Pressure Soft Measurement of Excavator Hydraulic System Based on Mutual Information and BO-LSTM

doi:10.3901/JME.2025.16.384

Abstract

Abstract: The pressure signal is a crucial parameter in the electro-hydraulic ratio control of excavators. Currently, pressure sensors are used to acquire pressure signals. However, these sensors are susceptible to failure when operating under harsh conditions, which can lead to control failure in excavators. Therefore, a pressure soft measurement model based on long-short term memory (LSTM) network is proposed. Firstly, the hydraulic system principle of the SY375IDS excavator is analyzed. Then, the input of the model is determined using mutual information(MI). To address the challenge of selecting appropriate hyperparameters for the LSTM network, the Bayesian optimization(BO) algorithm is employed for tuning. Finally, the effectiveness of the model is verified by using the measured pressure data of the excavator during the 90° digging and dumping operation. The experimental results demonstrate that the BO-LSTM-based pressure soft measurement model exhibits higher accuracy and stability when compared to other models. In order to address the issue of simultaneous failure of multiple sensors, we conducted further research and developed a multi-output traditional prediction(MOTP) model and a multi-output cyclic prediction(MOCP) model based on BO-LSTM. Experimental comparison revealed that the MOCP model exhibited better performance.

Key words: pressure sensors, failure, long-short term memory(LSTM), mutual information, Bayesian optimization

CLC Number:

TP212

TAN Lin, MA Wei, YIN Chenbo, YANG Zhongliang, WANG Dayu. Pressure Soft Measurement of Excavator Hydraulic System Based on Mutual Information and BO-LSTM[J]. Journal of Mechanical Engineering, 2025, 61(16): 384-393.

