Energy consumption prediction of buildings has drawn attention in the related literature since it is very complex and affected by various factors. Hence, a challenging work is accurately estimating the energy consumption of buildings and improving its efficiency. Therefore, effective energy management and energy consumption forecasting are now becoming very important in advocating energy conservation. Many researchers work on saving energy and increasing the utilization rate of energy. Prior works about the energy consumption prediction combine software and hardware to provide reasonable suggestions for users based on the analyzed results. In this paper, an innovative energy consumption prediction model is established to simulate and predict the electrical energy consumption of buildings. In the proposed model, the energy consumption data is more accurately predicted by using the gradient boosting regression tree algorithm. By comparing the performance index Root Mean Square Error of different prediction models through experiments it is shown that the proposed model obtains lower values on different testing data. More detailed comparison with other existing models through experiments show that the proposed prediction model is superior to other models in energy consumption prediction.

中文翻译：

---

基于梯度提升回归树的家庭能耗预测

建筑能耗预测由于其复杂性、受多种因素的影响而受到相关文献的关注。因此，准确估算建筑物的能耗并提高其效率是一项具有挑战性的工作。因此，有效的能源管理和能源消耗预测对于倡导节能现在变得非常重要。许多研究人员致力于节约能源和提高能源利用率。现有的能耗预测工作结合了软件和硬件，根据分析结果为用户提供合理的建议。本文建立了一种创新的能耗预测模型来模拟和预测建筑物的电能消耗。在所提出的模型中，利用梯度提升回归树算法更准确地预测能耗数据。通过实验比较不同预测模型的性能指标均方根误差，表明所提出的模型在不同的测试数据上获得较低的值。通过实验与其他现有模型进行更详细的比较表明，所提出的预测模型在能耗预测方面优于其他模型。