The system provides efficient monitoring and emission prediction by utilising available data and real-time data acquisition. It enables continuous improvement and enhancement without operational limitations, while also functioning as a standalone installation with a user-friendly interface.

Emission monitoring is often inaccurate because physical sensors in gas turbines are prone to drift and data spikes, resulting in unreliable CO and NOx readings that can compromise regulatory compliance. Conventional systems also incur high maintenance costs, as hardware sensors require frequent calibration and expensive cooling chambers to withstand high-temperature turbine environments. In addition, limited process visibility means operators lack real-time, AI-driven insights needed to optimise combustion efficiency and reduce environmental impact.

CONERS works by establishing a direct live data connection with industrial control systems, such as the MLNG PI system, to capture real-time operational data from gas turbine generators. This data is transmitted through a PI Connection Manager, enabling seamless integration with existing plant databases without the need for additional physical sensors. The system then uses a predictive engine to process the data in real time and generate CO and NOx emission estimates, which are displayed on a live dashboard for immediate operational visibility. It also exports results to local databases for reporting and long-term analysis, supporting regulatory compliance and process optimisation.

CONERS is an AI-based soft sensor that predicts NOx and CO emissions from gas turbines using real-time operational data, eliminating the need for unreliable physical sensors. It integrates with existing control systems to generate live emission forecasts, compliance checks, and trip probability predictions. This improves monitoring accuracy, reduces maintenance costs, and enhances operational visibility and efficiency.