References

[1] ZHANG Weiting , YANG Dong , WANG Hongchao. Data-driven methods for predictive maintenance of industrial equipment: A Survey[J]. IEEE Systems Journal, 2019, 13(3): 2213-2227.
[2] 彭宇,刘大同. 数据驱动故障预测和健康管理综述[J]. 仪器仪表学报, 2014, 35(3): 481-495. PENG Yu, LIU Datong. Data-driven prognostics and health management : A review of recent advances[J]. Chinese Journal of Scientific Instrument, 2014, 35(3): 481-495.
[3] 王丹,康琦,杨居衡,等. 基于 UKF 的海上天然气井数据驱动软测量方法[J]. 天然气工业, 2022, 42(9): 84-92. WANG Dan, KANG Qi, YANG Juheng, et al. UKF-based data-driven soft sensing for offshore gas wells[J]. Natural Gas Industry, 2022, 42(9): 84-92.
[4] WANG Lei. BP network implementation based on computer Matlab neural network toolbox[J]. Journal of Physics: Conference Series, 2020, 1648(2): 022134.
[5] HUANG Limin, JING Yu, CHEN Hangyu, et al. A regional wind wave prediction surrogate model based on CNN deep learning network[J]. Applied Ocean Research, 2022, 126: 103287.
[6] AMALOU I, MOUHNI N, ABDALI A. Multivariate time series prediction by RNN architectures for energy consumption forecasting[J]. Energy Reports, 2022, 8(9): 1084-1091.
[7] NGUYEN N D, NGUYEN V T. Development of ANN structural optimization framework for data-driven prediction of local two-phase flow parameters[J]. Progress in Nuclear Energy, 2022, 146: 104176.
[8] YUAN Xiaofeng, LI Lin, SHARDT Y A W, et al. Deep learning with spatiotemporal attention-based LSTM for industrial soft sensor model development[J]. IEEE Transactions on Industrial Electronics, 2020, 68(5): 4404-4414.
[9] ZHANG Junhui, WANG Di, XU Bing, et al. Flow control of a proportional directional valve without the flow meter[J]. Flow Measurement and Instrumentation, 2019, 67: 131-141.
[10] HUO Dongyang, CHEN Jinshi, ZHANG Han, et al. Intelligent prediction for digging load of hydraulic excavators based on RBF neural network[J]. Measurement, 2023, 206: 112210.
[11] KILIC E, DOLEN M, CALISKAN H, et al. Pressure prediction on a variable-speed pump controlled hydraulic system using structured recurrent neural networks[J]. Control Engineering Practice, 2014, 26: 51-71.
[12] WU Yuting, YUAN Mei, DONG Shaopeng, et al. Remaining useful life estimation of engineered systems using vanilla LSTM neural networks[J]. Neurocomputing, 2018, 275: 167-179.
[13] EISAID A, HIGGINS J, HIGGINS J, et al. Optimizing long short-term memory recurrent neural networks using ant colony optimization to predict turbine engine vibration[J]. Applied Soft Computing Journal, 2018, 73: 969-991.
[14] ZHANG Bin, ZHANG Shaohui, LI Weihua. Bearing performance degradation assessment using long short-term memory recurrent network[J]. Computers in Industry, 2019, 106: 14-29.
[15] CABRERA D, GUAMÁN A, ZHANG Shaohui, et al. Bayesian approach and time series dimensionality reduction to LSTM-based model-building for fault diagnosis of a reciprocating compressor[J]. Neurocomputing, 2019, 380: 51-66.
[16] 崔佳旭,杨博. 贝叶斯优化方法和应用综述[J]. 软件学报, 2018, 29(10): 3068-3090. CUI Jiaxu, YANG Bo. Survey on bayesian optimization methodology and applications[J]. Journal of Software, 2018, 29(10): 3068-3090.
[17] GUO Xiaohui , WANG Yuanfeng , MEI Shengqi. Monitoring and modelling of PM2.5 concentration at subway station construction based on IoT and LSTM algorithm optimization[J]. Journal of Cleaner Production, 2022, 360: 132179.
[18] 季天瑶,王挺韶. 基于词嵌入与卷积神经网络的建筑能耗预测[J]. 华南理工大学学报(自然科学版), 2021, 49(6): 40-48. JI Tianyao , WANG Tingshao. Building energy consumption prediction based on word embedding and convolutional neural network[J]. Journal of South China University of Technology(Natural Science Edition), 2021, 49(6): 40-48.
[19] 石怀涛,尚亚俊,白晓天,等. 基于贝叶斯优化的 SWDAE-LSTM 滚动轴承早期故障预测方法研究[J]. 振动与冲击, 2021, 40(18): 286-297. SHI Huaitao, SHANG Yajun, BAI Xiaotian, et al. Early fault prediction method combining SWDAE and LSTM for rolling bearings based on Bayesian optimization[J]. Journal of Vibration and Shock, 2021, 40(18): 286-297.
[20] 曹东辉,师建鹏,石向星,等. 电液比例挖掘机闭环控制技术思考[J]. 液压与气动, 2021, 45(6): 142-149. CAO Donghui, SHI Jianpeng, SHI Xiangxing, et al. Thinking of closed loop control technology of electrohydraulic proportional excavator[J]. Chinese Hydraulics ＆ Pneumatics, 2021, 45(6): 142-149.
[21] 刘鸿泉,陈少林,孙晓颖,等. 基于神经网络的核电厂设备易损性分析[J]. 力学学报, 2022, 54(7): 2059-2070. LIU Hongquan, CHEN Shaolin, SUN Xiaoying, et al. Vulnerability analysis of NPP equipment based on neural network[J]. Chinese Journal of Theoretical and Applied Mechanics, 2022, 54(7): 2059-2070.
[22] 王岩,陈耀然,韩兆龙,等. 基于互信息理论与递归神经网络的短期风速预测模型[J]. 上海交通大学学报, 2021, 55(9): 1080-1086. WANG Yan, CHEN Yaoran, HAN Zhaolong, et al. Short-term wind speed forecasting model based on mutual information and recursive neural network[J]. Journal of Shanghai Jiao Tong University, 2021, 55(9): 1080-1086.
[23] LI Daoqing, YU Xiaodong, LIU Shulin, et al. Wind power prediction based on PSO-Kalman[J]. Energy Reports, 2022, 4(8): 958-968.
[24] LIMA F T , SOUZA V. A large comparison of normalization methods on time series[J]. Big Data Research, 2023, 34: 100407.
[25] 李鹏,冯存前,许旭光,等. 一种利用贝叶斯优化的弹道目标微动分类网络[J]. 西安电子科技大学学报, 2021, 48(5): 139-148. LI Peng, FENG Cunqian, XU Xuguang, et al. Ballistic target fretting classification network based on Bayesian optimization[J]. Journal of Xidian University, 2021, 48(5): 139-148.