基于知识的神经网络推理机制的研究

在传统专家系统模糊知识规则库的基础上,给出模糊规则库规范化的处理方法,并将模糊规则通过正向推理处理以消解规则的冗余,最终以利于网络推理时的正确输入.给出了网络推理的原理、具体实现方法及网络推理的自组织学习算法.通过网络推理迭代次数的变化来改变学习率因子的方法,从而大大提高网络推理的效率.     

基于神经网络注塑成型工艺参数优化

采用BP神经网络建立注塑成型工艺参数与注塑制品收缩率之间的网络模型,并通过试验数据对成型工艺参数进行优化。     

塑料注射缺陷修正的模糊建模方法

依据塑料注射过程的系统特征和工艺人员的试模思路，结合模糊理论的特点和处理问题的优势，建立了基于专家知识的模糊推理模型来描述复杂的注射成形过程，并将模糊推理模型用于试模过程中缺陷的在线修正和工艺参数优化。详细介绍了输入输出变量的选择、正交性隶属函数的选定、模糊ifthen规则的形式、模糊推理机制、解模糊策略以及多缺陷反馈时工艺参数调整量的冲突消解等问题，并开发出了相应的软件系统。     

模糊推理技术在螺纹磨削工艺决策中的应用

螺纹磨削加工工艺参数的确定较为复杂,实际生产过程中对技术人员的经验水平有很高的要求。本文提出将模糊推理技术引入该领域,通过对实际加工数据的学习来生成模糊规则库,并以此推理工艺参数值,为加工工艺参数的确定提供有效的参考。     

基于Takagi-Sugeno模糊神经网络的欠驱动无人艇直线航迹跟踪控制

研究一类欠驱动无人艇的直线航迹跟踪控制问题,提出了一种自适应T-S(Takagi-Sugeno)模糊神经网络控制方法。首先在神经网络体系结构中设计前件网络匹配T-S模糊控制器的模糊规则前件,设计后件网络进行T-S模糊运算推理从而生成模糊规则后件;其次基于梯度下降法原理,设计了T-S模糊规则参数的优化学习算法;然后结合BP神经网络的误差反向传播原理和梯度下降法,设计了模糊神经网络体系误差的反向传播迭代算法,用于高斯隶属度函数参数的学习优化;最后设计了基于T-S模型的模糊神经网络控制器,并通过仿真实验验证了所提出方法和所设计控制器的有效性。     

遗传优化算法在压缩机故障诊断中的应用

设计了一种模糊神经网络推理系统诊断汽车空调压缩机故障，并且应用遗传算法对系统进行优化及对BP网络算法进行改进。这种方法能够优化模糊系统的参数和结构。并且能删除无用的模糊规则。结果证明这种推理系统具有收敛速度快，泛化能力好、推广性强的特点，极大提高了汽车空调压缩机故障诊断的效率和准确性。     

减速机故障诊断专家系统的关键技术研究

开发用于减速机故障诊断的专家系统,提供两种诊断方式:基于经验规则的感官现象诊断和基于检测数据的模糊诊断,从而满足不同场合下的故障诊断需要.根据不同产品型号故障特点,对知识库的采用"系列化"设计,并着重讨论采用故障树规则推理和模糊决策技术进行故障推理,发挥它们对不同知识表示时的推理优越性,体现了诊断方法的科学合理性.     

模糊神经网络在电火花线切割加工中的应用

分析了电火花线切割加工工艺特点,针对其加工过程的复杂,各工艺参数之间相互制约,难以用传统的数学方法解决,而提出了用模糊神经网络对电火花线切割加工过程建立模型,通过网络的训练,实现加工参数的优化选取和预测,并且提取出有效的规则集.实验结果说明了这一方法具有可行性和实用性,并取得了较好的效果.     

基于Alopex神经网络的旋转机械故障诊断专家系统

提出了一种可以减小网络规模的故障表示方法,并将Alopex算法引入神经网络模型的训练过程,将人工神经网络与规则推理相结合,建立了旋转机械故障诊断的神经网络专家系统。该系统充分利用了神经网络与规则推理的优点,采用正反向混合推理方式调用知识库中的各种知识进行诊断。采用二进制数码表示机械的各种故障,基于Alopex算法训练神经网络。建立的专家系统克服了基于规则专家系统的自学习困难问题和基于神经网络诊断系统的系统控制能力弱的缺点,具有较强的诊断能力。     

面向创新方法的切削工艺知识库研究

传统切削数据库通过数据查询,实例推理和规则推理来生成工艺方案,而面对大量的非典型切削工艺问题却无能为力。为了扩展切削数据库的应用范围,辅助工艺人员解决非典型切削工艺问题,基于混合推理和工艺创新设计方法,提出一种面向创新的切削工艺知识库模型。该知识库的关键技术在于:基于切削工艺实例的标准化表示构建实例库;基于切削工艺标准建立推理规则;基于工艺问题分类建立创新方法的策略。最后,基于上述模型,构建了切削知识库的结构、框架和原型系统。     

基于模糊神经网络的工艺设计和知识表达与推理

工艺设计知识的表达不存在普遍通用的表达模式,而必须根据各类工艺知识的结构特点选择适合的知识表达方法。文以模糊集理论和社会网络原理为基础,研究了工艺设计领域模糊性知识表达与推理的实现技术,提出了一种新的基于模糊神经网络的工艺设计知识表达与推理方法,并给出了一个具体的表达实例。     

Extraction Fuzzy Linguistic Rules from Neural Networks for Maximizing Tool Life in High-speed Milling Process

In metal cutting industry it is a common practice to search for optimal combination of cutting parameters in order to maximize the tool life for a fixed minimum value of material removal rate(MRR). After the advent of high-speed milling(HSM) pro cess, lots of experimental and theoretical researches have been done for this purpose which mainly emphasized on the optimization of the cutting parameters. It is highly beneficial to convert raw data into a comprehensive knowledge-based expert system using fuzzy logic as the reasoning mechanism. In this paper an attempt has been presented for the extraction of the rules from fuzzy neural network(FNN) so as to have the most effective knowledge-base for given set of data. Experiments were conducted to determine the best values of cutting speeds that can maximize tool life for different combinations of input parameters. A fuzzy neural network was constructed based on the fuzzification of input parameters and the cutting speed. After training process, raw rule sets were extracted and a rule pruning approach was proposed to obtain concise linguistic rules. The estimation process with fuzzy inference showed that the optimized combination of fuzzy rules provided the estimation error of only 6.34 m/min as compared to 314 m/min of that of randomized combination of rules.     

冲天炉网形图重构的自适应模糊推理方法

网形图是指导冲天炉熔炼操作的重要工具。综合自适应模糊推理的建模功能和神经网络的学习能力，直接从实验数据中提取推理规则，建立了基于网形图的冲天炉熔炼过程模型。模型具有较高的预测精度和泛化能力，可以帮助操作者认识熔炼规律，据此得出的新型网形图使用起来更为方便快捷。     

Experimental-based optimization of polymer injection molding process parameters using anfis-ga method

Compared with ordinary injection-molded parts, the slender, cantilevered, and thin-walled plastic parts are harsh on the injection molding process conditions. For complexity and particularity, it is difficult to form such parts. It is also more likely to cause excessive warpage deformation, affecting the molding quality and performance. The automobile audio shell is a typical slender, cantilevered, thin-walled plastic part. When the mold structure and material are determined, optimizing its injection molding process is the most economical and effective method to manufacture the products with the optimum properties. In order to minimize the warpage deformation, the adaptive network based fuzzy inference system (ANFIS) and genetic algorithm (GA) were adopted to optimize the injection molding process parameters. In particular, considering the high-dimensional nonlinear relationship between the process parameters and the warpage, the ANFIS is constructed as the prediction model of the warpage. Then, the GA is used to globally optimize the prediction model to determine the optimal process parameters. The results show that the optimization method based on ANFIS-GA has a good performance. The warpage is reduced to 0.0925 mm while reduced by 88.25 %. The optimal injection molding process parameters are used for simulation and manufacture, verifying the effectiveness and reliability of the optimization method.

     

The application of fuzzy theory for the control of weld line positions in injection-molded part

This research proposes the fuzzy theory for the control of weld lines in plastic injection molding. The weld line occurs as a result of geometrical changes in molded parts in the injection molding process. The weld line is one of the defects present in plastic injection-molded parts; the line affects the quality of parts as well as the strength of the products. In the present study, fuzzy theory was applied in the design of injection molding. First, expert experiences were transformed into IF∼THEN∼ rules to establish the knowledge base for developing fuzzy inference rules. The rules were then used to adjust the molding parameters, which in turn were applied to control the weld line position in the injection molding process. The results indicate that fuzzy theory exhibited favorable applicability in the control of the weld line as well as decreased the simulation time, thereby accelerating the design process of injection molding.     

