COMMUNICATIONS SUSTAINABILITY ANALYSIS
Domain consistent industrial decarbonisation of global coal power plants
This paper introduces a human-in-the-loop constraint-based optimisation framework for decarbonising thermal power plants. It integrates domain expertise with data-driven methods, showcasing effectiveness at improving thermal efficiency and reducing turbine heat rate in a 660 MW coal-fired plant. When extended to 56 similar plants worldwide, the model projects a cumulative lifetime mitigation of 60.2 million tons of carbon emissions. These results highlight the potential of ML and optimisation with human expertise for large-scale industrial decarbonisation.
Executive Impact at a Glance
Leverage our AI-powered insights to drive efficiency and sustainability in your operations.
0.6%
Mean Efficiency Gain
93 kJ/kWh
Turbine Heat Rate Reduction
60.2 Mt
Projected CO2 Mitigation
Deep Analysis & Enterprise Applications
Select a topic to dive deeper, then explore the specific findings from the research, rebuilt as interactive, enterprise-focused modules.
Our approach combines machine learning (ML) for predictive modeling with a human-in-the-loop (HITL) constraint-based optimization framework. This ensures that the generated solutions are not only mathematically optimal but also domain-consistent and implementable within existing industrial control systems. The HITL constraints, defined by human experts, prevent solutions from violating operational safety limits and structural dependencies between process variables.
Enterprise Process Flow
Start
→
Train ML models
→
Embed ML models in optimisation problem and solve the optimisation problem
→
Is the evaluation criteria satisfied?
→
End
Empirical tests on a 660 MW supercritical coal-fired power plant demonstrated significant improvements. The mean efficiency gain was 0.64%, directly translating to less fuel consumption per unit of power. Concurrently, the mean turbine heat rate was reduced by 93 kJ/kWh, indicating improved energy conversion efficiency.
0.64%
Mean Efficiency Gain
A critical aspect of our framework is ensuring domain consistency. Traditional ML-driven optimization often yields mathematically feasible but practically unworkable solutions. By embedding HITL constraints, we ensure that the estimated optimal operating variables maintain their structural relationships and stay within human-defined tolerance thresholds, thus preserving plant safety and operational integrity.
| Approach | Benefits |
|---|---|
| Traditional ML Optimization |
|
| HITL-MLOPT Framework |
|
Extending the model's application to 56 similar coal plants worldwide, our analysis projects a cumulative lifetime mitigation of 60.2 million tons of carbon emissions. This highlights the transformative potential of our framework for large-scale industrial decarbonization, offering a sustainable path for reducing the carbon footprint of existing fossil fuel infrastructure.
Cumulative CO2 Mitigation
Our model projects a significant reduction in CO2 emissions when applied globally. This is achieved by increasing thermal efficiency across operational coal power plants, reducing fuel consumption, and consequently, carbon output over their remaining operational lifetimes.
Value: 60.2 Million Tons CO2
Label: Projected Lifetime Mitigation
Advanced ROI Calculator
Estimate the potential return on investment for implementing AI-driven efficiency solutions in your enterprise.
Your AI Implementation Roadmap
A typical timeline for integrating advanced AI-driven optimization into your industrial operations, ensuring smooth adoption and measurable results.
Phase 01: Discovery & Assessment
Initial consultation, data infrastructure assessment, and definition of key performance indicators (KPIs) for AI integration.
Phase 02: Model Development & Customization
Training and fine-tuning ML models using your operational data, incorporating HITL constraints and domain expertise.
Phase 03: Pilot Implementation & Verification
Deploying the AI framework in a pilot environment, rigorously testing and verifying solutions against real-time operational data.
Phase 04: Full-Scale Deployment & Monitoring
Seamless integration into your existing control systems, continuous monitoring, and iterative refinement for sustained performance gains.
Ready to Transform Your Operations?
Connect with our AI specialists to discuss how domain-consistent industrial decarbonisation can benefit your enterprise. Book a free consultation today.
Ready to Get Started?
Book Your Free Consultation.
Let's Discuss Your AI Strategy!
Lets Discuss Your Needs
Select a Date
Sun
Mon
Tue
Wed
Thu
Fri
Sat