一种时空协同的图卷积长短期记忆网络及其工业软测量应用

软测量技术的发展有效解决了工业过程中对于难以直接测量的质量变量的感知困难,为过程的控制与优化提供了有力保障.通常在含有多个质量变量的过程中,样本间的时序关系和多个质量变量间相互影响的空间关系能够反映过程本身的特性,这种时空特性的挖掘有益于软测量模型性能的提升,而传统软测量方法往往局限于对时序关系的学习而并未考虑对质量变量间的空间关系进行有效利用.对此,提出一种时空协同的图卷积长短期记忆网络(graph convolution long short-term memory networks, GC-LSTM),并应用于工业软测量场景.采用多通道网络结构将图卷积网络的空间关系挖掘能力与长短期记忆网络的时序关系学习能力相结合,对过程进行时空协同学习以实现软测量应用.具体而言,每条通道用于对每种质量变量进行独立学习;对于过程的时序特性,利用各通道内的长短期记忆网络提取针对不同质量变量的时序特征;对于过程的空间特性,构建质量变量间空间关系的图结构,采用跨通道的图卷积运算将不同通道内不同质量变量的时序特征基于空间关系进行融合,得到兼具过程时空特性的特征,从而在软测量建模中实现过程时空协同学习与融合...     

Bayesian on-line spectral change point detection: a soft computing approach for on-line ASR

Current automatic speech recognition (ASR) works in off-line mode and needs prior knowledge of the stationary or quasi-stationary test conditions for expected word recognition accuracy. These requirements limit the application of ASR for real-world applications where test conditions are highly non-stationary and are not known a priori. This paper presents an innovative frame dynamic rapid adaptation and noise compensation technique for tracking highly non-stationary noises and its application for on-line ASR. The proposed algorithm is based on a soft computing model using Bayesian on-line inference for spectral change point detection (BOSCPD) in unknown non-stationary noises. BOSCPD is tested with the MCRA noise tracking technique for on-line rapid environmental change learning in different non-stationary noise scenarios. The test results show that the proposed BOSCPD technique reduces the delay in spectral change point detection significantly compared to the baseline MCRA and its derivatives. The proposed BOSCPD soft computing model is tested for joint additive and channel distortions compensation (JAC)-based on-line ASR in unknown test conditions using non-stationary noisy speech samples from the Aurora 2 speech database. The simulation results for the on-line AR show significant improvement in recognition accuracy compared to the baseline Aurora 2 distributed speech recognition (DSR) in batch-mode.     

Development of soft and distributed tactile sensors and the application to a humanoid robot

This paper describes a comprehensive tactile sensor system which can cover wide areas of full-body robots. Based on design criteria which are introduced from requirements, we develop two types of tactile sensor elements. One is a multi-valued touch sensor which has multi-level pressure thresholds. It is capable of covering wide areas of robot surfaces. The other is made of soft, conductive gel, which has the advantage of compliance compared with other sheet-type tactile sensors. With these two types sensors, we develop the tactile sensor system on the full-body robot 'H4'. Details of the sensor system on the robot and some experiments using tactile information are described.     

Robust semi-supervised mixture probabilistic principal component regression model development and application to soft sensors

Traditional data-based soft sensors are constructed with equal numbers of input and output data samples, meanwhile, these collected process data are assumed to be clean enough and no outliers are mixed. However, such assumptions are too strict in practice. On one hand, those easily collected input variables are sometimes corrupted with outliers. On the other hand, output variables, which also called quality variables, are usually difficult to obtain. These two problems make traditional soft sensors cumbersome. To deal with both issues, in this paper, the Student's t distributions are used during mixture probabilistic principal component regression modeling to tolerate outliers with regulated heavy tails. Furthermore, a semi-supervised mechanism is incorporated into traditional probabilistic regression so as to deal with the unbalanced modeling issue. For simulation, two case studies are provided to demonstrate robustness and reliability of the new method.     

Bayesian and non-bayesian evidential updating

Four main results are arrived at in this paper. (1) Closed convex sets of classical probability functions provide a representation of belief that includes the representations provided by Shafer probability mass functions as a special case. (2) The impact of “uncertain evidence” can be (formally) represented by Dempster conditioning, in Shafer's framework. (3) The impact of “uncertain evidence” can be (formally) represented in the framework of convex sets of classical probabilities by classical conditionalization. (4) The probability intervals that result from Dempster-Shafer updating on uncertain evidence are included in (and may be properly included in) the intervals that result from Bayesian updating on uncertain evidence.     

A model updating strategy for predicting time series with seasonal patterns

Traditional methodologies for time series prediction take the series to be predicted and split it into training, validation, and test sets. The first one serves to construct forecasting models, the second set for model selection, and the third one is used to evaluate the final model. Different time series approaches such as ARIMA and exponential smoothing, as well as regression techniques such as neural networks and support vector regression, have been successfully used to develop forecasting models. A problem that has not yet received proper attention, however, is how to update such forecasting models when new data arrives, i.e. when a new event of the considered time series occurs.This paper presents a strategy to update support vector regression based forecasting models for time series with seasonal patterns. The basic idea of this updating strategy is to add the most recent data to the training set every time a predefined number of observations takes place. This way, information in new data is taken into account in model construction. The proposed strategy outperforms the respective static version in almost all time series studied in this work, considering three different error measures.     

A successful application of a Test-Driven Development strategy in the industrial environment

Unit Test-Driven Development (UTDD) and Acceptance Test-Driven Development (ATDD) are software development techniques to incrementally develop software where the test cases, unit or acceptance tests respectively, are specified before the functional code. There are little empirical evidences supporting or refuting the utility of these techniques in an industrial context. Just a few case studies can be found in literature within the industrial environment and they show conflicting results (positive, negative and neutral). In this report, we present a successful application of UTDD in combination with ATDD in a commercial project. By successful we mean that the project goals are reached without an extra economic cost. All the UTDD and ATDD implementations are based on the same basic concepts, but they may differ in specific adaptations to each project or team. In the implementation presented here, the business requirements are specified by means of executable acceptance tests, which then are the input of a development process where the functional code is written in response to specific unit tests. Our goal is to share our successful experience in a specific project from an empirical point of view. We highlight the advantages and disadvantages of adopting UTDD and ATDD and identify some conditions that facilitate success. The main conclusions we draw from this project are that ATDD contributes to clearly capture and validate the business requirements, but it requires an extensive cooperation from the customer; and that UTDD has not a significant impact neither on productivity nor on software quality. These results cannot be generalized, but they point out that under some circumstances a test-driven development strategy can be a possible option to take into account by software professionals.     

Incremental Bayesian broad learning system and its industrial application

Artificial Intelligence Review - Broad learning system (BLS) is viewed as a class of neural networks with a broad structure, which exhibits an efficient training process through incremental...     

A decoupled multiple model approach for soft sensors design

This paper presents the application of a multiple model approach for the design of a soft sensor aiming at predicting the quality of the product of a separation unit in oil sands processing. The variable of interest here for the product quality is the percentage of water present in the product stream and the goal of the soft sensor is to provide an alternative mean for obtaining on-line measurements of the water-content value. The most reliable measurement of the water-content is obtained through laboratory analysis that introduces delays and long sampling time in the availability of the water-content data. These constraints generate a multi-rate sampling problem for the soft sensor design. The usage of a decoupled multiple model structure allows handling the multi-rate problem and designing a dynamical prediction model for the water-content. The effectiveness of the designed soft sensor in predicting the quality of the product stream is illustrated by on-line implementation results.     

New coordination strategy for on-line hierarchical optimizing control of large scale industrial processes

A new coordination strategy for hierarchical optimizing control is presented, Unlike the price coordination method this approach is not a primal and dual method and its convergence behaviour does not depend on the feature of the saddle point of the problem lagrangian. Compared with the price coordination method, this approach has two advantages. Firstly, its applicability conditions are easier to satisfy and secondly, its convergence behaviour is more desirable in terms of convergence rate and insensitivity of the hessian structure of the problem. A variable augmentation technique is employed to increase the flexibility of the iterative gain selection and to improve further the convergence behaviour of the method. Optimality and convergence analysis are provided for each different version of the algorithm presented. A comparative study between different versions of the algorithm presented and a single iterative integrated system optimization and parameter estimation (ISOPE) method with global feedback is also provided using computer simulation.     

贝叶斯分类器的关联向量机多模型软测量建模

为了改善软测量模型的估计精度,提出了一种基于贝叶斯分类算法和关联向量机的多模型软测量建模方法。采用贝叶斯分类器对样本数据集进行分类,并对不同类别的输入数据分别建立关联向量回归机子模型,用“切换开关”方式组合作为最终的软测量模型输出。将该方法应用于双酚A生产过程的质量指标软测量建模,仿真结果表明：与单模型支持向量机相比,该方法估计精度较高,具有一定的应用价值。     

Robust model updating with insufficient data

The increasing need of many industrial fields for highly accurate predictions of performance and reliability gives rise to the need for enhanced underlying mathematical models. The thereby available test data are usually rather limited due to the high costs of experimental measurements. Therefore, decisions have to be made based on limited, incomplete information, which poses a challenging problem.Recently, an approach for coping with insufficient data has been introduced that attempts to extract the information delivered by the data and processes it using few additional assumptions. The underlying distribution is based on an appropriate confidence level providing a safeguard against severe underestimation of the variability of the measured quantities. This method has been applied within the field of statics involving the stochastic identification of one single structural parameter. The present paper shows the extension of this approach to the field of dynamics. It is shown how to deal with insufficient information by applying kernel densities on the stochastic representation of modal data. In addition, the problem of correlation of the established multi-dimensional probability density function will be addressed. As a numerical example the structural dynamics application of the Validation Challenge Workshop has been chosen.     

