Back to Portfolio
Live in ProductionRenewable Energy / Solar
☀️

Solar Panel IoT Monitoring System — Performance Analytics & Fault Detection

A string-level IoT monitoring system for a 5MW solar farm — tracking performance ratio, detecting underperforming strings, and integrating with inverters via Modbus — increasing annual energy yield by 8%.

8%
Yield Increase
320
Strings
5d → 18h
MTTR↓

The Problem

The operator of a 5MW solar farm was monitoring performance at inverter level only — they could see total AC output but had no visibility into which strings or panels were underperforming. Shading, soiling, and module degradation caused string-level losses that were invisible until they accumulated to inverter-level impact. Annual energy yield was 8–12% below modeled values (PVSyst). Soiling detection was manual — operators scheduled cleaning based on calendar, not actual panel soiling.

💡Our Solution

We installed ESP32 Modbus gateways at each combiner box (16 strings per box, 20 combiner boxes total). Hall-effect current sensors measure current for each individual string. A pyranometer and module temperature sensor provide irradiance and temperature data for performance ratio calculation. Inverter Modbus data (AC power, MPPT voltage, efficiency) is also ingested. Lambda calculates performance ratio per string every 15 minutes, flagging strings below 85% of expected value. A soiling index is derived from irradiance vs. output degradation over time. The React heatmap shows every string color-coded — green (good), yellow (degraded), red (fault).

🔗System Architecture

String Sensors (Modbus) → ESP32 Gateways → MQTT → AWS IoT Core → InfluxDB → Lambda (PR calculation) → React Dashboard

Tech Stack

Hardware
  • ESP32 Modbus gateway per combiner box
  • Hall-effect current sensor per string (20A)
  • Voltage divider string voltage measurement
  • Pyranometer (irradiance)
  • Ambient + module temperature sensor
  • SMA / Fronius / Huawei inverter Modbus interface
Communication
  • Modbus RS485 (string to gateway)
  • MQTT to cloud via site Wi-Fi / 4G
  • Inverter Modbus integration
Cloud
  • AWS IoT Core
  • InfluxDB (string-level time-series)
  • Lambda (performance ratio + soiling calculation)
  • S3 (daily performance reports)
Frontend
  • React solar farm dashboard
  • String-level heatmap (color-coded performance)
  • Inverter MPPT efficiency chart
  • Automated O&M report generator

Key Features

String-level current and voltage monitoring (20A resolution)
Performance ratio per string, per inverter, per farm
Soiling index detection and cleaning schedule recommendation
Inverter MPPT efficiency and clipping analysis
Module temperature impact correction (temperature coefficient)
Fault detection: open circuit, partial shading, mismatch
Automated daily O&M PDF reports
Energy yield vs. PVSyst model variance tracking

Results Delivered

  • 8% increase in annual energy yield from fault detection + soiling management
  • 320 strings monitored across 20 combiner boxes
  • 15 underperforming strings identified in first month (shading and soiling)
  • Cleaning schedule optimized: 6 events/year → 3 targeted cleanings
  • Fault MTTR (mean time to repair) reduced from 5 days to 18 hours

Technologies

ESP32ModbusMQTTReactInfluxDBPyranometerInverter Integration

Who This Is For

Solar farm operators, EPC contractors, O&M service providers, rooftop solar installers

Need a similar solution?

We've built production IoT systems like this across 15+ countries. Let's talk about yours.

Let's Build Together

Got an IoT challenge?
We've shipped it.

Whether you need a fleet to track, a factory to monitor, or a farm to automate — our team has done it before and we'd love to build it with you. Typical response time: under 24 hours.

No upfront commitment99.9% uptime SLANDA on requestFixed-price options