MiroxMirox
  • Platform

    • Philosophy
    • Platform Overview
    • Platform Resources
  • Mirox-Cloud

    • Cloud Overview
    • Connected Microservices
  • Mirox-Agent

    • Agent Overview
    • Deployment Options
    • Data Scraper
    • Digital Twin
  • Technical Details

    • Metric Collection
  • Information

    • Supported Plants
  • Plant Types

    • Solar Plants
    • Wind Plants
    • Battery Storage
    • Alarm System
  • Monitoring & Visualization

    • Real-time Monitoring
    • Digital Twin
    • Component States
    • Inverter Status Codes
    • Inverter Events
    • Loss Detection
    • Power Limits & Curtailment
    • Efficiency Detection
    • KPI Dashboard
  • Data Management

    • Events
    • Tickets
    • Forecasts
    • Reports
  • Integration & Sharing

    • Cooperations
    • API Tokens
    • VPN
    • Proxy
  • AI

    • AI Assistant & Wizards
    • Agentic Access (MCP)
  • Billing

    • Market & Tariffs
    • Accounting & Billing
  • Collaboration

    • Invitations
  • Security

    • Authentication
    • Account Lockout
    • Permission System
    • Network Segmentation
    • Cooperation Restrictions
    • Access Audit Logging
    • Activity & Audit Trail
  • Nodes

    • mrxnode
  • Application

    • Door Control
    • Generic Relay
  • Edge Cluster

    • Orchestration
  • Getting Started

    • Onboarding
    • Setup
  • Personal

    • Using the VPN
    • Using the Proxy
    • Two-Factor Authentication
    • Sessions
    • API Tokens
    • Notifications
    • Connect Microsoft Teams
  • Per Park

    • Contacts
    • Network Devices
    • Data Loggers
    • Components
    • Direct VPN (per Agent)
    • Data Volume
    • History Import
  • Organization

    • Member Permissions
    • Cooperations
    • File Storage
    • VPN Services
  • Data Export

    • Export Metric API
    • MiroxQL Query Language
    • External Report Generation
    • Grafana
    • API Overview
  • Support

    • Request an Integration
  • mrxnode

    • Overview
    • How-To Guide
    • Container Deployment
    • Command Cheatsheet
    • Troubleshooting
  • Reporting

    • External Report Generator
    • Raw Data Export for Excel
  • Remote Access
  • AI in Mirox
  • History Import
  • English
  • Deutsch
  • Español
  • Français
  • Português
  • Italiano
  • English
  • Platform

    • Philosophy
    • Platform Overview
    • Platform Resources
  • Mirox-Cloud

    • Cloud Overview
    • Connected Microservices
  • Mirox-Agent

    • Agent Overview
    • Deployment Options
    • Data Scraper
    • Digital Twin
  • Technical Details

    • Metric Collection
  • Information

    • Supported Plants
  • Plant Types

    • Solar Plants
    • Wind Plants
    • Battery Storage
    • Alarm System
  • Monitoring & Visualization

    • Real-time Monitoring
    • Digital Twin
    • Component States
    • Inverter Status Codes
    • Inverter Events
    • Loss Detection
    • Power Limits & Curtailment
    • Efficiency Detection
    • KPI Dashboard
  • Data Management

    • Events
    • Tickets
    • Forecasts
    • Reports
  • Integration & Sharing

    • Cooperations
    • API Tokens
    • VPN
    • Proxy
  • AI

    • AI Assistant & Wizards
    • Agentic Access (MCP)
  • Billing

    • Market & Tariffs
    • Accounting & Billing
  • Collaboration

    • Invitations
  • Security

    • Authentication
    • Account Lockout
    • Permission System
    • Network Segmentation
    • Cooperation Restrictions
    • Access Audit Logging
    • Activity & Audit Trail
  • Nodes

    • mrxnode
  • Application

    • Door Control
    • Generic Relay
  • Edge Cluster

    • Orchestration
  • Getting Started

    • Onboarding
    • Setup
  • Personal

    • Using the VPN
    • Using the Proxy
    • Two-Factor Authentication
    • Sessions
    • API Tokens
    • Notifications
    • Connect Microsoft Teams
  • Per Park

    • Contacts
    • Network Devices
    • Data Loggers
    • Components
    • Direct VPN (per Agent)
    • Data Volume
    • History Import
  • Organization

    • Member Permissions
    • Cooperations
    • File Storage
    • VPN Services
  • Data Export

    • Export Metric API
    • MiroxQL Query Language
    • External Report Generation
    • Grafana
    • API Overview
  • Support

    • Request an Integration
  • mrxnode

    • Overview
    • How-To Guide
    • Container Deployment
    • Command Cheatsheet
    • Troubleshooting
  • Reporting

