Controllers Overview

Comprehensive process control system for wastewater treatment simulation

Introduction

ErgoPlanner provides a comprehensive controller framework for automatic process control in wastewater treatment simulations. Controllers adjust process parameters in real-time to maintain setpoints and optimize treatment performance.

Controller Categories

Base PID Controllers (Task 95)

Standard PID control for common wastewater parameters:

Controller	Application	Setpoint Unit	Manipulated Variable
DO	Dissolved Oxygen	mg/L	Airflow (m³/h)
SRT	Solids Retention Time	days	WAS Flow (m³/d)
RAS	Return Sludge	ratio	RAS Ratio
WAS	Waste Sludge	m³/d	WAS Flow
Flow	Flow Control	m³/d	Pump Flow
pH	pH Control	pH units	Chemical Dose (L/h)

Extended Controllers (Task 186)

Advanced control strategies for complex processes:

Controller	Strategy	Key Application
Cascade-Controller	Master-slave PID	Nested control loops
Deadband-Controller	On/off with hysteresis	Equipment protection
Switch-Controller	Smooth S-curve transition	State changes
TimeOnOff-Controller	Programmable cycling	Intermittent aeration
Ratio-Controller	Flow ratio with PID trim	RAS/WAS ratio
ContinuousP-Controller	Integrating proportional	Valve positioning
Feedforward-Controller	Disturbance-based	Load compensation

Control Theory Fundamentals

PID Control Equation

The standard PID algorithm implemented:

$$u(t) = K_p \cdot e(t) + K_i \int_0^t e(\tau) d\tau + K_d \frac{de(t)}{dt}$$

Where:

• $u(t)$ = Controller output
• $e(t)$ = Error (setpoint - process variable)
• $K_p$ = Proportional gain
• $K_i$ = Integral gain
• $K_d$ = Derivative gain

Discrete Implementation

The discrete form used in simulation:

$$u_k = K_p \cdot e_k + K_i \cdot \sum_{i=0}^{k} e_i \cdot \Delta t + K_d \cdot \frac{e_k - e_{k-1}}{\Delta t}$$

Features

Anti-Windup Protection

Prevents integral term accumulation during output saturation:

1. Integral Clamping: Limits integral term magnitude
2. Back-Calculation: Reduces integral when saturated

// Anti-windup implementation
const maxIntegral = ki > 0 ? (outputMax - outputMin) / ki : Infinity;
newIntegral = Math.max(-maxIntegral, Math.min(maxIntegral, newIntegral));

Derivative Filtering

Low-pass filter on derivative term to reduce noise sensitivity:

$$D_{filtered} = \alpha \cdot D_{raw} + (1-\alpha) \cdot D_{previous}$$

Where $\alpha$ is the filter coefficient (default: 0.1).

Bumpless Transfer

Smooth transitions when:

• Switching between manual and auto mode
• Changing controller setpoints
• Re-initializing controllers

Controller Service Architecture

graph TB
    subgraph ControllerService
        CI[Controller Initialization]
        CE[Controller Execution]
        CO[Output Application]
    end

    subgraph Controllers
        PID[PID Controllers]
        EXT[Extended Controllers]
    end

    subgraph Simulation
        Graph[Calculation Graph]
        Nodes[Process Units]
    end

    UI[User Configuration] --> CI
    CI --> PID
    CI --> EXT

    Graph --> CE
    PID --> CE
    EXT --> CE
    CE --> CO
    CO --> Nodes

Configuration

Base Controller Configuration

{
  "id": "do-controller-1",
  "type": "DO",
  "enabled": true,
  "setpoint": 2.0,
  "kp": 50.0,
  "ki": 10.0,
  "kd": 5.0,
  "targetNodeId": "aerobic-reactor-1",
  "manipulatedVariable": "airflow"
}

Extended Controller Configuration

{
  "id": "cascade-do-1",
  "type": "Cascade",
  "enabled": true,
  "master": {
    "setpoint": 3000,
    "kp": 1.0,
    "ki": 0.1,
    "kd": 0,
    "pvNodeId": "reactor-1",
    "pvVariable": "MLSS"
  },
  "slave": {
    "kp": 2.0,
    "ki": 0.5,
    "kd": 0.1,
    "pvNodeId": "reactor-1",
    "pvVariable": "DO",
    "targetNodeId": "reactor-1",
    "manipulatedVariable": "airflow"
  },
  "mode": "auto",
  "bumplessTransfer": true
}

Tuning Guidelines

General Principles

1. Start Conservative: Begin with low gains and increase gradually
2. Tune Sequential: For cascade, tune slave first, then master
3. Monitor Oscillations: Reduce Kp if oscillations occur
4. Integral Action: Ki should be 5-10× slower than Kp response

Typical Tuning Values

Application	Kp Range	Ki Range	Kd Range	Notes
DO Control	20-100	5-20	0-10	Fast response needed
SRT Control	50-200	0.5-5	0	Slow integration
RAS Ratio	0.2-1.0	0.02-0.1	0	Gentle control
pH Control	20-100	2-10	0.5-3	Nonlinear near 7.0
Flow Control	0.5-2.0	0.05-0.2	0-0.1	Quick response

Best Practices

1. Start with Manual Mode

• Verify process variable measurement
• Confirm output actuator response
• Check setpoint reasonableness

2. Monitor Controller Performance

• Track error magnitude over time
• Watch for oscillations
• Check for output saturation

3. Consider Process Dynamics

• Faster controllers for faster processes
• Longer integral times for sluggish processes
• Use derivative sparingly

4. Use Appropriate Controller Type

• Simple loops: Standard PID
• Nested loops: Cascade
• Equipment protection: Deadband
• Load changes: Feedforward

Troubleshooting

Issue	Likely Cause	Solution
Oscillation	Kp too high	Reduce Kp by 50%
Slow response	Kp too low	Increase Kp gradually
Offset	Ki too low	Increase Ki
Overshoot	Ki too high or Kd too low	Reduce Ki or add Kd
Noise amplification	Kd too high	Reduce Kd or increase filter
Saturation	Limits too tight	Check output limits

Related Topics

• Process-Units-Overview
• Simulation-Engine
• Calculation-Modes

References

1. Seborg, D.E., et al. "Process Dynamics and Control" (4th Edition)
2. Rieger, L., et al. "Guidelines for Using Activated Sludge Models" (IWA)
3. Olsson, G. "ICA and Me - A Subjective Review" (Water Research, 2012)