Development and application of soft sensor model for heterogeneous information of aluminum reduction cells

The soft sensor model for heterogeneous information is presented because of the difficulty of online acquiring heterogeneous information of aluminum reduction cells. Firstly many redundancy samples are optimized by Fuzzy C-Means in order to acquire classified samples. Then dynamic process of heterogeneous information of aluminum reduction cells is modeled by multiple neural networks. Finally soft sensor model for heterogeneous information of aluminum reduction cells is developed. The model was used in 320 KA prebaked aluminum reduction cells in Guangxi Branch of China Aluminum Corporation. The results indicate that there are three types of instabilities for aluminum reduction cells: single anode irrationality, parameters irrationality of heat balance and outside operations. Corresponding measures to eliminate the three types of instabilities for aluminum reduction cells are the following: raising the anode, adjusting the parameters of heat balance and improving the operation of changing anode and taping metal.     

A Bayesian approach to design of adaptive multi-model inferential sensors with application in oil sand industry

In the context of process industries, online monitoring of quality variables is often restricted by inadequacy of measurement techniques or low reliability of measuring devices. Therefore, there has been a growing interest in the development of inferential sensors to provide frequent online estimates of key process variables on the basis of their correlation with real-time process measurements. Representation of multi-modal processes is one of the challenging issues that may arise in the design of inferential sensors. In this paper, Bayesian procedures for the development and implementation of adaptive multi-model inferential sensors are presented. It is shown that the application of a Bayesian scheme allows for accommodating the overlapping operating modes and facilitating the inclusion of prior knowledge. The effectiveness of the proposed procedures are first demonstrated through a simulation case study. The efficacy of the method is further highlighted by a successful industrial application of an adaptive multi-model inferential sensor designed for real-time monitoring of a key quality variable in an oil sands processing unit.     

Bayesian statistical model checking with application to Stateflow/Simulink verification

We address the problem of model checking stochastic systems, i.e., checking whether a stochastic system satisfies a certain temporal property with a probability greater (or smaller) than a fixed threshold. In particular, we present a Statistical Model Checking (SMC) approach based on Bayesian statistics. We show that our approach is feasible for a certain class of hybrid systems with stochastic transitions, a generalization of Simulink/Stateflow models. Standard approaches to stochastic discrete systems require numerical solutions for large optimization problems and quickly become infeasible with larger state spaces. Generalizations of these techniques to hybrid systems with stochastic effects are even more challenging. The SMC approach was pioneered by Younes and Simmons in the discrete and non-Bayesian case. It solves the verification problem by combining randomized sampling of system traces (which is very efficient for Simulink/Stateflow) with hypothesis testing (i.e., testing against a probability threshold) or estimation (i.e., computing with high probability a value close to the true probability). We believe SMC is essential for scaling up to large Stateflow/Simulink models. While the answer to the verification problem is not guaranteed to be correct, we prove that Bayesian SMC can make the probability of giving a wrong answer arbitrarily small. The advantage is that answers can usually be obtained much faster than with standard, exhaustive model checking techniques. We apply our Bayesian SMC approach to a representative example of stochastic discrete-time hybrid system models in Stateflow/Simulink: a fuel control system featuring hybrid behavior and fault tolerance. We show that our technique enables faster verification than state-of-the-art statistical techniques. We emphasize that Bayesian SMC is by no means restricted to Stateflow/Simulink models. It is in principle applicable to a variety of stochastic models from other domains, e.g., systems biology.     

面向工业软测量应用的定制化生成对抗数据填补模型

在工业领域,数据缺失十分普遍,对解决下游任务(如软测量、异常检测)造成阻碍,这些任务大多依赖完整而高质量的数据集构造模型.现有缺失数据填补方法很少考虑数据填补后的具体下游任务(软测量).如何根据下游任务针对性地进行数据填补是当前研究中的挑战之一.为此,提出一种加入临时软测量模块的对抗生成数据填补模型(SSIGAN).与生成对抗数据填补模型(GAIN)相比,SSIGAN模型显式地考虑了软测量损失对数据填补模型的影响,通过临时软测量模块指导对质量相关变量的修复,实现数据填补的“定制化”,用于更精准的工业软测量建模.通过某工业炼钢过程中的终点成分软测量实验,验证了所提出方法对软测量质量相关变量缺失数据填补效果以及最终软测量效果的提升.     

Modeling for soft sensor systems and parameters updating online

Soft sensor technology is an important means to estimate important process variables in real-time. Modeling for soft sensor system is the core of this technology. Most nonlinear dynamic modeling methods integrate the processes of building the dynamic and static relationships between secondary and primary variables, which limits the estimation accuracy for primary variables. To avoid the problem, a kind of soft sensor model consisting of a dynamic model in cascade with a static one is proposed. The model identification and update online are conducted in substep way. In order to improve the model update efficiency, two improved Gauss–Newton recursive algorithms, which avoid nonsingular covariance matrix, are proposed for time-invariant and time-variant soft sensor systems. The uniform convergence for dynamic model parameter and the existence of estimation deviations for static model parameters are proved for time-invariant soft sensor system. The parameters of time-variant soft sensor system would be boundedly convergent. Case study confirms that, on the basis of the proposed model and recursive algorithms, the dynamic and static characteristics of soft sensor system can be described efficiently, and the primary variables are ensured to be estimated accurately.