Research on Real-time Prediction of $\text{CO}_{2}$ Emissions from Thermal Power Plants Based on Multilayer Perceptron

As China's demand for electricity continues to grow, thermal power generation, while securing energy supply, also brings about large amounts of CO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> emissions. An accurate and timely grasp of CO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> emissions from thermal power plants is important for emission reduction. In this paper, the multilayer perceptron (MLP) model is used to predict the CO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> emissions in real time by combining the actual data from thermal power plants. The maximum error between the prediction results and the raw data is only 4.98%, and the correlation coefficient is as high as 0.999. In addition, this paper also investigates the influence of different input parameters (power generation, coal consumption, stack flue gas <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\text{flow}$</tex> SO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> emissions and <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\text{NO}_{\mathrm{x}}$</tex> emissions) on the prediction results, and finds that the prediction of five inputs is significantly more effective compared to any four inputs. And the effects of SO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> and <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\text{NO}_{\mathrm{x}}$</tex> emissions on the prediction results are more significant. This study provides data support and theories for the development of emission control strategies for thermal power plants.