Research And Development Of High Efficiency Gas Stove Based On Machine Learning

In recent years,China has vigorously developed natural gas.One trend of energy structure adjustment is to increase the proportion of natural gas,and the production and consumption of natural gas show a continuous growth trend.At the same time,gas stoves have become a very important part of the family life.Therefore,the high efficiency gas stove not only carries the people’s longing for a better life,but also is a concrete embodiment of the theory of lucid waters and lush mountains.All well-known gas stove enterprises are vigorously carrying out research and development of high efficiency gas stoves.At present,the development of high efficiency gas stoves mainly depends on experience,resulting in long development cycle and high development costs.The training of the machine learning prediction model can be based on a large amount of experimental data of the enterprise.Its model construction does not depend on a complex physical model,and the prediction results can be obtained conveniently and quickly.Therefore,the application of machine learning in the development of gas appliances is expected to achieve better results.In this paper,two methods,random forest and support vector machine,were used to construct the burner model and the ejector model.The genetic algorithm were used to determine the model parameters.When designing the burner,a random forest model and a support vector machine model optimized by genetic algorithm were constructed to predict the thermal efficiency of the gas stove.Both models showed small errors in predicting thermal efficiency: the root mean square error was less than 1 and the average absolute percentage error was less than 1.3%.Both the random forest model and the support vector machine model showed good prediction performance when predicting the ejecting coefficient,and the support vector machine model had higher accuracy.The machine learning method can be effectively used in the research and development of high efficiency gas stoves.According to the optimal parameter combination of the burner and the ejector obtained by the two models,two high efficiency gas stoves were designed,and the predicted thermal efficiency values were 72.66% and 72.76%,respectively.According to the experimental method stipulated in the national standard,the actual thermal efficiency measured by the two gas stoves were 72.30% and 71.72% respectively.