    • External Report Generator
    • Raw Data Export for Excel
  • Remote Access
  • AI in Mirox
  • History Import
  • English
  • Deutsch
  • Español
  • Français
  • Português
  • Italiano
  • English
  • Getting Started

    • Onboarding
    • Setup
  • Personal

    • Using the VPN
    • Using the Proxy
    • Setting Up Two-Factor Authentication
    • Managing Your Sessions
    • API Tokens
    • Notifications
    • Connect Microsoft Teams
  • Per Park

    • Managing Plant Contacts
    • Managing Network Devices
    • Configuring Data Loggers
    • Configuring Components
    • Configuring VPN Servers per Agent (Direct VPN)
    • Data Volume per Plant
    • Importing a Plant's History
  • Organization

    • Managing Member Permissions
    • Creating Cooperations
    • Using File Storage
    • Organization VPN Services
  • Data Export

    • Export Metric API
    • MiroxQL Query Language
    • External Report Generation
    • Using Grafana as an External Read Platform
    • API Overview
  • Support

    • Request an Integration
  • mrxnode

    • mrxnode Overview
    • mrxnode How-To Guide
    • Container Deployment
    • mrxnode Command Cheatsheet
    • Troubleshooting

Configuring Components

The Components page is where you review and refine the physical make-up of a plant — its inverters, combiner boxes, strings, irradiation sensors, and feed-in meters. You do not build this inventory by hand: the platform generates it automatically from what the plant reports. Getting the configuration right is what turns raw measurements into a clear-sky simulation, a performance ratio, loss detection, and string-level analysis. This guide walks you through each task and explains what your edits actually drive.

Components create themselves — with two exceptions

Every inverter, combiner box, and string the data loggers report is discovered by the Mirox-Agent and appears on this page on its own — you never create monitored components manually. Manual creation exists for exactly two cases:

  1. Unmonitored strings — parts of the plant no logger reports. Add them so the digital twin covers the whole plant (see Add strings below).
  2. The feed-in meter — the platform always requires a feed-in meter for its further analysis. If the park has no monitored physical meter, create the virtual feed-in meter, which simulates the park's feed-in as a robust sum of all inverters (see Other Component Types).

Where it lives

Open your plant's Components section in Mirox (in the app: open a plant from your plants, then its Components section). The tabs across the top — Stromzähler (feed-in meter), Irradiation, Inverter, Combiner Box, and Strings — group every component type.

Why Configuration Matters

The Components page is not just an inventory. The numbers you enter here become the reference the Digital Twin compares live telemetry against:

  • Clear-sky simulation uses each string's panel datasheet, azimuth, tilt, and panel count to model the energy a string should produce under clear skies.
  • Performance ratio and degradation are measured against the configured peak power, so an under- or over-stated peak power skews every efficiency figure.
  • Loss detection attributes shortfalls to specific components — it can only do that if the component tree is complete and correctly mapped.
  • String analysis and the component states (producing normal, degraded, inferred, and so on) depend on the same configuration.

Keep configured power and string power in agreement

A power mismatch — where the plant's configured peak power does not equal the sum of its strings — degrades digital-twin accuracy. The Review Status card flags it, and you should resolve it before relying on the analysis (see Check the Review Status).

Read the Status Cards

Above the tabs, three cards summarize the plant's configuration health:

  • Peak Power — the configured peak production of the plant.
  • System Metrics — current system-level figures, including the logger count.
  • Review Status — how complete and consistent the configuration is (covered below).

Configure Inverters

Open the Inverter tab. Each row is an inverter the platform has discovered or that you added. Click a row to open its detail drawer, where you can:

  1. Assign a datasheet. Pick the inverter model from the Datasheet dropdown. The datasheet supplies the manufacturer, model, rated peak performance, and efficiency figures shown below the selector.
  2. Add notes. Use the Information field for any free-text context (service history, location, quirks).
  3. Mark as hidden. Toggle Mark as Hidden to remove a decommissioned or duplicate inverter from the main view. Use the Show Hidden switch on the tab to bring hidden items back temporarily.
  4. Click Save.

The drawer also shows read-only state — connected strings, combiner boxes, panel count, connected peak power, status, last-seen time, and discovery source. Use the Connected Components cards to jump straight to that inverter's combiner boxes or strings.

Auto-detection fills the name

An inverter with no name shows a "Waiting for Auto-Detection" placeholder and is named automatically once data arrives. Clearing a name with the eraser hands it back to auto-detection.

When an Inverter Is Replaced

When an inverter is physically swapped, its data logger gives the new device a different address. The platform therefore sees a second inverter rather than a change to the existing one: the retired unit stays in the list and never reports again, while the replacement arrives with no configuration at all.

Left alone, that costs you three things. The plant counts an inverter it no longer has, so its rated AC power is too high. The retired unit keeps an open "not communicating" event that nothing can ever close. And the machine that is actually producing has no orientation, module counts, or datasheet — so no simulation, performance ratio, or string analysis reaches it.

Confirming a replacement

Mirox looks for this on its own. When an inverter falls silent and exactly one newly appearing inverter carries the same name on the same data logger, the pair is flagged and you are notified. The inverter shows a Replacement? badge on the Inverter tab.

Click the badge to compare the two side by side — address, serial number, model, string channels, and the dates of their last and first measurements. The dialog also states exactly what confirming will do, including the corrected plant totals.

Nothing changes until you confirm. Mirox proposes the pair; you decide.

Confirming does all of the following in one step:

  1. Carries the configuration over — orientation, tilt, module counts, the parallel-string factor, and datasheets move to the new inverter's channels, matched channel by channel.
  2. Retires the old entry — the previous inverter and its channels are hidden so nothing is counted twice.
  3. Closes its open events — the retired unit can never report again, so nothing else could ever close them.
  4. Corrects the plant totals — inverter count and rated AC power drop back to what is physically installed.

If the pair is not a replacement, choose Not a replacement and it will not be suggested again.

Measurements are not lost

The retired inverter keeps its history. Its measurements simply end on the changeover date. In Analysis, the period before the swap is drawn as a separate, greyed-out line labelled (before replacement) — for the inverter and for each of its string channels — so you can see the whole season without the two machines being merged into one curve. Choose a date range that covers the period before the changeover to see it.

Finding the retired unit

Open the new inverter's detail drawer. Its Replacement section names the previous unit, its address, serial number, and last signal, and offers a button to open that retired inverter directly. This is the only route to it, since retired components are hidden from the lists.

If Mirox does not spot it

Detection needs the same name on the same data logger. If the device was renamed during the swap, or you already tidied the old entry away by hand, no suggestion appears — you can still record the replacement yourself from the new inverter and get exactly the same result.

Add Combiner Boxes

Monitored combiner boxes (Generatoranschlusskästen) are discovered automatically together with their strings. Add them manually only when you model an unmonitored part of the plant and want its strings grouped the way the plant is actually wired. The Combiner Box tab only appears once a plant has at least one.

  1. On the Inverter, Combiner Box, or Strings tab, click Add Combiner Box.
  2. Select the parent Inverter.
  3. Enter the Number of Combiner Boxes to create (1–50) and, optionally, a Name Prefix (for example GAK-).
  4. Click Save.

Add and Configure Strings

Strings carry the panel configuration that drives the clear-sky model, so this is the most impactful part of the page. Open the Strings tab (filter by inverter, and by combiner box where present, to narrow the list).

Add strings

Strings that report measurements are discovered automatically — you never add those by hand. Adding strings is for the unmonitored parts of the plant: the created strings complete the digital twin where no logger reports, and the analysis covers them through their parent inverter's measurements instead of direct string telemetry.

  1. Click Add String.
  2. Choose the target — To Inverter or To Combiner Box — and select one or more targets.
  3. Fill in the string configuration:
    • Number of Strings — how many identical strings to create per target.
    • Panel Datasheet — the panel model; its rated power feeds the peak-power calculation.
    • Azimuth — orientation in degrees (0–360, where 180 = South).
    • Tilt — angle in degrees (0–90).
    • Panels per Row and Parallel Rows — together these define how many panels each string carries.
  4. The dialog previews the estimated peak power you are about to add, computed from the panel rating and counts. Click Save.

Edit strings in bulk

  1. Select one or more strings in the table (or none, to apply to a wider scope).
  2. Click Edit.
  3. Adjust the panel datasheet, azimuth, tilt, or panels in series.
  4. Choose the scope under Apply changes to — Selected Strings, All Strings in Combiner Box, All Strings in Inverter, or All Strings in Park.
  5. Toggle Mark as Verified once a string's configuration is confirmed correct.
  6. Toggle Mark as reviewed to clear the review flag (needs attention) for the selection. This dismisses the flag, but it returns on the next analysis run if the underlying problem persists.
  7. Click Save.

Verify as you go

Marking strings verified is how you track which configuration you have reviewed. Verifying also copies any detected values into the manual fields you left empty, so the configuration you confirm matches what the analysis found, and it clears the review flag. Detection keeps running on verified strings as a cross-check — if a later run confidently contradicts a verified value, the string is flagged for attention while your manual value stays in effect. The Review Status card tracks your verification progress in its pie chart, independently of the warnings.

