kersulis

        - module MatpowerCases
        - using MAT
        - 
        - export loadcase,casenames
        - 
        - """ Return a Dict containing power system data
        - in MATPOWER's format.
        - Return this list of cases by running casenames().
        - """
        - function loadcase(caseName::ASCIIString; describe=true)

kersulis

using Requests, Graphs, ProgressMeter

function surfer(root,n=100)
    # Crawling mechanics: hyperlink regex and
    # filter used for skipping certain links
    rexp = r"https?://[^\"|<|>|;|'|)| |…]+"
    # Focus on links that contribute to the graph by skipping
    # all links that contain the following strings:
    skip = [".gif";".jpg";".jpeg";".pdf";".css";".js";".asp";".mwc";".ram";
        ".cgi";"lmscadsi";"cybernet";"w3.org";"google";"yahoo";

kersulis

"""
    createY(f,t,x [,r,b]) -> Y
Create an admittance matrix for AC power flow.
All inputs are real. The output matrix is real if no line
resistances are provided (DC case), and complex otherwise.

* `f`,`t`: vectors encoding all lines (fi,ti)
* `x`: per-unit reactance xi for all lines
* `r`: per-unit resistance ri for all lines
* `b`: per-unit susceptance bi for all lines

kersulis

"""
Calculate injection shift factor matrix.
Each row corresponds to a line in the network.
Each column corresponds to a node.
Credit to Jonathon Martin for derivation.

Inputs:
* `Y`: full admittance matrix
* `lines`: vector of tuples; each tuple encodes a line as (i,j)
* `ref`: index of angle reference bus

kersulis

"""
Naive LU decomposition (no pivoting) from
[Bindel's CS6210 class notes](http://www.cs.cornell.edu/~bindel/class/cs6210-f12/notes/lec09.pdf).

Beware A[1,1] = 0...
"""
function mylu(A)
    m,n = size(A)
    for j = 1:n-1
        A[j+1:n,j] = A[j+1:n,j]/A[j,j]

kersulis

"""
Given a permutation vector, return the corresponding permutation matrix such that
`A[p,:] = P*A`.
See [Wikipedia](https://en.wikipedia.org/wiki/Permutation_matrix#Definition).
"""
function pvec2mat(p)
    n = length(p)
    P = zeros(n,n)
    for i in 1:n
        P[i,:] = ej(n,p[i])

kersulis

"""
LTC model used by [Baghsorkhi2015](http://deepblue.lib.umich.edu/handle/2027.42/113584)
(see section 5.3.2).
"""
type DBODE
    "index of bus whose voltage is regulated"
    bus::Int64
    "index of line whose tap is influenced"
    line::Int64
    "voltage setpoint [pu]"

kersulis

import numpy as np

def procrustes(shapes):
    """
    Syntax:     ashapes, sm = procrustes(shapes)

    Inputs:     shapes is a n x p x d array, where n is the number of
                shapes, p is the number of points per shape, and d is
                the dimension of each point (d=2 for xy plane, etc.).
                Thus, shapes[0,5,:] contains the d coordinates for the

kersulis

# Sonic Pi v2.10
use_bpm 130

notes = (ring :b3, :fs3, :g3, :c4)
notes_sliced = (ring :b3, :as3, :a3, :as3)

define :long do
  4.times do
    with_fx :bitcrusher, mix: 0.2 do
      with_synth :hoover do

kersulis

using Optim
import Optim.retract!
import Optim.project_tangent!

struct SphereSubspaceComplementIntersection <: Manifold
    U::Matrix
    proj::Matrix
end

function SphereSubspaceComplementIntersection(U)