Solar Platform
What technology stack makes decentralized solar possible?
ABCD Stack
| Layer | Technology | Solar Application |
|---|---|---|
| A | AI/ML | Dispatch optimization, yield prediction, demand forecasting |
| B | Blockchain | Generation proofs, carbon credits, settlement |
| C | Crypto/Tokens | Operator rewards, GCC tokens, governance |
| D | DePIN - Data | Solar farms, inverters, meters, batteries |
The integration: DePIN deploys generation capacity → Blockchain verifies and records → Crypto aligns incentives → AI optimizes dispatch and rewards.
SaaS Stack
| Job To Be Done | Products | Cost Range |
|---|---|---|
| Site Assessment | Aurora Solar, HelioScope, EagleView, Nearmap | $150-500/mo |
| Yield Modeling | PVsyst, PlantPredict, SAM (free) | $0-2,500/yr |
| Proposal/Quote | SurgePV, OpenSolar, Enact | $100-300/mo |
| CRM/Sales | Salesforce Solar, SolarNexus | $50-300/user/mo |
| Project Mgmt | Scoop Solar, SiteCapture | $100-500/mo |
| Monitoring | SolarEdge, Enphase Enlighten, AlsoEnergy | Free-$50/site/mo |
| O&M/Asset Mgmt | PowerFactors, Raptor Maps, Zeitview | Enterprise |
| Interconnection | Interconnection.io, GridX | Emerging |
| Carbon/RECs | M-RETS, APX, Glow (GCC) | Transaction-based |
Aerial Intelligence
The site assessment bottleneck — measuring roofs, assessing shading, calculating usable area — is being eliminated by aerial imagery + AI. This is where Nearmap on OpenSolar leads.
How It Works
Address entered → Aerial imagery loaded (5-7 cm/px)
↓
ML segments roof boundary → Calculates area, slope, azimuth
↓
Shading analysis from oblique imagery → Identifies obstructions
↓
Auto-generates panel layout → Estimates kW capacity + annual kWh
↓
Proposal created with imagery + calculations → Customer signs
Platform Comparison
| Platform | Imagery Source | ML Capabilities | Design Tool | Pricing |
|---|---|---|---|---|
| Nearmap + OpenSolar | Own aerial fleet (5-7 cm/px) | Roof segmentation, obstruction detection | Full design-to-proposal | Subscription + free design tool |
| Aurora Solar | Google/Bing satellite + LIDAR | Shade analysis, auto-design | Full design + sales | $150-370/mo |
| HelioScope | Google satellite | Basic shading | Commercial-focused design | $179/mo |
| EagleView | Own aerial + satellite | Roof measurement reports | Report-only (no design) | Per-report |
| Google Solar API | Google imagery + LIDAR | Solar potential estimates | API-only (no UI) | Per-query |
Resolution Matters
| Source | Resolution | Use Case | Limitation |
|---|---|---|---|
| Google/Bing satellite | ~30 cm/px | General overview | Can't see roof fixtures |
| Nearmap aerial | 5-7 cm/px | Accurate roof measurement | Coverage limited to metros |
| Drone capture | 1-2 cm/px | Single-site detailed survey | Doesn't scale |
| LIDAR (Google) | Point cloud | 3D shading model | Coverage gaps, aging data |
Higher resolution → more accurate panel placement → fewer change orders → better unit economics.
DePIN Connection
Aerial intelligence feeds directly into the DePIN solar stack:
- Generation Verification: Satellite imagery confirms panel deployment matches on-chain claims
- Yield Prediction: Accurate roof measurements + shading analysis improve production forecasts
- Portfolio Monitoring: Detect panel degradation, vegetation overgrowth, or physical damage across fleet
- Carbon Credit Validation: Time-stamped imagery provides additionality evidence
See PropTech VSaaS for the broader aerial intelligence landscape across real estate.
Layer 1: DePIN
Physical Infrastructure: The hardware that generates and measures solar energy.
| Device Category | Examples | Function |
|---|---|---|
| Panels | Mono/Poly/Bifacial, 400W-700W | Convert sunlight to DC power |
| Inverters | String, Micro, Central | Convert DC to AC, grid synchronization |
| Meters | IoT smart meters | Measure generation, verify claims |
| Batteries | LFP, NMC storage | Time-shift generation, grid services |
| Trackers | Single/dual axis | Optimize panel orientation |
DePIN Solar Protocols:
- Glow — Industrial solar farms
- Daylight Energy — Consumer energy rewards
- Starpower — DER coordination
- Srcful — Grid edge intelligence
- React Protocol — Virtual power plants
See DePIN Tech for broader patterns.
