Self-Organizing Recurrent Stochastic Configuration Networks for Nonstationary Data Modelling

TL;DR

The paper addresses modeling nonstationary nonlinear dynamics in industrial data and proposes self-organizing recurrent stochastic configuration networks (SORSCNs) as an online, adaptive extension of RSCNs. The method blends online projection-based readout updates with a dynamic growing-pruning-growing reservoir strategy guided by an improved sensitivity analysis that incorporates correlation, and uses the RSC algorithm to adjust the reservoir structure as new data arrive. Two industrial datasets (debutanizer soft sensing and short-term power load) show that SORSCN2 outperforms ESN, RSCN, OSL-SCN, and SOMESN in both validation and testing, delivering better generalization with a more compact reservoir. The work demonstrates a practical approach for continual learning in nonstationary environments and outlines pathways for further enhancements in online adaptation and interpretability of randomized learners.

Abstract

Recurrent stochastic configuration networks (RSCNs) are a class of randomized learner models that have shown promise in modelling nonlinear dynamics. In many fields, however, the data generated by industry systems often exhibits nonstationary characteristics, leading to the built model performing well on the training data but struggling with the newly arriving data. This paper aims at developing a self-organizing version of RSCNs, termed as SORSCNs, to enhance the continuous learning ability of the network for modelling nonstationary data. SORSCNs can autonomously adjust the network parameters and reservoir structure according to the data streams acquired in real-time. The output weights are updated online using the projection algorithm, while the network structure is dynamically adjusted in the light of the recurrent stochastic configuration algorithm and an improved sensitivity analysis. Comprehensive comparisons among the echo state network (ESN), online self-learning stochastic configuration network (OSL-SCN), self-organizing modular ESN (SOMESN), RSCN, and SORSCN are carried out. Experimental results clearly demonstrate that the proposed SORSCNs outperform other models with sound generalization, indicating great potential in modelling nonlinear systems with nonstationary dynamics.

Figures (10)

• Figure 1: Architecture of the basic SORSCN.
• Figure 2: Flowchart of the debutanizer column process.
• Figure 3: Distribution of the five input variables (${u}_{1}$-${u}_{5}$) for the debutanizer column process.
• Figure 4: Prediction fitting curves and error values of different models for the debutanizer column.
• Figure 5: Testing NRMSE of SORSCN2 with different parameters (correlation coefficient $\alpha$ and the threshold of model scale adaptability $\gamma$ ) on the debutanizer column process.