LIVEdataset aec-bench@releasetasks 552models 18disciplines 5last submission … · built …

electricalboth

String Sizing

Calculates the maximum and minimum number of PV modules per string by applying temperature coefficients to open-circuit and maximum-power-point voltages at site extremes, per AS/NZS 5033 and IEC 62548. Cold-corrected Voc must not exceed the inverter maximum DC voltage, and hot-corrected Vmp must remain above the inverter minimum MPPT tracking voltage.

both: The model must reason numerically unaided.

Standards

AS/NZS 5033 IEC 62548

How this task is generated

One template produces many comparable benchmark tasks while keeping the scoring contract fixed.

01
Template
The reusable contract shown on this page.
02
Scenario
An archetype and site context are sampled.
03
Difficulty tier
Inputs may be hidden at harder tiers.
04
Task prompt
The model responds with the declared outputs.

Parameters

Inputs the model receives, and the outputs it is scored on.

Inputs

9 inputs

Always given

Included directly in every task prompt.

Show 7 inputs

Voc stc
voc_stc_v
PV module open-circuit voltage at STC (25 degC, 1000 W/m2)
20 – 60 V
Vmp stc
vmp_stc_v
PV module maximum power point voltage at STC
15 – 55 V
Temp coeff voc pct per c
temp_coeff_voc_pct_per_c
Temperature coefficient of Voc (negative value)
-0.45 – -0.2 %/degC
Temp coeff vmp pct per c
temp_coeff_vmp_pct_per_c
Temperature coefficient of Vmp (negative value)
-0.5 – -0.25 %/degC
Inverter max dc voltage
inverter_max_dc_voltage_v
Inverter maximum DC input voltage
450 – 1500 V
Inverter min mppt voltage
inverter_min_mppt_voltage_v
Inverter minimum MPPT tracking voltage
100 – 500 V
Inverter nominal mppt voltage
inverter_nominal_mppt_voltage_v
Inverter nominal (midpoint) MPPT voltage
200 – 900 V

Derived from scenario

Sampled from the scenario and inferable from its description.

Site min temp c
site_min_temp_c
Site minimum ambient temperature
Derived from the archetype scenario. Hidden at hard difficulty.
-10 – 10 degC
Site max temp c
site_max_temp_c
Site maximum ambient temperature
Derived from the archetype scenario. Hidden at hard difficulty.
30 – 50 degC

Scored outputs

4 outputs

Voc corrected cold

voc_corrected_cold_v

Temperature-corrected Voc at site minimum temperature (V)

Scores if within ±3% of the reference value.

Vmp corrected hot

vmp_corrected_hot_v

Temperature-corrected Vmp at site maximum temperature (V)

Scores if within ±3% of the reference value.

Max modules per string

max_modules_per_string

Maximum number of modules per string (integer, rounded down)

Scores if within ±1% of the reference value.

Min modules per string

min_modules_per_string

Minimum number of modules per string (integer, rounded up)

Scores if within ±1% of the reference value.

Difficulty

Each template is sampled at three tiers. Harder tiers may hide inputs, forcing the model to infer them from the scenario description.

easy

All inputs given

All parameters given, residential module with small inverter

Inverter max dc voltage restricted to: {'min': 450, 'max': 600}

Inverter min mppt voltage restricted to: {'min': 100, 'max': 200}

Inverter nominal mppt voltage restricted to: {'min': 200, 'max': 380}

medium

All inputs given

All parameters given, any module type and larger inverters

hard

Some inputs hidden

Site temperatures hidden, agent infers from location context

Hidden inputs

Site min temp csite_min_temp_c
Site max temp csite_max_temp_c

Prompt replacement text

The installation is a {{ archetype.description }} system at {{ archetype.site_context }}

Task bundle

The exact instruction and parameter contract used to generate this task, pinned to the published library source.

/workspace

instruction.md

Teal lines show Jinja input conditions, not task visibility policy. A line renders only when that input or tool is visible.

1You are a senior electrical engineer specializing in solar photovoltaic system design.2 3## Problem4 5Determine the maximum and minimum number of PV modules per string based on temperature-corrected voltage limits, in accordance with AS/NZS 5033 and IEC 62548.6 7## Given8 9| Parameter | Value | Unit |10|-----------|-------|------|11| Module Voc at STC | {{ voc_stc_v }} | V |12| Module Vmp at STC | {{ vmp_stc_v }} | V |13| Temperature coefficient of Voc | {{ temp_coeff_voc_pct_per_c }} | %/degC |14| Temperature coefficient of Vmp | {{ temp_coeff_vmp_pct_per_c }} | %/degC |15{% if site_min_temp_c is defined %}16| Site minimum temperature | {{ site_min_temp_c }} | degC |17{% endif %}18{% if site_max_temp_c is defined %}19| Site maximum temperature | {{ site_max_temp_c }} | degC |20{% endif %}21| Inverter max DC voltage | {{ inverter_max_dc_voltage_v }} | V |22| Inverter min MPPT voltage | {{ inverter_min_mppt_voltage_v }} | V |23| Inverter nominal MPPT voltage | {{ inverter_nominal_mppt_voltage_v }} | V |24{% if archetype_description is defined %}25 26### Site Context27 28{{ archetype_description }}29{% endif %}30 31{% if tool_available %}32## Available Tool33 34A string sizing calculation tool is available at `/workspace/string-sizing_calc.py`. Run it with:35 36```bash37python3 /workspace/string-sizing_calc.py --help38```39 40You may use this tool to verify your calculations or compute values directly.41{% endif %}42 43## Required44 45Calculate the following:46 471. Temperature-corrected Voc at site minimum temperature (V)482. Temperature-corrected Vmp at site maximum temperature (V)493. Maximum number of modules per string (limited by inverter max DC voltage at coldest conditions)504. Minimum number of modules per string (to maintain voltage above inverter min MPPT at hottest conditions)51 52## Applicable Standards53 54- AS/NZS 5033 — Installation and safety requirements for photovoltaic (PV) arrays55- IEC 62548 — Photovoltaic (PV) arrays — Design requirements56 57## Constraints58 59- No internet access is available. Work from engineering knowledge and the provided tool.60- Use the following temperature correction formulas per AS/NZS 5033:61  - Voc_corrected_cold = Voc_stc x (1 + (temp_coeff_voc / 100) x (T_min - 25))62  - Vmp_corrected_hot = Vmp_stc x (1 + (temp_coeff_vmp / 100) x (T_max - 25))63- Maximum modules per string = floor(inverter_max_dc_voltage / Voc_corrected_cold)64- Minimum modules per string = ceil(inverter_min_mppt_voltage / Vmp_corrected_hot)65- Temperature coefficients are expressed as %/degC (negative values, since voltage decreases with increasing temperature)66- STC reference temperature = 25 degC67 68## Output Format69 70Show your step-by-step working in Markdown, including the temperature corrections and string sizing calculations. At the end of your solution, include a JSON block with your final answers in exactly this format:71 72```json73{74  "voc_corrected_cold_v": <numeric_value>,75  "vmp_corrected_hot_v": <numeric_value>,76  "max_modules_per_string": <integer_value>,77  "min_modules_per_string": <integer_value>78}79```80 81Write your complete solution to `/workspace/output.md`.82