An Improved Rough Set Approach to Design of Gating Scheme for Injection Moulding

The gate is one of the most important functional structures in an injection mould, as it has a direct influence on the quality of the injection products. The design of a gating scheme includes the selection of the types of gate and calculation of the sizes and determination of the location, which depends heavily on prior experience and knowledge and involves a trial-and-error process. Due to the vagueness and uncertainty in the design of a gating scheme, classical rough set theory is not effective. In this paper, a fuzzy rough set model is proposed, which is not based on equivalent relationships but on fuzzy similarity relationships. An inductive learning algorithm based on the fuzzy rough set model (FRILA) is then presented. Compared to decision tree algorithms, the proposed algorithm can generate fewer classification rules; moreover, the generated rules are more concise. Finally, an intelligent prototype system for the design of a gating scheme based on an induced fuzzy knowledge base is developed. An illustrative example proves the effectiveness of the proposed method.     

智能注射成形中工艺参数的多目标自学习优化

注射成形工艺参数是保障产品质量的关键因素。传统试错法严重依赖工艺人员的试模经验,随着注射成形工艺广泛应用于电子、航空航天等国家战略领域,产品的高端化对工艺参数智能化设置水平提出更高的要求。由于成形产品存在多方面的质量要求,且不同质量指标间可能相互制约,因此亟需一种工艺参数多目标智能优化方法,以获得不同优化目标间的帕累托最优。已有学者利用智能优化方法,如非支配排序遗传算法等,对多目标优化问题进行求解,但是此类方法需大量样本数据对质量-参数关系进行建模,存在试验次数多、且对不同材料及模具的适应性较差等问题。为解决上述问题,提出一种注射成形工艺参数多目标自学习优化方法,在优化过程中实时计算并更新各个工艺参数的梯度,并由不同质量指标的多梯度下降算法对多个目标函数进行优化,在优化过程中实现各工艺参数对产品质量影响程度的自主学习,省去了采集大量数据来建立多个质量模型的过程,实现了注射成形工艺参数的高效智能优化。在基准测试函数实验中,所提方法的优化结果与理论解的相对误差小于2%。同时数值仿真与注射成形实验结果表明,所提方法能高效获得多个优化目标的帕累托最优。     

应用基于学习的模糊逻辑智能选择铣削加工参数

分析传统专家系统在铣削加工参数智能选择应用中存在的问题，提出一种可实现铣削用量智能选择的模糊逻辑推理方法。构造了以刀具直径、加工深度和材料硬度为输入，铣削进给量为输出的模糊推理模型，给出基于人工神经网络与k-means聚类相结合的机器学习方法，实现了推理规则知识的自动获取。通过手册数据与模型推理结果的对比实验，表明给出的方法具有较好的铣削用量智能匹配性能，研究成果为实现铣削用量在线智能选择提供了一种新方法。     

智能消除注塑制品缺陷的研究

研究了如何利用注塑专家的经验知识,采用基于规则的模糊逻辑推理,寻找无缺陷注塑过程参数的方法,尽量减少对操作者的专业知识的要求。首次将基于经验知识的模糊逻辑系统运用于注塑过程参数的自动设定,实现了初步的系统。实验结果表明该系统层次分明、结构简单、实用、高效。     

Fuzzy-sliding mode control of a polishing robot based on genetic algorithm

This paper proposes a fuzzy-sliding mode control which is designed by a self tuning fuzzy inference method based on a genetic algorithm. Using the method, the number of inference rules and the shape of the membership functions of the proposed fuzzy-sliding mode control are optimized without the aid of an expert in robotics. The fuzzy outputs of the consequent part are updated by the gradient descent method. It is further guaranteed that the selected solution becomes the global optimal solution by optimizing Akaike’s information criterion expressing the quality of the inference rules. In order to evaluate the learning performance of the proposed fuzzy-sliding mode control based on a genetic algorithm, a trajectory tracking simulation of the polishing robot is carried out. Simulation results show that the optimal fuzzy inference rules are automatically selected by the genetic algorithm and the trajectory control result is similar to the result of the fuzzy-sliding mode control which is selected through trial error by an expert. Therefore, a designer who does not have expert knowledge of robot systems can design the fuzzy-sliding mode controller using the proposed self tuning fuzzy inference method based on the genetic algorithm.