一种基于异构数据库间数据综合平台的构建

应用数据挖掘技术在异构数据库间建立数据挖掘平台 ,并使用相应的数据分析软件搭建基于B/S结构的数据分析平台 ,为DSS系统提供决策信息源 ,已成为电力调度决策支持系统中的常用模式。本文对实现数据综合平台的关键技术 :异构数据库间的数据挖掘、中间库的搭建、以及基于B/S数据分析等进行了深入研究 ,并提出了相应解决方案。     

基于Web的刀具全生命周期管理信息系统研究

为解决目前企业中刀具种类繁多、管理成本高、效率低等问题,建立了一种基于B/S结构的刀具全生命周期管理系统.首先以刀具的分类为依据,采用分类码与顺序码相结合的结构,用13位数字对刀具个体进行编码.其次根据扩展泰勒公式和损伤积累理论,建立了刀具寿命预测模型,同时将刀具耐用度公式和刀具个体的切削历史记录在数据库中,不仅为刀具的使用提供寿命预测,还能为刀具的修磨和报损预警提供决策支持.然后通过对企业中刀具的采购、库存、报废等实际需求的分析,基于B/S的三级体系结构,确定了刀具采购、刀具维护、刀具使用以及系统管理等功能模块.最后对刀具管理信息系统中不同用户进行了角色划分,确定了系统的工作流程.     

基于粗糙集的多源信息融合处理技术

基于粗糙集理论与融合分析评价提出了多源遥感信息粗糙决策级融合算法,有效地提高融合的速度和精度.基于粗糙集的决策级融合算法是一种最小算法,即基于不可分辨的思想和知识简化的方法,提出了融合处理中决策规则的最小化方法和基于粗糙集理论的规则生成技术,用来精化知识,删除冗余信息,克服信息的过分膨胀和低效现象.利用粗糙集进行信息融合能够方便地对不完整数据进行分析、推理,提取有用特征和简化信息处理,生成的融合决策规则简单易行,能够显著地提高融合速度,增强系统的决策能力.     

CAM环境下飞机结构件加工刀具及切削参数自动推送技术研究

加工刀具及切削参数选择作为工艺决策的重要环节,目前仍依赖人工经验,严重影响工艺决策质量和效率。针对上述问题,提出了一种基于模糊综合评判法的加工刀具及切削参数决策方法,基于历史加工数据从刀具属性、加工状态和加工效果三方面对数据库中加工刀具及对应的切削参数进行综合评价,从而实现加工刀具与切削参数的优选,通过CAA开发技术,实现了在CAM环境下的加工刀具及切削参数自动推送,有效提升了工艺设计效率和质量,通过飞机典型结构件应用案例表明了方法的有效性。     

基于Web的刀具全生命周期管理信息系统研究

为解决目前企业中刀具种类繁多、管理成本高、效率低等问题,建立了一种基于B／S结构的刀具全生命周期管理系统。首先以刀具的分类为依据,采用分类码与顺序码相结合的结构,用13位数字对刀具个体进行编码。其次根据扩展泰勒公式和损伤积累理论,建立了刀具寿命预测模型,同时将刀具耐用度公式和刀具个体的切削历史记录在数据库中,不仅为刀具的使用提供寿命预测,还能为刀具的修磨和报损预警提供决策支持。然后通过对企业中刀具的采购、库存、报废等实际需求的分析,基于B／S的三级体系结构,确定了刀具采购、刀具维护、刀具使用以及系统管理等功能模块。最后对刀具管理信息系统中不同用户进行了角色划分,确定了系统的工作流程。     

复合材料切削刀具数据库的设计与实现

复合材料切削加工数据库是实现复合材料高效、精密及智能加工重要支撑,本研究开发了一款基于B/S结构的复合材料切削刀具数据库系统,包括刀具和加工工艺的信息查询、混合推理和信息管理等功能。对复合材料切削刀具数据库进行信息与功能需求分析,完成了数据库系统的总体框架设计和功能模块开发,阐述了混合推理技术在数据库系统中实现刀具和加工工艺参数推荐的过程,为用户构建了一个高效、实用的复合材料切削刀具数据库系统。     

基于知识图谱的工艺推理系统

为了企业能够有效、准确地利用专家知识和数据,减少工艺人员的重复性劳动,开发了基于知识图谱的工艺自动推理(决策)系统。该系统利用知识图谱技术构建了三层次工艺信息模型,将工艺知识与设备资源组织在一起,实现了零件/特征级别的工艺自动决策。该系统采用B/S设计架构,能与Web环境中的其他数据主体无缝集成。通过某航空发动机锥齿轮轴在原型系统中的工艺决策,验证了系统的实用性。     

基于WEB的知识管理及其软件系统体系结构的设计

结合知识的社会化、外在化、综合化、内在化螺旋形动态转化过程,把知识的处理过程转化为知识获取、知识传递、知识利用、知识创新、知识评价、知识保护的这样一个动态循环过程,并设计了软件功能模块来实现这些过程;提出了基于WEB的浏览器/服务器(即B/S)模式的知识管理软件系统结构,简化了客户端软件;最后对系统开发的关键技术进行了探讨。     

模糊正交法在GCr15钢切削用量优化中的应用

模糊正交法是把正交试验的结果模糊化,以模糊数学的理论与方法处理试验数据,能在同样试验工作量情况下获得更多的信息,并将较复杂的问题简化。本例在建立表面粗糙度、刀具使用寿命和切削工时多目标隶属函数的基础上,以模糊综合评价值为目标函数,对GCr15轴承钢切削量进行优化,获得了满意的结果。     

基于切削任务的切削数据库系统研究

切削数据库技术已成为切削加工数据管理的重要手段。对现有切削数据库系统的发展和不足进行分析，根据实际切削加工的真实需求确定了系统主要功能和整体业务流程；在建立的切削数据多维描述模型和编码规则下，使用B/S网络结构和MySQL管理软件完成切削数据库系统的搭建；利用数据交换接口实现数据库系统与刀具管理平台的集成，并提出了基于粒子群算法的多目标优化方法，促进了切削数据的高效管理和应用。     

基于粗糙集的海量数据挖掘

文章结合粗糙集在处理海量数据的约简方面的优势和层次分析法的决策优势,建立了以决策属性为上层因素,紊件属性为下层因素的合理评价模型,并对提取的规则再次进行了约简。文中利用改进的粗糙集属性约简算法来降低海量数据的冗余度,提取约简后的规则,借助粗糙集的属性重要度理论．弥补了层次分析法中评价因子的主观因素。此算法模型省略对核的提取过程,对提取的规则进行了定量的分析,实现了海量数据在属性与规则上的约简。实例证明了算法的有效性。     

基于模糊理论的刀具寿命可靠性

分析刀具寿命合理值的模糊性,把刀具寿命合理值视为模糊变量。根据可能性理论,给出表征刀具寿命合理值的模糊约束和可能性分布函数。在考虑刀具寿命合理值存在的模糊性时,刀具寿命的清晰允许区间应该变为模糊允许区间。基于模糊集定义表示刀具寿命模糊允许区间的隶属函数。根据模糊概率理论,提出刀具寿命可靠性的计算公式。实例计算表明,基于模糊概率理论计算出的刀具寿命可靠性能充分体现刀具的模糊工作状态,可避免和减少通常强制换刀具产生的经济损失。     

CHMM for tool condition monitoring and remaining useful life prediction

To solve the problems of tool condition monitoring and prediction of remaining useful life, a method based on the Continuous Hidden Markov Model (CHMM) is presented. With milling as the research object, cutting force is taken as the monitoring signal, analyzed by wavelet packet theory to reduce noise and extract the energy feature of the signal as a basis for diagnosis. Then, CHMM is used to diagnose tool wear state. Finally, a Gaussian regression model is proposed to predict the milling tool’s remaining useful life after the test sample data are verified to be consistent with the Gaussian distribution based on a reliable identification of the milling tool wear state. The probability models of tool remaining useful life prediction could be established for tools with different initial states. For example, when an unknown state of milling force signal is delivered to the milling tool online diagnostic system, the state and the existing time of this state could be predicted by the established prediction model, and then, the average remaining useful life from the present state to the tool failure state could be obtained by analyzing the transfer time between each state in the CHMM. Compared to the traditional probabilistic model, which requires a large amount of test samples, the experimental cost is effectively reduced by applying the proposed method. The results from the experiment indicate that CHMM for tool condition monitoring has high sensitivity, requires less training samples and time, and produces results quickly. The method using the Gaussian process to accurately predict remaining life has ample potential for application to real situations.     

Acquisition and Active Navigation of Knowledge Particles throughout Product Variation Design Process

The variation design of complex products has such features as multivariate association, weak theory coupling and implicit knowledge iteration. However, present CAD soft wares are still restricted to making decisions only according to current design status in dynamic navigation which leads to the huge drain of the knowledge hidden in design process. In this paper, a method of acquisition and active navigation of knowledge particles throughout product variation design process is put forward. The multi-objective decision information model of the variation design is established via the definition of condition attribute set and decision attribute set in finite universe. The addition and retrieval of the variation semantics is achieved through bidirectional association between the transplantable structures and variation design semantics. The mapping relationships between the topology lapping geometry elements set and constraint relations set family is built by means of geometry feature analysis. The acquisition of knowledge particles is implemented by attribute reduction based on rough set theory to make multi-objective decision of variation design. The topology lapping status of transplantable substructures is known from DOF reduction. The active navigation of knowledge particles is realized through embedded event-condition-action(ECA) rules. The independent prototype system taking Alan, Charles, Ian's system(ACIS) as kernel has been developed to verify the proposed method by applying variation design of complex mechanical products. The test results demonstrate that the navigation decision basis can be successfully extended from static isolated design status to dynamic continuous design process so that it more flexibly adapts to the different designers and various variation design steps. It is of profound significance for enhancing system intelligence as well as improving design quality and efficiency.     

