Zeros of complete elliptic integrals and its application to Melnikov functions

Imagine you are a detective trying to solve a mystery about cycles. In the world of mathematics, specifically in the study of dynamic systems (like how a pendulum swings or how planets orbit), a "cycle" is a path that repeats itself forever.

Sometimes, we want to know: If we slightly nudge a system, how many new repeating loops (cycles) can appear?

This paper is a guidebook for detectives trying to count these potential new loops in a very specific, tricky scenario. Here is the breakdown in simple terms:

1. The Setting: A Bumpy Road with a Switch

Imagine a car driving on a smooth, circular track (a perfect orbit). This is our "Hamiltonian system."

The Nudge: Now, imagine we add a tiny bit of wind or a slight bump to the road (this is the "perturbation" or $\epsilon$ ).
The Switch: The track has a special rule: if the car is on the left side of a line, the road feels one way; if it's on the right, it feels a different way. This is a "piecewise smooth" system. It's like driving on a road that changes its surface texture exactly when you cross a painted line.

The detective's job is to figure out: After the nudge, how many new loops can the car get stuck in?

2. The Clue: The "Melnikov Function"

To solve this, mathematicians use a special tool called the Melnikov function. Think of this function as a metal detector.

When you sweep the metal detector over the ground (the possible paths), it beeps when it finds metal.
In math, the "beep" is a zero (a point where the function equals zero).
The Rule: Every time the metal detector beeps (finds a zero), it means a new repeating loop (limit cycle) might exist.
The Goal: We want to know the maximum number of times this detector can beep. If we know the max, we know the max number of new loops.

3. The Problem: The "Three Magic Numbers"

The metal detector in this specific case doesn't just give a simple number. Its signal is made up of a complex recipe involving three special mathematical ingredients, known as Complete Elliptic Integrals (let's call them K, E, and $\Pi$ ).

Think of K, E, and $\Pi$ as three very stubborn, complex flavors in a smoothie.

K is like a strong coffee.
E is like a rich chocolate.
$\Pi$ is like a spicy chili.

The Melnikov function is a smoothie made by mixing these three flavors with different amounts of sugar (polynomials). The question is: How many times can this specific smoothie taste exactly like water (zero)?

For a long time, mathematicians only knew how to count the zeros if the smoothie only had two flavors (K and E). But this paper tackles the much harder case where all three flavors are mixed together.

4. The Solution: The "Recipe Book"

The author, Jihua Yang, acts like a master chef who has figured out the rules of this kitchen.

Step 1: The Linear Independence Check. First, he proves that these three flavors (K, E, $\Pi$ ) are truly distinct. You can't make the "Chili" flavor just by mixing "Coffee" and "Chocolate." They are independent. This is crucial because if they weren't, the counting would be impossible.
Step 2: The Counting Formula. He derives a new formula (a "Recipe Book") that tells you the maximum number of beeps based on how much of each ingredient you used.
- If your "sugar" (the polynomials) is degree $n$ , he gives you a specific number (like $5.5n + 43$) that represents the absolute maximum number of zeros.
- It's like saying: "If you use up to 10 cups of sugar, your smoothie can never taste like water more than 60 times."

5. The Application: The Triangle Trap

Finally, he applies this new counting rule to a specific shape: a Triangle.

Imagine a triangular room where a particle bounces around.
There is a line cutting through the triangle where the physics changes.
Using his new "Recipe Book," he calculates exactly how many new loops could possibly form in this triangular room when the system is slightly disturbed.

The Big Picture

Why does this matter?
This relates to the famous Hilbert's 16th Problem, which is one of the hardest unsolved puzzles in math. It asks: "What is the maximum number of loops a system can have?"

This paper doesn't solve the whole puzzle, but it builds a very strong, new bridge across a difficult part of the river. It gives mathematicians a precise tool to count the possibilities in complex, switching systems, ensuring they don't miss any potential loops and don't overcount them.

In a nutshell:
The author invented a new mathematical ruler to measure the complexity of a specific type of "smoothie" (a mix of three hard-to-calculate integrals). He used this ruler to prove exactly how many "loops" a specific triangular system can create when nudged, solving a piece of a century-old mathematical mystery.

Here is a detailed technical summary of the paper "Zeros of complete elliptic integrals and its application to Melnikov functions" by Jihua Yang.

1. Problem Statement

The paper addresses a specific aspect of the Weak Hilbert's 16th Problem, which seeks to determine the maximum number of limit cycles that can bifurcate from a period annulus in a perturbed planar Hamiltonian system.

Context: The study focuses on perturbed Hamiltonian systems of the form $\dot{x} = H_y + \varepsilon f, \dot{y} = -H_x + \varepsilon g$ , where the perturbation involves piecewise smooth polynomials separated by a straight line.
Core Challenge: The number of limit cycles is bounded by the number of isolated zeros of the first-order Melnikov function $I(h)$ . For systems with switching lines, this function often reduces to a linear combination of complete elliptic integrals of the first ( $K$ ), second ( $E$ ), and third ( $\Pi$ ) kinds.
Specific Gap: While the number of zeros for combinations of only $K$ and $E$ (i.e., $r(k)=0$ ) was established by Gasull et al. [20], the case involving all three types ( $r(k) \neq 0$ ) with polynomial or rational coefficients remained an open problem. The paper aims to establish an upper bound for the number of zeros of the function:
$I(k) = p(k)K(k) + q(k)E(k) + r(k)\Pi(\mu(k), k)$
where $p, q, r$ are real polynomials and $\mu(k)$ is a real polynomial or rational function.

