Vectors: Definitions, Spaces, and Linear Transformations

 A vector is a quantity that has both magnitude and direction. It is typically represented by an arrow whose length indicates the magnitude and whose orientation in space indicates the direction. Vectors are used in various fields of science and engineering to describe phenomena that have a direction associated with them, such as velocity, force, and displacement. Unlike scalars, which have only magnitude, vectors provide a more complete representation of directional quantities.

Linear Combination of Vectors:

Measurement Equations: State estimation involves determining the voltage magnitude and phase angle at various buses in a power system. The measurements (such as power injections, power flows, and voltage magnitudes) can be represented as linear combinations of the state variables. Solving these linear combinations provides an estimate of the system state.

Sensitivity Analysis: In contingency analysis, the impact of potential failures (e.g., line outages, generator failures) on the power system is studied. The changes in power flows and voltages due to these contingencies can be expressed as linear combinations of pre-contingency states and sensitivity factors.

Aggregate Load Representation: The total load on a power system can be represented as a linear combination of individual loads. This aggregation is essential for load forecasting and demand-side management.

Linearly Dependent Vectors: 

Vectors are said to be linearly dependent if there exist scalars, not all zero, such that a linear combination of these vectors results in the zero vector. Mathematically, vectors $\mathbf{v}_1, \mathbf{v}_2, \dots, \mathbf{v}_n$ are linearly dependent if there exist scalars $c_1, c_2, \dots, c_n$ not all zero, such that:

$$c_1 \mathbf{v}_1 + c_2 \mathbf{v}_2 + \cdots + c_n \mathbf{v}_n = \mathbf{0}$$

Linearly Independent Vectors:

Vectors are linearly independent if the only way a linear combination of these vectors can result in the zero vector is if all the scalars are zero. In other words, vectors $\mathbf{v}_1, \mathbf{v}_2, \dots, \mathbf{v}_n$are linearly independent if the equation: 

$$$$c1v1+c2v2+⋯+cnvn=0

implies that:

$$c_1=c_2=\cdots =c_n=0$$

A set of orthogonal vectors is linearly independent but not vice-versa.

Linear Independent and dependent Vectors applications in power system:

Jacobian Matrix: In power flow analysis (load flow studies), the Jacobian matrix, which represents partial derivatives of power equations with respect to voltage magnitudes and angles, must be non-singular for the Newton-Raphson method to converge. The non-singularity implies that the rows and columns of the Jacobian matrix are linearly independent, ensuring a unique solution for the system of equations.

Observability: State estimation involves determining the voltage magnitude and phase angle at various buses in a power system. For a system to be fully observable, the measurement matrix must have full rank, meaning its rows (or columns) must be linearly independent. This ensures that the system states can be uniquely determined from the available measurements.

Vector Space:

A vector space (also known as a linear space) is a collection of vectors, which are objects that can be added together and multiplied ("scaled") by numbers, called scalars in this context. Scalars are often taken from the field of real numbers, but they can also be complex numbers or any other field.

Mathematically, a vector space $V$ over a field $F$ is defined by the following axioms. These axioms must hold for all elements $\mathbf{u}, \mathbf{v}, \mathbf{w}$ in $V$ and all scalars $a, b$ in $F$:

1. Additive Associativity:

   $$$$(u+v)+w=u+(v+w)

2. Additive Commutativity:

   $$$$u+v=v+u

3. Additive Identity: There exists an element $\mathbf{0}$ in $V$, called the zero vector, such that:

   $$$$u+0=u

4. Additive Inverse: For every $\mathbf{u}$ in $V$, there exists an element $-\mathbf{u}$ in $V$ such that:

   $$$$u+(−u)=0

5. Scalar Multiplication Associativity:

   $$$$a(bu)=(ab)u

6. Distributivity of Scalar Multiplication with respect to Vector Addition:

   $$$$a(u+v)=au+av

7. Distributivity of Scalar Multiplication with respect to Field Addition:

   $$$$(a+b)u=au+bu

8. Multiplicative Identity: There exists an element $1$ in $F$ (the multiplicative identity of the field), such that:

   $$$$1u=u

These axioms ensure that vector spaces can be used to generalize a wide range of mathematical and physical concepts.

