Fully Distributed Learning for Deep Random Vector Functional-Link Networks

In the contemporary era, the proliferation of information technology has led to an unprecedented surge in data generation, with this data being dispersed across a multitude of mobile devices. Facing these situations and the training of deep learning model that needs great computing power support, the distributed algorithm that can carry out multi-party joint modeling has attracted everyone’s attention. The distributed training mode relieves the huge pressure of centralized model on computer computing power and communication. However, most distributed algorithms currently work in a master-slave mode, often including a central server for coordination, which to some extent will cause communication pressure, data leakage, privacy violations and other issues. To solve these problems, a decentralized fully distributed algorithm based on deep random weight neural network is proposed. The algorithm decomposes the original objective function into several sub-problems under consistency constraints, combines the decentralized average consensus (DAC) and alternating direction method of multipliers (ADMM), and achieves the goal of joint modeling and training through local calculation and communication of each node. Finally, we compare the proposed decentralized algorithm with several centralized deep neural networks with random weights, and experimental results demonstrate the effectiveness of the proposed algorithm.

基于M-SVR与RVFLNs的高炉十字测温中心温度估计

由于高炉中心温度较高,十字测温中心位置传感器极易损坏,并且更换周期长,因而导致无法及时判断炉顶煤气流分布.采用多输出支持向量回归(M-SVR)和随机权神经网络(RVFLNs)两种数据驱动智能建模方法建立高炉十字测温中心带温度估计模型,并基于实际工业数据对建立的模型进行验证和比较分析.结果表明,在样本数量较小时,M-SVR模型和RVFLNs模型都具有较好的温度估计效果,但当样本数量充足时,M-SVR模型的泛化性能和估计精度更优于RVFLNs模型.

Multivariable Dynamic Modeling for Molten Iron Quality Using Incremental Random Vector Functional-link Networks

Molten iron temperature as well as Si, P, and S contents is the most essential molten iron quality （MIQ） indices in the blast furnace （BF） ironmaking, which requires strict monitoring during the whole ironmaking production. However, these MIQ parameters are difficult to be directly measured online, and large-time delay exists in off-line analysis through laboratory sampling. Focusing on the practical challenge, a data-driven modeling method was presented for the prediction of MIQ using the improved muhivariable incremental random vector functional-link net- works （M-I-RVFLNs）. Compared with the conventional random vector functional-link networks （RVFLNs） and the online sequential RVFLNs, the M-I-RVFLNs have solved the problem of deciding the optimal number of hidden nodes and overcome the overfitting problems. Moreover, the proposed M I RVFLNs model has exhibited the potential for multivariable prediction of the MIQ and improved the terminal condition for the multiple-input multiple-out- put （MIMO） dynamic system, which is suitable for the BF ironmaking process in practice. Ultimately, industrial experiments and contrastive researches have been conducted on the BF No. 2 in Liuzhou Iron and Steel Group Co. Ltd. of China using the proposed method, and the results demonstrate that the established model produces better estima ting accuracy than other MIQ modeling methods.

基于稀疏表示剪枝集成建模的烧结终点位置智能预测

烧结终点位置(BTP)是烧结过程至关重要的参数,直接决定着最终烧结矿的质量.由于BTP难以直接在线检测,因此,通过智能学习建模来实现BTP的在线预测并在此基础上进行操作参数调节对提高烧结矿质量具有重要意义.针对这一实际工程问题,首先提出一种基于遗传优化的Wrapper特征选择方法,可选取使后续预测建模性能最优的特征组合;在此基础上,为了解决单一学习器容易过拟合的问题,提出了基于随机权神经网络(RVFLNs)的稀疏表示剪枝(SRP)集成建模算法,即SRP-ERVFLNs算法.所提算法采用建模速度快、泛化性能好的RVFLNs作为个体基学习器,采用对基学习器基函数与隐层节点数等参数进行扰动的方式来增加集成学习子模型间的差异性;同时,为了进一步提高集成模型的泛化性能与计算效率,引入稀疏表示剪枝算法,实现对集成模型的高效剪枝;最后,将所提算法用于烧结过程BTP的预测建模.工业数据实验表明,所提方法相比于其他方法具有更好的预测精度、泛化性能和计算效率.

