Simulation of Optimization Decision Model for Low-carbon Economy Schemes Based on Machine Learning Algorithms

Abstract

Low-carbon economy (LCE) is a shift from high-carbon intensive industries and activities to low-emission or zero-emission industries and activities, which requires a lot of investment in new technologies and infrastructure, as well as changes in policies and regulations to promote sustainability and reduce emissions. In this article, the back propagation neural network (BPNN) algorithm in machine learning (ML) is applied to the construction of low-carbon economic scheme optimization decision model, and the performance of the model is simulated and tested. The results show that BPNN algorithm has higher accuracy and reliability in predicting the development level of LCE, which is 23.8% higher than Support Vector Machine (SVM) algorithm. Through the comparative study of BPNN algorithm and SVM algorithm in forecasting the development level of LCE, the advantages of BPNN algorithm in forecasting accuracy and application value are verified. BPNN algorithm can better adapt to and deal with the complex data patterns in the field of LCE, improve the accuracy and reliability of prediction, and provide strong technical support and guarantee for the growth of LCE.