基于突变论的刀具寿命建模与预报的研究

介绍了基于突变论的方法建立刀具磨损与破损失效寿命模型的可行性的研究结果。借助于最小二乘回归，建立了变参数三次多项式车／立铣刀具磨／破损融合模型。利用这一模型完成了一种车／立铣监视系统。实验结果表明，监视刀具磨损和破损的成功率≥９５％，刀具磨损值的预报误差≤１０％。叙词     

THE ADAPTABILITY OF A TOOL WEAR MONITORING SYSTEM UNDER CHANGING CUTTING CONDITIONS

The process of metal cutting is a complex phenomenon that has been researched for many years but the aim of practical cutting tool condition monitoring has yet to be achieved. Previous work by the current authors using two neural networks (to classify acquired data) moderated by an Expert System (based on Taylor's tool life equation) has shown that it is possible to accurately monitor tool wear with a single machine/tool/material/cutting condition combination and to identify any inconsistencies between the predictions of the neural networks and engineering practice. This paper investigates the effects that minor inconsistencies in cutting conditions might have on such a system by determining the ‘zone of influence’ of this working system by systematically varying the cutting conditions whilst keeping all other variables fixed. The investigation has found that the zone of influence is small but usable, and an approach to the utilisation of the system in a machine shop is suggested.     

灰色GM(1,1)模型在齿轮疲劳寿命预测中的应用

在齿轮疲劳寿命预测中,需要大量的齿轮疲劳试验数据.运用灰色GM(1,1)模型预测疲劳试验数据,可以大幅度缩短试验时间,节约试验费用,快速获得齿轮可靠性指标.文中简要介绍了灰色GM(1,1)模型的原理,通过具体实例建立了齿轮疲劳寿命GM(1,1)模型.预测结果表明,将灰色系统理论用于齿轮的疲劳寿命试验数据预测是可行的,理论上为有效缩短齿轮疲劳寿命试验时间提供了一个快速有效的方法.     

Condition-based tool management for small batch production

This paper presents a novel tool management concept for cutting processes which integrates tool relevant information, such as distribution data, tool orders, tool condition, and allocation data, within a centralized information cycle. The developed tool management approach uses decentralized identification and storage technologies, enabling an autonomous cooperation of tools and machine tools within a production. The first part of the paper is focused on the assessment of tool condition in a flexible job shop production. A tool wear monitoring system based on cutting force coefficients is developed and demonstrated by an exemplary milling operation. Thereby, it is shown that cutting force coefficients are suitable for wear monitoring and prediction, even for varying cutting conditions. For the online assessment of the current tool condition and for the prediction of residual tool life, an empirical tool wear model is demonstrated. This is applied to a novel condition-based tool management strategy which enables the optimum exploitation of the life time and performance of the cutting tool. The developed condition-based tool management concept is finally demonstrated by a software demonstrator.     

变精度粗糙集在齿轮故障诊断中的应用

在二级齿轮箱的变负载过程中,为了有效地处理非平稳信号,采用小波包提取特征参量(条件属性值);为了有效地处理带噪声的数据,将变精度粗糙集理论引入到齿轮的故障诊断中,提出了一种条件属性约简方法.首先对连续属性进行离散化;然后定义集合M,根据实际情况,选取不同的正确分类率β,利用变精度粗糙集的近似分类质量进行条件属性约简,并与加入噪声数据后所得的约简结果进行了对比;最后通过齿轮故障实例验证了此方法的有效性和实用性.     

Reliability analysis of the control system of large-scale vertical mixing equipment

The control system of vertical mixing equipment is a concentrate distributed monitoring system (CDMS). A reliability analysis model was built and its analysis was conducted based on reliability modeling theories such as the graph theory, Markov process, and redundancy theory. Analysis and operational results show that the control system can meet all technical requirements for high energy composite solid propellant manufacturing. The reliability performance of the control system can be considerably improved by adopting a control strategy combined with the hot spared redundancy of the primary system and the cold spared redundancy of the emergent one. The reliability performance of the control system can be also improved by adopting the redundancy strategy or improving the quality of each component and cable of the system.