RVFLN-based online adaptive semi-supervised learning algorithm with application to product quality estimation of industrial processes

Direct online measurement on product quality of industrial processes is difficult to be realized,which leads to a large number of unlabeled samples in modeling data.Therefore,it needs to employ semi-supervised learning(SSL)method to establish the soft sensor model of product quality.Considering the slow time-varying characteristic of industrial processes,the model parameters should be updated smoothly.According to this characteristic,this paper proposes an online adaptive semi-supervised learning algorithm based on random vector functional link network(RVFLN),denoted as OAS-RVFLN.By introducing a L2-fusion term that can be seen a weight deviation constraint,the proposed algorithm unifies the offline and online learning,and achieves smoothness of model parameter update.Empirical evaluations both on benchmark testing functions and datasets reveal that the proposed OAS-RVFLN can outperform the conventional methods in learning speed and accuracy.Finally,the OAS-RVFLN is applied to the coal dense medium separation process in coal industry to estimate the ash content of coal product,which further verifies its effectiveness and potential of industrial application.

A Machine Learning Approach for Artifact Removal from Brain Signal

Electroencephalography(EEG),helps to analyze the neuronal activity of a human brain in the form of electrical signals with high temporal resolution in the millisecond range.To extract clean clinical information from EEG signals,it is essential to remove unwanted artifacts that are due to different causes including at the time of acquisition.In this piece of work,the authors considered the EEG signal contaminated with Electrocardiogram(ECG)artifacts that occurs mostly in cardiac patients.The clean EEG is taken from the openly available Mendeley database whereas the ECG signal is collected from the Physionet database to create artifacts in the EEG signal and verify the proposed algorithm.Being the artifactual signal is non-linear and non-stationary the Random Vector Functional Link Network(RVFLN)model is used in this case.The Machine Learning approach has taken a leading role in every field of current research and RVFLN is one of them.For the proof of adaptive nature,the model is designed with EEG as a reference and artifactual EEG as input.The peaks of ECG signals are evaluated for artifact estimation as the amplitude is higher than the EEG signal.To vary the weight and reduce the error,an exponentially weighted Recursive Least Square(RLS)algorithm is used to design the adaptive filter with the novel RVFLN model.The random vectors are considered in this model with a radial basis function to satisfy the required signal experimentation.It is found that the result is excellent in terms of Mean Square Error(MSE),Normalized Mean Square Error(NMSE),Relative Error(RE),Gain in Signal to Artifact Ratio(GSAR),Signal Noise Ratio(SNR),Information Quantity(IQ),and Improvement in Normalized Power Spectrum(INPS).Also,the proposed method is compared with the earlier methods to show its efficacy.

高炉铁水质量鲁棒正则化随机权神经网络建模

高炉炼铁过程运行优化与控制依赖于可靠、稳定的难测铁水质量(Molten iron quality, MIQ)指标模型.针对现有MIQ建模方法的不足,本文提出一种新型的数据驱动鲁棒正则化随机权神经网络(Random vector functional-link networks,RVFLNs)算法,用于实现MIQ指标在线估计的鲁棒建模.首先,为了提高建模效率和降低计算复杂度,采用数据驱动典型相关性分析方法从众多变量中提取与MIQ相关性最强的变量作为建模输入变量;其次,由于传统RVFLNs网络的输出权值由最小二乘估计获得,易受离群数据影响而鲁棒性差,引入基于Gaussian分布加权的M估计技术,提出新型鲁棒RVFLNs算法建立多元MIQ指标的鲁棒模型;同时,在鲁棒加权后的最小二乘损失函数基础上,进一步引入L1和L2两个正则化项以构成优化目标函数的Elastic net,用于稀疏化RVFLNs网络的输出权值矩阵,解决RVFLNs网络多重共线性和过拟合的问题.最后,基于某大型高炉工业数据,进行充分数据实验,结果表明所提方法具有更高的建模与估计精度以及较强的鲁棒性能.

磨浆过程输出纤维长度随机分布预测PDF控制

磨浆过程作为制浆和造纸工业最为重要的生产环节之一,其输出纤维长度随机分布(Fiber length stochastic distribution,FLSD)形状直接决定着后续造纸过程的能耗和纸品质量.针对传统的均值和方差难以描述输出FLSD特征,即具有非高斯分布特性,本文提出一种磨浆过程输出FLSD的预测概率密度函数(Probability density function,PDF)控制方法.首先,采用径向基函数(Radical basis function,RBF)神经网络逼近输出FLSD的PDF,在采用迭代学习方法完成基函数参数整定基础上对相应权值向量进行估计.其次,针对权值之间存在强耦合特点,利用随机权神经网络(Random vector functional-networks,RVFLNs)建立表征输入变量和权值之间关系的预测模型.最后,基于建立的输出FLSD模型设计预测PDF控制器,最终实现对期望输出PDF的跟踪控制.基于工业数据实验验证了所提方法的有效性,为磨浆过程优化运行和控制提供理论依据.