2. Methodology

The authors employ a rigorous analytical approach combining differential equations, linear algebra, and algebraic geometry.

A. Linear Independence and Differential Equations

Picard-Fuchs Equations: The authors derive the first and second-order linear differential equations satisfied by the vector $V(k) = (K(k), E(k), \Pi(\mu(k), k))^T$ . This allows them to express derivatives of the elliptic integrals in terms of the integrals themselves.
Linear Independence: Using symmetric elliptic integrals (Carlson and Gustafson's results), the paper proves that $K(k)$ , $E(k)$ , and $\Pi(\mu(k), k)$ are linearly independent with respect to rational function coefficients, provided $\mu(k)$ is not degenerate (specifically $\mu(k) \neq k^2$ and $\mu'(0) \neq 0$ ).

B. Reduction Strategy (Theorem 1.2)

To estimate the zeros of $I(k)$ , the authors introduce a transformation to eliminate the third-kind integral $\Pi$ .

They define a change of variable $Z(k) = X(k)\Pi(\mu(k), k)$ using an integrating factor $X(k)$ .
By differentiating the scaled function $\frac{I(k)X(k)}{r(k)}$ , they transform the problem into finding the zeros of a new function of the form:
$M(k)K(k) + N(k)E(k)$
where $M(k)$ and $N(k)$ are derived from the original polynomials $p, q, r, \mu$ and their derivatives.
This reduction allows the application of existing bounds for the two-integral case (Gasull et al.) to the three-integral case via Rolle's Theorem.

C. Application to Piecewise Smooth Systems

The authors analyze a specific Hamiltonian triangle system with three invariant straight lines.
They derive the algebraic structure of the Melnikov function for this system, showing it can be expressed as a linear combination of generator integrals ( $I_{i,j}, J_{i,j}$ ).
Using recurrence relations and variable transformations, these generator integrals are reduced to combinations of $K, E, \Pi$ with specific polynomial coefficients dependent on the perturbation degree $n$ .

3. Key Contributions and Results

A. General Upper Bounds (Theorems 1.3 & 1.4)

The paper provides explicit upper bounds for the number of zeros of $I(k)$ based on the degrees of the polynomials $p, q, r$ (degrees $m, n, l$ ) and $\mu$ (degree $s$ ).

Theorem 1.3 (Polynomial $\mu$ ):
The number of zeros is bounded by a function $\psi(m, n, l, s)$ . The bound varies depending on the relationship between $s$ (degree of $\mu$ ) and $l$ (degree of $r$ ).
- Example: If $s \ge 2$ and $l \le \max\{m,n\} + s$ , the bound is $2\max{m,n} + 3l + 6s + 7$.
- The paper also proves the existence of polynomials achieving $m+n+l+2$ zeros, showing the bound is relatively tight.
Theorem 1.4 (Rational $\mu$ ):
For the specific rational case $\mu(k) = \frac{2k^2}{1+k^2}$ , the bound is refined to:
$\psi(m, n, l) = \begin{cases} 2\max\{m,n\} + 3l + 11 & \text{if } l \le \max\{m,n\} + 2 \\ 5l + 7 & \text{if } l > \max\{m,n\} + 2 \end{cases}$

B. Application to Hamiltonian Triangle (Theorem 1.5)

Applying the general theory to the piecewise smooth perturbed Hamiltonian triangle with polynomial perturbations of degree $n$ :

The authors derive an explicit upper bound for the number of zeros of the Melnikov function $I(h)$ .
Result: The number of zeros is bounded by $\lfloor \frac{11n}{2} \rfloor + 43$ .
This result is significant because it handles the complexity of the switching line and the specific geometry of the triangle, which generates a non-trivial combination of all three elliptic integrals.

4. Significance

Extension of Theory: The work significantly extends the known results on the zeros of Melnikov functions by including the third kind of elliptic integral ( $\Pi$ ), which was previously excluded or treated as a degenerate case in many studies.
Computable Estimates: It provides concrete, computable upper bounds for the number of limit cycles in piecewise smooth systems, a notoriously difficult area in bifurcation theory.
Methodological Advancement: The technique of using Picard-Fuchs equations to reduce the dimension of the problem (from 3 integrals to 2) and then applying Rolle's theorem offers a robust framework for analyzing complex integral functions.
Weak Hilbert's 16th Problem: The paper contributes directly to the weak Hilbert's 16th problem by solving the zero-counting problem for a specific class of non-smooth perturbations, a step toward understanding the global behavior of limit cycles in such systems.

5. Limitations and Future Work

Lower Bounds: The paper focuses exclusively on upper bounds. It notes that establishing a lower bound (proving the existence of that many limit cycles) is difficult due to the complexity of verifying the independence of the coefficients in the polynomial expressions.
General Rational Functions: While Theorem 1.4 covers a specific rational $\mu$ , Remark 1.1 suggests that general rational $\mu(k) = Q(k)/P(k)$ can be handled similarly, though the calculations become significantly more complex.

In summary, this paper provides a definitive analytical framework for counting the zeros of Melnikov functions involving all three complete elliptic integrals, offering new quantitative results for the bifurcation of limit cycles in piecewise smooth Hamiltonian systems.