Vector Sub-Space: 

Sub Space means a space that can be covered by adding any two or more vectors or by multiplying with scaler. These two original vectors (vectors being used for the operation) should also belong to the same sub space and the result of the operation should also be within the space. If I represent this in mathematical symbols, it would be as follows. To be a Subspace, it must meet all these criteria. If only one of the criterial is not met, it cannot be a Sub space.

The mathematical definition of a subspace $W$ of a vector space $V$ over a field $F$ requires $W$ to satisfy three key conditions:

1. Non-empty: $W$ must contain at least the zero vector $\mathbf{0}$ of $V$. This ensures that $W$ is non-empty.

2. Closed under Addition: For all vectors $\mathbf{u}, \mathbf{v}$ in $W$, the sum $\mathbf{u} + \mathbf{v}$ must also be in $W$. Mathematically, if $\mathbf{u}, \mathbf{v} \in W$, then $\mathbf{u} + \mathbf{v} \in W$.

3. Closed under Scalar Multiplication: For every vector $\mathbf{u}$ in $W$ and every scalar $a$ in $F$, the scalar product $a\mathbf{u}$ must also be in $W$. That is, if $\mathbf{u} \in W$ and $a \in F$, then $a\mathbf{u} \in W$.

These conditions ensure that $W$ itself functions as a vector space, inheriting the structure and properties of $V$, but possibly with fewer dimensions or restricted to a certain subset of $V$'s elements. Subspaces are fundamental in various areas of mathematics and applied disciplines, providing a framework for constructing more complex vector spaces from simpler ones.

Basis and Dimension:

Basis of a Vector Space

A basis of a vector space is a set of vectors that satisfies two key properties:

1. Linear Independence: The vectors in the basis must be linearly independent, meaning no vector in the set can be written as a linear combination of the others.
2. Spanning or Generating: The set of vectors must span the entire vector space, meaning any vector in the space can be expressed as a linear combination of the vectors in the basis.

Mathematically, a set of vectors $\{\mathbf{v}_1, \mathbf{v}_2, \dots, \mathbf{v}_n\}$ is a basis for a vector space $V$ if every element $\mathbf{u}$ in $V$ can be uniquely expressed as:

$$$$u=c1v1+c2v2+⋯+cnvn​

where $c_1, c_2, \dots, c_n$ are scalars from the underlying field.

Dimension of a Vector Space

The dimension of a vector space is defined as the number of vectors in any basis of the space, assuming all bases of the space have the same number of vectors (this is guaranteed by the Steinitz exchange lemma). The dimension provides a measure of the size of the vector space in terms of the minimum number of independent directions needed to span the space.

a) For a finite-dimensional vector space, the dimension is simply the count of vectors in its basis.

b) For an infinite-dimensional space, the dimension can be defined but involves more complex set-theoretical concepts.

Linear Transformations:

A linear transformation is a function between two vector spaces that preserves the operations of vector addition and scalar multiplication. It's a fundamental concept in linear algebra, underlying many areas such as geometry, differential equations, and quantum mechanics.

Let $V$ and $W$ be vector spaces over the same field $F$. A function $T: V \rightarrow W$ is called a linear transformation if the following two properties hold for all vectors $\mathbf{u}, \mathbf{v}$ in $V$ and all scalars $a$ in $F$:

1. Additivity:

   $$$$T(u+v)=T(u)+T(v)

2. Homogeneity (or Scalar Compatibility):

   $$$$T(au)=aT(u)
   

   
   
   
   
   

Implications and Properties

These properties imply that linear transformations are completely determined by their effect on a basis of the domain space $V$. If $\{\mathbf{v}_1, \mathbf{v}_2, \dots, \mathbf{v}_n\}$ is a basis for $V$, then knowing $T(\mathbf{v}_1), T(\mathbf{v}_2), \dots, T(\mathbf{v}_n)$ in $W$ allows you to compute the image of any vector in $V$ under $T$.

Matrix Representation

If $V$ and $W$ are finite-dimensional, we can represent $T$ by a matrix once bases for $V$ and $W$ are chosen. If $V$ has dimension $n$ and $W$ has dimension $m$, and if $\{{\mathbf{v}}_1,\dots ,{\mathbf{v}}_n\}\; is\; a\; basis\; for$$V$ and $\{\mathbf{w}_1, \dots, \mathbf{w}_m\}$ is a basis for $W$, then the action of $T$ on any vector $\mathbf{u} \in V$ can be expressed as:

$$\mathbf{u}=x_1{\mathbf{v}}_1+\cdots +x_n{\mathbf{v}}_{\mathrm{n}}$$$$T(\mathbf{u}) = x_1 T(\mathbf{v}_1) + \dots + x_n T(\mathbf{v}_n)$$

Where $T(\mathbf{v}_i)$ can be expressed as a linear combination of the basis vectors of $W$, leading to a matrix representation of $T$.

