Abstract
Filtering neural networks (FNNs) are popular computing frameworks for process system modeling. However, they are vulnerable to non-Gaussian noise and consequently may suffer from low filtering accuracy. To overcome the problem, in this paper, a novel model construction algorithm by combining the improved clustering kernel function smoothing technique and the particle filter neural network (ICKS-PFNN) is proposed. Specifically, ICKS-PFNN firstly presents a construction framework for particle filter neural network (PFNN), which utilizes the dynamic approximation of particles to adjust the NN’s weights and thresholds in real time. Then, the proposed model uses kernel fuzzy C-means algorithm to uncover clusters in the particles of PFNN. A novel proportional distribution sampling strategy is adopted to maintain the diversity in particle clusters, through merging the inferior and superior particles to generate new particles based on the set proportional factors, rather than directly eliminating particles. At last, the estimation of the PFNN model is achieved by utilizing a kernel function smoothing method to update the particles in each cluster. The proposed model has been tested on the real-world system for aluminium electrolysis manufacturing and compared with several closely related frameworks. The experimental results show ICKS-PFNN obtains a superb performance when compared with other baselines. ICKS-PFNN is able to tackle noise and improve the prediction accuracy when dealing with non-Gaussian systems. Successfully applying the proposed framework in aluminium electrolysis manufacturing broadens the practical impact of FNN systems.
Similar content being viewed by others
Abbreviations
- NN:
-
Neural network
- PF:
-
Particle filter
- EKF:
-
Extended Kalman filter
- UKF:
-
Unscented Kalman filter
- FNN:
-
Filtering neural network
- FCM:
-
Fuzzy C-mean clustering
- MLR:
-
Multiple linear regression
- SSE:
-
Sum squared error
- MAE:
-
Mean absolute error
- RMSE:
-
Root mean squared error
- AEM:
-
Aluminium electrolysis manufacturing
- ICKS:
-
Improved clustering kernel function smoothing
- CKS-PFNN:
-
Particle filter neural network model based on the clustering kernel function smoothing method
- ICS-PFNN:
-
Particle filter neural network model based on the improved clustering smoothing method
- ICKS-PFNN:
-
Particle filter neural network model based on the improved clustering kernel function smoothing method
- EKFNN:
-
Extended Kalman filter neural network
- UKFNN:
-
Unscented Kalman filter neural network
- BPNN:
-
Back-propagation neural network
- PFNN:
-
Particle filter neural network
- PDF:
-
Probability density function
- KFCM:
-
Fuzzy kernel C-mean clustering
- NLMR:
-
Multiple nonlinear regression
- MSE:
-
Mean squared error
- MRE:
-
Mean relative error
- R:
-
Correlation coefficient
References
Gui W, Yue W, Xie Y, Zhang H, Yang C (2018) A review of intelligent optimal manufacturing for aluminum reduction production. Zidonghua Xuebao/Acta Automat Sin 44(11):1957–1970
Yi J, Huang D, Siyao F, He H, Li T (2016) Multi-objective bacterial foraging optimization algorithm based on parallel cell entropy for aluminum electrolysis production process. IEEE Trans Ind Electron 63(4):2488–2500
Zhou W, Shi J, Yin G, He W, Yi J (2020) Optimal control for aluminum electrolysis process using adaptive dynamic programming. IEEE Access 8(1–1):12
Yang C, Zhou L, Huang K, Ji H, Long C, Chen X, Xie Y (2019) Multimode process monitoring based on robust dictionary learning with application to aluminium electrolysis process. Neurocomputing 332:305–319
Yi W, Li W, Wang Y, Zhang K (2019) Remaining useful life prediction of lithium-ion batteries using neural network and bat-based particle filter. IEEE Access 7:54843–54854
Lebreux M, Desilets M, Allard F, Micheau P, Blais A (2020) An on-line estimation tool for predicting the time-varying ledge profile inside aluminum electrolysis cells. Numer Heat Transf Part A Appl 77(2):134–161
Bustillo A, Urbikain G, Perez JM, Pereira OM, Luis N, de Lacalle L (2018) Smart optimization of a friction-drilling process based on boosting ensembles. J Manuf Syst 48:108–121
