Geometry- and position-aware analysis on a monthly resource model with PVWatts v8 benchmark continuity. Aggregate loss assumptions, resource provenance confidence, and the explainable PV Opportunity Score. Built to rank quickly across zones, buildings and portfolios.
Platform
Transparent assumptions. Reproducible results.
A solar number earns trust by showing its work. This page describes how SolaMaxa calculates, how methodology is governed, and how every material figure in a report is designed to be traced and independently checked.
Calculation fidelity
Two tiers, clearly separated
SolaMaxa runs two clearly separate analysis modes. Every run states which tier produced it and what that tier's inputs support.
Hourly (or sub-hourly) resource from imported, provenance-tracked weather sources; anisotropic sky irradiance; an explicit sequential loss waterfall; mounting-aware temperature; inverter behaviour; and deterministic modelled uncertainty ranges. Every default comes from a versioned register; overrides carry a reason.
Fidelity is stated per run — screening, detailed (estimated inputs), or detailed (validated inputs). Measured-calibrated fidelity is reserved for runs that actually include measured data and calibration; it is never claimed without them. Detailed analysis does not replace final professional electrical, façade, structural, fire or compliance design.
Methodology governance
Versioned, immutable, parity-tested
Formulas with stable identity
Material formulas carry stable identifiers and versions in a human-readable register — symbolic form, variable definitions with units, applicability and source references. The register text stands on its own.
Immutable history
Changing a formula creates a new version; historical versions remain intact and are referenced by the reports and workpapers that used them. History is never rewritten in place.
Code-to-register parity
Automated parity tests assert that the registered formula's behaviour matches the production implementation. When either side changes, the test fails until a new version is registered.
Calculation assurance
Workpapers, reconciliation, re-performance
Assurance is a design property of the platform, not a badge. Three mechanisms work together:
Calculation workpapers
Every material calculated value resolves to an immutable workpaper: inputs, units, assumptions, the human-readable formula, the values substituted, each material intermediate result, and provenance. Workpapers are hash-verified and version-referenced.
Report reconciliation
Material report figures are mapped to their workpaper steps before a final report issues. Presentation rounding is a declared transformation — a rounding difference is labelled as rounding, never presented as physics.
Designed for independent checking
Detailed-PV workpapers carry interval-level evidence and a worked manual re-performance example, with reconciliation from interval to monthly to annual under declared tolerances — so a technically competent reviewer can re-derive material results without SolaMaxa source code.
Assumptions & data
Defaults are visible. Gaps stay gaps.
You can expect
- Every default with an identifier, version, value, conditions, source and rationale — inspectable before you rely on it
- Overrides stored with a reason and a user
- Resource data with provider, dataset, period, resolution and licence provenance
- Assumption confidence labels — verified, supplied, estimated, default
- Unknown values reported as unknown — never zero-filled
You should not expect
- Engineering certification or sign-off — professional review is a separate discipline
- Probabilistic P50/P90 statements — outputs are deterministic modelled ranges
- Measured-calibrated fidelity without measured data and calibration
- Silent interpolation or silent equivalence between standards
Validation
Tiered, and honest about its reach
The screening tier is validated for continuity against PVWatts v8 and regional benchmarks. The detailed engine runs against a tiered validation framework — reference cases, benchmark continuity and independent re-performance of the calculation chain. Validation supports the modelling; it is not a claim of certification, accreditation or engineering sign-off.
Where a capability has known limitations — synthetic test weather, literature-based mounting parameters, bounded-out effects — the limitation is stated on the run and in the report, not buried in a manual.
See it on a building you know
Interrogate the methodology on your own building.
We'll walk the fidelity tiers, the assumption register and a workpaper — including which assumptions move the result and how to challenge them.
SolaMaxa is pre-beta. We are currently selecting the first architectural, BIPV, engineering and property teams.