External inflows#

Note

Engine: OpenSWMM 6 — refactored.

External inflows, dry-weather flows, RDII, unit hydrographs, and inflow-area decay all live behind solver.inflows.

Reference: openswmm_inflows.h.

Quickstart#

from openswmm.engine import Solver

with Solver("model.inp") as s:
    # External inflow — node accepts id or index.
    s.inflows.add_external("J1", "FLOW", ts_name="rain1")

    # Dry-weather flow.
    s.inflows.add_dwf(
        "J1", "FLOW",
        avg_value=0.5,
        hourly_pattern="DLY1",
    )

    # RDII inflow with unit hydrograph.
    s.inflows.add_rdii("J1", uh_name="UH1", area=2.5)

    # Inspect counts.
    print(s.inflows.external_count, s.inflows.dwf_count, s.inflows.rdii_count)

    # Per-row read (RDII, hydrographs, decay — C API supports get).
    rdii = s.inflows.get_rdii(0)
    print(rdii.node_index, rdii.uh_name, rdii.area)

Note

The ts_name / *_pattern arguments must reference a time series or pattern that already exists in the model. Adding an inflow is allowed in any lifecycle state, but creating the referenced time series or pattern is only valid in BUILDING or OPENED state. If you need to create them, open the solver explicitly and add them before initialize() / start() — see the lifecycle note in Tables (time series, curves, patterns).

Methods#

External inflows ([INFLOWS])#

  • add_external(node, constituent, *, ts_name, type, m_factor, s_factor, baseline, pattern)

  • external_count property

Dry-weather flow ([DWF])#

  • add_dwf(node, constituent, *, avg_value, monthly_pattern, daily_pattern, hourly_pattern, weekend_pattern)

  • dwf_count property

RDII ([RDII])#

  • add_rdii(node, uh_name, area)

  • get_rdii(idx) -> RDIIEntry

  • rdii_count property

Unit hydrographs ([HYDROGRAPHS])#

  • add_hydrograph(uh_name, month, response, r, t, k, *, dmax, drecov, dinit)

  • get_hydrograph(idx) -> HydrographEntry

  • hydrograph_count property

  • add_hydrograph_gage(uh_name, gage_name) / get_hydrograph_gage(idx) -> HydrographGageEntry

  • hydrograph_gage_count / hydrograph_group_count properties

  • get_hydrograph_group_id(idx)

RDII decay ([RDII_DECAY])#

  • add_rdii_decay(uh_name, response, k_dep, k_0, k_T, T_ref, theta_rec, T_freeze)

  • get_rdii_decay(idx) -> RDIIDecayEntry

  • rdii_decay_count property

All node/link/subcatchment/gage arguments accept int | str.

C API constraint#

The C side only exposes add + count for external inflows and DWF — there is no per-row delete / set / get. The Python view doesn’t pretend to be a MutableSequence for those families. RDII, hydrographs, and RDII decay all have get accessors so per-row reading works.

Routing interface files ([FILES])#

Two separate models can be chained through legacy SWMM routing interface files: an upstream model declares SAVE OUTFLOWS "file" in its [FILES] section and writes one row per outlet node per reporting step; a downstream model declares USE INFLOWS "file" and receives those flows (and pollutant loads) as node lateral inflows, interpolated between file periods. Node names in the file must match node ids in the receiving model; flows are converted from the file’s declared units. Paths can also be set programmatically via ModelBuilder.files_set() with the "INFLOWS_PATH" / "OUTFLOWS_PATH" keys. A missing or malformed inflows file causes Solver.start() to raise (legacy errors 351/353/357).

The other [FILES] slots behave like legacy SWMM: SAVE RUNOFF exports each runoff substep to a binary runoff interface file and USE RUNOFF replays it in place of the runoff computation; SAVE RDII exports the computed RDII inflows (legacy SWMM5-RDII binary) and USE RDII overrides the internal unit-hydrograph computation entirely — the file’s flows (binary or legacy text format) become the RDII inflows. USE/SAVE RAINFALL (the collated binary rain file) is not implemented; the engine warns and reads gage data files directly.

See also#