Investigate alarms by installation and configure the appropriate monitoring rule.
Open Alarms → Alarm management for the installation-oriented monitoring workspace. The overview groups signals and assets by site so you can move from a portfolio issue to the affected equipment.
Open an installation requiring attention to inspect its rule and affected equipment. Fictional demonstration data.
A portfolio with several hundred installations is easiest to review by exception. The fictional account below contains 360 sites with one PV-status rule per site. Its daytime snapshot has four distinct groups:
| PV-status outcome | Sites | What to do next |
|---|---|---|
| A production-related alarm is active | 12 | Inspect the affected inverter and compare fresh measurements with its peers. |
| Recent measurements are unavailable | 6 | Investigate data availability before drawing a conclusion about production. |
| The PV-status check is healthy | 338 | Leave these outside the immediate investigation queue unless another rule or observation requires attention. |
| The PV-status rule is paused | 4 | Check why monitoring was paused; a paused rule is not evidence that the installation is healthy. |
| Total | 360 | 18 sites need investigation; 4 have a separate monitoring exception. |
The real alarm workspace with a fictional 360-site account. Start with the installations requiring attention, then investigate one site in detail.
These counts describe the PV-status rule only. A site can have several rules and appear in more than one category. Do not add unrelated tile counts together as if each tile represented a different installation. This example also contains one custom rule, so 360 installations do not mean exactly 360 rules.
For a first pass:
The Running fine group is useful context, not a guarantee that every possible issue has been checked. It reflects the rules and measurements available for those installations.
A no-data condition and a confirmed production problem are different. Check the last measurement time before interpreting an alarm. Filtering the overview changes what you see; it does not disable the underlying rule.
At Demo Solar Park 001, eight inverters are monitored. Seven are producing, but inverter 01 reports 0 kW while the others continue producing. The example is evaluated at 14:30 in the Europe/Brussels time zone, with its latest measurements at 14:29.
Open the rule's Actions menu and choose View analysis or Analysis. The illustrated chart compares inverter 01 with two of its producing peers over the same daytime interval. The other five inverters are still part of the site's monitoring; the three chart lines are not the site's complete equipment list.
Inverter 01 drops to zero while two comparison inverters continue producing. These are invented measurements, displayed in the actual analysis dialog.
Read the evidence in order:
The alarm is a reason to investigate. The chart alone does not identify a failed component or prove an electrical fault. By contrast, for the six sites with missing measurements, start with the last data timestamp and connection: no data is not the same measurement as 0 kW.
Open the incident's Not being handled or Being handled control. Choose I'll handle this to take ownership, or inspect the colleague already named there. Add a comment recording the observed interval, comparison and next step. Take over and Stop handling change responsibility; they do not repair the site or change the rule's measured condition.
The incident history records ownership and comments alongside the observed fault. Fictional identities and incident history.
In this example the operator records that inverter 01 remains at zero, notes that peers are producing, and schedules a connection check. If a later check is needed, choose Create follow-up task and set the reminder time. The creator becomes that task's owner; another authorized user can take it over.
Keep the three outcomes separate:
Use Previous incident when available to inspect an earlier occurrence. A recurrence is a new incident: its handling and comments start separately from the earlier occurrence. See Follow-up actions for the distinction between these incident controls and the separate work-planning board.
Choose Add alarm and select the available type. Performance ratio (PR) monitors production relative to irradiation; performance index (PI) uses its configured comparison. A custom alarm is available where the account supports it. Select the installation and devices, fill in the rule-specific conditions and save.
PR monitoring can be enabled from the rule workflow by users with the relevant rights. The calculation still needs suitable PV configuration and measurements. Enabling PR without those inputs does not make its result reliable.
Choose the rule type that matches the condition you need to monitor. Available choices depend on your access. Fictional demonstration data.
The fictional account also contains Daytime production below 5 kW, a custom rule for Demo Solar Park 001. Its example settings are:
| Setting | Example value |
|---|---|
| Measurement | Production |
| Condition | Below 5 kW |
| Continuous duration | 30 minutes |
| Active days and time window | Monday–Friday, 10:00–15:00 |
| Time zone | Europe/Brussels |
| Recipient | The selected operations recipient |
| Recovery notification | Enabled |
These values illustrate the controls, not a recommended threshold for every PV site. A fixed clock window is not a sunrise/sunset calculation, and low production can have weather, operating or maintenance explanations. This rule checks total site production; it can remain healthy while a separate PV-status rule finds one underperforming inverter.
The continuous duration is not the evaluation interval or an email-delivery promise. This example's evaluator checks every 15 minutes; the condition must still meet its 30-minute requirement, and delivery follows the recipient's schedule. Gaps in the measurements cannot prove a continuous violation or a recovery.
When applying a custom rule to a portfolio selection, review the selected installations and their available measurements. The selection is a snapshot: later changes to portfolio membership do not automatically update that custom-rule selection. This is separate from the new-installation monitoring described below.
Accounts with active alarms are automatically followed up for new installations daily unless automatic monitoring was deliberately disabled. New eligible installations are added within the account's site limit. Existing, skipped and deliberately removed alarms are preserved. Automatic configuration does not add PR rules.
The daily discovery task is separate from the evaluation frequency of individual alarm rules. It does not mean every alarm is evaluated only once a day.
Full alarm management requires the relevant account access and alarm/email permissions. A restricted user may receive a read-only or narrower view. Email recipients are managed separately from the conditions that trigger an alarm.