METHODS & SOURCES
Show the working.
A useful estimate tells you what went in, how it was calculated and what it cannot answer.
Three types of information
- User data: bills, load profiles, tariffs and equipment specifications you supply. Their accuracy is not verified by the browser.
- Teaching assumptions: clearly labelled hypothetical values used in examples. These are not market quotations or measured projects.
- External references: primary sources linked for technical context. A reference does not certify the site or automatically validate a default.
Model boundaries
| Tool | Method | Important boundary |
|---|---|---|
| Solar | Annual kWh ÷ (365 × sun hours × performance ratio); round to panels | No hourly PV/load matching or tariff settlement |
| BESS sizing | Energy derates and discrete series/parallel topology; separate PCS power | No certified equipment combination or protection design |
| Dispatch | Interval energy balance, efficiency and SOC clipping | Synthetic load is illustrative; no tariff optimisation |
| EV charging | Driving energy, charging losses, equipment/site limits and time windows | No detailed thermal model or charge taper |
| Planning notebook | Small, fully stated energy-balance comparisons | No location lookup, current price feed or equipment approval |
How the new examples are checked
The notebook formulas have automated checks for the published examples, zero demand, invalid efficiencies, whole-panel rounding and insufficient charging time. Static checks verify local page links and sitemap coverage. Browser checks exercise navigation, form changes, scenario storage and export. These are software checks, not an independent engineering certification.
The inherited calculators have their own model limits. The new guides document those limits without presenting every inherited feature as newly validated. Current incentives, tariffs and vehicle specifications must be checked with their issuing authority or manufacturer.
Source and correction practice
When a model changes, update the associated example, test the expected result and record the material change. When a cited source changes, inspect the effect on the content rather than simply updating a date. For errors, record the page, inputs, expected result and observed result using the issue-report worksheet.
Suggested ongoing review: check broken links and reader-reported issues monthly; review policy-sensitive defaults before each publication; review a guide when its model changes. This is a maintenance practice to follow, not a claim of historical reviews.
AI and editorial responsibility
AI assisted the preparation of this version’s explanations and code. Equations, example values and sources are exposed for inspection. The examples do not claim first-hand installation experience. The public author is Soheb Zaidi. Attribution does not imply an independent technical certification of the models.
Publication record
See the dated change log for this version’s changes and known limitations. Dates represent content changes, not automatic renewals of source validity.