How values are prioritised

For every string property — azimuth, tilt, panels per row, parallel rows, panel count, peak power — the analysis uses a clear priority: manual before detected before default. Configure only what you are sure of; the analysis fills the gaps with the values it detects, and falls back to a default only when neither exists. A manual value is never overwritten by a detection. Instead, when a confident detection contradicts a manual value, the string is flagged for attention so you can decide — your value stays in effect until you change it.

Inspect a single string

Click a string row to open its detail drawer:

  • Configuration shows each property as Configured vs Detected side by side, plus the effective totals (total panels, peak power) actually in use.
  • Analysis Metrics compares Expected vs Measured: the expected degradation from the panel datasheet's aging at the string's installation age — with the corresponding power loss (W) and voltage loss (V) — next to what the analysis measured. Both columns are cumulative since installation, so they compare directly; a measured value below expected means the string ages better than its datasheet predicts. The performance ratio and the installation age complete the picture.
  • The installation date defaults to the park's commissioning date; set the string's replacement date (pencil icon) to override it for repowered strings — degradation is computed against this date.
  • Information holds free-text notes; the connected components cards jump to the parent inverter or combiner box.

Check the Review Status

The Review Status card tells you whether the configuration is complete and self-consistent:

  • A verified percentage and pie chart show how many visible strings you have confirmed.
  • A power comparison shows the configured Expected peak power against the Strings total, and the Offset between them. When they match, it reads Power Matching.
  • A needs-attention counter reports how many strings are flagged for review — because a value (azimuth, tilt, panels per row, parallel rows) is unknown on both sides (neither configured nor detected), a string was detected as not connected, a measurement looks defective, or a confident detection conflicts with a configured value. Strings whose configuration is complete — from manual or detected values — are never counted here, even while still unverified.
  • When strings are flagged, the card shows a Review button. Clicking it filters the table to exactly the strings the counter reports — the number on the card always equals the rows you see. Resolve them and mark them verified or reviewed. If the totals are correct but the expected peak power is wrong, adjust the expected peak power in the plant's core data settings instead.

Once the power matches and nothing is flagged, the card reads All Good and offers a Deep Analysis action that asks the agent to re-evaluate the plant with the corrected configuration. Verification progress is tracked separately by the pie — unverified strings with a complete configuration are fine and raise no warning.

While an analysis runs, you can open the live deep-analysis protocol to follow it in real time: the start and mode, the day-quality verdict (detected curtailment or cloudy periods and what was excluded), per-string progress, a degradation summary, "results saved", and the final peak-power comparison of detected vs configured vs nameplate. It is the quickest way to see what the analysis concluded and why.

In the strings table, problems are also shown inline so you can spot them without opening each drawer:

  • Where a confident detection disagrees with a configured value, the detected value appears as a small badge right next to the configured one (for example 162° → 102°). The configured value is deliberately not marked as wrong: as the cell's info icon explains, a mismatch has two possible causes — the string is not configured the way it was actually built, or the configuration is correct and the plant really behaves differently (defect strings, unknown obstacles, or terrain and shading the analysis cannot fully compensate). If you know the applied configuration is correct, simply leave it — persistent problems are picked up by the health watchdog and appear as events on the plant's monitor page.
  • The Analysis column condenses each string's state into a single shield — the most severe open issue wins: green only when the string is verified and every cross-check agrees, red for a confident conflict, amber for detected as not connected, an orange question mark for values unknown on both sides, and a dash for not verified yet. The info icon in the column header carries this legend.

Other Component Types

  • Irradiation — the irradiation sensors used as the production reference. Click a row to view sensor details.
  • Stromzähler (Feed-in Meter) — the meters that measure energy fed to the grid. The platform always requires a feed-in meter for its further analysis. If the park has no monitored physical meter, click Add Virtual Feed-in Meter: the virtual meter simulates the park's feed-in as a robust sum of all inverters. It is an estimate for the analysis — never a revenue-grade or billing measurement. Click a row to view meter details.

Related Guides

  • Setup — the short end-to-end workflow: configure what you know, let the analysis detect the rest, review and verify
  • Component States — the operational states the Digital Twin assigns to the components you configure here
  • Digital Twin — the engine that turns this configuration plus live telemetry into monitoring insight
  • Loss Detection — how a complete, correct component tree lets shortfalls be attributed to a specific string or inverter
  • Solar Plants — the supported solar configurations and the parameters Mirox monitors
  • Configuring Data Loggers — map discovered devices so components start reporting data
Prev
Configuring Data Loggers
Next
Configuring VPN Servers per Agent (Direct VPN)
MIT Licensed | Copyright 2026 Mirox Verwaltungs GmbH | Privacy