面向海量不平衡数据的轨道电路故障诊断方法

为解决轨道电路故障诊断任务中监测数据类别不平衡导致诊断模型决策边界产生偏移,以及数据海量导致训练速度慢的问题,提出数据重采样与集成学习算法相结合的轨道电路故障诊断方法。首先,通过特征合成和数据重采样处理不平衡数据,重采样包括随机降采样和合成少数类过采样技术(SMOTE);然后,利用训练高效的轻量梯度提升机(LightGBM)算法构建面向海量监测数据的故障诊断模块,并设计训练及诊断流程,以网格搜索和交叉验证方法调整关键参数;最后,引入不易受数据不平衡影响的Macro-F₁值作为故障诊断模型评价指标。结果表明:特征合成、数据重采样对不平衡数据下的各故障诊断模型的综合表现均有不同程度的提升,LightGBM相较于其他算法在综合表现和训练时间上都是最佳,可保障准确性和面对海量数据时的快速性。     

针对安全生产隐患的智能搜索引擎系统研究

为解决用户在现有搜索引擎上难以检索到专业和全面的安全生产隐患知识的问题,提出一种安全生产隐患数据的挖掘方法和针对安全生产隐患的智能搜索引擎系统的设计方法。结果表明:对于用户的搜索,系统能够直接反馈智能化查询结果,包括:隐患内容、隐患来源、隐患的行业,隐患场所、法规标准、事故信息等数据以及数据之间的关联关系。给出的示例和结果图表,说明了这种数据挖掘方法的有效性;对相关智能搜索引擎系统的组成、工作原理及实现方式的阐述,说明了系统设计方法的有效性,可为安全生产隐患智能搜索引擎系统的设计与开发提供参考。     

基于大数据分析的大气网格化监测质控技术研究

基于对现阶段中国环境空气的污染特征及监测手段进行分析,研究了如何采用传感器技术进行大气污染防治网格化监测。对标物校准、训化校准、自适应校准、传递校准等校准质控技术进行了充分研究并基于此建立了自主学习神经网络算法的校准体系。为解决传感器应用过程中零点漂移、温度和湿度漂移、时间漂移等问题,利用大数据、基因算法成功开发出了智能数据修正模型,实现了低成本、高稳定、具有相当精度且可自动化运行的网格化监测体系。     

Measures of preferred direction for environmental and ecological circular data

Circular or angular variables indicating direction or cyclical time can be of great interest to scientists studying ecology, biology or environmental issues. A common problem of interest in circular data is estimating a preferred direction and its corresponding distribution. This problem is complicated by the so-called “wrap-around effect” on the circle, which exists because there is no natural minimum or maximum. The usual statistics employed for linear data are inappropriate for directional data, as they do not account for its circular nature. Three choices for summarizing the preferred direction (the sample circular mean, sample circular median and a circular analog of the Hodges–Lehmann estimator) are discussed, with examples from environmental and ecological applications. Similar to the linear data case, the relative emphases of different methods sometimes yield different measures of preferred direction in the presence of outliers or lack of symmetry in the original data. Received: November 2003 / Revised: June 2004     

Modelling the response of tree growth to temperature and CO2 elevation as related to the fertility and current temperature sum of a site

A methodology for simulating climate change impacts on tree growth was introduced into a statistical growth and yield model in relation to variations in site fertility and location implemented with current temperature sum. This was based on a procedure in which the relative enhancement in stem volume growth was calculated from short-term runs of a physiological simulation model for Scots pine (Pinus sylvestris L.), Norway spruce (Picea abies (L.) Karst.) and silver birch (Betula pendula Roth.) stands. These simulations were made for a set of stands with species-specific variations in stand characteristics, location and fertility type first in current climatic conditions and then in different combinations of CO₂ and temperature elevations. Based on these simulations, the relative enhancement of volume growth induced by the climate change (relative scenario effect, RSEv) was calculated and modelled in relation to: (i) CO₂ and temperature elevation, stand density and the competition status of the tree in its stand, and (ii) variations in site fertility type and current temperature sum of a stand. Finally, these transfer functions for RSEv were applied to adapt the stem volume growth in the statistical growth and yield model to reflect the response to climate change.     

基于Profibus-DP的废钢破碎线控制系统设计与实现

江苏鼎豪再生资源设备有限公司2012年成功开发了大型废钢破碎线,采用PLC并基于Profibus-DP总线构成监督控制与采集（SCADA）系统。用模糊控制思想实现了破碎机的自动加料控制,并采用组态软件WINCC开发了操作、监控界面,在控制室实现了对整条破碎生产线进行控制与监控。     

Recursive Operability Analysis as a decision support tool for Risk-Based Maintenance

This work was developed with the support of MEMC, one of the most important producers of ultra-pure silicon wafers for electronic applications throughout the world. The availability of ultra-pure water is of prime importance in the silicon production process. In order to maximize the availability of UP water, MEMC has developed a preventive maintenance program and a detailed record of each maintenance intervention is recorded. This has allowed a complete failure rate data bank to be developed. In order to optimize the maintenance intervention, a Recursive Operability Analysis (ROA) has been used as a decision support tool. The results of the ROA, coupled with the failure rates, have made it possible to calculate the expected number of events (ENE) of various top events (TEs). The magnitude of each TE has been estimated on the basis of the monetary losses provoked by each event. The risk then has been calculated and the events ranked on this basis. Maintenance policies have been optimized with the aim of reducing the risk of the top ranked events.     

一次特大沙尘暴对乌鲁木齐市环境空气质量的影响分析

通过对2001年5月11日乌鲁木齐市发生的特大沙尘暴及同期有关监测数据进行分析比较,指出乌鲁木齐市沙尘暴中的主要污染物为空气中的可吸入颗粒物,其成因除自然环境因素处,还与城市扬尘有关。     

试论环境监测数据审核机制的建立及运作

环境监测数据审核是环境监测质量管理的重要一环。在对环境监测数据审核重要性的认识，及其内容，依据分析解释的基础上建立了环境监测数据审核运行机制。并提出了几点措施，确保环境监测数据在获取过程中得到了有效监控，把监测数据差错率消除在监测过程的各个环节。