Skip to content

Latest commit

 

History

History
204 lines (171 loc) · 9.88 KB

MATPOWER-Release-Notes-8.0b1.md

File metadata and controls

204 lines (171 loc) · 9.88 KB
Raw

What's New in MATPOWER 8.0b1

Released Dec 22, 2022

Below is a summary of the changes since version 7.1 of MATPOWER. See the CHANGES.md file for all the gory details. For release notes for previous versions, see Appendix H of the MATPOWER User's Manual.

Major Redesign:

MATPOWERÊ8 introduces a major redesign and rewrite of all of the MATPOWER internals in the form of the flexible, all-new MATPOWER object-oriented core architecture (MP-Core) and new two user-level frameworks to access it. (Previously developed under the name MP-Element in a separate repository at https://github.com/MATPOWER/mp-element.)

  • MP-Core -- Provides unparalleled flexibility and customization capabilities at all levels. Built around an explicit three-layer modeling structure, with data, network and mathematical model objects and a task object to manage them. Facilitates new modeling (e.g. unbalanced multiphase elements, FACTS devices, etc.), new controls (e.g. optimization of transformer taps, PAR angles, etc.), new problem formulations, and more.
  • Flexible Framework -- Provides new top-level functions (note underscores) run_pf(), run_cpf(), run_opf() for running power flow (PF), continuation power flow (CPF) and optimal power flow (OPF), along with new MATPOWER Extension API for user access to the full customization capability of MP-Core.
  • Legacy Framework -- Allows MP-Core modeling to be used internally by legacy runpf(), runcpf(), runopf(), etc. Facilitates use of legacy test suite, but is restricted to legacy customization mechanisms.

See the new MATPOWER Developer's Manual and MATPOWER Technical NoteÊ5 for details of the new architecture. The User's Manual has not yet been updated for the flexible framework.

The features based on MP-Core are available under MATLAB 9.1 or Octave 6.2 or newer, where the legacy framework uses MP-Core's new modeling by default for:

  • rundcpf -- DC power flow
  • rundcopf -- DC optimal power flow
  • runpf -- AC power flow, for all except radial and hybrid Newton-Raphson formulations/solvers
  • runcpf -- AC continuation power flow
  • runopf -- AC OPF, for solvers MIPS, fmincon, IPOPT, and Artelys Knitro, for all formulations

Under older versions of MATLAB or Octave, MATPOWER automatically reverts to the legacy core code, which is still included. The legacy core can also be selected manually on newer versions with the 'exp.use_legacy_core' option or by disabling MP-Core in the legacy framework via have_feature('mp_core', 0).

New Three-Phase and Hybrid Proof-of-Concept Examples

Prototype examples of PF, CPF, and OPF, built using the new MATPOWER Extension API, for:

  • unbalanced three-phase models
  • hybrid transmission (balanced single-phase) / distribution (unbalanced three-phase) models
  • multiple problem formulations

Note: These are proof-of-concept only, with much work remaining to define a full set of three-phase model elements and their respective parameters and data formats. But they do work!

New Features:

  • MP-Test 8.0b1 includes new functions for testing string values and text file contents. For details, see the MP-Test 8.0b1 release notes.

  • MIPS 1.5 adds to mplinsolve() the ability to save an LU factorization and reuse it to solve for additional right-hand sides. For details, see the MIPS 1.5 release notes (also in Appendix C in the MIPS User's Manual).

  • MP-Opt-Model 4 adds support for:

    • New class of problems -- parameterized nonlinear equations (PNE), providing generalized continuation power flow (CPF) capabilities to MP-Core.
      • Predictor/corrector numerical continuation method for tracing solution curves for PNE problems.
      • Plotting of solution curves.
      • User-defined event functions and callback functions.
      • Warm-start capabilities.
    • Artelys Knitro 13.1 and more.

    For details, see the release notes for MP-Opt-Model 4.0 and 4.1 (also in Appendix C in the MP-Opt-Model User's Manual).

  • MOST 1.2 adds calculation of expected temporal locational marginal price (TLMP), includes transitions into first period in ramping reserves, and more. For details, see the MOST 1.2 release notes (also in Appendix B in the MOST User's Manual).

