基于AHPSO-SVM的农产品冷链物流质量安全预警模型

目的为了实时监测农产品始终处于低温、适宜湿度等条件下,实现农产品质量信息的及时反馈和预警。方法针对农产品在冷链物流中的质量安全问题,首先分析影响其质量安全的因素,整合供应链上的追溯信息和监测信息,构建农产品的质量安全预警指标体系。然后设计结合交叉变异算子的自适应混合粒子群算法(AHPSO)优化支持向量机(SVM)参数,以此建立基于AHPSO-SVM的农产品冷链物流质量安全预警模型。结果以苹果的预警指标体系为例,经过模型训练和预测后,预测输出曲线与期望输出曲线均能较好拟合,误差值小。结论该方法较传统的BP神经网络与支持向量机方法,在解决实际问题中预测结果精度更高,可以有效提高农产品冷链物流中质量安全预警的准确性。     

随机需求下半开放式冷链物流车辆路径优化

目的 降低冷链物流配送成本的同时,保证客户体验及碳排放达到企业要求。方法 综合考虑运输距离、客户软时间窗约束、碳排放、生鲜变质等因素,以制冷成本及对配送时间惩罚成本在内的总成本最低、碳排放量最低、生鲜产品新鲜度最高为目标,建立多目标生鲜配送路径优化模型,并设计模拟退火算法,以北京某冷链物流企业为例进行求解验证。结果 得出生鲜配送方案,通过配送模式对比表明多中心半开放式在降本和缩短路径长度方面更具优势,其中运输总费用和车辆行驶总距离相比于单中心独立配送模式分别降低了8.41%和36.36%。结论 需求不确定下,合理决策路由可在达到企业对生鲜产品新鲜度及碳排放标准的同时,有效降低配送成本。     

A method to evaluate direct and indirect design dependencies between components in a product architecture

Methods for evaluating the strength of design dependencies in a product architecture have been widely studied in the literature; however, evaluating the effects of direct and indirect interactions between components/modules remains a challenge. In fact, indirect connections between components/modules are often overlooked in many cases when evaluating design dependencies. Having a more consistent way of defining a product architecture that considers both its direct and indirect connections is important, especially when analyzing redesign complexity and change propagation. In this study, we propose a systematic method to evaluate direct and indirect design dependencies between components in product architectures. Interfaces are classified into six different types based on a thorough review of the literature, and a method for evaluating design dependencies is introduced to estimate the relative importance of interfaces directly from a set of comparable products. Using an electrical circuit analogy, the proposed method can quantify both direct and indirect design dependencies between components within a product architecture. We compare design dependency results for different wireless computer mice to validate the effectiveness of the proposed method. The results show that using the proposed design dependency measure including direct and indirect effects provides more reliable design dependency results.     

An integrated approach to product family redesign using commonality and variety metrics

Redesigning a product family entails carefully balancing the trade-offs between commonality and differentiation that are governed by the underlying platform architecture. Numerous metrics for commonality and variety exist to support product family and product platform design; however, rarely are they used in concert to help redesign platforms and families of products effectively. In this paper, we introduce an integrated approach that uses multiple product family metrics to establish an effective platform redesign strategy. Specifically, we present a detailed procedure to integrate the generational variety index, product line commonality index, and design structure matrix to prioritize components for redesign based on variety and commonality needs in a family of products. While all three of these tools exist in the literature and have been used extensively to support product family design, the novelty in our work lies in their integration to establish a redesign strategy for platform architectures that achieves a better balance between the commonality and variety within a product family. To demonstrate the proposed approach, case studies involving two generations of wireless computer mice and two families of dishwashers are presented. Ongoing and future work is also discussed.     

基于DSP的智能家居无线网络节点的设计

 将无线传感器网络技术融入智能家居系统中,设计合理的网络节点成为智能家居系统的核心问题．在分析无线传感器网络节点结构基础上,提出以DSP 芯片TMS320F206为处理器、AT86RF230为无线通信芯片的无线传感器网络节点．充分考虑节能和扩展性的需求,采用节点定时打开与关闭的协议且打开与关闭的时间比为1∶99．试验表明该无线网络节点能满足低功耗、长传输距离、准确定位、抗干扰的要求．     

Functional characterisation of mechanical joints to facilitate its selection during the design of open architecture products

New trends in product design require the use of modularity as key feature aimed to improve functional performance and the generation of open architecture products. For mechanical systems, one of the challenges during early design stages of these products involves the proper selection of joining methods among their constructive components. A robust joint selection process must consider product requirements, life cycle analysis and eventual procedures for assembly and disassembly. However, the general approach towards a Design-for-Assembly (DFA)/Design-for-Disassembly (DFD) only considers design, manufacturing and in some cases final disposal stage. Additionally, most of the works found in the literature are merely focused on assembly operations, disregarding economic and environmental benefits from optimising disassembly complexity. Herein, a functional characterisation of mechanical joint methods for the assembly and disassembly activities that take place throughout the product life cycle is proposed, focusing on open architecture products. Additionally, a classification of joining methods, a joint complexity metric valuation and a selection process are proposed for the conceptual design stage. The approach integrates both DFA and DFD principles in a formal methodology. The proposed selection roadmap can be implemented to increase product sustainability positively regarding resources optimisation, operational time and costs in reuse, remanufacturing and recycling tasks.     

MAC-layer channel utilisation enhancements for wireless mesh networks

The authors focus on a wireless mesh network, that is, an ad hoc IEEE 802.11-based network whose nodes are either user devices or Access Points providing access to the mesh network or to the Internet. By relying on some work done within the IEEE 802.11s TG, the network nodes can use one control channel and one or more data channels, each on separate frequencies. Then, some problems related to channel access are identified and a MAC scheme is proposed that specifically addresses the problem of hidden terminals and the problem of coexisting control and data traffic on different frequency channels. An analytical model of the MAC scheme is presented and validated by using the Omnet++ simulator. Through the developed model, we show that our solution achieves very good performance both in regular and in very fragmented mesh topologies, and it significantly outperforms the standard 802.11 solution.     

Development of a metric to assess the complexity of assembly/disassembly tasks in open architecture products

This paper describes two metrics to measure the complexity involved in assembly and disassembly tasks for open architecture products during its use phase. The approach proposed is based on the summary of all tasks required to assemble and disassemble a predefined set of modules to generate several product variants, which comprise different working levels or functionalities. The aim of the method is to provide a useful tool to designers in the analysis of product complexity regarding use and further phases in which the assembly and disassembly of modules are required. The benefits and usefulness of the metrics are oriented to enhance the sustainability performance of products through the measurement of complexity in modular systems for the decision-making during the design stage. The reduction of complexity involves significant benefits in all lifecycle phases of product, especially when the user or customer is responsible for many related tasks (maintenance, upgrading, reconfiguration and final disposal of modules). The metrics and their calculation process are illustrated using two case study products.     

考虑消费者辨识能力的冷链博弈分析

消费者辨识能力是指消费者对所购产品或服务品质的辨别能力。针对忽略消费者辨识能力将得出不准确结论，给决策者的抉择造成误导的问题，采用完全信息静态博弈方法，分别构建了政府、冷链产品企业和消费者之间的不考虑消费者辨识能力及考虑消费者辨识能力的混合策略博弈模型。通过分析Nash均衡解及期望收益，对比发现，忽略消费者辨识能力将低估政府部门的期望收益，同时对企业和消费者的期望收益也预估不准确。因此，在实际问题中应关注消费者对产品的辨识能力。     

Energy-efficient directional routing between partitioned actors in wireless sensor and actor networks

Actor?actor communication is an important part of the functioning of wireless sensor?actor networks and enables the actor nodes to take coordinated action on a given event. Owing to various reasons such as actor mobility and low actor density, the actor network tends to get partitioned. The authors propose to use the underlying sensor nodes, which are more densely deployed, to heal these partitions. In order to maximise the utilisation of the limited energy available with the sensor nodes, a new routing protocol for actor?actor communication using directional antennas on the actor nodes is proposed. The authors contribution is threefold. First, using simulations they show that the problem of partitioning in the actor networks is significant and propose an architecture with directional antennas on actor nodes and sensor bridges to heal these partitions. Second, they identify the routing problem for this architecture based on a theoretical framework and propose centralised as well as distributed solutions to it. Third, they develop a routing protocol based on the distributed solution and show, using network simulations, that the proposed protocol not only heals the network partitions successfully, but also achieves high throughput and fairness across different flows, in addition to maximising the network lifetime.     

无线传感器网络中基于信誉的恶意节点检测

提出一种基于信誉的恶意节点检测方法——RMDMN,在分簇的网络结构基础上,对节点的行为属性(如丢包率、转发率、位置匹配信息等)和网络攻击进行建模,结合阈值比较法动态地更新节点信誉值并进行恶意节点判断.实验仿真显示,该方法具有一定的恶意节点检测能力.     

Highway Bridge Assessment Using an Adaptive Real-Time Wireless Sensor Network

A real-time wireless sensor network platform capable of maintaining lossless data transmission over several minutes of continuous, high-rate sampling is presented in this paper. The platform was designed specifically to provide the capability to enable expeditious system identification, as well as load rating of highway bridges without compromising the typical data acquisition parameters employed in comparable cable-based tests. Consequently, the hardware signal conditioning interface permits data collection from a variety of sensors typical to structural health monitoring, including accelerometers, strain transducers, and temperature sensors. The embedded software features a proprietary network transmission protocol capable of lossless, real-time delivery of up to 40 measurement channels at an effective sampling rate of 128 samples per second per channel. Documented in this paper is a field study on an end-of-service highway bridge in which ambient vibration monitoring was performed using 60 accelerometers interfaced with 30 wireless sensor nodes operating within one of two simultaneously operating star topology networks. In addition, an experimental load rating of the entire structure was performed through large-scale strain measurement facilitated by the same wireless sensor network platform.     

开放式结构产品的模块规划

 产品的多元化和市场竞争使客户对产品的个性化需求不断增加,传统的产品结构难以满足低成本和个性化的需求．开放式结构产品使得产品既拥有大批量定制的优点又能满足客户个性化的需求．开放式结构产品将产品的模块分为通用模块、定制模块和个性化模块．模块划分和模块类型规划是开放式结构产品设计的重要步骤．针对开放式结构产品的设计,提出了一种以公理设计为分析框架、以质量功能展开（QFD）为基本模型的模块划分和分类方法．该方法采用公理设计建立功能需求与概念结构之间的映射关系,通过分析概念结构之间的功能、空间和信息流关系建立了设计关联矩阵,运用系统聚类法对设计关联矩阵进行聚类模块划分,利用QFD规划产品中零部件的类型并建立模块类型规划准则,最后根据零部件类型确定了开放式结构产品的模块类型．应用所提出的方法对玩具喷漆机进行了模块规划,得出了开放式结构玩具喷漆机的结构方案,验证了该方法的有效性和可行性．     

Microstructure and composition of SHS products of powder multicomponent oxide mixtures with aluminum and graphite

The microstructure and composition of products of SHS combustion of two types of multicomponent oxide mixtures with aluminum and graphite powders, differing in the presence or absence of a thermite additive, are studied. The combustion product—oxide carbide ceramic based on Al₂O₃/TiC—serves for immobilization of graphite-containing radioactive wastes. It is proposed to use results of the studies of “undercombusted” samples of such composites for analysis of the initial stages of the process of their front combustion. Procedures of combustion of the composites with the presence or absence of a thermite additive are specified. The problems of distribution in the combustion products of the impurities introduced into the original system, which simulate the most characteristic composition of nuclides in nuclear-fuel degradation products and in radiation-exposed graphite, are studied     

Convolutional Neural Network for Histopathological Osteosarcoma Image Classification

Osteosarcoma is one of the most widespread causes of bone cancer globally and has a high mortality rate. Early diagnosis may increase the chances of treatment and survival however the process is time-consuming (reliability and complexity involved to extract the hand-crafted features) and largely depends on pathologists’ experience. Convolutional Neural Network (CNN—an end-to-end model) is known to be an alternative to overcome the aforesaid problems. Therefore, this work proposes a compact CNN architecture that has been rigorously explored on a Small Osteosarcoma histology Image Dataaseet (a high-class imbalanced dataset). Though, during training, class-imbalanced data can negatively affect the performance of CNN. Therefore, an oversampling technique has been proposed to overcome the aforesaid issue and improve generalization performance. In this process, a hierarchical CNN model is designed, in which the former model is non-regularized (due to dense architecture) and the later one is regularized, specifically designed for small histopathology images. Moreover, the regularized model is integrated with CNN’s basic architecture to reduce overfitting. Experimental results demonstrate that oversampling might be an effective way to address the imbalanced class problem during training. The training and testing accuracies of the non-regularized CNN model are 98% & 78% with an imbalanced dataset and 96% & 81% with a balanced dataset, respectively. The regularized CNN model training and testing accuracies are 84% & 75% for an imbalanced dataset and 87% & 86% for a balanced dataset.     