Scenario archetypes

Each generated task is drawn from one of these realistic scenario bands.

Site contexts ground each scenario in a real locale the model can use to infer hidden values.

Residential mono perc

residential_mono_perc

Residential rooftop monocrystalline PERC module (~400 W)

sydney-suburban-rooftopmelbourne-suburban-rooftopbrisbane-suburban-rooftop

Parameter ranges

voc_stc_v: 37 – 42
vmp_stc_v: 31 – 35
temp_coeff_voc_pct_per_c: -0.3 – -0.25
temp_coeff_vmp_pct_per_c: -0.38 – -0.3

Commercial bifacial

commercial_bifacial

Commercial bifacial monocrystalline module (~550 W) in Sydney/Melbourne temperate climate

sydney-commercial-rooftopperth-commercial-rooftop

Parameter ranges

voc_stc_v: 45 – 52
vmp_stc_v: 38 – 44
temp_coeff_voc_pct_per_c: -0.3 – -0.25
temp_coeff_vmp_pct_per_c: -0.35 – -0.28
site_min_temp_c: 0 – 5
site_max_temp_c: 35 – 42

Utility scale hjt

utility_scale_hjt

Utility-scale heterojunction (HJT) module (~600 W) in arid inland Australia (Alice Springs/Broken Hill climate)

western-nsw-ground-mountqueensland-ground-mountsouth-australia-ground-mount

Parameter ranges

voc_stc_v: 48 – 55
vmp_stc_v: 40 – 47
temp_coeff_voc_pct_per_c: -0.28 – -0.22
temp_coeff_vmp_pct_per_c: -0.32 – -0.26
site_min_temp_c: -5 – 5
site_max_temp_c: 40 – 48

Thin film cdte

thin_film_cdte

Thin-film cadmium telluride (CdTe) module (~450 W) in tropical northern Australia (Darwin/Cairns climate)

darwin-commercial-rooftoptownsville-ground-mount

Parameter ranges

voc_stc_v: 24 – 30
vmp_stc_v: 16 – 22
temp_coeff_voc_pct_per_c: -0.35 – -0.28
temp_coeff_vmp_pct_per_c: -0.4 – -0.32
site_min_temp_c: 15 – 20
site_max_temp_c: 35 – 40

Example task

sydney-commercial-rooftop-commercial-bifacial-preview — hard difficulty, some inputs hidden.

Commercial bifacial monocrystalline module (~550 W) in Sydney/Melbourne temperate climate. sydney-commercial-rooftop. Required outputs: voc_corrected_cold_v, vmp_corrected_hot_v, max_modules_per_string, min_modules_per_string

The model sees

Scenario context and visible inputs.

voc_stc_v: 45 to 52 V
vmp_stc_v: 38 to 44 V
temp_coeff_voc_pct_per_c: -0.3 to -0.25 %/degC
temp_coeff_vmp_pct_per_c: -0.35 to -0.28 %/degC
inverter_max_dc_voltage_v: 450 to 1500 V
inverter_min_mppt_voltage_v: 100 to 500 V
inverter_nominal_mppt_voltage_v: 200 to 900 V

Executable tool: string-sizing_calc.py

The model must infer

Inputs withheld at this difficulty.

Site min temp c
site_min_temp_c
Site max temp c
site_max_temp_c

Stand-in text in the prompt

The installation is a {{ archetype.description }} system at {{ archetype.site_context }}

The model must produce

The scored JSON answer schema.

{
  "voc_corrected_cold_v": <number>,
  "vmp_corrected_hot_v": <number>,
  "max_modules_per_string": <number>,
  "min_modules_per_string": <number>
}

voc_corrected_cold_v · scored within ±3%
vmp_corrected_hot_v · scored within ±3%
max_modules_per_string · scored within ±1%
min_modules_per_string · scored within ±1%

String Sizing

How this task is generated

Template

Scenario

Difficulty tier

Task prompt

Parameters

Inputs

Scored outputs

Voc corrected cold

Vmp corrected hot

Max modules per string

Min modules per string

Difficulty

easy

medium

hard

Task bundle

Scenario archetypes

Residential mono perc

Commercial bifacial

Utility scale hjt

Thin film cdte

Example task

The model sees

The model must infer

The model must produce