Layer 2: Blockchain (Trust Infrastructure)
The immutable layer that records generation proofs and carbon credits.
| Function | Traditional | Blockchain-Enabled |
|---|---|---|
| Generation Verification | Manual audits, periodic | IoT + satellite + on-chain proofs |
| Carbon Credit Issuance | Registry-based, opaque | Tokenized GCC, transparent |
| Settlement | Monthly billing cycles | Real-time atomic settlement |
| Additionality Proof | Trust-based claims | Cryptographic verification |
Platform Options:
- Solana — High throughput, Glow's choice
- Ethereum/L2s — Carbon market integration
- Custom — Purpose-built for energy data
Layer 3: Crypto/Tokens (Coordination Infrastructure)
The economic layer that aligns incentives across the network.
| Token Type | Function | Example |
|---|---|---|
| Protocol Tokens | Governance, staking, operator rewards | GLW |
| Carbon Tokens | Represent verified carbon avoidance | GCC |
| Governance Tokens | Protocol decision making | veGLW |
| Equipment NFTs | Represent deployed farm stake | Farm tokens |
Token Flow (Glow):
Protocol collects all electricity revenue
↓
Revenue funds recursive deployment
↓
Operators earn GLW for verified generation
↓
GCC issued for carbon avoidance
↓
GCC sold to carbon buyers
↓
Proceeds fund more infrastructure
Layer 4: AI/ML (Intelligence Infrastructure)
The models that turn data into decisions.
| Application | Input | Output | Status |
|---|---|---|---|
| Yield Prediction | Weather, historical, equipment | Production forecasts | Mature |
| Dispatch Optimization | Price signals, demand, storage | Optimal generation schedule | Emerging |
| Maintenance Prediction | Performance data, degradation | Preventive maintenance alerts | Growing |
| Grid Integration | Load patterns, frequency | Demand response participation | Emerging |
| Fraud Detection | Generation claims, satellite | Invalid claim flagging | Growing |
The Data Flywheel:
Farms generate verified production data
↓
Data trains better prediction models
↓
Better predictions improve yields
↓
Higher yields attract more farms
↓
More farms generate more data
Tools
See Principles for the data model (how solar systems work).
| Tool | Purpose | Use Case |
|---|---|---|
| PVWatts | Quick yield estimates | Initial feasibility |
| HelioScope | Commercial design + shading | Project proposals |
| PVsyst | Bankable energy modeling | Financing |
| Aurora Solar | Residential sales + design | Consumer market |
| SurgePV | Proposal automation | Sales workflow |
| PlantPredict | Utility-scale optimization | Large projects |
| OpenSolar | Free design-to-proposal | Residential sales |
| Nearmap | Aerial imagery + ML roof analysis | Site assessment |
Data Sources: NREL NSRDB, SolarAnywhere API, Solargis, Nearmap Aerial Imagery, Google Solar API
Standards: UL 1703 (modules), UL 1741 & IEEE 1547 (inverters), AS/NZS 4777.1:2024
Platform Maturity Assessment
| Component | Maturity | Key Players | Gap |
|---|---|---|---|
| Farm Deployment | Growing | Glow, EPCs | Interconnection queues |
| Generation Verification | Growing | IoT + satellite | Gaming prevention |
| Carbon Tokenization | Growing | GCC | Market adoption |
| Dispatch Optimization | Nascent | No dominant player | AI opportunity |
| Grid Integration | Nascent | Utilities, ISOs | Regulatory friction |
| Token Economics | Growing | GLW/GCC | Sustainability debate |
Build vs Buy
| Need | Build | Buy/Partner |
|---|---|---|
| Solar Farms | ✓ For DePIN model | |
| Generation Hardware | ✓ Commodity panels, inverters | |
| Blockchain Infrastructure | ✓ Use Solana/existing | |
| Verification System | ✓ Competitive advantage | |
| Token Design | ✓ Core to model | |
| Grid Interconnection | ✓ Work with utilities |
Context
- DePIN — Physical infrastructure patterns
- Blockchain — Trust layer options
- Solana — Glow's blockchain
- AI — Intelligence layer capabilities
- Energy KPIs — Metrics
Questions
Where does the intelligence layer create more value than the hardware layer?
- Which gap in the maturity table is the highest-leverage entry point for a new player?
- If aerial imagery eliminates the site assessment bottleneck, what becomes the next bottleneck?
- When does the data flywheel generate enough signal to make dispatch optimization bankable?