmodified.
Built to order and sized to the tag count, scan rate, and retention period of the target site or fleet. Final hardware selection, operating system baseline, and analysis component versions are recorded on the as-built configuration record delivered with the system.
| Part Number | Description | Scope |
|---|---|---|
EDY-VPO-100 | ENERDYNE Virtual Plant Operator — single-site license | One facility |
EDY-VPO-500 | ENERDYNE Virtual Plant Operator — fleet license | Multiple facilities, quantity per agreement |
EDY-VPO-100-R | Retroactive analysis engagement — historical data study | Per study |
EDY-VPO-100-C | Commissioning, integration, and operator handover | Per deployment |
EDY-VPO-100-S | Support and tuning term | Per term |
On-premises operational data historian and analysis platform for solar and battery energy storage facilities. Provides unidirectional read-only capture of plant telemetry into a central historian, retention of fleet operating history, and pattern-based analysis of historical equipment faults to support maintenance planning and operator decision support. Advisory output only; no supervisory control and no write path to plant control systems. Integrated in the United States by Enerdyne LLC, Montana.
A retroactive analysis engagement (EDY-VPO-100-R) requires no
installation on the control network and no forward-looking claim. Enerdyne is supplied with an
export of existing historical plant data and returns an analysis of faults that already occurred,
including the precursors visible in the data beforehand. It validates the engine against outcomes
the customer already knows, and it is the lowest-risk way to establish whether the approach
earns a permanent deployment.
The ENERDYNE Log File Server (EDY-LFS-100) establishes and proves the integrity of the data path feeding this system — useful where data quality is itself in question. The Virtual Plant Operator consumes trustworthy plant data and adds retention and fault analysis on top of it.
Enerdyne is separately developing a hardened platform layer — SynthOS and the NexusFabric card — as a future target environment for this product. That work is research and development and is not a prerequisite for deployment. The Virtual Plant Operator runs on conventional hardened server hardware today.
Configuration, pricing, and a site-specific data-flow diagram are prepared per facility.
Contact Enerdyne