Application of data mining algorithm in energy-saving renovation prediction of urban landscape buildings

Abstract

In order to solve the problem of the heavy burden of electricity and energy consumption of urban landscape buildings, a prediction model for energy conservation and reconstruction of urban landscape buildings was established by combining data mining algorithms. Firstly, the energy demand and consumption of urban landscape buildings are analyzed, and the energy consumption of buildings is predicted by relevant mathematical calculation methods. Then, combined with data mining technology, effective information is extracted from the basic building information of urban landscape, daily energy consumption, operation data and other aspects. Finally, the prediction model of building energy conservation transformation based on data mining algorithm is constructed, and the Bayesian energy model is used for parameter correction. Test the performance of the model and find that under the single influence factor of different energy consumption, the change trend of total energy consumption is different. Among them, lighting power density factor has the greatest impact on energy consumption, and its annual energy consumption change rate can reach about 0.35. Applying the prediction model to the energy consumption prediction of 15 urban single buildings, it was found that the total energy consumption of the buildings before the transformation was much higher than the total energy consumption after the transformation, and the energy saving rate of the whole observation sample building group was as high as 18.5%, while the highest energy saving rate of the single buildings reached 30.1%. To sum up, the model has good prediction ability. Applying it to the energy conservation prediction of urban landscape buildings can better complete the energy prediction task and achieve the energy conservation goal.