Multivariate statistic validation of pH and ORP data as control inputs for biological nitrogen removal at full scale

Area of Science:

• Environmental Engineering
• Water Treatment Technologies
• Process Control

Background:

• Biological nitrogen removal is crucial for wastewater treatment.
• Optimizing aeration energy demand is a key challenge.
• Advanced control strategies are needed for efficient nutrient removal.

Purpose of the Study:

• To validate oxidation-reduction potential (ORP) and pH as input data for advanced control strategies.
• To optimize biological nitrogen removal while minimizing aeration energy.
• To assess the correlation between ORP/pH data and nitrogen removal efficiency.

Main Methods:

• Applied statistical multivariate projection to on-line ORP and pH sensor data.
• Installed sensors in a full-scale plug-flow reactor.
• Correlated derivative ORP and pH signals with nitrogen-based sensor data.

Main Results:

• pH and ORP derivative signals show strong correlations with nitrogen-based sensor data.
• Derivative signals are effective control inputs for nitrification, SND, and denitrification.
• Controllers based on derivative signals demonstrated enhanced resilience to sensor faults and data biases.

Conclusions:

• ORP and pH derivative signals are reliable inputs for advanced wastewater treatment control.
• This approach optimizes nitrogen removal and reduces energy consumption.
• Derivative signal-based control enhances process stability and sensor fault tolerance.