UESGraphs pandapipes simulation Pipeline

Architecture Documentation

Version: 2.2.0 Last Updated: April 2026

Table of Contents

  1. Overview

  2. Pipeline Architecture

  3. Core Entry Point

  4. Parameter Assignment Flow

  5. Demand and Time-Series Handling

  6. pandapipes Model Generation

  7. File and Folder Structure

  8. Validation and Error Handling

  9. Extension Guide

Overview

The UESGraphs pandapipes simulation pipeline enables the execution of thermo-hydraulic network simulations based on the graph-structure.

Key Design Principles

  1. Graph as Single Source of Truth: All topology, parameters and time-series references are stored on the UESGraph before model generation

  2. Excel-based Configuration: Uniform parameters (pipes, supply, demand, simulation setup) are read from Excel sheets

  3. Fail Early: Missing files, invalid graphs or inconsistent parameters abort the pipeline with explicit error messages

  4. Traceability: Every major processing step persists intermediate graph states to JSON

Technology Stack

  • Graph Representation: NetworkX-based UESGraph

  • Simulation Engine: pandapipes

  • Configuration: Excel (openpyxl)

  • Time-Series Data: CSV files

Pipeline Architecture

High-Level Flow

UESGraph / JSON
     ↓
uesgraph_to_pandapipes
     ↓
┌─────────────────────────────────┐
│ 1. Validate input paths         │
│ 2. Load simulation settings     │
│ 3. Normalize UESGraph           │
│ 4. Assign demands               │
│ 5. Assign pipe parameters       │
│ 6. Assign supply parameters     │
│ 7. Assign demand parameters     │
│ 8. Load ground temperature      │
│ 9. Generate pandapipes model    │
└─────────────────────────────────┘
     ↓
pandapipes Results

Core Entry Point

### uesgraph_to_pandapipes(…)

Location: systemmodels_pp/utilities.py Responsibility: End-to-end orchestration of pandapipes simulations Inputs: UESGraph or JSON path, Excel config, demand CSVs, ground temperature data Outputs: Timestamped directory with pandapipes files

Parameter Assignment Flow

### Execution Order

1. Demand data assignment
2. Pipe parameters (Excel: "Pipes")
3. Supply parameters (Excel: "Supply")
4. Demand parameters (Excel: "Demands")

—

### Pipe Parameters

Assigned to edges:

  • dIns

  • kIns

  • roughness

  • ground_depth

  • sections

Source: Excel sheet “Pipes” Missing values fall back to hardcoded defaults.

—

### Supply Parameters

Applied to building nodes flagged as supply:

`python is_supply_heating == True `

Assigned attributes:

  • TIn

  • TReturn

  • dpIn

  • pReturn

  • dpFlow

Unit conversions (Pa → bar) are handled explicitly in the pipeline.

The total number of supplies is stored as:

`python uesgraph.graph["number_of_supplies"] `

—

### Demand Parameters

Applied to non-supply building nodes:

  • dTDesign

  • TReturn

  • cp_default

Additionally, network-wide defaults are stored on graph level:

`python uesgraph.graph["dT_Net"] uesgraph.graph["cp_default"] `

—

Demand and Time-Series Handling

### Demand Assignment

Demand CSVs are mapped via:

`python assign_demand_data(uesgraph, {"heating": input_heating, "cooling": input_cooling, "dhw": input_dhw}) `

Result:

  • Time-series are attached as graph node attributes

  • Naming conventions are preserved for downstream processing

### Ground Temperature

Ground temperature CSV is loaded once and later sliced by ground depth, defined in the simulation Excel:

`python ground_temp_list = ground_temp_df[ground_depth].tolist() `

—

Pandapipes Model Generation

### Model Directory Creation

Each simulation run creates a timestamped folder:

workspace/
  └── models/
      └── SimYYYYMMDD_HHMMSS/

### Model Generation

The final pandapipes model is created via:

`python generate_simulation_model(uesgraph, sim_name, sim_params, ground_temp_list, sim_model_dir) `

See also

  • Depending on the simulation mode (static, dynamic, transient), the simulation will be executed differently. For more details on the different simulation modes, see Dynamic Simulation with pandapipes

Responsibilities of `generate_simulation_model`:

  • Convert UESGraph → pandapipes net

  • Create pipes, junctions, sinks, sources

  • Assign thermal & hydraulic parameters

  • Attach time-series data

  • Execute simulation

Used components:

  • pandapipes.pipe

  • pandapipes.heat_consumer

  • pandapipes.circ_pump_pressure

  • pandapipes.circ_pump_mass (if multiple supplies)

  • pandapipes.junction

Solver Settings for static simulation:

Solver Settings for dynamic simulation:

  • mode: ‘hydraulics’

  • iter: 100

  • For more detail on dynamic simulation using pandapipes, see Dynamic Simulation with pandapipes - Technical details on dynamic simulations with pandapipes.

Solver Settings for transient simulation:

—

File and Folder Structure

workspace/
  ├── uesgraphs_origin.json
  ├── uesgraphs_with_demand.json
  ├── uesgraphs_simplified.json
  └── models/
      └── Sim20250115_101530_MySimulation/
          ├── uesgraphs.json
          ├── uesgraphs_return.json
          ├── [component models]
          └── setup_params.csv   (optional)

—

Validation and Error Handling

### Validation Levels

  1. File System - Missing files → abort

  2. Graph Consistency - Missing edge names → auto-generated - Invalid graph type → abort

  3. Excel Parsing - Missing sheets → abort - Missing values → defaults or error (depending on context)

  4. Simulation Generation - pandapipes errors are logged and re-raised

### Logging Strategy

  • DEBUG: Detailed processing steps

  • INFO: Pipeline milestones

  • ERROR: Abort conditions

—

Extension Guide

### Adding New Parameters

  1. Add attribute to UESGraph (node or edge)

  2. Reference or override via Excel

  3. Ensure generate_simulation_model consumes the parameter

—

This document describes the architecture of the UESGraphs pandapipes simulation pipeline.