Similarity transformation

A similarity transformation changes the coordinate basis of a state-space model using an invertible matrix, so the circuit looks simpler but keeps the same eigenvalues and dynamics.

What is similarity transformation?

A similarity transformation in Electrical Circuits and Systems II is a change of coordinates for a state-space model. You keep the same physical system, but you rewrite the state vector with an invertible matrix so the equations are easier to work with.

If the original state matrix is A and you choose a new state vector x = Tz, then the transformed system uses T^{-1}AT as the new state matrix. That new matrix represents the same system in a different basis. The current, voltage, or energy behavior has not changed, but the math may become cleaner.

That is why similarity transformations show up right next to state equations, matrix exponentials, and modal analysis. A messy matrix can sometimes be turned into a diagonal or nearly diagonal form, especially if the system has enough independent eigenvectors. When that happens, each state component evolves more separately, which makes solving the differential equations much easier.

The biggest thing to remember is what does not change. Similarity transformations preserve eigenvalues, so they preserve the system’s natural modes and stability picture. If the original A matrix has poles or eigenvalues in the left half-plane, the transformed matrix will have the same ones. What changes is the representation, not the underlying circuit behavior.

A quick way to see the benefit is to think about a coupled RLC network written in a hard-to-read coordinate system. In one basis, the state equations may be tangled together. After a similarity transformation, you might reveal modal coordinates that separate the dynamics into simpler first-order pieces or a block form. That is why this tool matters when you are trying to solve state equations by hand, check stability, or prepare a model for control design.

One common mistake is to treat similarity as if it changes the system itself. It does not. If two matrices are similar, they describe the same linear system from two different coordinate views, so you should expect the same characteristic behavior even if the formulas look very different.

Why similarity transformation matters in Electrical Circuits and Systems II

Similarity transformation shows up whenever a Circuits II problem moves from raw state equations to a cleaner matrix form. It is the bridge between a complicated state matrix and a version you can actually solve, interpret, or compare with a physical circuit.

This matters most in solution of state equations. If you can transform A into diagonal form, or at least into a form with simpler blocks, then computing e^{At} becomes much easier. That makes initial-condition problems less painful and helps you trace how the circuit’s natural response is built from its modes.

It also connects directly to stability and mode shape. Because similar matrices share eigenvalues, you can change coordinates without changing whether the system is stable or unstable. That makes similarity transformations a safe algebraic tool for simplifying a model before you talk about poles, modes, or controller design.

In class problems, this often shows up as a step before modal analysis or a check after finding eigenvectors. If your matrix is hard to interpret in the original circuit variables, a similarity transformation can reveal which combinations of voltages and currents actually decouple the dynamics.

How similarity transformation connects across the course

State Matrix

The state matrix is the matrix you transform. Similarity transformations rewrite that matrix in a different coordinate basis, usually to make the entries easier to interpret or solve. In a state-space problem, you are often taking a messy A matrix and replacing it with T^{-1}AT so the new system has cleaner dynamics.

Eigenvalues

Eigenvalues stay the same under similarity transformation, which is why the system’s stability and natural modes do not change. If you find the eigenvalues of the original matrix, you already know the eigenvalues of any similar matrix. That is the main reason similarity is useful in state-space work.

Modal Analysis

Modal analysis uses similarity ideas to separate a system into independent or weakly coupled modes. When the transformation succeeds, each mode can be studied on its own instead of inside one tangled set of equations. That makes it easier to see which parts of the circuit die out quickly and which ones dominate the response.

State Space Representation

A similarity transformation does not leave state-space representation behind, it changes the coordinates inside it. The physical circuit is still the same, but the chosen state variables may be different. That is why two state-space models can look unrelated at first and still describe the same system.

Is similarity transformation on the Electrical Circuits and Systems II exam?

A quiz or problem-set question usually asks you to show that two state-space matrices are similar, find the change-of-basis matrix T, or use a transformed A matrix to simplify e^{At}. You may also be asked to explain why the eigenvalues do not change after the transformation. The move is to identify the invertible matrix, apply T^{-1}AT, and then use the cleaner form to solve state equations or check stability.

If the problem gives a diagonalizable matrix, the fastest path is often to find eigenvectors, build T from them, and rewrite the system in modal coordinates. If the matrix is not diagonalizable, you may still use a similarity transformation to reach a simpler Jordan or block form. The grading usually cares about whether you kept the eigenvalues and the system meaning intact while changing the coordinates correctly.

Key things to remember about similarity transformation

• A similarity transformation changes the coordinate system of a state-space model, not the underlying circuit behavior.

• The transformed matrix is found with an invertible change-of-basis matrix, usually written as T^{-1}AT.

• Similar matrices have the same eigenvalues, so stability and natural modes stay the same.

• The main payoff in Circuits II is simpler state equations, especially when you want to compute the matrix exponential or separate modes.

• If the new matrix looks easier to solve, that is the point. If the eigenvalues changed, the transformation was done incorrectly.

Frequently asked questions about similarity transformation

What is a similarity transformation in Electrical Circuits and Systems II?

It is a change of coordinates for a state-space model using an invertible matrix. The new matrix looks different, but it represents the same linear system and keeps the same eigenvalues. In practice, you use it to make state equations easier to solve.

Does a similarity transformation change the eigenvalues?

No. Similar matrices always have the same eigenvalues, which is why the system’s stability does not change under the transformation. What changes is the representation, not the underlying dynamics.

Why would I use a similarity transformation on a circuit model?

You use it to simplify a hard state matrix, often before solving the differential equations or doing modal analysis. If the transformed matrix is diagonal or close to it, you can separate the system into simpler pieces and work faster.

Is similarity transformation the same as diagonalization?

Not exactly. Diagonalization is one special result you may get from a similarity transformation when the matrix has enough independent eigenvectors. A matrix can still be similar to another form, like a block or Jordan form, even when it cannot be diagonalized.