com.interpss.core.sparse.impl

Class SparseEqnDoubleImpl

java.lang.Object

com.interpss.core.sparse.impl.AbstractSparseEquation

com.interpss.core.sparse.impl.SparseEqnDoubleImpl

All Implemented Interfaces:

ISparseEquation, ISparseEqnDouble

Direct Known Subclasses:

SparseEqnMatrix2x2Impl

public class SparseEqnDoubleImpl
extends AbstractSparseEquation
implements ISparseEqnDouble

Sparse Equation of data type double for solving the [A]X=B problem. To outside, the index number is from 0 to n-1. When [A] is a square matrix, LU decomposition method is used to solve the eqn When [A] is a non-square matrix, QR decomposition method is used to provide a Least Square or Weighted Least Square solution

Nested Class Summary

Nested classes/interfaces inherited from interface org.interpss.numeric.sparse.base.ISparseEquation

ISparseEquation.IndexType

Constructor Summary

Constructors

Constructor and Description
SparseEqnDoubleImpl(int n)
SparseEqnDoubleImpl(int rows, int cols, int order)
SparseEqnDoubleImpl(int rows, int cols, int order, boolean weightedLS)

Method Summary

Modifier and Type	Method and Description
void	addToAij(double x, int i, int j) Add the aij element to the matrix.
void	addToBi(double bi, int i) Set bi element.
double	err(double[] x)
boolean	factorization() Decomposition of the matrix.
boolean	factorization(double tolerance) Decomposition of the matrix.
double	getAij(int i, int j) Get the aij element.
SparseEqnDataType.DblAii	getElem(int i)
int	getTotalElements() Get the total elements in the matrix
double	getXi(int i) Get the bi element.
double[]	getXVector() Get the b/x vector
void	increaseDimension() Increase matrix dimension by 1.
void	reset() Reset the matrix to zero - set all aii and bi to 0.0 and clear the aijList
void	setAij(double x, int i, int j) Set the aij element.
void	setB2Unity(int i) Set all b elements to 0.0 and bi = 1.0, a unit vector.
void	setB2Zero() Set all b elements to 0.0 and bi = 1.0, a unit vector.
void	setBi(double bi, int i) Set bi element.
void	setBVector(double[] b) Set b/x vector.
void	setElem(int i, SparseEqnDataType.DblAii aii)
void	setToZero() Reset all matrix element to zero, preserve the matrix structure.
void	solveEqn() Solve the [A]X = B problem
double[]	solveEqn(double[] b) Solve the [A]X = B problem
java.lang.String	toString() Get the string representation of the matrix object

Methods inherited from class com.interpss.core.sparse.impl.AbstractSparseEquation

getBusId, getDimension, getDimension, getZeroAii_row, isFactorized, isSquareMatrix, setBusId, setFactorized, setZeroAii_row, solveEqn

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, wait, wait, wait

Methods inherited from interface org.interpss.numeric.sparse.base.ISparseEquation

getBusId, getDimension, getDimension, getZeroAii_row, isFactorized, isSquareMatrix, setBusId, setFactorized, solveEqn

Constructor Detail

SparseEqnDoubleImpl

public SparseEqnDoubleImpl(int n)

SparseEqnDoubleImpl

public SparseEqnDoubleImpl(int rows,
                           int cols,
                           int order)

SparseEqnDoubleImpl

public SparseEqnDoubleImpl(int rows,
                           int cols,
                           int order,
                           boolean weightedLS)

Method Detail

increaseDimension

public void increaseDimension()

Description copied from interface: ISparseEquation

Increase matrix dimension by 1. B[n] = 0.0, aii(n)=1.0, aij(n)=0.0

Specified by:

increaseDimension in interface ISparseEquation

Parameters:

n - matrix dimension

addToAij

public void addToAij(double x,
                     int i,
                     int j)

Add the aij element to the matrix. If aij exists, aij += x.

Specified by:

addToAij in interface ISparseEqnDouble

Parameters:

x - the aij element

i - the element row number

j - the element column number

getElem

public SparseEqnDataType.DblAii getElem(int i)

setElem

public void setElem(int i,
                    SparseEqnDataType.DblAii aii)

setAij

public void setAij(double x,
                   int i,
                   int j)

Set the aij element.

Specified by:

setAij in interface ISparseEqnDouble

Parameters:

x - the aij element

i - the element row number (1-n convention)

j - the element column number (1-n convention)

getAij

public double getAij(int i,
                     int j)

Description copied from interface: ISparseEqnDouble

Get the aij element.

Specified by:

getAij in interface ISparseEqnDouble

Parameters:

i - the element row number

j - the element column number

setB2Zero

public void setB2Zero()

Set all b elements to 0.0 and bi = 1.0, a unit vector.

Specified by:

setB2Zero in interface ISparseEquation

Overrides:

setB2Zero in class AbstractSparseEquation

Parameters:

i - the element row number

setB2Unity

public void setB2Unity(int i)

Set all b elements to 0.0 and bi = 1.0, a unit vector.

Specified by:

setB2Unity in interface ISparseEquation

Parameters:

i - the element row number

getXi

public double getXi(int i)

Get the bi element.

Specified by:

getXi in interface ISparseEqnDouble

Parameters:

i - the element row number

Returns:

the bi element

getXVector

public double[] getXVector()

Get the b/x vector

Parameters:

i - the element row number

Returns:

the bi element

setBVector

public void setBVector(double[] b)

Set b/x vector.

Parameters:

b - the b vector

setBi

public void setBi(double bi,
                  int i)

Set bi element.

Specified by:

setBi in interface ISparseEqnDouble

Parameters:

bi - the bi element

i - row number (1-n convention)

addToBi

public void addToBi(double bi,
                    int i)

Set bi element.

Specified by:

addToBi in interface ISparseEqnDouble

Parameters:

bi - the bi element

i - row number

reset

public void reset()

Reset the matrix to zero - set all aii and bi to 0.0 and clear the aijList

Specified by:

reset in interface ISparseEquation

setToZero

public void setToZero()

Reset all matrix element to zero, preserve the matrix structure.

Specified by:

setToZero in interface ISparseEquation

solveEqn

public void solveEqn()
              throws IpssNumericException

Solve the [A]X = B problem

Specified by:

solveEqn in interface ISparseEquation

Throws:

IpssNumericException

solveEqn

public double[] solveEqn(double[] b)
                  throws IpssNumericException

Solve the [A]X = B problem

Specified by:

solveEqn in interface ISparseEqnDouble

Throws:

IpssNumericException

factorization

public boolean factorization()
                      throws IpssNumericException

Decomposition of the matrix.

Specified by:

factorization in interface ISparseEquation

Overrides:

factorization in class AbstractSparseEquation

Parameters:

tolerance - the tolerance for matrix singular detection

Returns:

if succeed return true.

Throws:

IpssNumericException

factorization

public boolean factorization(double tolerance)
                      throws IpssNumericException

Decomposition of the matrix.

Specified by:

factorization in interface ISparseEquation

Overrides:

factorization in class AbstractSparseEquation

Parameters:

tolerance - the tolerance for matrix singular detection

Returns:

if succeed return true.

Throws:

IpssNumericException

getTotalElements

public int getTotalElements()

Get the total elements in the matrix

Specified by:

getTotalElements in interface ISparseEquation

Returns:

the total elements

toString

public java.lang.String toString()

Get the string representation of the matrix object

Overrides:

toString in class java.lang.Object

Returns:

the string representation

err

public double err(double[] x)