py_pidx

⚠️ Note: This repository is a mirror of the original Python PID controller package developed and maintained by Mehrab Mahdian.
For the latest updates, issue tracking, and contributions, please visit the original repo:
https://github.com/mehrabmahdian/py_pidx

py_pidx is an advanced Python library for PID control that provides robust, real-world-ready features, including:

• 🧠 Gain scheduling
• 🧾 Deadband handling (error & output)
• 🚦 Anti-windup
• 🔁 Slew rate limiting
• 🌀 Derivative filtering
• 🧮 Feedforward term
• 🧪 Precision control with dynamic time steps
• 🔄 Integral and derivative-on-measurement option

---

🔧 Installation

pip install py_pidx

Feature	Description
Kp, Ki, Kd	Tunable proportional, integral, and derivative gains
GainScheduler	Dynamically adjust PID gains at runtime
Deadbands	Zero out small control efforts or errors to reduce actuator jitter
Anti-windup	Prevents runaway integral values during saturation
Derivative on Measurement	Optionally compute D-term from PV instead of error
Integral on Measurement	Optionally compute I-term on process variable instead of error
Low-pass Filtering	Smooths noisy derivative term using alpha filtering
Feedforward	Optional term to improve open-loop response
Slew Rate Limiting	Prevents rapid output changes that may shock actuators

🔧 Basic Usage

from py_pidx import PID

pid = PID(Kp=1.2, Ki=0.5, Kd=0.05, setpoint=100)

while True:
    pv = read_sensor()  # Your process variable
    control_signal = pid.run(pv)
    send_to_actuator(control_signal)
    time.sleep(0.01)

🎯 Advanced Features

Gain Scheduling

from py_pidx.gain_scheduler import GainScheduler

class MyScheduler(GainScheduler):
    def get_gains(self, pv):
        if pv < 10:
            return (2.0, 0.1, 0.01)
        elif pv < 20:
            return (1.5, 0.05, 0.005)
        else:
            return (1.0, 0.01, 0.001)

pid = PID(gain_scheduler=MyScheduler())

Deadbands

pid = PID(
    output_deadband_limits=(-0.1, 0.1),
    error_deadband_limits=(-0.05, 0.05)
)

Anti-Windup and Slew Rate

pid = PID(
    Ki=0.3,
    anti_windup=True,
    max_output_rate=10.0  # Max 10 units/sec change
)

Derivative Filtering and Feedforward

pid = PID(
    Kd=0.2,
    derivative_filter=0.3,  # Low-pass smoothing factor (0.0–1.0)
    feedforward=1.5
)

Testing / Resetting / Toggling

pid.set_auto_mode(False)    # Pause control
pid.set_auto_mode(True)     # Resume control
pid.reset()                 # Clear internal state

PID Controller Methods

Method	Description
run(pv, delta_time=None)	Compute the control output
reset()	Clear internal PID state
set_auto_mode(enabled)	Enable/disable PID control loop
set_output_limits(min, max)	Clamp controller output
__repr__()	Debug string with current gains and output

📄 Documentation

👉 Full API Reference and usage examples: 📖 https://mehrabmahdian.github.io/py_pidx/

🧠 Why Use py_pidx?

• Designed with real control systems in mind
• Great for robotics, automation, process control, and simulations
• Extensible and highly configurable
• Clean, readable, and well-documented

📃 License

This project is licensed under the terms of the MIT license.

🤝 Contributing

Pull requests and feature ideas are welcome! Please open an issue for bugs or enhancement suggestions.

⭐️ Show Your Support

If you find this library useful, consider starring the repo or sharing it with others in the controls/automation community!

📬 Contact

Maintained by Mehrab Mahdian.

If you have questions, feature requests, or would like to collaborate, feel free to reach out:

📧 Email:
memahdian@outlook.com
mehrab.mahdian@taltech.ee

🧑‍💻 GitHub: @mehrabmahdian
💼 LinkedIn: Mehrab Mahdian

I’m always happy to connect with fellow developers, engineers, and researchers.