Many industrial thermal processes are large-scale time-varying nonlinear distributed parameter systems (DPSs). To effectively model such systems, dual extreme learning machine based online spatiotemporal modeling with...
详细信息
Many industrial thermal processes are large-scale time-varying nonlinear distributed parameter systems (DPSs). To effectively model such systems, dual extreme learning machine based online spatiotemporal modeling with adaptive forgetting factor (AFFD-ELM) is proposed in this paper. This method can recursively update the parameters of the low-order temporal model by using newly arriving data under Karhunen- Loeve (KL) based space/time separation. In this way, the time-varying dynamics can be tracked real-time very well as output data increases over time. Besides, since the training samples are usually timeliness, adaptive forgetting factor (AFF) is also embedded in this method to improve the onlinelearning effects by adding a reasonable weight to previous data. This onlinelearning strategy makes the process promising for online modeling under continuously samples environment. The proposed method is utilized for online temperature prediction of the curing oven. Simulation results verify the efficiency and viability of the online spatiotemporal model.
An ensemble of neural networks has been proved to be an effective machine learning framework. However, very limited studies in the current literature examined the neural network ensemble for online regression;furtherm...
详细信息
An ensemble of neural networks has been proved to be an effective machine learning framework. However, very limited studies in the current literature examined the neural network ensemble for online regression;furthermore, these methods were combination of online individual models and did not consider the ensemble diversity. In this paper, a novel online sequential learning algorithm for neural network ensembles for online regression is proposed. The algorithm is built upon the decorrelated neural network ensembles (DNNE) and thus referred to as online-DNNE;so it uses single-hidden layer feed-forward neural networks with random hidden nodes' parameters as ensemble components and introduces negative correlation learning to train base models simultaneously in a cooperative manner which can effectively maintain the ensemble diversity. The online-DNNE only learns the newly arrived data, and the computation complexity is thus reduced. The results of the experiments with benchmarks show the effectiveness and significant advantages of the proposed approach.
Sewage treatment process has the following characteristics: nonlinear, delay etc, and is very complicated to establish the model for its control process. A reasonable model is set up for elaborate prediction effluent ...
详细信息
ISBN:
(纸本)9781538660577
Sewage treatment process has the following characteristics: nonlinear, delay etc, and is very complicated to establish the model for its control process. A reasonable model is set up for elaborate prediction effluent quality, which can satisfy the standard of the effluent water and requirements of energy saving simultaneously. Extreme learning machine (ELM), i.e the machine learning method that new lately developed has high accuracy, reliability and outstanding performance in prediction To get higher prediction effect, in this paper, there are two ways are proposed to improve the ELM, (1) Optimizing the parameters. The ELM whose input weights and bias threshold are optimized by particle swarm optimization algorithm (PSO) and genetic algorithm (GA), respectively;(2) Changing learning mode. To develop an online sequential learning algorithm (OS) for the ELM with additive or radial basis function (RBF) hidden nodes in a unified framework. Therefore, the several comparison approaches refer to optimize the ELM, e.g., PSO-ELM, GA-ELM, OS-ELM are applied to effluent quality prediction, and chemical oxygen demand (COD) is taken as examples in this paper. The results show that PSO-ELM model has remarkably superior performance on effluent quality prediction than peer models in terms of mean absolute error, mean absolute percentage error, root mean square error, and coefficient of determination
In this paper, an Analog-Digital Mixing Network (ADMN) is advanced for simultaneously collecting data and classifying the Power Quality (PQ) events. Based on recently developed Compressed Sampling (CS) theory, power s...
详细信息
In this paper, an Analog-Digital Mixing Network (ADMN) is advanced for simultaneously collecting data and classifying the Power Quality (PQ) events. Based on recently developed Compressed Sampling (CS) theory, power signals are sampled via a new robust and semi-supervised compressive sampling scheme, and then the recorded data are directly used as features for the subsequent classification. Moreover, an online sequential learning algorithm (OSLA) is proposed to learn the training data one-by-one or chunk by chunk, and discard them as long as the training procedure is completed to keep the memory bounded in onlinelearning. Consequently, ADMN can collect data streams and classify them sequentially, which provides a promising way to deal with the "big data". Some experiments are taken on the classification of real PQ events, and the experimental results show the efficiency and superiority of our proposed method to its counterparts. (C) 2015 Elsevier B.V. All rights reserved.
暂无评论