Yi J, Bai J, Zhou W, He H, Yao L (2018) Operating parameters optimization for the aluminum electrolysis process using an improved quantum-behaved particle swarm algorithm. IEEE Trans Ind Inf 14(8):3405–3415
Huang K, Wen H, Ji H, Cen L, Chen X, Yang C (2019) Nonlinear process monitoring using kernel dictionary learning with application to aluminum electrolysis process. Control Eng Pract 89:94–102
Chen Z, Li Y, Chen X, Yang C, Gui W (2017) Semantic network based on intuitionistic fuzzy directed hyper-graphs and application to aluminum electrolysis cell condition identification. IEEE Access 5:20145–20156
Yue W, Chen X, Gui W, Xie Y, Zhang H (2017) A knowledge reasoning fuzzy-Bayesian network for root cause analysis of abnormal aluminum electrolysis cell condition. Front Chem Eng China 11(3):414–428
Huang K, Yiming W, Yang C, Peng G, Shen W (2020) Structure dictionary learning-based multimode process monitoring and its application to aluminum electrolysis process. IEEE Trans Autom Sci Eng 17(4):1989–2003
Khera N, Khan SA (2017) Prognostics of aluminum electrolytic capacitors using artificial neural network approach. Microelectron Reliab 81(81):328–336
Chenhua X, Wang L, Lin X, Li Z, Xin Yu (2016) Intelligent optimization of cell voltage for energy saving in process of electrolytic aluminum. J Adv Comput Intell Intell Inf 20(2):231–237
Li J, Zhou P, Pian J (2014) Multi-fault diagnosis of aluminum electrolysis based on modular fuzzy neural networks. Asian J Chem 26(11):3339–3343
Zeng S, Bing L (2016) Application of genetic neural network for diagnosis of anode anomaly and metal wave in aluminum electrolysis. Int Conf Artif Intell: Technol Appl 2016:325–328
Bak C, Roy AG, Son H (2021) Quality prediction for aluminum diecasting process based on shallow neural network and data feature selection technique. CIRP J Manuf Sci Technol 33:327–338
Ding W, Yao L, Li Y, Long W, Yi J, He T (2021) Dynamic evolutionary model based on a multi-sampling inherited hapfnn for an aluminium electrolysis manufacturing system. Appl Soft Comput 99:106925
Acosta M, Kanarachos S (2018) Tire lateral force estimation and grip potential identification using neural networks, extended kalman filter, and recursive least squares. Neural Comput Appl 30(11):3445–3465
Pesce V, So Silvestrini, Lavagna M (2020) Radial basis function neural network aided adaptive extended Kalman filter for spacecraft relative navigation. Aerosp Sci Technol 96:105527
Sassan Goleijani and Mohammad Taghi Ameli (2019) An agent-based approach to power system dynamic state estimation through dual unscented Kalman filter and artificial neural network. Soft Comput 23(23):12585–12606
Wang Y, Chai S, Nguyen HD (2019) Unscented Kalman filter trained neural network control design for ship autopilot with experimental and numerical approaches. Appl Ocean Res 85:162–172
Yao L, Li T, Li Y, Long W, Yi J (2019) An improved feed-forward neural network based on ukf and strong tracking filtering to establish energy consumption model for aluminum electrolysis process. Neural Comput Appl 31(8):4271–4285
Peng K, Jiao R, Dong J, Pi Y (2019) A deep belief network based health indicator construction and remaining useful life prediction using improved particle filter. Neurocomputing 361:19–28
Zhang X, Liu D, Lei B, Liang J, Ji R (2021) An intelligent particle filter with resampling of multi-population cooperation. Digit Signal Process 115:103084
Qin W, Lv H, Liu C, Nirmalya D, Jahanshahi P (2019) Remaining useful life prediction for lithium-ion batteries using particle filter and artificial neural network. Ind Manag Data Syst 120(2):312–328
Cadini F, Sbarufatti C, Corbetta M, Cancelliere F, Giglio M (2019) Particle filtering-based adaptive training of neural networks for real-time structural damage diagnosis and prognosis. Struct Control Health Monitor 26:12
Kasantikul K, Yang D, Wang Q, Lwin A (2018) A novel wind speed estimation based on the integration of an artificial neural network and a particle filter using beidou geo reflectometry. Sensors 18(10):3350
Wang D, Yang F, Tsui K, Zhou Q, Bae SJ (2016) Remaining useful life prediction of lithium-ion batteries based on spherical cubature particle filter. IEEE Trans Instrum Meas 65(6):1282–1291
Wei W, Gao S, Zhong Y, Chengfan G, Gaoge H (2018) Adaptive square-root unscented particle filtering algorithm for dynamic navigation. Sensors 18(7):2337
Taifu LI (2014) Kalman artificial neural network with measurable noise estimation by gamma test for dynamic industrial process modeling. J Mech Eng 50(18):29
Tai-Fu LI, Yao LZ, Jun YI, Wen-Jin HU, Ying-Ying SU, Jia W (2014) An improved ukfnn based on square root filter and strong tracking filter for dynamic evolutionary modeling of aluminum reduction cell. Acta Autom Sin 40(3):522–530
Fazli S, Ghiri SF (2013) Robust fuzzy c-means clustering with spatial information for segmentation of brain magnetic resonance images. Int J Sci Eng Investig 2:12
Wang Y, Zhen W, Xia A, Guo C, Chen Y, Yang Y, Tang Z (2019) Energy management strategy for hev based on kfcm and neural network. Concurr Comput Pract Exp 31(10):e4838
Khanlari M, Ehsanian M (2017) An improved kfcm clustering method used for multiple fault diagnosis of analog circuits. Circuits Syst Signal Process 36(9):3491–3513
Zhang H, Wang S, Xu X, Chow TWS, Jonathan Wu QM (2018) Tree2vector: learning a vectorial representation for tree-structured data. IEEE Trans Neural Netw Learn Syst 29(11):5304–5318
Liu Y, Ma S, Du X (2020) A Novel Effective Distance Measure and a Relevant Algorithm for Optimizing the Initial Cluster Centroids of K-means. IEEE Access 2020:3044069
Yang X, Zhang G, Jie L, Ma J (2011) A kernel fuzzy c-means clustering-based fuzzy support vector machine algorithm for classification problems with outliers or noises. IEEE Trans Fuzzy Syst 19(1):105–115
Zhao F, Jiao L, Liu H (2013) Kernel generalized fuzzy c-means clustering with spatial information for image segmentation. Digital Signal Process 23(1):184–199
Lin HY (2020) Feature clustering and feature discretization assisting gene selection for molecular classification using fuzzy c-means and expectation maximization algorithm. J Supercomput 77:1–17
Zhang D, Chen S (2003) Clustering incomplete data using kernel-based fuzzy c-means algorithm. Neural Process Lett 18(3):155–162
Liu Y, Li Z, Xiong H, Gao X, Wu J (2010) Understanding of internal clustering validation measures, pp 911–916
Zhe Z, Xiyu L, Lin W (2020) Spectral clustering algorithm based on improved gaussian kernel function and beetle antennae search with damping factor. Comput Intell Neurosci. https://doi.org/10.1186/s12859-018-2402-0
Fashoto SG, Mbunge E, Ogunleye G, Burg J (2021) Implementation of machine learning for predicting maize crop yields using multiple linear regression and backward elimination. Malays J Comput 6(1):679–697
El Genidy M (2019) Multiple nonlinear regression of the Markovian arrival process for estimating the daily global solar radiation. Commun Stat Theory Methods 48(22):5427–5444
Moayad A, Weishi S, Xumin L, Qi Y (2020) A bayesian learning model for design-phase service mashup popularity prediction. Expert Syst Appl 149:113231
Gu Y, Lu W, Xu X, Qin L, Zhang H (2019) An improved Bayesian combination model for short-term traffic prediction with deep learning. IEEE Trans Intell Transp Syst 99:1–11
Aladag CH, Yolcu U, Egrioglu E (2013) A new multiplicative seasonal neural network model based on particle swarm optimization. Neural Process Lett 37(3):251–262
Li W, Kong D, Jinran W (2017) A novel hybrid model based on extreme learning machine, k-nearest neighbor regression and wavelet denoising applied to short-term electric load forecasting. Energies 10(5):694
Acknowledgements
This study was supported by the National Natural Science Foundation of China (No.518 05059), Chongqing Research Program of Basic Research and Frontier Technology under grant (cstc2018jcyjA X0350 and cstc2018jcyjA1663) and Special Project of Technological Innovation and Application Development in Chongqing (No. cstc2019jscx-msxmX0054).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no conflict of interest.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Ding, W., Yao, L., Li, Y. et al. Incremental learning model based on an improved CKS-PFNN for aluminium electrolysis manufacturing. Neural Comput & Applic 34, 2083–2102 (2022). https://doi.org/10.1007/s00521-021-06530-5
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00521-021-06530-5