  • New options:

    • New AC power flow solver based on fsolve() function, selected by setting 'pf.alg' option to 'FSOLVE'. (fsolve() is part of the MATLAB Optimization Toolbox and included in Octave. Supported via MP-Core only.)
    • New Implicit Z-bus Gauss method power flow for distribution systems, selected by setting pf.alg option to 'ZG'. (Support for PV buses is very limited.)
    • New 'exp.use_legacy_core' option to bypass MP-Core and force use of legacy core code for runpf(), runcpf(), runopf().

  • New functions/methods:

    • have_feature_mp_core - Determines availability of MP-Core.
    • run_mp - Top-level function for running any task (PF, CPF, OPF) with the new MP-Core and flexible framework.
    • run_pf - Wrapper around run_mp for running PF.
    • run_cpf - Wrapper around run_mp for running CPF.
    • run_opf - Wrapper around run_mp for running OPF.
    • MP-Opt-Model 4.1
      • pnes_master() unified interface for parameterized nonlinear equation (PNE) solvers
      • mpopt2pneopt() creates or modifies an options struct for pnes_master() from a MATPOWER options struct
      • other functions used to implment PNE functionality
    • MP-Test 8.0b1
      • t_str_match() -- test that a string matches expected value
      • t_file_match() -- test that the content of a text files matches that of another file

New Case Files:

  • Two new European case files. Thanks to Florin Capitanescu.
    • case60nordic -- 60-bus Nordic case
    • case8387pegase -- 8387-bus PEGASE case

New Documentation:

Other Improvements:

  • MP-Opt-Model object is now used for power flow as well as OPF and is added as om field to power flow results struct.
  • New MATPOWER Docker image (named matpower/matpower) is based on the official GNU Octave image (gnuoctave/octave) and is available for multiple MATPOWER and Octave versions.
  • Each power flow is initialized with the solved voltages of the previous one when changing PV buses to PQ during Q limit enforcement. Thanks to Tostado-VŽliz, Kamel, Jurado.
  • The makePTDF() function can now optionally use a different slack distribution for each bus by specifying the slack input as a matrix. Thanks to Jon Martinez Corral.
  • Automatically reduce order of polynomial generator costs with higher order coefficients equal to zero. Allows the DC OPF to solve cases, e.g. with cubic costs where the 3rd order term is 0, such as cases exported by PowerWorld. Thanks to Rajesh Mookerjee.

Bugs Fixed:

  • Fatal bug in int2ext() when called with mpopt and an int2ext user callback function.
  • Bug affecting AC OPF initialization for cases with piecewise linear costs and cartesian voltage representations. Fix results in improved convergence for affected cases.
  • Fix generator voltage set points in case9target to match case9.
  • A vector-valued label passed to apply_changes() now throws a useful error.
  • Fix bug in previously undocumented feature of makePTDF() where the slack input is a matrix. Thanks to Jon Martinez Corral.

Incompatible Changes:

  • Removed unused mpopt argument from opf_gen_cost_fcn() inputs.
  • Remove deprecated functions:
    • d2AIbr_dV2() -- use dA2br_dV2() instead.
    • d2ASbr_dV2() -- use dA2br_dV2() instead.
    • Deprecated methods of @opf_model:
      • add_constraints() -- use the corresponding one of the following methods instead: add_lin_constraint(), add_nln_constraint(), or init_indexed_name().
      • add_costs() -- use the corresponding one of the following methods instead: add_quad_cost(), add_nln_cost(), add_legacy_cost(), or init_indexed_name().
      • add_vars() -- use the corresponding one of the following methods instead: add_var(), or init_indexed_name().
      • build_cost_params() -- no longer needed, incorporated into params_legacy_cost().
      • get_cost_params() -- use params_legacy_cost() instead.
      • getv() -- use params_var() instead.
      • linear_constraints() -- use params_lin_constraint() instead.
  • Remove deprecated option:
    • 'opf.init_from_mpc' -- use 'opf.start' instead.