Enhancing interface adaptability of open architecture products

Open architecture products are built with the platform and open interfaces where different add-on modules from various sources can be connected to meet individualized customers’ requirements. Since add-on modules from different vendors are interacted with the product platform through the interfaces, enhancing the interface adaptability to facilitate interactions between the platform and add-on modules is critical for the development of open architecture products. In this work, an approach of open interface modeling is proposed based on parts and their assembling relationships. The quantification of the open interface is introduced with functional, structural, manufacturing, and operational adaptabilities. The developed analytical approach can enhance the interface adaptability of open architecture products. An example is provided to demonstrate the effectiveness of the developed method.     

A Mobile Host Approach for Wireless Powering and Interrogation of Structural Health Monitoring Sensor Networks

Wireless sensor networks (WSNs) for structural health monitoring (SHM) applications can provide the data collection necessary for rapid structural assessment after an event such as a natural disaster puts the reliability of civil infrastructure in question. Technical challenges affecting deployment of such a network include ensuring power is maintained at the sensor nodes, reducing installation and maintenance costs, and automating the collection and analysis of data provided by a wireless sensor network. In this work, a new "mobile host" WSN paradigm is presented. This architecture utilizes nodes that are deployed without resident power. The associated sensors operate on a mechanical memory principle. A mobile host, such as a robot or unmanned aerial vehicle, is used on an as-needed basis to charge the node by wireless power delivery and subsequently retrieve the data by wireless interrogation. The mobile host may be guided in turn to any deployed node that requires interrogation. The contribution of this work is the first field demonstration of a mobile host wireless sensor network. The sensor node, referred to as THINNER, capable of collecting data wirelessly in the absence of electrical power was developed. A peak displacement sensor capable of interfacing with the THINNER sensor node was also designed and tested. A wireless energy delivery package capable of being carried by an airborne mobile host was developed. Finally, the system engineering required to implement the overall sensor network was carried out. The field demonstration took place on an out-of-service, full-scale bridge near Truth-or-Consequences, NM.     

Efficient power-consumption-based load-sharing topology control protocol for harsh environments in wireless sensor networks

Energy conservation and network performance are critical issues in wireless sensor networks.The authors present a novel efficient communication topology control protocol, called quorum-based load-sharing control protocol (QLSCP). QLSCP is a quorum-based communication protocol, which chooses appropriate communication nodes, adjusts the service loads of critical nodes and performs adaptive sleep management. QLSCP is suitable for harsh environments without a central control server calculating the locations of sensors, using the factor of the remaining power to build the system topology. The proposed protocol divides the topology operation into topology formation, adjustment and execution phases. The topology formation phase builds the backbone of an enhanced tree. In the topology adjustment phase, the enhanced tree is adjusted by an optimal balance of critical nodes in the backbone. In the topology execution phase, the efficient surplus-energy consuming (ESC) mechanism is proposed to efficiently exhaust the energy of each node. The ESC mechanism is designed to efficiently exhaust the latest remaining electronic power of sensor nodes. Simulation results of QLSCP demonstrate that it efficiently achieves to prolong system lifetime in harsh environments.     

The future of computing--new architectures and new technologies

All modern computers are designed using the 'von Neumann' architecture and built using silicon transistor technology. Both architecture and technology have been remarkably successful. Yet there are a range of problems for which this conventional architecture is not particularly well adapted, and new architectures are being proposed to solve these problems, in particular based on insight from nature. Transistor technology has enjoyed 50 years of continuing progress. However, the laws of physics dictate that within a relatively short time period this progress will come to an end. New technologies, based on molecular and biological sciences as well as quantum physics, are vying to replace silicon, or at least coexist with it and extend its capability. The paper describes these novel architectures and technologies, places them in the context of the kinds of problems they might help to solve, and predicts their possible manner and time of adoption. Finally it describes some key questions and research problems associated with their use.