Kernel and Image

• Kernel (Null space) of $T$: The set of all vectors in $V$ that map to the zero vector in $W$, denoted as $\text{Ker}(T)$.
• Image of $T$: The set of all vectors in $W$ that can be obtained by applying $T$ to some vector in $V$, denoted as $\text{Im}(T)$.

Null Space:

The null space (or kernel) of a linear transformation is a fundamental concept in linear algebra that describes the set of all vectors that are mapped to the zero vector under a given linear transformation. Mathematically, the null space of a linear transformation $T$ from a vector space $V$ to another vector space $W$ (both over the same field $F$) is defined as follows:

Mathematical Definition

Let $T: V \rightarrow W$ be a linear transformation. The null space of $T$, denoted $\text{Ker}(T)$, is given by:

$$$$Ker(T)={v∈V:T(v)=0}

Here, $\mathbf{0}$ represents the zero vector in $W$.

Properties of the Null Space

1. Subspace: The null space $\text{Ker}(T)$ is a subspace of $V$. This means it is closed under vector addition and scalar multiplication, and contains the zero vector of $V$.
2. Dimension: The dimension of the null space of $T$ is referred to as the nullity of $T$. It indicates the number of linearly independent vectors in $V$ that are mapped to the zero vector in $W$.
3. Solution to Homogeneous Equations: When $T$ is represented by a matrix $A$, the null space of $T$ (or $A$) corresponds to the solution set of the homogeneous linear equation $A\mathbf{x}=\mathbf{0.}$

Metric Spaces Examples:

1. Euclidean Space

The most familiar example of a metric space is the $n$-dimensional Euclidean space $\mathbb{R}^n$, with the Euclidean distance metric given by:

$$d(\mathbf{x}, \mathbf{y}) = \sqrt{(x_1 - y_1)^2 + (x_2 - y_2)^2 + \cdots + (x_n - y_n)^2}$$

where $\mathbf{x} = (x_1, x_2, \ldots, x_n)$and $\mathbf{y} = (y_1, y_2, \ldots, y_n)$ are points in $\mathbb{R}^n$.

2. Discrete Metric Space

Any set can be turned into a metric space by defining the discrete metric:

$$d(x, y) = \begin{cases} 0 & \text{if } x = y \\ 1 & \text{if } x \neq y \end{cases}$$This metric satisfies all the metric space properties and is particularly useful in theoretical computer science and topology.

3. Manhattan Metric

In $\mathbb{R}^n$, the Manhattan metric (or taxicab metric) is another common metric defined as::

$$$$d(x,y)=∣x1−y1∣+∣x2−y2∣+⋯+∣xn−yn∣

This metric is used in various applications such as urban planning, where distances are measured along grid-like paths rather than "as the crow flies."

4. Space of Continuous Functions

Consider the space $C([a,b])\; of\; all\; continuous\; real-valued\; functions\; on\; the\; interval$$[a,b].\; This\; space\; can\; be\; turned\; into\; a\; metric\; space\; using\; the\; supremum\; (or\; uniform)\; metric:$

$$d(f, g) = \sup_{x \in [a, b]} |f(x) - g(x)|$$

This metric is central in analysis, particularly in studying the convergence of sequences of functions.

5. Hamming Distance on Strings

In information theory and computer science, the set of all binary strings of fixed length $n$ can be considered a metric space with the Hamming distance defined as:

$$$$d(x,y)=the number of positions at which the corresponding symbols are different

This metric is crucial in coding theory for error detection and correction.

6. Complex Plane

The complex numbers $\mathbb{C}$ with the usual distance metric given by:

$$$$d(z,w)=∣z−w∣

where $|z|$is the modulus of $z$. This makes $\mathbb{C}$ analogous to $\mathbb{R}^2$ under the Euclidean metric.

Norm of a real vector spaces:

Examples: