Battery energy storage systems generate a continuous stream of operational data, including alarms, warnings, and system signals. Every inverter warning, BMS notification, and communication interruption becomes part of the system’s history.

In theory, this data should make it easier to understand system behavior. In practice, it often slows teams down.

As portfolios grow, alarm streams and system logs quickly reach hundreds of thousands of entries per site. At that point, the challenge is no longer collecting data, but understanding what actually matters.

In one observed case, a single utility-scale site generated over 120,000 system records within seven months, including thousands of critical errors.

Operators spend time:

• searching for relevant signals

• checking if an issue has occurred before

• trying to understand whether an alarm had real operational impact

This directly affects battery fault investigation. What should take minutes can take significantly longer when relevant issues are buried in large datasets.

The issue is not the volume of data, but how it is structured and interpreted.

System signals and alarms are the most granular representation of system behavior, but on their own, they are not designed for decision-making. Most tools still require operators to work directly with this raw layer.

The same dataset can either overwhelm operators or help answer practical questions:

• Which issues require immediate attention?

• Has this fault occurred before?

• How much downtime did this issue cause?

The latest updates to the Cellect platform focus on structuring operational data so it can be explored, prioritized, and connected to real performance impact.

The Cellect platform improves how operational data is explored and analyzed across BESS assets, including updates to the events view.

Key capabilities include:

• Real-time search by component or alarm name

• Advanced filtering by time range, priority, and status

• Duration visibility to assess impact

• Structured severity levels to support prioritization

• Export of filtered datasets for reporting and deeper analysis

With these capabilities, operators can move from reviewing individual alarms to understanding system behavior, fault patterns, and performance impact over time.

1. Prioritize what matters

Not every signal requires action.

By structuring alarms by severity, you can quickly identify which issues need attention. This reduces time spent reviewing low-priority alerts and helps focus on faults that affect operations.

2. Quantify what matters

A signal becomes useful when you understand its impact.

With duration visibility available, you can assess:

• how long a component was affected

• whether the issue resulted in downtime

• whether follow-up action is needed

This connects operational data directly to system performance.

3. Recognize recurring issues

Single alarms rarely explain system behavior.

By filtering and reviewing historical data, you can identify:

• repeated faults on the same component

• patterns over time

• issues that persist after maintenance

This helps shift from reacting to alerts to improving reliability across the system.

When an alarm appears, the first question is whether it matters.

Typical questions include:

• How often does this alarm occur?

• How long does it last?

• Does it affect availability?

Typical workflow in the Cellect platform

1. Search for the alarm or event name

2. Filter results by time range

3. Sort results to identify repeated occurrences

4. Review duration

5. Export results if further analysis is needed

This allows you to move from an alert to a clear understanding of its operational impact.

When investigating a fault, you often need to understand whether it has happened before.

Typical workflow in the Cellect platform

1. Search for the component name

2. Filter by priority to focus on relevant signals

3. Adjust the time range

4. Sort results to identify recurring issues

5. Export findings if needed

This helps determine whether a failure is isolated or part of a recurring issue.

As BESS portfolios grow, operational data becomes more complex and harder to navigate.

Improving how this data is structured and interpreted is not just a usability improvement. It directly affects how quickly teams can perform fault analysis and how well they understand system behavior.

When you can:

• prioritize critical issues

• quantify their impact

• recognize recurring faults

you can reduce investigation time and make more informed decisions.

The value of operational data depends on how easily it can be explored, interpreted, and connected to real system outcomes. By structuring raw signals and alarms within the Cellect platform, operators can turn large volumes of data into insight that supports day-to-day operations and long-term asset performance.

If you want to explore how this works in practice, get in touch with our team. We can walk through typical investigation workflows and show how operational data can be analyzed more efficiently across your sites.