A Novel alpha beta-Approximation Method Based on Numerical Integration for Discretizing Continuous Systems

summary

The gist

In this article, we propose a novel discretization method based on numerical integration for discretizing continuous systems and demonstrate that the proposed method is superior to all typical…

In short

The paper proposes a new discretization method called Scalable Bilinear Transformation (SBT) using numerical integration based on a Scalable Hexagonal Approximation (SHA). This method uses shape factor ($\alpha$) and time factor ($eta$) to discretize continuous systems, outperforming typical methods by achieving a 25% reduction in root-meansquare error and superior performance in resonant controllers.

Key concepts

Scalable Bilinear Transformation (SBT)
The SBT is the proposed method consisting of two factors: a shape factor ($\alpha$) and a time factor ($\beta$). It is derived from numerical integration using the Scalable Hexagonal Approximation (SHA) to transform continuous systems into discrete ones, aiming for high accuracy in control applications.
Shape Factor ($\alpha$)
The shape factor $\alpha$ is defined by the ratio of two quantities: $S_{bw rec}$ and $(S_{bw rec} + S_{fw rec})$. This factor helps define one aspect of the transformation, influencing how the system's characteristics are mapped during discretization.
Time Factor ($\beta$)
The time factor $\beta$ is defined as the ratio of $T'$ to $T$. This factor relates to the time domain scaling in the transformation. Together with $\alpha$, it forms the SBT, which is used to map continuous system dynamics into a discrete form.
Scalable Hexagonal Approximation (SHA)
The SHA is a novel technique used in numerical integration that approximates the area enclosed by an error function with a scalable hexagonal shape. This approximation forms the basis for deriving the SBT method, allowing for accurate discretization of continuous systems.

Terminology used across episodes

This episode discusses

The paper

A Novel alpha beta-Approximation Method Based on Numerical Integration for Discretizing Continuous Systems · Read on arXiv

XiŁan Jiaotong University · ¦n Jiao Tong University of Water Resources and Electric Power, Zhejiang University of Water Resources and Electric Power, SolaX Power Network Technology (Zhejiang) Co. · State Key Laboratory of Electrical Insulation and Power Equipment

DOI: 10.1109/TPEL.2026.3732662

Transcript

Introduction to the show: ident: Robotics Radio. Generated commentary on the latest robotics and control papers.

Rosa: We're looking at the paper: "A Novel alpha beta-Approximation Method Based on Numerical Integration for Discretizing Continuous Systems".

Dev: Here's the gist of the paper. In this article, we propose a novel discretization method based on numerical integration for discretizing continuous systems and demonstrate that the proposed method is superior to all typical methods.

Rosa: The Proposed Method The proposed method is termed the αβ-approximation or Scalable Bilinear Transformation (SBT).

Dev: In contrast to existing methods, the proposed method consists of two factors, i.e., shape factor (α) and time factor (β).

Rosa: Depending on the discretization technique applied, we identify two primary distortion modes in discrete resonant controllers: frequency warping and resonance damping.

Paper summary: Rosa: We talked about the paper "A Novel alpha beta-Approximation Method Based on Numerical Integration for Discretizing Continuous Systems" and how it introduces this new method called the Scalable Bilinear Transformation, or SBT. The core idea is that instead of just using one standard way to discretize things, you use two factors, a shape factor alpha and a time factor beta.

Dev: And they claim these factors allow them to identify two main distortion modes in discrete resonant controllers: frequency warping and resonance damping, and they offer a theoretical explanation for those distortions.

Taro: I'm thinking about how this relates to real-world failure modes we see when systems start behaving unpredictably outside of the lab environment.

Rosa: Right, and the paper states that they demonstrate that their proposed method is superior to all typical methods by achieving a twenty-five percent reduction in root-meansquare error compared to the State Of The Art method.

Dev: That twenty-five percent reduction is significant when you're looking at control accuracy, so it's not just a small tweak. How do they manage that level of improvement across different discretization techniques?

Taro: It seems like the mechanism relies on using numerical integration, specifically something called the Scalable Hexagonal Approximation, or SHA, to define those factors alpha and beta.

Rosa: That's right, and they define alpha as being based on backward rectangular area divided by the sum of that backward rectangular area and forward rectangular area. And beta is defined by scaling the sampling time from T to T prime.

Dev: So it's basically a way to map out the error curve in a scaled way, and they show how this leads to the SBT being denoted as one over beta times T z minus one over alpha z plus one minus alpha.

Taro: I'm curious about that transformation equation itself, because if we want to understand why it works, we need to look at how the mapping from the s plane to the z plane is done.

Rosa: They map s and z by substituting their expressions into a specific expression involving sigma s plus j omega s and gamma z plus j zeta z.

Dev: And they show that this mapping is stable unless the left half of the s plane gets mapped inside the unit circle of the z plane, which means sigma s has to be less than or equal to zero.

Taro: That stability check is crucial for any real-time system, so if we want a safety case, we need to know how this method behaves when things start going sideways.

Rosa: They also compared it against methods like Tustin with pre-warping and a two integrator based scheme for voltage-controlled voltage-source inverters.

Dev: And they found that the Tustin method with pre-warping introduced no additional phase lag, whereas the other two methods incurred an extra time delay, which is a key difference they are highlighting.

Taro: So it's not just about one method being better; it's about understanding which distortion modes—like frequency warping versus resonance damping—are present in the system you're trying to model.

Rosa: Exactly, and the paper provides that theoretical explanation for those modes, showing why this method is the way forward for discretization.

Conclusion: Dev: So we've gone through this whole thing on "A Novel alpha beta-Approximation Method Based on Numerical Integration for Discretizing Continuous Systems," and the authors essentially argue that this SBT method provides an optimal way to handle these errors in discretizing continuous systems.

Rosa: And they conclude that by using the Scalable Hexagonal Approximation, they can achieve this optimal performance by carefully designing those shape factor alpha and time factor beta using a straightforward or an optimal approach.

Taro: So what does this mean for us when we look at the title and who wrote it? Who are these folks? Are they experts in real-time control or just theoretical mathematicians?

Dev: They seem to be a mix of both, which is why they're interesting; they’re not just one type of expert.

Rosa: They are researchers who have put this out as a unified framework for numerical integration that tackles the discretization problem head-on.

Taro: So in simple terms, what's the big picture here for someone just listening to the show? What does it actually change for us?

Dev: It means we have a tool that can give us better control system performance without needing a completely different hardware setup.

Rosa: It’s about getting better results by improving the accuracy of how we discretize the system, and this paper shows this is achievable at a computational cost equivalent to simpler methods.

Taro: So basically, what's the final word on this SBT method?

Dev: The final word is that it's a highly effective way to discretizing resonant controllers because it delivers optimal performance when compared against the Euler, Tustin, and SOTA methods.

More episodes

← Home