The world installed a record 511 GW of solar in 2025, according to IRENA's Renewable Capacity Statistics 2026, and cumulative solar capacity now sits near 2.4 TW. Fleets of this size generate far more data than any operator can read by hand. The 2026 Solar Risk Assessment from kWh Analytics, which brings together data from several industry contributors, shows why the data matters. SmartHelio's analysis of 6.5 GW of global assets found that inverter shutdowns drive 28% of recoverable performance risk. Above, Surveying's thermal data from more than 3,000 assets shows degradation accelerating after year seven.

Operators who track the right signals find these losses early. Operators who track too many signals, or the wrong ones, drown in alarms. This article covers what a solar management dashboard should show, why each metric matters, and how to keep the screen useful for daily work.

Why Solar Operators Need One Dashboard for Every Asset

A solar plant rarely fails all at once. Output drifts down through small faults that sit in different systems, so a single view is the fastest way to connect them.

Most plants already collect data. Inverters report to vendor portals, weather stations log to SCADA, and maintenance teams track work in a separate tool. Each system answers part of the question. None of them answers whether the plant produced what it should have, and if not, why.

A solar management dashboard joins these sources and compares actual output against expected output in near real time. It also ranks problems by lost energy, so crews fix the costliest fault first. Without that ranking, a failed combiner box and a minor communication drop look equally urgent.

Track Energy Output Against Expected Production

Raw generation figures mislead because sunshine changes every hour. The dashboard must compare output with what the weather allowed.

Start with these core yield metrics.

• Energy produced by plant, block, inverter and string, at 5 to 15 minute resolution
• Specific yield in kWh per kWp, which lets you compare plants of different sizes
• Performance ratio (PR), which divides actual energy by the energy an ideal plant would produce under the measured irradiation
• Weather-adjusted expected energy, built from plane-of-array irradiance and module temperature

Performance ratio deserves careful handling. A plain PR drops in hot summer months because module temperature lowers output, which can trigger false concern. Use a temperature-corrected PR for trend analysis and keep the plain PR for contractual reporting.

Show the gap between expected and actual energy in kWh and in currency. A line on a chart gets ignored, while a figure labeled "12 MWh lost yesterday" gets a work order.

Watch Inverter Health to Catch the Largest Recoverable Losses

Inverters convert every kilowatt the plant sells, so one inverter problem affects a whole block. That makes inverter monitoring the highest-value section of the dashboard.

Track status, active power, efficiency, internal temperature, and fan or cooling alarms for each unit. Compare each inverter against its peers on the same site, since identical units under identical sun should perform alike. A unit that lags its neighbors by several percent points to a derating setting, a thermal issue, or a string problem.

Watch for these patterns in particular.

• Repeated short trips that clear on their own and never reach an alarm log
• Power limits caused by configuration errors or thermal derating
• Efficiency drift on a single unit while the rest hold steady
• Restart delays after grid events

Vendor-published case studies from Delfos Energy describe several inverter issues that standard alarms missed, including a configuration limit and a fan-related derating. Treat those as vendor reports, not independent research, but the pattern is consistent with kWh Analytics' finding that shutdowns carry most of the recoverable risk.

Check String and DC-Side Performance

Many losses start on the DC side, where inverter-level views cannot see them. String-level data narrows a problem from a block of panels to a specific circuit.

Monitor string current, combiner box current, DC voltage and insulation resistance where the hardware supports it. Compare strings that share an inverter input and flag the ones that fall below their peers. A dead string, a blown fuse, a failed connector and a ground fault all show up as a current gap.

Add thermal and inspection data when you have it. Drone-based infrared surveys reveal hot spots, bypass diode failures and delaminated modules. Feed those findings into the dashboard as tagged assets, so the next performance dip links back to a known defect.

Degradation tracking also belongs here. Plot yearly specific yield and PR per block, then watch for acceleration. Surveying's finding of faster degradation after year seven suggests operators should review older blocks more closely.

Measure Weather, Irradiance and Soiling Accurately

Every performance metric depends on the irradiance measurement, so bad sensor data quietly corrupts the whole dashboard. Operators should treat weather instruments as critical assets.

Measure plane-of-array irradiance with reference cells or pyranometers, plus module back-surface temperature, ambient temperature, wind speed, and humidity. Compare readings across stations on large sites, and flag any sensor that drifts from its neighbors.

Soiling needs its own indicator. A soiling ratio compares a clean reference cell against a dirty one, and the dashboard can then estimate the revenue lost to dust and set cleaning schedules against real cost. In snowy regions, add a snow-cover indicator. In windy regions, track tracker stow events, since kWh Analytics' 2026 report cites Array Technologies data showing that standard hourly wind modeling can underpredict energy losses from tracker wind stow by up to 4%.

Follow Availability, Downtime and Curtailment

Availability tells you how long the plant could produce, and downtime tells you what stopped it. Both feed contracts and insurance conversations.

Report availability two ways. Time-based availability counts hours when equipment ran, while energy-based availability weights each hour by the sun available, so a midday outage costs more than a dawn outage. ICF's analysis of 130 US projects, cited in an earlier kWh Analytics report, found median annual availability of 97.7%, with the best projects reaching 99% or more through faster response times and more frequent preventive maintenance.

