Multi-step temperature prediction for farms based on iTransformer and LSTM model fusion

doi:10.14088/j.cnki.issn0439-8114.2025.05.021

Abstract

Abstract: To address the nonlinear and complex characteristics of farm temperature data, based on meteorological station data from Huaxing Farm in Changji City, Xinjiang Uygur Autonomous Region,seven features including temperature, ground infrared temperature, dew point temperature, relative humidity, vapor pressure, station pressure, and sea-level pressure were selected as model input features through Spearman correlation analysis,and comparative analysis was conducted among the iTransformer-LSTM model, Transformer model, LSTM model, iTransformer model, and Transformer-LSTM model. The results showed that the iTransformer-LSTM model achieved the best performance. Compared with the optimal baseline model iTransformer, this model reduced the root mean square error (RMSE) by 13.72%, mean absolute error (MAE) by 14.12%, and mean absolute percentage error (MAPE) by 13.61%.The iTransformer-LSTM model could effectively extract time-series feature representations, capture long-term dependencies, and characterize global features and contextual information, making it suitable for multi-feature multi-step time-series temperature prediction tasks.

Key words: iTransformer, LSTM, model fusion, multi-feature, farm temperature, multi-step prediction

CLC Number:

TP183

XIE Qi, ZHANG Tai-hong, LIU Hai-peng. Multi-step temperature prediction for farms based on iTransformer and LSTM model fusion[J]. HUBEI AGRICULTURAL SCIENCES, 2025, 64(5): 134-140.

References

[1] 王春乙, 张继权, 霍治国,等. 农业气象灾害风险评估研究进展与展望[J].气象学报, 2015, 73(1): 1-19.
[2] 李勇,童绍玉. 气候变化对我国农业的影响及其适应研究[J].安徽农学通报, 2023, 29(5): 151-157,175.
[3] 肖风劲, 张海东, 王春乙,等. 气候变化对我国农业的可能影响及适应性对策[J].自然灾害学报, 2006(S1): 327-331.
[4] 王怡, 普运伟. 基于CNN-BiLSTM-Attention融合神经网络的大气温度预测[J].中国水运(下半月), 2023, 23(1): 25-27.
[5] 王强,秦华旺,齐春帅,等.基于自适应SA-PSO改进的XGBoost气温预测方法[J].电子测量技术,2023,46(7): 67-72.
[6] 陈岚,文斌,贺南,等.基于融合模型动态权值的气温预测[J].电子测量技术,2022,45(15): 68-74.
[7] VASWANI A, SHAZEER N, PARMAR N, et al.Attention is all you need[J].Advances in neural information processing systems, 2017, 30: 5998-6008.
[8] LIN T, WANG Y, LIU X, et al.A survey of transformers[EB/OL].(2021-6-15).https://doi.org/10.48550/arXiv.2106.04554.
[9] ZENG A, CHEN M, ZHANG L, et al.Are transformers effective for time series forecasting[J]Proceedings of the AAAI conference on artificial intelligence, 2023, 37(9): 11121-11128.
[10] LIU Y,HU T G, ZHANG H R, et al. iTransformer: Inverted transformers are effective for time series forecasting[EB/OL].(2024-03-14). https://arxiv.org/abs/2310.06625.
[11] HOCHREITER S,SCHMIDHUBER J.Long short-term memory[J].Neural computation, 1997, 9(8), 1735-1780.
[12] CAO K, ZHANG T, HUANG J.Advanced hybrid LSTM-transformer architecture for real-time multi-task prediction in engineering systems[J].Scientific reports , 2024, 14(1): 4890.
[13] SHI J, WANG S, QU P, et al.Time series prediction model using LSTM-Transformer neural network for mine water inflow[J].Scientific reports, 2024, 14(1): 18284.
[14] HAN L, ABDEL-ATY M, YU R, et al.LSTM + Transformer real-time crash risk evaluation using traffic flow and risky driving behavior data[J].IEEE transactions on intelligent transportation systems, 2024, 25(11): 1-13.
[15] MA S, ZHANG T, ZHAO Y, et al.TCLN: A Transformer-based CONV-LSTM network for multivariate time series forecasting[J].Applied intelligence, 2023, 53(23): 28401-28417.
[16] CAI P X, FAN Y C, LEU F Y.Compare encoder-decoder, encoder-only, and decoder-only architectures for text generation on low-resource datasets[A].Advances on broad-band wireless computing, communication and applications: Proceedings of the 16th international conference on broad-band wireless computing[C].Communication and applications, 2021.216-225.
[17] 张驰, 郭媛, 黎明. 人工神经网络模型发展及应用综述[J].计算机工程与应用, 2021, 57(11): 57-69.
[18] 常禧龙, 梁琨, 李文涛. 深度学习优化器进展综述[J].计算机工程与应用, 2024, 60(7): 1-12.