Solar asset intelligence
Intelligent inspection. Spatiotemporal twins.
Robotic terahertz diagnostics assess subsurface condition. Digital twins track change across sites and time, guiding decisions to lower lifecycle costs.
Intelligent robotic inspection
Drone-mounted edge AI screens RGB and thermal imagery. Targeted terahertz scans assess subsurface material and interface condition.
Spatiotemporal digital twins
Models link material condition, environmental exposure and generation history, updating with new evidence to track change across sites and time.
Lifecycle decision support
Compare interventions by expected net benefit. Use measured outcomes to refine maintenance, material selection and site-specific design.
How it works
From inspection to informed action.
Intelligent screening
Drone-mounted edge AI checks RGB and thermal image quality, maps observations to individual modules and prioritises anomalies for targeted inspection.
Targeted THz diagnostics
The wheel–rotor platform scans priority modules and healthy controls, using terahertz to assess encapsulation layers and buried interfaces in compatible module structures.
Spatiotemporal modelling
Inspection, exposure and electrical data update module–string–plant models. ST-Twins track condition across sites and time, identifying where further measurements could improve decisions.
Decisions and feedback
Owners compare maintenance and replacement options by expected net benefit. Actual costs and generation responses refine the models, improving future recommendations and inspection priorities.
- 01Screen
Aerial screening
Scheduled and event-triggered RGB and thermal surveys identify and locate anomalies. Module-level mapping, terrain and environmental data provide spatial context and reveal changes over time.
- 02Diagnose
Targeted diagnostics
Robotic close-range inspections examine selected anomalies alongside healthy reference modules. Where suitable, terahertz sensing adds subsurface evidence beyond conventional optical and thermal imaging.
- 03Model
Spatiotemporal digital twin
Inspection records, environmental exposure, electrical data and maintenance history are connected through asset IDs and timestamps to track degradation, estimate energy losses and quantify uncertainty.
- 04Optimise
Decision intelligence
Cleaning, reinspection, repair, replacement and deferral are evaluated against recoverable energy, cost and risk. Measured intervention outcomes refine future predictions and operating decisions.
Verified outcomes continuously refine the system
Why it matters
Modules from the same factory can look identical on day one and age very differently in the field. The difference is traceable, if the data is connected.
Who we work with
Three decision-makers, one condition and degradation database
Asset owners & O&M
Operating plants
Schedule cleaning, repairs and replacements on what they recover, and stop paying for low-value work.
Learn moreDevelopers & EPCs
Deploying plants
Rank layout, tilt, row spacing, mounting height and module choice by whole-of-life cost, not just first-year yield.
Learn moreModule manufacturers
Manufacturing plants
Trace field failure modes back to bills of materials, supplier batches and lamination windows before warranty costs arrive.
Learn moreOutcomes
Lower lifecycle cost. More delivered energy. Longer useful life.
Targeted inspections
Fleet-wide screening directs detailed diagnostics where they matter most.
Value-driven maintenance
Prioritise interventions by recoverable energy, cost and risk.
Evidence-led prevention
Trace degradation and intervention outcomes to inform future design and maintenance.
About Metrifield
Solar asset intelligence, grounded in engineering.
Founded in Sydney by Harrison Su, Sohan Domingo and Laurens Willems van Beveren, Metrifield Dynamics brings together expertise in materials science, field inspection and edge AI to connect multiscale diagnostics with energy performance and lifecycle economics. Meet the team.
Cut avoidable costs across the solar lifecycle.
Identify recoverable energy losses, unnecessary maintenance and recurring defects. Turn module-level evidence into better operating decisions and future designs.