基于深度案例推理的城市固废焚烧过程故障诊断方法

Fault Diagnosis Method of Municipal Solid Waste Incineration Process Based on Deep Case-based Reasoning

摘要

针对城市固废焚烧(municipal solid waste incineration, MSWI)过程设备众多、工况多变导致故障诊断准确性不高的问题, 提出一种深度案例推理方法, 以实现故障的检测、定位与隔离。在案例检索环节, 通过孪生神经网络(Siamese neural networks, SNN)和自注意力机制对案例之间的相似度进行深度评估, 实现了故障检测与故障定位; 在案例改编环节, 采用基于集成深度Q网络(deep Q network, DQN)的案例差异启发式改编方法获得多目标故障解决方案, 实现了故障隔离。实验结果表明, 提出的方法能够准确地检测故障、判断故障类型并给出解决方案, 可以促进MSWI过程的安全、平稳运行。

Abstract

To address the challenge that the traditional methods are difficult to achieve effective fault diagnosis accuracy due to the large number of equipment and variable working conditions in the municipal solid waste incineration (MSWI) process, a deep case-based reasoning method is proposed to realize fault detection, location and isolation. For case retrieval, the similarity between cases was deeply evaluated by Siamese neural network (SNN) and self-attention mechanism to realize fault detection and fault location. For case adaptation, a case difference heuristic adaptation method based on integrated deep Q network (DQN) was used to obtain a multi-objective fault solution and realize fault isolation. Results show that the method can accurately detect faults, identify fault types and give solutions, which can promote the safe and stable operation of the MSWI process.

关键词

故障诊断 / 案例推理 / 相似性度量 / 孪生神经网络(Siamese neural networks, SNN) / 自注意力机制 / 深度Q网络(deep Q network, DQN)

Key words

fault diagnosis / case-based reasoning / similarity measurement / Siamese neural network (SNN) / self-attention mechanisms / deep Q network (DQN)

参考文献

[1] 汤健, 夏恒, 余文, 等 . 城市固废焚烧过程智能优化控制研究现状与展望[J]. 自动化学报, 2023, 49(10): 2019-2059.

[2] Tang J, Xia H, Yu W, et al. Research status and prospects of intelligent optimization control for municipal solid waste incineration process[J]. Acta Automatica Sinica, 2023, 49(10): 2019-2059.

[3] Gao Z W, Cecati C, Ding S . A survey of fault diagnosis and fault—tolerant techniques part Ⅱ: fault diagnosis with knowledge—based and hybrid/active approaches[J]. IEEE Transactions on Industrial Electronics, 2015, 62(6): 3768-3774.

[4] Qin S J . Survey on data—driven industrial process monitoring and diagnosis[J]. Annual Reviews in Control, 2012, 36(2): 220-234.

[5] Zhao J S, Huang J C, Sun W . On—line early fault detection and diagnosis of municipal solid waste incinerators[J]. Waste Management, 2008, 28(11): 2406-2414.

[6] Chen J C, Lin K Y . Diagnosis for monitoring system of municipal solid waste incineration plant[J]. Expert Systems with Applications, 2008, 34(1): 247-255.

[7] Tavares G, Zsigraiová Z, Semiao V, et al. Monitoring, fault detection and operation prediction of MSW incinerators using multivariate statistical methods[J]. Waste Management, 2011, 31(7): 1635-1644.

[8] Ding C X, Yan A J . Fault detection in the MSW incineration process using stochastic configuration networks and case—based reasoning[J]. Sensors, 2021, 21(21): 7356.

[9] Wang S Q, Zhou C C, Cao Y, et al. Multiblock dynamic enhanced canonical correlation analysis for industrial MSW combustion state monitoring[J]. Control Engineering Practice, 2023, 138: 105612.

[10] 赵健程, 冯良骏, 岳嘉祺, 等 . 从零样本学习理论模型到工业应用: 动机、演变与挑战[J]. 控制与决策, 2024, 39(9): 2833-2857.

[11] Zhao J C, Feng L J, Yue J Q, et al. From zero—shot learning theoretical model to its industrial application: motivation, evolution and challenges[J]. Control and Decision, 2024, 39(9): 2833-2857. (in Chinese)

[12] Ding S S, Chen R W, Liu H, et al. ASG—HOMGAT: a high—order multi—head graph attention network with adaptive small graph structure for rolling bearing fault diagnosis[J]. Measurement Science and Technology, 2024, 35(6): 065018.

[13] Wu W Q, Song C Y, Zhao J, et al. Iterative deep graph learning and embedding for industrial fault diagnosis with limited labeled data[J]. IEEE Transactions on Instrumentation and Measurement, 2024, 73: 3501715.

[14] 金江涛, 许子非, 李春, 等 . 基于深度学习与混沌特征融合的滚动轴承故障诊断[J]. 控制理论与应用, 2022, 39(1): 109-116.

[15] Jin J T, Xu Z F, Li C, et al. Rolling bearing fault diagnosis based on deep learning and chaotic feature fusion[J]. Control Theory & Applications, 2022, 39(1): 109-116. (in Chinese)

[16] Xiang L, Zhang X Y, Zhang Y, et al. A novel method for rotor fault diagnosis based on deep transfer learning with simulated samples[J]. Measurement, 2023, 207: 112350.

[17] Hakim M, Omran A A B, Ahmed A N, et al. A systematic review of rolling bearing fault diagnoses based on deep learning and transfer learning: taxonomy, overview, application, open challenges, weaknesses and recommendations[J]. Ain Shams Engineering Journal, 2023, 14(4): 101945.

[18] Aamodt A, Plaza E . Case—based reasoning: foundational issues, methodological variations, and system approaches[J]. AI Communications, 1994, 7(1): 39-59.

[19] Homem T P D, Santos P E, Costa A H R, et al. Qualitative case—based reasoning and learning[J]. Artificial Intelligence, 2020, 283: 103258.

[20] Bromley J, Bentz J W, Bottou L, et al. Signature verification using a “Siamese” time delay neural network[J]. International Journal of Pattern Recognition and Artificial Intelligence, 1993, 7(4): 669-688.

[21] Vaswani A, Shazeer N, Parmar N, et al. Attention is all you need[C]// 2017 31st Annual Conference on Neural Information Processing Systems. Red Hook, NY: Curran Associates, Inc., 2017: 5998-6008.

[22] Ye X M, Leake D, Jalali V, et al. Learning adaptations for case—based classification: a neural network approach[M]// Case—based Reasoning Research and Development. Cham: Springer, 2021: 279-293.

[23] Mnih V, Kavukcuoglu K, Silver D, et al. Human—level control through deep reinforcement learning[J]. Nature, 2015, 518(7540): 529-533.

基金资助

国家自然科学基金资助项目(62373017)

北京市自然科学基金资助项目(4212032)