Importing a Plant's History
When a plant moves to Mirox, its measurements did not start on the day the Mirox-Agent was connected. The history import brings the years recorded by the previous monitoring system into the plant, so production graphs, performance ratios, curtailment analyses and exports reach back to the first day the plant fed in. The imported measurements sit next to the live ones in the same plant and are marked as imported, so they can be told apart at any time and removed again in one step.
Where it lives
Open the plant's Data Flow section and choose the Import tab (in the app: open a plant from your plants, then Data Flow ▸ Import). The tab is available to operators of the plant.
What you need
- An export of the previous monitoring system as one ZIP archive. Mirox recognises the archive format of supported providers on its own; you do not choose a provider, and you do not unpack or rename anything.
- A plant in Mirox whose components already exist, so the import knows where the measurements belong. Inverters, strings, feed-in meters, irradiation sensors and the control unit are matched against the components of the plant.
- A connected Mirox-Agent for the plant. The agent writes the history into the plant's data store; the Import tab shows whether it is connected before you start.
Archives may be large (several gigabytes are normal). The upload is resumable: if the connection drops or the browser is closed, pick the same file again and it continues where it stopped.
The five steps
- Upload — drop the archive onto the Import tab. The upload starts at once and the archive is verified and read on the platform. This takes a few minutes for a large export.
- Automatic mapping — every device found in the archive is matched to a component of the plant by serial number, then by name. The summary tells you in one line whether everything matched.
- Review, if something needs attention — the component groups (control unit, feed-in meters, irradiation sensors, inverters, strings) each show how many of their devices are mapped. Open a group only when you want to change something: which data columns are written, or where a single device goes.
- Start — one click hands the import to the plant's agent. Nothing needs to be configured for a normal import.
- Watch — the progress bar shows the days written, the points stored and the remaining time. The history of the import lists every day and every remark the platform recorded.
Mapped, unknown source, or left out
Every device of the archive ends up in one of three states, and the best guess is made for you:
| State | Meaning | When it is chosen automatically |
|---|---|---|
| Mapped | The measurements are written onto a component of the plant, exactly like live data. | A component with the same serial number or name exists in the plant. |
| Unknown source | The measurements are written into the plant, marked as coming from an unknown source, but not tied to a component. | The device carries data but no component of the plant matches it. Control signals, irradiance and temperature are worth having even without a component. |
| Left out | Nothing of this device is imported. | The archive holds no measurements for the device, or you chose to leave it out. |
You can change any device's state in its group: pick the component it belongs to, choose unknown source, or leave it out. Every change is saved automatically.
What is written, and what is not
- Live data is never overwritten. The import stops at the moment the plant's own monitoring in Mirox began, so imported and live measurements never overlap. You can set an earlier cutoff or narrow the day range under Import range.
- Every day is checked against the plant's live data before it is written. A day that already has live measurements is skipped.
- Data columns are chosen per component group. The defaults fit the archive; switch off what the plant should not receive.
- Strings are imported together with their inverter, including the calculated string power and energy the analysis pages read.
While it runs
The import runs on the plant's agent, day by day, at a gentle pace so the plant's live monitoring is never disturbed. While the agent runs one of its own analyses, the import pauses by itself and continues the moment that work is done; the progress bar says so. It survives restarts of the agent and interruptions of the connection: the platform keeps the position of the last day written, and the agent continues from there as soon as it is back.
- Pause stops after the current day; Resume continues from the same day.
- Stop ends the run; what was written stays until you purge it.
- The Purge data action removes every measurement the import wrote, in one step, and keeps the archive and the mapping so the import can be run again.
- Delete removes the import and its archive from the platform. It is refused while imported measurements may still exist; purge them first.
- Re-import days writes a range of days again from the same archive and mapping. Use it when a day is missing or incomplete, for example after a disturbance of the time-series store during the run: pick the first and last day, and the plant's agent writes them once more. Days that were already complete are simply written again with the same values, so a generous range does no harm. Start the range with the last complete day before the gap: the energy of a day that was only partly written is then credited to the right day.
After the import
Imported measurements appear on the plant's pages like any other data: production and period graphs, the performance ratio, curtailment analyses, exports and reports all reach back to the first imported day. Every imported series carries a mark that names it as imported, so a later purge affects only the imported data, never the live measurements.