Correction model for CO detection in the coal combustion loss process in mines based on GWO-SVM

Accurate detection of CO gas is crucial to the prevention of coal combustion. Tuneable diode laser absorption spectroscopy (TDLAS) is a reliable method for CO detection during coal combustion. The influences of temperature and pressure cause changes in the line strength and linewidth of the index gases’ absorption spectra, leading to sizable measurement errors. To correct the distortion of the CO absorption spectrum caused by temperature and pressure fluctuation, a compensation model based on the grey wolf optimizer–support vector machine (GWO–SVM) was proposed. The results were compared with those of the single SVM, the back propagation neural network (BPNN), and multiple regression analysis (MRA). MRA was revealed to result in the lowest accuracy, which indicated that MRA is not ideal for compensation in TDLAS. The hyperparameter selection of the SVM had the disadvantages of randomness and blindness, which led to instability and large errors. The BPNN achieved better correction in the training stage, but severe overfitting occurred in the testing stage. The modified results revealed that the GWO–SVM model had higher accuracy and stability than the other models. It effectively inhibited the effects of temperature and pressure on the measured concentration and greatly improved the measurement accuracy. The equipment is thus suitable for CO gas detection with the aim to preventing coal combustion loss, and it can be further applied to loss prevention in other process fields.     

单矢量水听器抗混响方法研究

混响是主动声纳的主要干扰。矢量水听器是较新型的水声测量设备,其接收的混响和目标信号之间存在相位差异。基于这些差异,探索了矢量声能流方法用于抗混响处理的可行性,在理论上得到了较高的空间处理增益和时间处理增益。计算结果和仿真结果表明,相对于常规的声压平方积分器,该方法具有很好的抗混响效果。     

SVM与ANN在湖泊富营养化评价中的对比研究

支持向量机是由Vapnik等提出的建立在统计学习理论基础上的一种新的机器学习方法,由于其使用结构风险最小化原则代替经验风险最小化原则,又由于其应用了核函数思想,它可以较好地解决非线性问题;人工神经网络(ANN)已经较成功解决模式识别和任意非线性函数回归问题,但是存在训练样本不足,并可能出现过拟合现象。SVM的结构风险最小化算法引起了科学界的关注,对传统基于经验风险最小化的神经网络算法提出了挑战,文章介绍了SVM和ANN的基本原理,并对二者在巢湖富营养化水平评价上做对比研究,结果表明,ANN比较容易陷入局部最优,支持向量机评价结果更加符合实际。     

2004年几次台风暴雨Q矢量诊断的比较分析

利用非地转ω方程说明了Q矢量散度的物理意义,并通过计算2004年对浙江省造成重大影响的几个主要台风影响过程中的Q矢量散度场,分析比较了Q矢量散度场在不同等压面和垂直方向上的分布特点及其与台风、暴雨的关系,指出低层等压面上Q矢量散度场的演变对台风暴雨预报的作用.     

支持向量机回归在臭氧预报中的应用

采用南京工业区2016年5月20日~8月15日这一高臭氧（O₃）期的O₃、O₃前体物和常规气象资料数据,利用支持向量机回归（SVMr）方法分别预报O₃的小时值、日最大值和最大8 h滑动平均值.结果表明,O₃小时值预报的相关系数（R²）为0.84,平均绝对误差（MAE）和平均绝对百分误差（MAPE）分别为3.44×10^-9和24.48,O₃前期浓度、紫外B波段辐射（UVB）和NO₂浓度是关键因子.O₃日最大值预报的主要因子是NO_x在07：00的浓度和UVB.预报O₃ 8 h时UVB和气温起重要作用.加入前体物项能够使O₃的预报精度提升10%~28%.与多元线性回归方法相比,SVMr对O₃浓度的预报有明显优势.     

