Component States

The Digital Twin evaluates every component of your plant and assigns it a clear operational state, so you can tell at a glance whether a part is healthy, underperforming, offline, or simply not reporting data. These states turn raw telemetry into actionable insight about each component's condition.

State Overview

Component states are determined in two ways:

Measured: Derived directly from component telemetry data
Inferred: Deduced from parent component analysis when direct data is unavailable

Available States

Performance States

State	Description	Source
Producing Normal	Component production is within the expected range, indicating healthy operation	Measured
Degraded Performance	Production is significantly below the expected threshold, indicating a performance issue that requires investigation	Measured
Overproducing	Production exceeds the expected threshold. This typically indicates a measurement error or configuration issue rather than actual overperformance	Measured

Inference States

State	Description	Source
Suspected Not Producing	The parent component shows a production deficit that matches this component's expected power contribution	Inferred
Inferred Producing	The parent component is producing normally despite missing data from this child component, suggesting the component is likely operational	Inferred
Inferred Degraded	The parent component shows a partial deficit that may be attributed to this component's reduced performance	Inferred
Inference Uncertain	Available evidence from the parent component is insufficient to reliably determine this component's status	Inferred

Data Quality States

State	Description	Source
No Measured Data	No timeseries data is available for this component. This could indicate a communication issue or that monitoring has not been configured	Measured
Inactive Component	The component consistently reports zero or -1, indicating the endpoint is not in use or the component is intentionally disabled	Measured
Logger Stuck	The data logger is reporting a constant non-zero value, indicating a hardware or communication malfunction	Measured
Measurement Conflict	The measured value significantly conflicts with the aggregated values from child components, suggesting a measurement or configuration error	Measured

Service States

Beside the states above, the Health column on the analysis page can show a blue service state: the component is producing normally, but something needs a visit.

These do not come from the Digital Twin's per-slot analysis at all. They are the open component-health events an inverter is carrying, shown in the same Health column — which is why they carry no Measured or Inferred source: the evidence behind each one is on the event itself.

State	Description
Device fault	The inverter itself reports a fault, in a function Mirox has no measurement of its own for. The event carries how often it was raised and the wording the device used
Safety alarm	The inverter reports an arc fault or a residual-current condition about itself. Mirox has no measurement of its own for these either, and somebody has to walk up to the machine to clear one
Elevated temperature, Insulation weakening, Phase imbalance, Efficiency declining	A wear pattern Mirox measured itself over days or months

On the station row these are no longer summed into one glyph. An inverter carrying exactly one open service item shows that item's own icon — temperature, insulation, phase, efficiency, device fault or safety alarm; a measurement conflict keeps its familiar orange gauge. An inverter carrying two or more shows a single blue wrench, and the wrench's number is the count of those multi-issue inverters. An outage never joins them: an inverter that is running hot and dark shows the temperature icon and the outage icon, and one that is hot, imbalanced and dark shows the wrench and the outage icon. Whichever icon you click — a single finding, the wrench, the orange gauge or one of the production icons — the same Open issues dialog opens: every inverter of the station that has anything open at all, each with its full list of open events, which is why that list is longer than the number on the icon you clicked. The icon only decides the tooltip you saw. See Inverter Events for what opens and closes each one.

The Health column is not limited to one state either. On the analysis inverter tab every open condition is stacked, so an inverter that is dark and running hot shows both, the most serious one first — with the same rule for service items: one of them is drawn as itself, several collapse into a single wrench whose tooltip names each finding. The row's colour follows the most serious condition of the stack. When an inverter carries two or more of them, that stacked entry reads Service items (N), and clicking it opens the same kind of list for that one inverter — every open event on a line of its own — from which you pick the event to open.

Degraded Performance: Actual performance problem
Suspected Not Producing: Likely outage
Inferred Degraded: Possible performance issue
Measurement Conflict: Data quality problem

Configuration or Hardware Issues

These states typically indicate technical problems rather than operational issues:

Overproducing: Check measurement calibration and configuration
Logger Stuck: Check data logger hardware and communications
Inactive Component: Verify component is expected to be active
No Measured Data: Check communication and monitoring configuration

Insufficient Information

Inference Uncertain: More data is needed for reliable status determination

Inverter Events — the events behind each inverter's health and service states
Digital Twin — the monitoring engine that detects and assigns these component states
Loss Detection — how non-producing states translate into quantified, confidence-rated energy loss
Efficiency Detection — performance-ratio analysis behind the degraded and overproducing states

Technical Implementation

For the engineering detail behind component evaluation, see the Digital Twin Architecture.