| 381 | 1 | 139 |
| 下载次数 | 被引频次 | 阅读次数 |
针对当前湿法冶金设备控制和智能检测模型较为简单、泛化能力较弱等问题，提出了一种基于深度学习的湿法冶金设备智能控制和故障检测算法模型，首先利用基于SAC深度强化学习算法对湿法冶金设备进行智能控制，再根据智能控制的近期历史数据，采用改进ARIMA算法对设备进行故障检测。为了进一步提升算法的实时性，引入LoRA微调网络对模型进行低参数微调和加速，LoRA微调网络对模型进行低参数微调和加速。该模型对设备智能化控制精度达93.24%,故障检测准确率达91.34%,实际应用效果较好。
Abstract：Aiming at the problems such as relatively simple control and intelligent detection model of hydrometallurgical equipment and weak generalization ability, an algorithm model for intelligent control and fault detection of hydrometallurgical equipment based on deep learning was proposed.Firstly, SAC deep reinforcement learning algorithm was used to perform intelligent control of hydrometallurgical equipment.The improved ARIMA algorithm is used to detect the fault of the equipment.In order to further improve the real-time performance of the algorithm, LoRA fine-tuning network is introduced to fine-tune and accelerate the model with low parameters, and LoRA fine-tuning network to fine-tune and accelerate the model with low parameters.The accuracy of the model is 93.24% and the accuracy of fault detection is 91.34%.The practical application effect is good.
[1] 彭姿馀，王高才，农望.基于SAC的多服务移动边缘计算中任务卸载和资源配置算法[J].计算机应用研究，2023,40(5):1497-1503.PENG Ziyu,WANG Gaocai,NONG Wang.SAC based algorithm for task offloading and resource provisioning in multiple-services mobile edge computing[J].Application Research of Computers,2023,40(5):1497-1503.
[2] 冯剑，姚罕琦，黄啸虎，等.ARIMA算法在工业控制器故障预测的应用[J].自动化仪表，2022,43(11):62-67.FENG Jian,YAO Hanqi,HUANG Xiaohu,et al.Application of ARIMA algorithm to industrial controller fault prediction[J].Process Automation Instrumentation,2022,43(11):62-67.
[3] 邓文艳.湿法冶金设备运行状态的多特征融合检测优化仿真[J].湿法冶金，2023,42(6):666-670.DENG Wenyan.Optimization simulation of multi-feature fusion detection for the operating state of hydrometallurgical equipment[J].Hydrometallurgy of China,2023,42(6):666-670.
[4] 李伟.基于改进PSO-SQP算法优化控制金氰化浸出过程[J].湿法冶金，2023,42(3):317-321.LI Wei.Optimization control of gold cyanide leaching process based on improved PSO-SQP algorithm[J].Hydrometallurgy of China,2023,42(3):317-321.
[5] 薛海莉，李晓敏.基于改进动态因果图的湿法冶金浸出过程故障诊断方法研究[J].湿法冶金，2024,43(2):201-205.XUE Haili,LI Xiaomin.Fault diagnosis method for hydrometallurgical leaching process based on improved dynamic causal diagram[J].Hydrometallurgy of China,2024,43(2):201-205.
[6] 李乐，舒越超，吴健鹏，等.基于PSO-BP神经网络湿式摩擦元件损伤预测模型[J].北京理工大学学报，2022,42(12):1246-1255.LI Le,SHU Yuechao,WU Jianpeng,et al.A damage prediction model of wet friction elements based on PSO-BP neural network[J].Transactions of Beijing institute of Technology,2022,42(12):1246-1255.
[7] 谢思静，文鼎柱.基于联邦分割学习与低秩适应的RoBERTa预训练模型微调方法[J].数据采集与处理，2024,39(3):577-587.XIE Sijing,WEN Dingzhu.Fine-tuning method for pre-trained model RoBERTa based on Federated split learning and Low-rank adaptation[J].Journal of Data Acquisition and Processing,2024,39(3):577-587.
[8] 何龙，张冉阳，赵刚要，等.基于BP神经网络的 GH5188 高温合金本构模型[J].特种铸造及有色合金，2021,41(2):223-226.HE Long,ZHANG Ranyang,ZHAO Gangyou,et al.Constitutive model of GH5188 superalloy based on BP neural network[J].Special Casting & Nonferrous Alloys,2021,41(2):223-226.
[9] 郁洲，毕敬，苑海涛.基于改进 DQN 算法的复杂海战场路径规划方法[J].智能科学与技术学报，2022,4(3):418-425.YU Zhou,BI Jing,YUAN Haitao.A path planning method for complex naval battle field based on an improved DQN algorithm[J].Chinese Journal of Intelligent Science and Technology,2022,4(3):418-425.
[10] 李国梁，李峭，徐亚军，等.基于 DDQN 的片上网络混合关键性消息调度方法[J].北京航空航天大学学报，2021,48(7):1233-1241.LI Guoliang,LI Qiao,XU Yajun,et al.A DDQN-based mixed-criticality messages scheduling method for network-on-chip[J].Journal of Beijing University of Aeronautics and Astronautics,2021,48(7):1233-1241.
[11] 郭佳丽，邢双云，栾昊，等.基于改进的LSTM算法的时间序列流量预测[J].南京信息工程大学学报(自然科学版),2021,13(5):571-575.GUO Jiali,XING Shuangyun,LUAN Hao,et al.Prediction of time series traffic based on improved LSTM algorithm[J].Journal of Nanjing University of Information Science & Technology(Natural Science Edition),2021,13(5):571-575.
[12] 袁利伟，龙皓楠，李斌，等.基于HMM算法的矿山水灾害链复杂演化网络模型构建及应用[J].化工矿物与加工，2023,52(8):47-55.YUAN Liwei,LONG Haonan,LI Bin,et al.Construction and application of complex evolutionary network model of mine water disaster chain based on HMM algorithm[J].Industrial Minerals & Processing,2023,52(8):47-55.
[13] 郭一阳，于炯，杜旭升，等.基于自编码器与集成学习的离群点检测算法[J].计算机应用，2022,42(7):2078-2087.GUO Yiyang,YU Jiong,DU Xusheng,et al.Outlier detection algorithm based on autoencoder and ensemble learning [J].Journal of Computer Applications,2022,42(7):2078-2087.
基本信息:
DOI：10.13355/j.cnki.sfyj.2025.02.011
中图分类号:TP181;TF30
引用信息:
[1]赵铮.基于实时机器学习算法的湿法冶金设备智能控制与故障检测模型设计研究[J].湿法冶金,2025,44(02):222-229.DOI:10.13355/j.cnki.sfyj.2025.02.011.
基金信息: