Glossary
Short definitions of terms used across ArrayTrace.
Reviewed
What this is for
The words ArrayTrace uses on screens and on reports, so a number and its name stay the same thing. These are the product's terms, not a physics textbook.
Steps
Look a word up below. Related articles go deeper.
Autonomy
Hours of cover during an outage, from a full bank. The storage screen reports it as Outage cover, the typical figure, and Worst outage, the least the bank carried. A year with no interruption reports a dash, not 0 h - there was nothing to cover.
Characterisation
The numbered datasheet model of a module, inverter or battery that a simulation actually runs. A catalogue row can hold several; one is the default. Without a characterisation, nothing can be simulated with that hardware.
Condition set
What the world does to an array: weather, soiling, losses, models. It carries no geometry. One design can be run against several sets.
DC:AC
The target ratio of DC nameplate to inverter AC rating. Inverter count on a wiring zone follows it; the achieved ratio is shown beside it.
Design
What would be built: field segments, keepouts, wiring. It carries no weather.
Field segment
The polygon you trace on the imagery. Modules tile the sloped plane inside it, respecting setbacks, row spacing and keepouts.
Keepout
Anything that blocks placement or casts shade: an area, a line, or a tree. Keepouts are drawn and listed on the designer's Obstructions rail.
kWh/kWp
Specific yield: how many kilowatt-hours a year each kilowatt of panel produces. A small array and a large one can be compared on the same footing.
Loss waterfall
Fifteen stored steps that decompose the solved year for a report. It is not a chain of derates applied to a guess. When a client asks why the performance ratio is what it is, the answer comes from those stored numbers.
Nameplate
STC capacity of the modules that have been laid out, in kWp. A new project reads 0 kWp until modules exist.
Never reviewed
Nobody has checked the figures. Catalogue rows and tariff prices render this rather than a date they did not earn. Indicative, not a basis for a quote you will defend as measured.
P50 / P90
Production probability. P90 means expected to be exceeded in nine years out of ten, so it is a lower figure than P50. Banks lend against P90; sales quotes P50. The exceedance is the modelled P50 read against a stated uncertainty, not a second simulation.
Performance ratio
PR: the share of specific yield a real installation keeps once shade, heat, wiring and downtime have taken theirs. A healthy roof lands near 0.80. Anything much under it is worth opening the loss waterfall.
Project
One physical site. Designs, condition sets, consumption, simulations, money and paperwork all hang off it.
Proposal freeze
Generating a document freezes the layout, the values and the branding at that moment. Nothing edits it afterwards. Regenerating makes another document beside it, so a customer holding a printed copy is holding what the system still says.
Self-consumption
The share of what the array made that did not leave the site.
Self-sufficiency
The share of what the site used that was served without the grid. A tiny array on a large load can have high self-consumption and almost no self-sufficiency. They are different numbers.
Simulation
The pairing of one design and one condition set. Results is a matrix of those pairings. The run happens off the page; the cell is the report.
Wiring zone
A group of field segments that share an inverter and string sizing. Generate wiring turns that into strings, hardware and conductor runs.
Tips and common mistakes
- A dash is not a zero. Unmeasured, not modelled, and "no outage in the year" all render a dash.
0is a measurement that came out at nothing. - A design carries no weather and a condition set carries no geometry. Mixing them is Design vs condition set.
- Self-consumption is about generation. Self-sufficiency is about load. Autonomy is about hours when the grid is away.