TOPICS
Search

Stable Polynomial


A stable polynomial is a real polynomial P for which all its roots lie in the left half-plane. The term "stable" is used to describe such a polynomial because, in the theory of linear servomechanisms, a system exhibits unforced time-dependent motion of the form e^(st), where s is the root of a certain real polynomial P(s)=0. A system is therefore mechanically stable iff P is a stable polynomial.

The polynomial x+a is stable iff a>0, and the polynomial x^2+ax+b is stable iff both a and b are greater than zero. The Routh-Hurwitz theorem can be used to determine if a polynomial is stable.

Given two real polynomials P and Q, if P and Q are stable, then so is their product PQ, and vice versa (Séroul 2000, p. 280). It therefore follows that the coefficients of stable real polynomials are either all positive or all negative (although this is not a sufficient condition, as shown with the counterexample x^3+x^2+x+1). Furthermore, the values of a stable polynomial are never zero for x>=0 and have the same sign as its coefficients. A stable polynomial with positive leading coefficient is a Hurwitz polynomial.

Root location can also be tested by quantifier elimination. For a real polynomial P, stability is equivalent to

  forall x,y in R ([Re[P(x+iy)]=0 ^ Im[P(x+iy)]=0]==>x<0).
(1)

Here Re and Im denote the real part and imaginary part, respectively. If the coefficients depend polynomially on real parameters, eliminating x and y gives a condition solely on those parameters (Hong et al. 1997, Trott 2006).

It is possible to decide if a polynomial is stable without first knowing its roots using the following theorem due to Strelitz (1977). Let A=x^n+a_(n-1)x^(n-1)+...+a_0 be a real polynomial with roots alpha_1, ..., alpha_n, and construct B=x^m+b_(m-1)x^(m-1)+...+b_0 as the monic polynomial with real coefficients of polynomial degree m=n(n-1)/2 having roots alpha_i+alpha_j for 1<=i<j<=n. Then A is stable iff all coefficients of A and B are positive (Séroul 2000, p. 281).

For example, given the cubic polynomial A=x^3+ax^2+bx+c, the sum-of-roots polynomial B is given by

 B=x^3+2ax^2+(a^2+b)x+(ab-c).
(2)

Resolving the inequalities given by requiring that each coefficient of A and B be greater than zero then gives the conditions for A to be stable as a>0, b>0, 0<c<ab.

Similarly, for the polynomial A=x^4+ax^3+bx^2+cx+d of polynomial degree 4, the sum-of-roots polynomial is

 x^6+3ax^5+(3a^2+2b)x^4+(a^3+4ab)x^3+(2a^2b+b^2+ac-4d)x^2+(ab^2+a^2c-4ad)x+(abc-c^2-a^2d),
(3)

so the condition for A to be stable can be resolved to a>0, b>0, 0<c<ab, 0<d<(abc-c^2)/a^2.

For A=x^5+ax^4+bx^3+cx^2+dx+e, the sum-of-roots polynomial is

 x^(10)+4ax^9+(6a^2+3b)x^8+(4a^3+9ab+c)x^7+(a^4+9a^2b+3b^2+4ac-3d)x^6+(3a^3b+6ab^2+5a^2c+2bc-5ad-11e)x^5+(3a^2b^2+b^3+2a^3c+6abc-c^2-2a^2d-2bd-22ae)x^4+(ab^3+4a^2bc+b^2c-4cd-16a^2e-4be)x^3+(2ab^2c+a^2c^2-bc^2+a^2bd+b^2d-3acd-4d^2-4a^3e-9abe+7ce)x^2+(abc^2-c^3+ab^2d-4ad^2-4a^2be-b^2e+4ace+4de)x+(abcd-c^2d-a^2d^2-ab^2e+bce+2ade-e^2).
(4)

The following Wolfram Language code computes the sum-of-roots polynomial B and inequalities obtained from the coefficients:

  RootSumPolynomial[r_List, x_]:=Module[
      {n = Length[r], i, j},
      RootReduce@Collect[Expand[
        Times@@((x - #)&/@Flatten[
          Table[r[[i]] + r[[j]], {i, n},
            {j, i+1, n}]])
      ], x]
  ]
  RootSumPolynomial[p_?PolynomialQ, x_]:=
    RootSumPolynomial[RootList[p, x], x]
  RootList[p_?PolynomialQ, x_]:=
    x /. {ToRules[Roots[p==0, x,
      Cubics -> False, Quartics -> False
    ]]}
  RootSumInequalities[p_?PolynomialQ, x_]:=
    And @@ (# > 0& /@
      Flatten[CoefficientList[#, x]& /@
        {RootSumPolynomial[p, x], p}])

while the following reduces the inequalities to a minimal set in the cubic case:

  Resolve[Exists[x, Element[(a | b | c | x), Reals],
    RootSumInequalities[x^3 + a x^2 + b x + c, x]
  ], {a, b, c}]

See also

Hurwitz Polynomial, Left Half-Plane, Routh-Hurwitz Theorem

Explore with Wolfram|Alpha

References

Hong, H.; Liska, R.; and Steinberg, S. "Testing Stability by Quantifier Elimination." J. Symb. Comput. 24, 161-187, 1997. https://doi.org/10.1006/jsco.1997.0121.Munro, N. (Ed.). Symbolic Methods in Control System Analysis and Design. London: Institution of Electrical Engineers, 1999. https://doi.org/10.1049/PBCE056E.Séroul, R. "Stable Polynomials." §10.13 in Programming for Mathematicians. Berlin, Germany: Springer-Verlag, pp. 280-286, 2000.Strelitz, S. "On the Routh-Hurwitz Problem." Amer. Math. Monthly 84, 542-544, 1977.Tóth, J.; Szili, L.; and Zachár, A. "Stability of Polynomials." Mathematica Educ. Res. 7, 5-12, 1998.Trott, M. "Polynomials in Inequalities." §1.2.3 in The Mathematica GuideBook for Symbolics. New York: Springer-Verlag, 2006. https://doi.org/10.1007/0-387-28815-5. https://www.mathematicaguidebooks.org/.

Referenced on Wolfram|Alpha

Stable Polynomial

Cite this as:

Weisstein, Eric W. "Stable Polynomial." From MathWorld--A Wolfram Resource. https://mathworld.wolfram.com/StablePolynomial.html

Subject classifications