Issue: Ground coupling and floor h_tr_floor calculation for FF cases

[anchapin]

Problem Description

Ground coupling and floor thermal conductance calculation may contribute to excessive temperatures in free-floating cases.

Root Cause

src/sim/thermal_model_core.rs:704-712 - h_tr_floor calculation:

let h_tr_floor_val = if spec.case_id == "195" {
    0.039 * zone_floor_area
} else if is_900_series_hvac {
    floor_u * zone_floor_area * 1.2  // 900-series WITH HVAC
} else {
    floor_u * zone_floor_area          // 900FF gets NO multiplier!
};

The 1.2 multiplier is NOT applied to free-floating cases:

• is_900_series_hvac is false when case_id.contains("FF") (line 700-702)
• 900FF: h_tr_floor = 0.190 * 48 * 1.0 = 9.12 W/K
• 900 with HVAC: h_tr_floor = 0.190 * 48 * 1.2 = 10.94 W/K

Floor Construction

Both insulated_floor() and high_mass_floor() have the same U-value (0.190 W/m²K) but different thermal mass:

• High mass floor: ~200,000 J/m²K per layer (concrete)
• Insulated floor: ~20,000 J/m²K per layer (wood/fiberglass)

Ground Temperature

src/sim/boundary.rs:85-147 - Constant 10°C ground temperature:

pub struct ConstantGroundTemperature {
    temperature: f64,  // Default: 10°C
}

Investigation Findings

Ground coupling is NOT the primary cause - at ~9 W/K conductance with 10°C ground, it provides at most ~273W of heat loss, insufficient to cause 80°C+ temperatures.

However, the floor thermal mass may be double-counted:

• Floor thermal mass is included in total_thermal_cap (line 747-757)
• Floor may also be double-counted in h_tr_ms calculation (line 804) using the "half-insulation rule"

Proposed Fix

1. Verify floor thermal mass is not double-counted in total_thermal_cap and h_tr_ms
2. Consider whether 900FF should get the 1.2 multiplier (but this would be case-specific)

Files to Investigate

• src/sim/thermal_model_core.rs:704-712 - h_tr_floor calculation
• src/sim/thermal_model_core.rs:747-757 - total_thermal_cap calculation
• src/sim/thermal_model_core.rs:759-804 - h_tr_ms physics calculation