基于Kinect的地铁乘客不安全行为识别方法与实验*

为解决地铁视频监控技术对乘客不安全行为只记录不识别且较少考虑识别精确度的问题,提出1种基于Kinect传感器的高效识别方法。以Pelvis为向量起点和动作活动高频关节为终点构建识别特征向量;运用余弦定理获得标准动作与测试动作关节的最大角度差序列;以最大角度差为动作特征量建立相似度计算模型,运用动态时间规整算法（DTW）将初始结果转换为动作相似度。以相连关节法为对照组开展对比实验,结果表明:前者在抽烟、挥拳、挥手呼救等行为识别的准确度分别为91.7%,86.9%,89.2%,平均比对照组高4%以上,显著提高了地铁乘客不安全行为的识别率,可为地铁智能安全管控提供理论与技术依据。     

基于KPCA-LSSVM的矿井工作面动力环境安全评价模型

为实现对采前工作面所处动力环境的客观、准确评价,选取9个直接影响工作面 动力环境的指标因素构建安全评价指标体系,建立基于核主成分分析(KPCA)和最小二乘 支持向量机(LSSVM)的工作面动力环境多因素耦合安全评价智能模型。首先根据KPCA理 论对评价指标施行简约化处理,剔除冗余信息,得出6个简约后的评价指标并输入LSSVM 模型中训练学习,最后得到评价模型。选取从平顶山矿区和大同矿区搜集到的30组工作 面历史数据,按照20∶10的比例对模型进行训练和测试,并将测试结果与其他四种模型 结果进行了对比,结果表明:KPCA方法可有效减少数据信息冗余,利用KPCA优化的 LSSVM模型可准确评价工作面动力环境,误判率为0。     

荧光假单胞菌高效广谱载体的构建及分析

为促进高GC含量基因在荧光假单胞菌(Pseudomonas fluorescens)中表达效果更加理想、操作更加简便,本研究首先采用不依赖基因序列和连接反应的克隆(Sequence and ligation independent cloning,SLIC)方法将载体pCIBhis上与复制相关的序列和标记基因片段构建成克隆载体pCIBS1.然后优化荧光假单胞菌转化方法,用电转化法将pCIBS1导入荧光假单胞菌BL915中,随后又将T7和tac基因启动子分别插入pCIBS1中,成功构建了表达载体pCIBS3和pCIBS2.研究发现载体pCIBS1在大肠杆菌和荧光假单胞菌中均较为稳定,并且将绿色荧光蛋白基因插入表达载体中,在大肠杆菌BL21(DE3)中获得表达,验证了表达载体功能.本研究构建的表达载体和建立的荧光假单胞菌BL915电转化方法,为高GC含量基因在荧光假单胞菌中的表达奠定了基础.     

Long-term prediction model of rockburst in underground openings using heuristic algorithms and support vector machines

Rockburst possibility prediction is an important activity in many underground openings design and construction as well as mining production. Due to the complex features of rockburst hazard assessment systems, such as multivariables, strong coupling and strong interference, this study employs support vector machines (SVMs) for the determination of classification of long-term rockburst for underground openings. SVMs is firmly based on the theory of statistical learning algorithms, uses classification technique by introducing radial basis function (RBF) kernel function. The inputs of models are buried depth H, rocks’ maximum tangential stress σ_θ, rocks’ uniaxial compressive strength σ_c, rocks’ uniaxial tensile strength σ_t, stress coefficient σ_θ/σ_c, rock brittleness coefficient σ_c/σ_t and elastic energy index W_et. In order to improve predictive accuracy and generalization ability, the heuristic algorithms of genetic algorithm (GA) and particle swarm optimization algorithm (PSO) are adopted to automatically determine the optimal hyper-parameters for SVMs. The performance of hybrid models (GA + SVMs = GA-SVMs) and (PSO + SVMs = PSO-SVMs) have been compared with the grid search method of support vector machines (GSM-SVMs) model and the experimental values. It also gives variance of predicted data. A rockburst dataset, which consists of 132 samples, was employed to evaluate the current method for predicting rockburst grade, and the good results of overall success rate were obtained. The results indicated that the heuristic algorithms of GA and PSO can speed up SVMs parameter optimization search, the proposed method is robust model and might hold a high potential to become a useful tool in rockburst prediction research.     

基于支持向量机的采空区稳定性分级

针对采空区稳定性分级的影响因素众多且关系复杂的特点,提出采用支持向量机理论对采空区稳定性进行分级。根据分级评价指标选取原则,选取岩体结构、地质构造、岩石的质量指标、地下可见水、地下水体、周边开采的影响、相邻空区的情况、工程布置、跨度、面积、高度、矿柱的尺寸及布置、埋藏深度和采空区的规格14个影响因子,建立了采空区稳定性评价指标体系,引入支持向量机理论,选择有向无环图方式构造多类分类器,得到采空区稳定性分级的支持向量机模型。将该模型用于山东黄金矿业西山矿区的25个采空区进行分级,并与未确知测度方法的分级情况