NGSAII-GPR模型在碳排放短期预测中的应用
The Application of NGSA II-GPR Model in Short-Term Carbon Emission Forecasting
DOI: 10.12677/CSA.2018.811195, PDF,   
作者: 石达顺*：深圳市中金岭南有色金属股份有限公司，广东 韶关；唐朝晖, 王 阳, 牛亚辉：中南大学信息科学与工程学院，湖南 长沙
关键词: 灰色理论；聚类分析；关联性分析；GPR；NGSAII；Grey Theory； Clustering Analysis； Relational Analysis； Gaussian Process Regression； NGSAII
摘要: 针对于采矿过程中以电机为研究对象的碳排放来源的复杂性以及其影响因素的多样性引起的碳排放短期预测精度不高的问题，结合灰色理论提出一种基于NGSAII-GPR模型的铅锌矿采矿过程碳排放预测方法。首先，对碳排放来源及其影响因素进行分析，采用灰色理论进行聚类分析以归并同类因素；其次，根据灰色关联性分析得到主要影响因素；最后，为解决超参数优化确定问题，将带精英策略的非支配排序遗传算法(NGSAII)引入到高斯过程回归(GPR)模型，提出了一种基于NGSAII-GPR的预测模型。经实验证明，相较于其他超参数优化确定方法，NGSAII能更好地对超参数进行优化确定，且相较于其他常规预测模型，NGSAII-GPR能更精确的预测铅锌矿采矿过程的碳排放量，其预测误差更小。
Abstract: Considering the low forecasting accuracy problem caused by the complexity of the carbon emission sources from the motor and the diversity of its impacts during the lead-zinc mine mining process, a carbon emission forecasting method for lead-zinc mine mining process is proposed based on improved Gaussian process regression model combined with the grey theory. Firstly, the sources of carbon emission and their impacts are analyzed and the grey theory is used to cluster and merge the similar impacts. Then, the grey relational analysis is applied to obtain the main impacts. Finally, In order to solve the problem of hyperparameter optimization, the non-dominated sorting genetic algorithm (NGSA II) with elite strategy (NGSA II) is introduced into the Gauss process regression (GPR). Meanwhile the NGSAII-GPR Model is proposed. The result shows that NGSA II can better optimize the hyperparameter when compared with other methods. In addition NGSAII-GPR Model can be used to forecast the short-term carbon emission of lead-zinc mine mining process with high accuracy and minimum error compared with other forecasting models.
文章引用：石达顺, 唐朝晖, 王阳, 牛亚辉. NGSAII-GPR模型在碳排放短期预测中的应用[J]. 计算机科学与应用, 2018, 8(11): 1762-1772. https://doi.org/10.12677/CSA.2018.811195

参考文献

[1] 武娴, 关伟, 许淑婷. 环渤海地区碳排放的省域比较与因素分解[J]. 资源与环境, 2015, 31(9): 1100-1104.
[2] 王少剑, 刘艳艳, 方创琳. 能源消费CO2排放研究综述[J]. 地理科学进展, 2015, 34(2): 151-164.
[3] Khosravi, A., Andrew, L.L.H. and Buyya, R. (2017) Dynamic VM Placement Method for Minimizing Energy and Carbon Cost in Geographically Distributed Cloud Data Centers. IEEE Transaction on Sustainable Computing, 2, 183-196.
[4] Li, X., Zhang, P., Li, S.D., et al. (2016) Study on Calculation Method of Carbon Emission in Utilization of ACCC Conductor in New or Modified Power Lines. China International Conference on Electrici-ty Distribution, Xi’an, 10-13 August 2016.
[5] Diabat, A., Abdallah, T., Al-Refai, A., et al. (2013) Strategic Closed-Loop Facility Location Problem with Carbon Market Trading. IEEE Transactions on Engineering Management, 60, 398-408. [Google Scholar] [CrossRef] 
[6] Liu, X., Moreno, B. and García, A.S. (2016) A grey Neural Network and In-put-Output Combined Forecasting Model Primary Energy Consumption Forecasts in Spanish Economic Sectors. Energy, 115, 1042-1054. [Google Scholar] [CrossRef] 
[7] 张俊深, 袁程炜. 基于BP神经网络与修正GM(1,1)模型的能源消费组合预测[J]. 统计与决策, 2016(5): 90-93.
[8] Perez-Suarez, R. and Lopez-Menendez, A.J. (2015) Growing Green Forecasting Kuznets Curves and Logistic Growth Models CO2 Emissions with Environmental. Environmental Science & Policy, 54, 428-437. [Google Scholar] [CrossRef] 
[9] 方德斌, 董博. 基于GPR模型的中国“十三五”时期碳排放趋势预测[J]. 技术经济, 2015, 34(6): 106-113.
[10] Kong, D.D., Chen, Y.J. and Li, N. (2018) Gaussian Process Regression for Tool Wear Prediction. Mechanical Systems and Signal Processing, 104, 556-574. [Google Scholar] [CrossRef] 
[11] Fang, D.B., Zhang, X.L., Yu, Q., et al. (2018) A Novel Method for Carbon Dioxide Emission Forecasting Based on Improved Gaussian Processes Re-gression. Journal of Cleaner Production, 173, 143-150. [Google Scholar] [CrossRef] 
[12] 甘迪, 柯德平, 孙元章, 等. 基于集合经验模式分解和遗传-高斯过程回归的短期风速概率预测[J]. 电工技术学报, 2015, 30(11): 138-147.
[13] Liu, K., Shen, X.L., Tan, Z.F., et al. (2012) Grey Clustering Analysis Method for Overseas Energy Project Investment Risk Decision. Systems Engineering Procedia, 3, 55-62. [Google Scholar] [CrossRef] 
[14] Li, X.M., Hipel, K.W. and Dang, Y.G. (2015) An Improved Grey Relational Analysis Approach for Panel Data Clusting. Expert System with Application, 42, 9105-9116. [Google Scholar] [CrossRef] 
[15] 王鑫, 李红丽. 台风最大风速预测的高斯过程回归模型[J]. 计算机应用研究, 2015, 32(1): 59-62.
[16] Zhang, J.X., Ma, Y.Z., Yang, T.H., et al. (2017) Estimation of the Pareto Front in Stochastic Simulation through Stochastic Kriging. Simulation Modelling Practice and Theory, 79, 69-86. [Google Scholar] [CrossRef] 
[17] Xue, X.S. and Wang, Y.P. (2017) Improving the Efficiency of NSGA-II Based Ontology Aligning Technology. Data & Knowledge Engineering, 108, 1-14. [Google Scholar] [CrossRef] 
[18] 姜惠兰, 安星, 王亚微, 等. 基于改进NSGA2算法的考虑风机接入电能质量的多目标电网规划[J]. 中国电机工程学报, 2015, 35(21): 5405-5411.
[19] 陈婕, 熊盛武, 林婉如. NSGA-II算法的改进策略研究[J]. 计算机工程与应用, 2011, 47(19): 42-45.
[20] Yang, Y., Cao, L.C., Zhou, Q., et al. (2018) Multi-Objective Process Parameters Optimization of Laser-Magnetic Hybrid Welding Combining Kriging and NSGA-II. Robotics and Computer-Integrated Manufacturing, 49, 253-262. [Google Scholar] [CrossRef]