汽车覆盖件自由曲面特征识别算法研究

特征技术应用于汽车覆盖件冲压工艺CAPP系统被认为是提高汽车覆盖件制造水平,并逐步实现CIMS的关键.应用特征识别技术,从CAD系统中识别出CAPP系统所需要的信息,将为工艺设计提供丰富的信息.通过汽车覆盖件自由曲面特征识别技术研究,提出了一种新的基于规则的特征识别算法,实现了从自由曲面模型识别出两种最基本的特征,孔类和翻边类特征.以UGII软件为平台,开发了汽车覆盖件零件模型自动特征识别功能,基本上解决了基于特征的覆盖件冲压工艺CAPP系统信息输入问题.     

Stacked denoising autoencoder‐based feature learning for out‐of‐control source recognition in multivariate manufacturing process

In multivariate statistical process control (MSPC), regular multivariate control charts (eg, T²) are shown to be effective in detecting out‐of‐control signals based upon an overall statistic. But these charts do not relieve the need for multivariate process pattern recognition (MPPR). MPPR would be very useful for quality operators to locate the assignable causes that give rise to out‐of‐control situation in multivariate manufacturing process. Deep learning has been widely applied and obtained many successes in image and visual analysis. This paper presents an effective and reliable deep learning method known as stacked denoising autoencoder (SDAE) for MPPR in manufacturing processes. This study will concentrate on developing a SDAE model to learn effective discriminative features from the process signals through deep network architectures. Feature visualization is performed to explicitly present feature representations of the proposed SDAE model. The experimental results illustrate that the proposed SDAE model is capable of implementing detection and recognition of various process patterns in complicated multivariate processes. Analysis from this study provides the guideline in developing deep learning‐based MSPC systems.     

A new heuristic for integrated process planning and scheduling in reconfigurable manufacturing systems

Integrated process planning and scheduling (IPPS) is a manufacturing strategy that considers process planning and scheduling as an integrated function rather than two separated functions performed sequentially. In this paper, we propose a new heuristic to IPPS problem for reconfigurable manufacturing systems (RMS). An RMS consists mainly of reconfigurable machine tools (RMTs), each with multiple configurations, and can perform different operations with different capacities. The proposed heuristic takes into account the multi-configuration nature of machines to integrate both process planning and scheduling. To illustrate the applicability and the efficiency of the proposed heuristic, a numerical example is presented where the heuristic is compared to a classical sequential process planning and scheduling strategy using a discrete-event simulation framework. The results show an advantage of the proposed heuristic over the sequential process planning and scheduling strategy.     

Feature-based geometric reasoning for process planning

We present a framework based on Domain Independent Form (DIF) features for automatic evaluation of manufacturability and process planning for machining. The framework enables interpretation of a common product model with respect to each task in the transition from design to manufacture. A key idea here is to generate the interpretation suitable for each task in two steps. In the first step, DIF features that are defined through feature enumeration are automatically extracted from the geometric model. The extracted DIF features are then mapped into features meaningful for individual tasks through geometric reasoning based on domain dependent knowledge. The formal approach to feature definitions and separation of the domain specific reasoning from the general geometric reasoning enable us to overcome the bottlenecks reported in features technology. Work reported in this paper has been funded in part by grants from Aeronautical Development Agency and the Department of Science and Technology.     

GA-based adaptive setup planning toward process planning and scheduling integration

Setup planning of a part for more than one available machine is a typical combinatorial optimisation problem under certain constraints. It has significant impact not only on the whole process planning but also on scheduling, as well as on the integration of process planning and scheduling. Targeting the potential adaptability of process plans associated with setups, a cross-machine setup planning approach using genetic algorithms (GA) for machines with different configurations is presented in this paper. First, based on tool accessibility analysis of different machine configurations, partially sequenced machining features can be grouped into certain setups; then by responding to the requirements from a scheduling system, optimal or near-optimal setup plans are selected for certain criteria, such as cost, makespan and/or machine utilisation. GA is adopted for the combinatorial optimisation, which includes gene pool generation based on tool accessibility examination, setup plan encoding and fitness evaluation, and optimal setup plan selection through GA operations. The proposed approach is implemented in a GA toolbox, and tested using a sample part. The results demonstrate that the proposed approach is applicable to machines with varying configurations, and adaptive to different setup requirements from a scheduling system due to machine availability changes. It is expected that this approach can contribute to process planning and scheduling integration when a process plan is combined with setups for alternative machines during adaptive setup planning.     

基于多通道Gabor滤波和特征融合的虹膜识别方法

多通道Gabor滤波器提取的虹膜特征具有冗余信息并存在部分非有效特征,针对此问题提出了改进方法。对同尺度不同方向的Gabor特征,利用幅值信息进行融合,对融合后特征进行相位编码,并运用海明距离匹配。这样,既保证了高识别性能,又将虹膜特征码压缩为传统方法的1／2,可提高匹配速度,并节约存储空间。还提出一种虹膜图像质量评价方法,可有效鉴别不适于识别的低质量虹膜图像。在CASIA和UBIRIS虹膜库的实验结果表明该方法是有效的。     

基于KBE的汽车覆盖件冲压工艺方案设计

汽车覆盖件的冲压工艺方案设计是一个复杂且经验性强的创造性过程,需要一种支持多种知识表达模式和多智能决策技术的方法,解决工艺方案决策问题.通过以工艺特征为核心来表达知识,建立了基于KBE的覆盖件工艺知识模型,采用基于知识驱动决策树表达模式的特征转换机制来实现设计特征到工艺特征的映射,在工艺知识库的支持下完成工艺方案设计.实现了知识的多种描述形式及其应用,有效地解决了工艺方案设计中基于图形、规则及经验知识的决策判断,提高了汽车覆盖件工艺方案设计的智能化程度.     

The application of STEP-NC using agent-based process planning

Computer aided part programming is an information intensive activity, and provides an interesting area for the application of artificial intelligence, knowledge-based systems and recently for agent technologies. Intelligent software agents can behave and respond autonomously to set criteria. The flexibility of agents and their ability to make autonomous/cooperative decisions makes them an excellent tool in the development of future computer aided process planning systems. In addition, a new ISO standard for NC machine tool programming is emerging called ISO 14649, informally known as STEP-NC. This standard is aimed to replace the existing ISO 6983 (G&M code) standard, which has been in use since the first NC machine tools were developed in the 1950s. STEP-NC does not describe tool movements but uses a higher-level language to define operations known as Workingsteps. A new breed of intelligent controller is envisaged to determine optimized tool trajectories allowing advanced inter-operability and a bi-directional information flow which allows edits made at the machine to be fed back to the CAD/CAM model. This paper presents a STEP-NC compliant computational environment used to demonstrate agent-based process planning, resulting in the generation of STEP-NC code. The system comprises a multi-agent framework, where agents represent the individual features of the component and work independently and cooperatively to generate process plans for discrete component manufacture. This is demonstrated through a series of case study components which evaluate various feature interaction scenarios.     

Reducing tardy jobs by integrating process planning and scheduling functions

This paper presents a dynamic approach to reduce tardy jobs through the integration of process planning and scheduling in a batch-manufacturing environment. The developed method aims at re-generating a schedule with fewer tardy jobs, step by step, by exploring the process plan solution space of the tardy jobs. The integrated system comprises a process planning module, a scheduling module, and an integrator module. The process planning module employs an optimisation approach in which the entire plan solution space is first generated and a search algorithm is then used to find the optimal plan, while the scheduling module is based on commonly used heuristics. Based on the job tardiness information of the generated schedule, the integrator module automatically issues a modification order to the process plan solution space of the tardy jobs. The process planning and scheduling modules are then re-run to generate a new plan/schedule solution. Through this iterative process, a satisfactory schedule can be gradually achieved. The uniqueness of this approach is characterised by the flexibility of the process planning strategy, which makes full use of the plan solution space intuitively to achieve a satisfactory schedule. Several examples are presented to confirm the efficacy and the effectiveness of the developed integration system.     

A CBFSA approach to resolve the distributed manufacturing process planning problem in a supply chain environment

In this competitive world cost and lead time reduction are of prime concern for manufacturing firms. To achieve this objective manufacturing entities are adopting several management philosophies such as Total Quality Management (TQM), just-in-time (JIT), and theory of constraints (TOC). The present paper addresses the advanced computer-aided process planning (ACAPP) problem in a distributed manufacturing supply chain environment and aims at cost and lead times reduction under several technological constraints. To deal with the complexity of the problem the constraint based fast simulated annealing (CBFSA) algorithm has been explored in this article. Extensive computations have been performed over the well-known benchmarks of advanced planning problems and the results obtained prove the superiority of the proposed algorithm over the prior approaches.     

A hint-based machining feature recognition system for 2.5D parts

The recognition of machining features in a part with arbitrary feature interactions is a difficult task. In this paper a hint-based machining feature recognition system for 2.5D parts with arbitrary feature interactions is presented. A pre-processor is used to screen out invalid 2.5D parts automatically and calculate the possible machining directions. The feature recognition starts with identifying the traces of features present in the part. Once a hint is found, geometrical reasoning is then used to compute other parameters to complete a maximally extended feature instance. The system handles hole, linear slot and circular slot features in the first pass. Various possibilities, like through or blind hole, through or blind slot, floorless or multi-floor slots are instantiated after their testing and repairing. The system then, likewise, handles floor-based and floorless pockets, in the second pass. The recognition process is tightly coupled with the machining environment to ensure machinability of the recognized features. A prototype of the system is implemented in visual C++ and ACIS® 3D modelling toolkit.     

一种视听融合的水下目标识别方法研究

特征提取是水下目标识别研究中最为关键的技术之一,特征参数的优劣将直接决定分类识别系统的性能。将声信号的听觉与视觉感知特征结合,应用于水下目标识别,通过实验得出如下结论,相比于单独应用听觉特征,融合特征的平均识别率能提高4%~6%以上,特别是将听觉特征与声谱图的Gabor小波变换特征、灰度-梯度共生特征进行融合后,分类性能较好,平均达到87%以上。     

多层并行分布式CAPP和PPC集成方法

针对多品种小批量车间作业型企业的工艺计划和生产计划的整体优化，提出了支持功能集成的多级并行分布式ＣＡＰＰ和ＰＰＣ集成的体系结构，给出了一种基于集成平台的系统实现方法和产品层原型系统的具体实现。     

一种有效降维的特征选择方法及其在水声目标识别中的应用

为在保证目标识别准确率基础上进行有效特征降维,文章以目标识别准确率为特征选择准则,提出一种支持向量机递归特征消除（Support Vector Machine Recursive Feature Elimination,SVM-RFE）快速筛选出部分优质特征子集与猫群算法（Cat Swarm Algorithm,CSO）迭代寻优结合的特征选择方法,并将该方法应用于水声目标识别的特征选择。实验数据处理结果表明：相比SVM-RFE和CSO特征选择算法,文中提出的方法在平均特征维数降低8%的基础上,平均目标识别率提高了1.88%,能够实现有效降维的目的。该方法对判断特征是否适合用于特定的目标识别也有一定应用价值。     

Integrated process planning and scheduling using an imperialist competitive algorithm

Effective performance of modern manufacturing systems requires integrating process planning and scheduling more tightly, which is consistently challenged by the intrinsic interrelation and intractability of these two problems. Traditionally, these two problems are treated sequentially or separately. Integration of process planning and scheduling (IPPS) provides a valuable approach to improve system performance. However, IPPS is more complex than job shop scheduling or process planning. IPPS is strongly NP-hard in that, compared to an NP-hard job shop scheduling problem with a determined process plan, the process plan for each job in IPPS is also to be optimised. So, an imperialist competitive algorithm (ICA) is proposed to address the IPPS problem with an objective of makespan minimisation. An extended operation-based representation scheme is presented to include information on various flexibilities of process planning with respect to determined job shop scheduling. The main steps of the proposed ICA, including empires construction, assimilation, imperialistic competition, revolution and elimination, are elaborated using an illustrative example. Performance of the proposed ICA was evaluated on four sets of experiments taken from the literature. Computational results of the ICA were compared with that of some existing algorithms developed for IPPS, which validates the efficiency and effectiveness of the ICA in solving the IPPS problem.     

More MILP models for integrated process planning and scheduling

The integration of process planning and scheduling is important for an efficient utilisation of manufacturing resources. In general, there are two types of models for this problem. Although some MILP models have been reported, most existing models belong to the first type and they cannot realise a true integration of process planning and scheduling. Especially, they are completely powerless to deal with the cases where jobs are expressed by network graphs because generating all the process plans from a network graph is difficult and inefficient. The network graph-specific models belong to the other type, and they have seldom been deliberated on. In this research, some novel MILP models for integrated process planning and scheduling in a job shop flexible manufacturing system are developed. By introducing some network graph-oriented constraints to accommodate different operation permutations, the proposed models are able to express and utilise flexibilities contained in network graphs, and hence have the power to solve network graph-based instances. The established models have been tested on typical test bed instances to verify their correctness. Computational results show that this research achieves the anticipant purpose: the proposed models are capable of solving network graph-based instances.     

An agent-based negotiation approach to integrate process planning and scheduling

This paper presents the development of an agent-based negotiation approach to integrate process planning and scheduling (IPPS) in a job shop kind of flexible manufacturing environment. The agent-based system comprises two types of agents, part agents and machine agents, to represent parts and machines respectively. For each part, all feasible manufacturing processes and routings are recorded as alternative process plans. Similarly, alternative machines for an operation are also considered. With regard to the scheduling requirements and the alternative process plans of a part, the proposed agent-based IPPS system aims to specify the process routing and to assign the manufacturing resources effectively. To establish task allocations, the part and machine agents have to engage in bidding. Bids are evaluated in accordance with a currency function which considers an agent's multi-objectives and IPPS parameters. A negotiation protocol is developed for negotiations between the part agents and the machine agents. The protocol is modified from the contract net protocol to cater for the multiple-task and many-to-many negotiations in this paper. An agent-based framework is established to simulate the proposed IPPS approach. Experiments are conducted to evaluate the performance of the proposed system. The performance measures, including makespan and flowtime, are compared with those of a search technique based on a co-evolutionary algorithm.     

水下目标空间源数特征提取与识别方法

针对水下目标识别正横阵位问题,并进一步扩大目标识别的阵位范围。给出了目标识别的阵列形式。用水平线列阵进行目标亮点水平方位的源数估计。提取并找出了能反映水下潜艇目标空间特征的特征量。提出了基于信噪比监督的空间源数识别方法,给出了相关矩阵的空间平滑的特征值。还提出了基于信噪比和盖尔圆半径监督的目标特征的自动分类方法。对本文提出的方法进行了大量计算机仿真,仿真结果表明。该方法可在正横阵位有效识别目标。     

Multi-objective process planning and scheduling using controlled elitist non-dominated sorting genetic algorithm

The integration of process planning and scheduling is considered as a critical component in manufacturing systems. In this paper, a multi-objective approach is used to solve the planning and scheduling problem. Three different objectives considered in this work are minimisation of makespan, machining cost and idle time of machines. To solve this integration problem, we propose an improved controlled elitist non-dominated sorting genetic algorithm (NSGA) to take into account the computational intractability of the problem. An illustrative example and five test cases have been taken to demonstrate the capability of the proposed model. The results confirm that the proposed multi-objective optimisation model gives optimal and robust solutions. A comparative study between proposed algorithm, controlled elitist NSGA and NSGA-II show that proposed algorithm significantly reduces scheduling objectives like makespan, cost and idle time, and is computationally more efficient.