GA-based learning for rule identification in fuzzy neural networks

Employing an effective learning process is a critical topic in designing a fuzzy neural network, especially when expert knowledge is not available. This paper presents a genetic algorithm (GA) based learning approach for a specific type of fuzzy neural network. The proposed learning approach consists of three stages. In the first stage the membership functions of both input and output variables are initialized by determining their centers and widths using a self-organizing algorithm. The second stage employs the proposed GA based learning algorithm to identify the fuzzy rules while the final stage tunes the derived structure and parameters using a back-propagation learning algorithm. The capabilities of the proposed GA-based learning approach are evaluated using a well-examined benchmark example and its effectiveness is analyzed by means of a comparative study with other approaches. The usefulness of the proposed GA-based learning approach is also illustrated in a practical case study where it is used to predict the performance of road traffic control actions. Results from the benchmarking exercise and case study effectively demonstrate the ability of the proposed three stages learning approach to identify relevant fuzzy rules from a training data set with a higher prediction accuracy than alternative approaches.     

Non-kinematic calibration of a six-axis serial robot using planar constraints

This paper describes a non-kinematic calibration method developed to improve the accuracy of a six-axis serial robot, in a specific target workspace, using planar constraints. Simulation confirms that the stiffness of the robot, as well as its kinematic parameters, can be identified. An experimental validation shows that the robot's accuracy inside the target workspace is significantly enhanced by reducing the maximum distance errors from 1.321 mm to 0.274 mm. The experimental data are collected using a precision touch probe, which is mounted on the flange of a FANUC LR Mate 200iC industrial robot, and a high precision 9-in. granite cube. The calibration method makes use of a linear optimization model based on the closed-loop calibration approach using multi-planar constraints. A practical validation approach designed to reliably evaluate the robot's accuracy after calibration is also proposed.     

2014年我国冷却设备行业发展现状及展望

对2014年我国冷却设备行业的发展状况进行了较为全面的叙述,对目前行业发展面临的一些问题做了比较客观的概括和总结,对行业发展趋势提出了一些观点及看法。     

融合分词隐层特征的汉语基本块识别

该文以字为基本标注单位,构建了一种汉语基本块识别的神经网络学习模型。模型联合分词任务的神经网络学习模型与基本块识别任务模型,将分词任务模型中学习得到的隐层特征融入基本块识别的模型中,两模型相互交替优化学习模型参数,并实现了以整句似然函数(而非单字似然函数)作为优化目标的算法。实验结果表明:1)以整句似然函数为优化目标的基本块识别的F值比单字似然情形要高出1.33%,特别是在多字块识别中,其召回率比单字似然情形要高出4.68%;2)融合分词任务模型中的隐层特征的汉语基本块识别模型的结果比不做融合的模型要高出2.17%,说明融合分词隐层特征的交替联合学习方法是有效的。     

Experimental evaluation of MPC-based anti-surge and process control for electric driven centrifugal gas compressors

The present work concerns model predictive control (MPC) of centrifugal gas compressors and describes the development of an MPC application for the tasks of anti-surge and process control. More specifically, the MPC formulation focuses on the question of how the transient manipulation of driver torque can be used to improve the performance of anti-surge and process control. For the purpose of testing and validating the proposed control algorithm, an experimental compressor test rig is presented, which is designed to mimic a typical centrifugal compressor application in the oil and gas industry. Modeling and parameter identification of the experimental setup is followed by the realization of the MPC solution on an embedded system to comply with the stringent real-time requirements for anti-surge control. Testing is performed with experiments using suction and discharge side disturbances, which are created by rapid valve closures. For comparison the same tests are repeated with conventional control approaches. The test results indicate improvements in maintaining the distance to surge by up to 11%, while at the same time reducing the process control settling time by up to 50%.     

A robust method for axis identification

The present paper proposes a new method for axis identification in discrete axially symmetrical geometric models. This method is based on-a-never-used-before property of the axially symmetrical surfaces for which the symmetry line of any section curve of the surface (or of a portion of it in the case of an incomplete axially symmetrical surface) always intersects the axis of symmetry of the surface. Thus the working principle of the method makes it very robust to local defectiveness, measurement noise and outliers.In order to compare it with the most cited methods presented in literature, several types of tests have been designed and performed. The robustness of those methods, on the one hand, has been evaluated by defining the Statistical Confidence Boundary at 1σ confidence level. The trueness of the method, on the other hand, has been evaluated on geometric models obtained by measuring real objects. The high robustness, which characterizes the proposed method, makes it particularly suitable for product geometric inspection where high accuracy is required.     

Combining Acoustic Emission and Vibration Technologies Harbors Huge Potential for Material Characterization

The demand for raw materials is growing worldwide, in step with increasing world population and rising living standards of emerging countries. It is becoming more difficult to exploit primary raw materials because of declining ore grades and increasingly complicated mineralogies. Thus, the efficient and environmentally sustainable exploitation of primary resources using innovative technologies has emerged as an important research field to address these issues. An innovative approach is introduced to optimize process and quality control by combining acoustic emission (AE) and vibration technology. The general feasibility of these two technologies for material recognition and characterization is examined. The new approach was tested to determine whether coal could be distinguished from waste rock.     

Bayesian system identification of dynamical systems using large sets of training data: A MCMC solution

In the last 20 years the applicability of Bayesian inference to the system identification of structurally dynamical systems has been helped considerably by the emergence of Markov chain Monte Carlo (MCMC) algorithms – stochastic simulation methods which alleviate the need to evaluate the intractable integrals which often arise during Bayesian analysis. In this paper specific attention is given to the situation where, with the aim of performing Bayesian system identification, one is presented with very large sets of training data. Building on previous work by the author, an MCMC algorithm is presented which, through combing Data Annealing with the concept of ‘highly informative training data’, can be used to analyse large sets of data in a computationally cheap manner. The new algorithm is called Smooth Data Annealing.     

突发疫情对染料行业影响及对策

一场新型冠状病毒肺炎疫情的爆发给染料行业的发展带来了影响,既有负面消极,也有积极影响。疫情将促进染料行业绿色发展,推动染料行业走向信息化、智能化时代,走向高质量发展,指出只要坚持绿色、创新和智能化,就能战胜疫情,推动中国染料行业稳健发展。     

核燃料循环设施安全关键岗位识别方法初探

研究建立了基于岗位的工作责任、安全风险和专业技能等特性指标的核燃料循环设施安全关键岗位识别方法,给出了核燃料循环设施安全关键岗位的定义、识别指标体系、识别原则、识别评价流程和后处理设施应用案例。