PSI - Issue 77

Michal Holubčík et al. / Procedia Structural Integrity 77 (2026) 413– 423 "Michal Holubčík" / Structural Integrity Procedia 00 (2026) 000 – 000

421

9

The intelligent regulation system is based on a cost function that quantifies combustion quality. This function integrates three weighted criteria: the error of thermal power, the deviation of CO concentration from the limit value, and fuel consumption. Each criterion is normalized and multiplied by a weight, allowing flexible prioritization, such as 10:0:0 for power-only control or 1:10:0 for emission minimization. The MATLAB optimization algorithm fmincon continuously evaluates the cost function and searches for the lowest value by adjusting feeder and fan frequencies. Optimization is executed every two seconds, ensuring real-time adaptation to combustion dynamics. Čajová Kantová et al. (2021) demonstrated that simulations of biomass combustion with modified flue gas tracts enable optimization of combustion efficiency, providing a practical tool for emission reduction strategies [10].

Figure 6 : Detailed flowchart of the objective function definition of combustion quality and optimization cycle determining the ideal air flow and fuel flow speeds.

The results clearly demonstrate the effectiveness of intelligent control when balancing thermal output stability with active suppression of carbon monoxide emissions. In the first scenario, equal weights were assigned to thermal power and CO reduction (1:1 :0). The system successfully stabilized output at the 9 kW set point while reducing CO concentration from 443 mg/m³ to 378 mg/m³, a decrease of 17.2%. In the second scenario, the weight on CO suppression was increased to 10 (1:10:0). This prioritization achieved a stronger effect, lowering CO emissions from 443 mg/m³ to 315 mg/m³, equivalent to a 34% reduction. Importantly, thermal output remained stable in both cases, confirming the capability of the optimization algorithm to balance energy performance with environmental goals. Conclusion The study demonstrated the successful integration of a wireless monitoring system, intelligent predictive models, and a frequency converter module, enabling smooth and adaptive combustion control. Multi-criteria optimization reduced CO concentrations by up to 34% while maintaining stable boiler output, proving the potential of smart regulation in small-scale biomass systems. Future research should focus on improving controller responsiveness under rapid load fluctuations, integrating additional emission indicators such as particulate matter, and