基于灰色神经网络模型的企业碳排放峰值预测

Grey Neural Network Model for Prediction of Carbon Emissions[JZ)][HS)]

摘要:

为预测企业碳排放峰值,帮助企业设计碳排放的减排路径,需要对企业碳排放峰值预测方法进行研究;当前采用基于TFDI模型的预测模型对企业碳排放峰值进行预测,预测过程中无法全面考虑企业碳排放影响因素,导致预测企业碳排放峰值出现误差;为此,提出一种基于灰色神经网络模型的企业碳排放峰值预测模型;该模型是以灰色模型为基础,与神经网络相融合构建的灰色神经网络,将模型中企业碳排放原数据进行叠加,并用微分方程表示,将VSTE算法作为灰色神经网络模型预测的基础算法,计算企业碳排放路径碳排放值,满足高斯分布随机函数,以此进行企业碳排放峰值的预测;实验结果证明,所提模型可以准确预测企业碳排放峰值,有效帮助企业设计碳排放减排路径。

Abstract:

In order to predict the peak of carbon emissions, and to help enterprises design the path of carbon emission reduction, it is necessary to study the prediction method of carbon emissions. At present, the prediction model based on TFDI model is used to predict the peak of carbon emissions, which can not fully consider the influence factors of carbon emissions in the process of prediction, leading to the prediction of the peak value of carbon emissions. Therefore, this paper puts forward a new model of carbon emission prediction based on grey neural network model. The model is based on the grey model, and neural network by combining grey neural network model, the corporate carbon raw data are superimposed, and is represented by differential equations, the VSTE algorithm as the basic algorithm grey neural network prediction model, the calculation of corporate carbon emissions path of carbon emissions, to meet the random Gauss distribution function in order to forecast, corporate carbon emissions to peak. The experimental results show that the proposed model can accurately predict the peak of carbon emissions, and help enterprises to design a path of carbon emission reduction.