Embodiments of the present invention disclose a boiler coal saving control method including a linear relationship model establishing step, an optimization target determination step, and a machine learning step.The linear relational model establishment step is used to establish a Gree level mechanism of the multilevel model, and establishes a linear relational model by the established settlement mechanism.The multi level model's Gree mechanism uses the three characteristic values of boiler load, coal quality, and ambient temperature in boiler basic working conditions as a Gree index to produce primary sequencing.Then, the secondary loading is performed with a boiler load.The optimization target determination step is used to determine the optimization target of the boiler, and the optimization target includes combustion efficiency of the boiler and control of the flue gas nitrification compound compound concentration.The machine learning step performs machine learning based on the data source, and specifically includes a model coding sub step, a knowledge ontology decision sub step, and a target optimization sub step.Since the control method does not need to change the combustion structure and principle of the boiler and does not need to be added, it is possible to provide a safe and reasonable operation proposal through the machine learning method, and to save coal and to increase the efficiency.Diagram
展开▼