An echo-state network (ESN) is an effective alternative to gradient methods for training recurrent neural network. However, it difficult determine the structure (mainly reservoir) of ESN match with given application. In this paper, a growing (GESN) proposed design size and topology reservoir automatically. First, GESN makes use block matrix theory add hidden units existing group by group, which leads multiple subreservoirs. Second, every subreservoir weight in created predefined singular...

A nonlinear model predictive control (NMPC) scheme is developed in this paper based on a self-organizing recurrent radial basis function (SR-RBF) neural network, whose structure and parameters are adjusted concurrently the training process. The proposed SR-RBF network represented general form for predicting future dynamic behaviors of systems. To improve modeling accuracy, spiking-based growing pruning algorithm an adaptive learning to tune respectively. Meanwhile, problem, improved gradient...

A novel growing-and-pruning (GP) approach is proposed, which optimizes the structure of a fuzzy neural network (FNN). This GP-FNN based on radial basis function neurons, have center and width vectors. The structure-learning phase parameter-training are performed concurrently. relies sensitivity analysis output. set rules can be inserted or reduced during learning process. algorithm implemented using supervised gradient decent method. convergence GP-FNN-learning process also discussed in this...

Because of their complex behavior, wastewater treatment processes (WWTPs) are very difficult to control. In this paper, the design and implementation a nonlinear model-predictive control (NMPC) system discussed. The proposed NMPC comprises self-organizing radial basis function neural network (SORBFNN) identifier multiobjective optimization method. SORBFNN with concurrent structure parameter learning is developed as model for approximating online states dynamic systems. Then, solution...

In this paper, a self-organizing radial basis function (SORBF) neural network is designed to improve both accuracy and parsimony with the aid of adaptive particle swarm optimization (APSO). proposed APSO algorithm, avoid being trapped into local optimal values, nonlinear regressive developed adjust inertia weight. Furthermore, algorithm can optimize size parameters an RBF simultaneously. As result, APSO-SORBF effectively generate model compact structure high accuracy. Moreover, analysis...

Multiobjective particle swarm optimization (MOPSO) algorithms have attracted much attention for their promising performance in solving multiobjective problems (MOPs). In this paper, an adaptive MOPSO (AMOPSO) algorithm, based on a hybrid framework of the solution distribution entropy and population spacing (SP) information, is developed to improve search terms convergent speed precision. First, global best (gBest) selection mechanism, entropy, introduced analyze evolutionary tendency balance...

To comply with the effluent standards and growing demands for safety reliability, operation of wastewater treatment processes (WWTPs) has been considered as a multiobjective control problem. In this article, data-driven predictive (MOPC) method is developed to deal conflicting objectives improve performance WWTPs. The main contributions MOPC are three folds: first, strategy in design MOPC. And an adaptive fuzzy neural network identifier, using relevant process data, designed catch nonlinear...

An adaptive gradient multiobjective particle swarm optimization (AGMOPSO) algorithm, based on a (stocktickerMOG) method and self-adaptive flight parameters mechanism, is developed to improve the computation performance in this paper. In AGMOPSO stocktickerMOG devised update archive convergence speed local exploitation evolutionary process. Meanwhile, according diversity information of particles, then established balance AGMOPSO. Attributed algorithm not only has faster higher accuracy, but...

Limited operating data resulting from complex and changeable working conditions significantly undermines the performance of deep learning-based methods for rolling bearing fault diagnosis. Generally, this problem can be solved by using generative adversarial network (GAN) to augment data. However, most GAN-based seldom comprehensively consider global interactions local dependencies in raw vibration signals during generation, leading a decline quality generated compromising diagnostic...

Along with the development of machine learning, deep and large language models (LLMs) such as GPT-4 (GPT: Generative Pre-Trained Transformer), artificial intelligence (AI) tools have been playing an increasingly important role in chemical material research to facilitate screening design. Despite exciting progress based AI assistance, open-source LLMs not gained much attention from scientific community. This work primarily focused on metal–organic frameworks (MOFs) a subdomain chemistry...

In this paper, a self-organizing fuzzy-neural-network with adaptive computation algorithm (SOFNN-ACA) is proposed for modeling class of nonlinear systems. This SOFNN-ACA constructed online via simultaneous structure and parameter learning processes. learning, set fuzzy rules can be self-designed using an information-theoretic methodology. The high spiking intensities (SI) are divided into new ones. And the small relative mutual information (RMI) value will pruned in order to simplify FNN...

One of the major obstacles in using radial basis function (RBF) neural networks is convergence toward local minima instead global minima. For this reason, an adaptive gradient multiobjective particle swarm optimization (AGMOPSO) algorithm designed to optimize both structure and parameters RBF paper. First, AGMOPSO algorithm, based on a method self-adaptive flight mechanism, developed improve computation performance. Second, AGMOPSO-based self-organizing network (AGMOPSO-SORBF) can (centers,...