Why polynomials appear in engineering
Polynomials appear in many engineering models: characteristic equations in control systems,
design equations in mechanics, calibration curves in instrumentation, pump curves in fluids,
and approximations to measured experimental data.
\[
p(x)=a_nx^n+a_{n-1}x^{n-1}+\cdots+a_1x+a_0.
\]
A root is a value of \(x\) that makes the polynomial equal to zero.
\[
p(x)=0.
\]
Engineering meaning of roots
The meaning of a root depends on the problem. In geometry or design, a positive real root may represent
a physical dimension. In control systems, roots of the characteristic equation determine stability.
\[
p(s)=0
\]
The roots of this equation are also called poles when \(p(s)\) is a system characteristic polynomial.
Continuous-time stability
For a continuous-time linear system, the system is asymptotically stable when every root has negative real part.
\[
\operatorname{Re}(s_i)<0
\qquad
\text{for every root }s_i.
\]
A root with positive real part indicates instability. A root on the imaginary axis usually requires special caution.
Discrete-time stability
For a discrete-time system, the roots must lie inside the unit circle.
\[
|z_i|<1
\qquad
\text{for every root }z_i.
\]
A root outside the unit circle indicates an unstable discrete-time response.
Quadratic focus
A second-degree polynomial has the form
\[
ax^2+bx+c=0.
\]
Its roots are found from the quadratic formula:
\[
x=\frac{-b\pm\sqrt{b^2-4ac}}{2a}.
\]
The discriminant
\[
\Delta=b^2-4ac
\]
tells whether the roots are real and distinct, repeated, or complex.
Cubic focus
Cubic polynomials appear in control, geometry, and design calculations:
\[
ax^3+bx^2+cx+d=0.
\]
Cubic equations may have three real roots or one real root with a complex-conjugate pair.
Numerical root-finding is often more practical than using the long exact cubic formula.
Newton's method
Newton's method starts from an initial guess and repeatedly uses the tangent line to improve the estimate.
\[
x_{k+1}
=
x_k
-
\frac{p(x_k)}{p'(x_k)}.
\]
The method is fast when the starting guess is close to a simple root. It may fail or jump away if
\(p'(x_k)\) is close to zero or the starting value is poor.
Derivative used in Newton's method
For a polynomial
\[
p(x)=a_nx^n+a_{n-1}x^{n-1}+\cdots+a_1x+a_0,
\]
the derivative is
\[
p'(x)=na_nx^{n-1}+(n-1)a_{n-1}x^{n-2}+\cdots+a_1.
\]
Newton's method uses this derivative at every step.
Polynomial fitting to engineering data
In data fitting mode, the calculator finds a polynomial curve that best matches measured points.
\[
\hat y
=
c_mx^m+c_{m-1}x^{m-1}+\cdots+c_1x+c_0.
\]
The fitted values \(\hat y_i\) are compared with the measured values \(y_i\).
Least-squares principle
Least squares chooses the coefficients that minimize the sum of squared residuals.
\[
\min
\sum_i
\left(y_i-\hat y_i\right)^2.
\]
If weights are used, larger weights make the corresponding points more important in the fit.
\[
\min
\sum_i
w_i
\left(y_i-\hat y_i\right)^2.
\]
Residuals
A residual is the difference between a measured value and the fitted value.
\[
r_i=y_i-\hat y_i.
\]
Small, randomly scattered residuals usually mean the model fits well. Large patterns in residuals often mean
the polynomial degree is wrong or the model is not physically appropriate.
Fit quality
The root mean square error gives a typical residual size.
\[
\mathrm{RMSE}
=
\sqrt{
\frac{\sum_i\left(y_i-\hat y_i\right)^2}{N-m-1}
}.
\]
The coefficient of determination \(R^2\) shows how much of the variation in the data is explained by the fit.
\[
R^2
=
1
-
\frac{\sum_i\left(y_i-\hat y_i\right)^2}
{\sum_i\left(y_i-\bar y\right)^2}.
\]
Using fitted roots carefully
A fitted polynomial may have roots that are mathematically correct but not physically meaningful.
Before using a fitted root in an engineering decision, check:
- whether the root lies inside the measured data range,
- whether the residuals are acceptable,
- whether the root satisfies physical constraints,
- whether a polynomial model is appropriate for the system,
- whether extrapolation is being used.
Common mistakes
- Using a high-degree polynomial fit just because it gives a larger \(R^2\).
- Ignoring complex roots in control-system stability analysis.
- Assuming every real root is physically possible.
- Using Newton's method without checking whether the derivative is near zero.
- Forgetting that repeated roots can be difficult for Newton's method.
- Trusting fitted roots far outside the measured data range.
- Ignoring residual plots when evaluating engineering data fits.
