Medium and Short-Term Electricity Coal Price Forecast Based on Feature Space Transformation and LSTM

doi:10.12068/j.issn.1005-3026.2021.04.005

Abstract

Abstract: Aiming at the problem that there are many influencing factors of electricity coal prices and the nonlinearity and multiple time lags are difficult to model， a data-driven medium and short-term electricity coal price forecasting method based on feature transformation and LSTM was proposed. In order to fully excavate the electricity coal price law contained in the massive data， the feature transformation method of different time scale granularity information was proposed. In order to solve the problem of over-fitting caused by multiple variables and small samples， the main influencing factors of medium and short-term coal prices were screened based on chi-square analysis and correlation coefficient. Based on the LSTM neural network， the feature translation correlation analysis method was used to determine the hysteresis of different influencing feature sequences， and the information redundancy in the model was optimized by the main analytic method， and on this basis， a deep learning model based on feature trends was formed. The years of historical data and comparative analysis with various models show the effectiveness and accuracy of this model.

Key words: coal price; continuous forecasting; feature transformation; long short neural network; data driven; LSTM(long short-term memory)

CLC Number:

LIAO Zhi-wei， CHEN Lin-tao， HUANG Jie-dong， ZHUANG Jing. Medium and Short-Term Electricity Coal Price Forecast Based on Feature Space Transformation and LSTM[J]. Journal of Northeastern University(Natural Science), 2021, 42(4): 483-493.

References

[1]Massimo P，Mauro P，Fabio R.A stochastic network equilibrium model for electric power markets with uncertain demand［J］.Optimization，2020，69:7-8.
[2]Zhu Y Y，Zhu C L，Wu Y X，et al.Research on the influence of coal price fluctuation on electricity price［C］//2019 4th International Conference on Intelligent Green Building and Smart Grid(IGBSG).Yichang，2019:581-585.
[3]Jiang S M，Yang C，Guo J T，et al.ARIMA forecasting of China’s coal consumption，price and investment by 2030［J］.Energy Sources，Part B:Economics，Planning，and Policy，2018，13(3):190-195.
[4]Zhao C H，Dassau E，Zisser H C，et al.Online prediction of subcutaneous glucose concentration for type 1 diabetes using empirical models and frequency-band separation［J］.AIChE Journal，2014，60(2)，574-584.
[5]Arunraj N S，Ahrens D.A hybrid seasonal autoregressive integrated moving average and quantile regression for daily food sales forecasting ［J］.International Journal of Production Economics，2015，170，321-335.
[6]Zheng K，Cai Y，Sun Y.Analysis on influencing factors of coal price in China［C］//11th International Conference on Service Systems and Service Management(ICSSSM).Beijing，2014:1-4.
[7]Mu C L，Liu S，Cheng C.Price transmission effect of coal and electricity linkage:a partial transmission input-output analysis［C］//2014 Seventh International Joint Conference on Computational Sciences and Optimization.Beijing，2014:351-355.
[8]Yuan C Q，Liu S F，Guo B H.The relationship among non-fossil energy consumption，economic growth and coal prices in China［C］// Proceedings of 2011 IEEE International Conference on Grey Systems and Intelligent Services.Nanjing，2011:339-342.
[9]Zhang B，Ma J H.Coal price index forecast by a new partial least-squares regression［J］.Procedia Engineering，2011(15):5025-5029.
[10]Ho S L，Xie M，Goh T N.A comparative study of neural network and Box-Jenkins ARIMA modeling in time series prediction［J］.Computers & Industrial Engineering，2002，42(2/3/4):371-375.
[11]Liu S，Huang W J，Zhang G H.Research and prediction of coal price in China［C］// International Conference on Risk Management & Engineering Management.Beijing，2008:255-259.
[12]王明月.GMDH算法的改进及其在煤炭价格系统中的应用［D］.南京:南京航空航天大学，2016.(Wang Ming-yue.The improvements of GMDH algorithm and its application in coal price system［D］.Nanjing:Nanjing University of Aeronautics and Astronautics，2016.)
[13]Zhang Y A，Yan B B，Memon A.A novel deep learning framework:prediction and analysis of financial time series using CEEMD and LSTM［J］.Expert Systems with Applications，2020，159:113609.
[14]Feng L J，Zhao C H.Dual attention based encoder-decoder:a customized sequence to sequence learning for the soft sensing of multiple quality variables［J］. IEEE Transactions on Neural Networks and Learning Systems，DOI:10.1109/TNNLS.2020.3015929.
[15]Feng Y Z，Zhao H W，Chen Y，et al.Price forecasting algorithm for coal and electricity based on PSO and RBF neural network［C］// 2009 IEEE International Conference on Control and Automation.Christchurch，2009:1365-1369.
[16]Fan X H，Wang L，Li S S.Predicting chaotic coal prices using a multi-layer perceptron network model［J］.Resources Policy，2016，50:86-92.
[17]Olivia B，Lailil M.Comparison of Gaussian and ANOVA kernel in support vector regression for predicting coal price［C］// 2018 International Conference on Sustainable Information Engineering and Technology(SIET).Malang，Indonesia，2018:147-150.
[18]El-HashashE F，El-Absy K M.Methods for determining the tetrachoric correlation coefficient for binary variables［J］.Asian Journal of Probability and Statistics，2018(1):1-12.
[19]鲍硕超，王清，鲍新华，等.辽阳硅石矿岩质边坡岩体裂隙宽度概率统计［J］.东北大学学报(自然科学版)，2014，35(6):885-889.(Bao Shuo-chao，Wang Qing，Bao Xin-hua，et al.Probability statistics analysis on rock fractures width at Liaoyang dians rock slope［J］.Journal of Northeastern University(Natural Science)，2014，35(6):885-889.)
[20]王斐，魏晓童，秦皞.基于sEMG和LSTM的下肢连续运动估计［J］.东北大学学报(自然科学版)，2020，41(3):305-310，342.(Wang Fei，Wei Xiao-tong，Qin Hao.Estimation of lower limb continuous movements based on sEMG and LSTM［J］.Journal of Northeastern University(Natural Science)，2020，41(3):305-310，342.)
[21]Li X B，Jiang B.Minimum system sensitivity study of linear discrete time systems for fault detection［J］.Mathematical Problems in Engineering，2013(1):276987.
[22]Luis E S N，Nathalia D A A，Layzon A L S，et al.Phytochemical profile of different anatomical parts of Jambu(Acmella Oleracea(L.)R.K.Jansen):a comparison between hydroponic and conventional cultivation using PCA and cluster analysis［J］.Food Chemistry，2020，332:127393.(上接第482页)2) 本文方法在Defend_the_center和Health_gathering两个场景下，都表现出了优于Arnold模型的训练过程和测试结果，可以更快、更好地训练出适用于VIZDOOM平台下的三维虚拟射击游戏的智能体.