Pair availability with these operational metrics.

• Mean time to detect a fault and mean time to repair it
• Open alarms by age and severity
• Grid outage and curtailment hours, shown separately from equipment faults
• Warranty and contract KPIs, with the clock running on response commitments

Separating curtailment from faults prevents false blame. A plant told to reduce output by the grid operator did not underperform.

Include Grid, Storage and Revenue Signals

Operators earn money at the point of interconnection, so the dashboard must follow energy beyond the inverter. Revenue context turns technical data into business data.

Track export power, power factor, voltage and frequency response at the meter. Compare meter readings against inverter totals to catch metering gaps and auxiliary losses. For plants with batteries, add state of charge, round-trip efficiency, cycle counts and degradation indicators.

Link production to prices and contracts. Show realized revenue against a P50 forecast, and display the effect of negative-price hours or contract caps. Operators managing merchant exposure can then decide when to curtail voluntarily or shift storage dispatch.

Add Safety, Fire and Security Alerts

Safety signals belong on the main screen, not in a separate system. They carry the highest consequence and the shortest response window.

The kWh Analytics 2026 report notes that only 4% of PV fire loss events occur in high wildfire risk areas, while 84% of fire events trace back to equipment-driven brushfire. The practical lesson for operators is to watch equipment condition, since most fires start inside the plant. Track connector and combiner temperatures, ground fault alarms, arc fault events and vegetation height near equipment.

Include perimeter security and camera alarms, transformer temperature and oil level, and tracker fault states. Route high-severity alerts to an on-call phone, and keep a clear acknowledgement trail.

Turn Alerts Into Work Orders

A dashboard that only displays problems leaves the work undone. The strongest designs connect each alert to a maintenance action.

Set alert rules by expected loss. A fault costing 20 MWh a day should page someone, while a minor communication drop should wait for the next shift. Group related alarms into one incident so a single failed inverter does not create fifty tickets.

Link the alert to your maintenance system. The incident should carry the asset, the fault time, the estimated lost energy and the last repair on that unit. After the crew closes the ticket, the dashboard should confirm that output recovered. That closed loop proves whether the fix worked.

Real-World Example: A 150 MWdc ERCOT Site

A utility-scale owner faced escalating DC health problems at a 150 MWdc site in ERCOT, according to a Raptor Maps case study. Power and revenue ran far below modeled expectations. Combiner, inverter-level, and tracker anomalies drove most of the lost capacity.

The owner used Raptor Maps' Sentry platform to detect and rank those anomalies, and it reported a 75% decrease in underperformance within 12 months. Raptor Maps puts the recovered annualized revenue at about US$ 2 million.ThatworksouttoroughlyUS13,000 per MWdc per year. The figures come from a vendor case study, not an independent audit, so treat them as a reference point and not a guarantee.

The example shows the value of ranking faults by revenue and by component. The owner did not need new panels. It needed to see which combiners, inverters, and trackers cost the most, and fix those first.

Plan Solar Management Dashboard Development Around Data Quality

Good screens cannot fix bad data. Any team scoping Solar Management Dashboard development should settle data questions before designing a single chart.

Decide which protocols feed the system, such as Modbus, OPC UA, IEC 61850 or vendor APIs. Normalize tag names across inverter brands so a "DC" field means the same thing everywhere. Choose a sampling rate that matches the fault you want to catch, since 15-minute data hides short trips that 1-minute data reveals.

Build validation into the pipeline. Flag missing readings, frozen values, and sensor drift, and show data quality on the dashboard itself so operators know which numbers deserve doubt. Add role-based views as well. Control room staff need live alarms, asset managers need monthly trends, and investors need portfolio-level KPIs.

Measure the ROI and Business Impact

The return comes from recovered energy, faster repairs, and less manual reporting. Operators can measure each one before and after launch.

Here is an illustration with assumed figures. A 100 MWac plant with a 25% capacity factor produces about 219,000 MWh a year. At an assumed US40perMWh,annualrevenueisaboutUS8.76 million. Each 1% of recovered production is worth roughly US$87,600 a year. Use your own capacity factor and price to rebuild the estimate.

Track these indicators against a baseline.

• Performance ratio and specific yield by block
• Mean time to detect and repair faults
• Energy-based availability
• Revenue recovered from corrected faults
• Analyst hours spent building reports

The 150 MWdc ERCOT example above shows the upper range of what ranked, component-level visibility can recover. Smaller plants will see smaller gains, but the logic holds.

Final Thoughts on Choosing What to Monitor

A solar management dashboard works best when it answers one question clearly. Did the plant produce what the weather allowed, and if not, what failed and what did it cost? Output against expected energy, inverter and string health, trustworthy weather data, availability, grid and revenue signals, and safety alerts cover that question from every angle.

Start with a short list of metrics, validate the data feeding them, and connect each alert to a maintenance action. Add detail as the team proves it uses what it already has. Operators who build the habit of ranking losses in kilowatt-hours and dollars will find problems sooner and fix them in the right order.