PowerFlow Module

Overview

The PowerFlow module provides comprehensive power flow analysis capabilities for electrical power systems. It includes implementations of various power flow algorithms including AC Newton-Raphson, DC power flow, and hybrid power flow for integrated AC/DC systems. The module supports both balanced and unbalanced power flow analysis with various features for handling different system components and control modes.

Flow Chart

PowerFlow flow chart

Core Features

AC power flow using Newton-Raphson method
DC power flow analysis
Hybrid power flow for integrated AC/DC systems
Unbalanced power flow analysis
GPU-accelerated computations
Support for ZIP load models
PV system integration
Comprehensive results processing and reporting

Main Functions

Power Flow Execution

`runpf(case, opt)`

Main function to execute AC power flow calculations.

Arguments:

case: Power system case data
opt: Options dictionary containing power flow parameters

Returns:

Results structure containing power flow solution

`rundcpf(case, opt)`

Executes DC power flow analysis.

Arguments:

case: Power system case data
opt: Options dictionary containing DC power flow parameters

Returns:

Results structure containing DC power flow solution

`runhpf(jpc, opt)`

Runs hybrid power flow calculation for integrated AC/DC systems.

Arguments:

jpc: MATPOWER-style power flow case structure
opt: Options dictionary containing hybrid power flow parameters

Returns:

Results structure containing hybrid power flow solution

`runupf(case, jpc_3ph, gs_eg, bs_eg, opt)`

Executes unbalanced power flow analysis on unsymmetrical load nodes.

Arguments:

case: Power system case data
jpc_3ph: Three-phase power flow case structure
gs_eg, bs_eg: Shunt conductance and susceptance
opt: Options dictionary

Returns:

Results structure containing unbalanced power flow solution

Solution Algorithms

`newtonpf(baseMVA, bus, gen, branch, Ybus, Yf, Yt, V0, ref, pv, pq, mpopt)`

Solves power flow using Newton-Raphson method.

Arguments:

baseMVA: Base MVA
bus, gen, branch: System component data
Ybus, Yf, Yt: Admittance matrices
V0: Initial voltage vector
ref, pv, pq: Bus type indices
mpopt: MATPOWER options structure

Returns:

Voltage solution and convergence information

`newtondcpf(baseMVA, bus, branch, Bbus, Bf, Pbusinj, Pfinj, Va0, ref, pv, pq, mpopt)`

Solves DC power flow using Newton method.

`adaptive_damped_newton(baseMVA, bus, gen, branch, Ybus, Yf, Yt, V0, ref, pv, pq, mpopt)`

Implements adaptive damped Newton method for improved convergence.

`currentinjectionpf(baseMVA, bus, gen, branch, Ybus, Yf, Yt, V0, ref, pv, pq, mpopt)`

Solves power flow using current injection method.

System Construction

`makeYbus(baseMVA, bus, branch)`

Builds the bus admittance matrix (Ybus) and branch admittance matrices (Yf, Yt).

Arguments:

baseMVA: Base MVA
bus: Bus data matrix
branch: Branch data matrix

Returns:

Ybus: Bus admittance matrix
Yf: From-bus branch admittance matrix
Yt: To-bus branch admittance matrix

`makeBdc(baseMVA, bus, branch)`

Builds the DC power flow matrices.

Arguments:

baseMVA: Base MVA
bus: Bus data matrix
branch: Branch data matrix

Returns:

Bbus: Nodal susceptance matrix
Bf: Branch susceptance matrix for line flows
Other DC model matrices

`makeSbus(baseMVA, bus, gen, Vm, load, pvarray; dc=false, Sg=nothing, return_derivative=false)`

Builds the vector of complex bus power injections.

Arguments:

baseMVA: Base MVA
bus: Bus data matrix
gen: Generator data matrix
Vm: Voltage magnitude vector
load: Load data matrix
pvarray: PV array data
Additional optional parameters

Returns:

Vector of complex bus power injections

`calculate_line_parameter(net, jpc, sequence, opt)`

Calculates line parameters for the branch matrix based on the specified sequence.

Arguments:

net: Network data structure
jpc: MATPOWER-style power flow case structure
sequence: Sequence type (1 for positive, 2 for negative, 0 for zero)
opt: Options dictionary

Returns:

Updated jpc structure with branch parameters

Bus and Generator Construction

`calculate_bus(net, jpc, sequence, slack_bus, opt)`

Builds the bus matrix for power flow calculations.

`build_gen(net, jpc)`

Builds the generator matrix for power flow calculations.

Bus Type Identification

`bustypes(bus)`

Identifies bus types (reference, PV, PQ) for AC power flow.

`dcbustypes(bus)`

Identifies bus types for DC power flow.

Jacobian Calculation

`dSbus_dV(Ybus, V, vcart)`

Computes partial derivatives of power injection w.r.t. voltage.

Arguments:

Ybus: Bus admittance matrix
V: Complex bus voltage vector
vcart: Boolean flag for cartesian coordinates

Returns:

Partial derivatives of power injection

Solution Processing

`pfsoln(baseMVA, bus, gen, branch, Ybus, Yf, Yt, V, ref, pv, pq, opt)`

Updates power flow solution with computed voltage values and calculates generator outputs and branch flows.

`dcpfsoln(baseMVA, bus0, gen0, branch0, Bbus, Bf, Pbusinj, Pfinj, Va, success, et, opt)`

Updates DC power flow solution.

`merge_results(results)`

Merges power flow results from multiple calculations.

`process_result(results, case, opt)`

Processes power flow calculation results and generates summary.

`process_pv_acsystem(case, results, opt)`

Processes PV AC systems and integrates them into the power flow solution.

Utility Functions

`julinsolve(A, b, solver)`

Solves linear system Ax = b using specified solver.

`total_load(bus, load)`

Calculates total load in the system.

`eliminate_element(mpc, element_type, element_id)`

Eliminates a specific element from the power system model.

Results Reporting

`pf_summary`

Generates comprehensive power flow summary reports including:

System summary
Bus data
Branch data
Generator data
Voltage violations
Branch violations
System losses

`compare_voltage_results(results, case, reference_file; tolerance_mag=1e-4, tolerance_ang=1e-3)`

Compares power flow calculation results with reference values from external tools like ETAP.

GPU Support

The module includes GPU-accelerated versions of key functions for improved performance on compatible hardware:

newtonpf_gpu
makeSbus_gpu
makeSdzip_gpu

Settings and Options

`settings()`

Defines default settings for power flow calculations including:

Convergence tolerance
Maximum iterations
Algorithm selection
Output verbosity
Enforcement options
Initialization methods

Usage Example

using HyDistFlow.PowerFlow

# Load case data
case = case3()

# Set options
opt = PowerFlow.settings()
opt[:pf_alg] = :NR  # Newton-Raphson algorithm
opt[:verbose] = 1

# Run power flow
results = PowerFlow.runpf(case, opt)

# Process results
PowerFlow.process_result(results, case, opt)

Advanced Features

Hybrid Power Flow

The module supports hybrid power flow for integrated AC/DC systems with various converter control modes:

Vθ control
PQ control
PV control
Vdc-Q control
Vdc-Vac control
Droop control

Unbalanced Power Flow

Unbalanced power flow analysis is supported for unsymmetrical load nodes:

Identifies unbalanced nodes in the system
Finds interface branches between balanced and unbalanced nodes
Solves the unbalanced power flow

ZIP Load Models

The module supports ZIP (constant impedance, constant current, constant power) load models through the makeSdzip function.