【Control】🧯 08. The Rationale Behind the “10% Rule” for Introducing FSM into PID Control

topics: [“control”, “PID”, “FSM”, “reliability design”, “industrial control”]

---

Introduction

Designs that overlay an FSM (Finite State Machine) on top of PID control
often look theoretically sound.

However, in real products and industrial systems,
there are many documented failure cases.

This article explains:

• Why continuously active FSMs are dangerous
• Why 10% deviation is a rational trigger threshold
• Why a Reliability Guard (B-Type) is indispensable

—all from the perspective of commercially viable control systems.

---

🧱 Fundamental Policy (Get This Wrong and Everything Breaks)

Control layers must be strictly separated:

1. PID: Primary control (always active)
2. FSM: Exception handling (emergency only)
3. Reliability Guard: Suppression of FSM overreach

FSM is not “smart control.”
FSM is insurance.

---

❌ Why FSM-First Designs Fail

When dealing with friction aging or long-term degradation:

• FSM switching continuously
• State transitions triggered by small fluctuations

such designs almost inevitably cause:

• Waveform discontinuities
• Overcompensation
• State-transition chatter
• Customer complaints like:

  “Something feels… off.”

The end result is ironic:

The FSM damages the system more than the degradation itself.

---

⏱ When Should FSM Activate?

The answer is simple:

Only when PID can be quantitatively proven
to no longer behave like the original PID.

This requires numerical triggers, not intuition.

---

📉 Why “10%”?

A Boundary That Emerged Naturally from the Degradation Model

In the friction-aging models used throughout this series:

• Continuous operation with PID alone
• Approximately 5 years of equivalent degradation
• Produced about 10% deviation from the initial waveform

Clear divergence appeared in at least one of:

• Amplitude
• Phase (time delay Δt)
• Control energy

This 10% threshold is:

• Not a rule of thumb
• Not a convenient heuristic

It is a boundary that emerged naturally from simulation results.

---

📏 10% of What, Exactly?

FSM triggering is defined using an OR condition.

FSM activates if any of the following exceeds
10% relative to the initial PID response:

• Amplitude deviation
• Phase / timing deviation (Δt)
• Control energy deviation

Why OR, Not AND?

Because AND conditions are too slow in real systems.

---

⚠ Reality Degrades Faster Than Models Predict

In actual equipment:

• Wear
• Contamination
• Temperature variation
• Partial damage

can compress what looks like
“5 years of model degradation”
into a few months.

Without OR conditions,
FSM intervention comes too late.

---

🛑 The Role of B-Type (Reliability Guard)

FSM is not infallible.

FSM intervention can easily:

• Worsen control quality
• Destabilize output
• Cause excessive current or voltage

This is not rare.

Therefore, B-Type introduces:

• Limits on FSM authority
• Continuous monitoring of effort and efficiency
• Rollback to PID dominance if reliability degrades

In other words:

FSM itself must not be trusted blindly.

---

🧾 The “10% Rule” from a Product Perspective

From a customer’s point of view:

• Years of stable PID behavior → no issue
• Correction only after visible degradation → acceptable
• Behavior changing from day one → distrust

The 10% rule is critical because it is:

• Explainable
• Documentable in specifications
• Defensible in customer support

---

Summary

• PID is the main actor
• FSM is for emergencies
• 10% is a model-derived, realistic threshold
• FSM triggers use OR conditions
• Reliability Guard constrains FSM behavior

FSM is not intelligence.
FSM is insurance — the last line of defense.

---

Closing Remarks

If FSM is active from day zero,
the PID design has likely failed.

If FSM activates only after a 10% deviation,
that behavior is correct for a commercial product.