A Surrogate Model for a CAES Radial Inflow Turbine with Test Data-Based MLP Neural Network Algorithm

It is usually to conduct a full-scale three-dimensional flow analysis for a radial turbine to find a way to increase the efficiency of a Compressed Air Energy Storage(CAES)system.However,long solving time and huge consumption of computing resources become a major obstacle to the analysis.Therefore,in present study,a surrogate model with test data-based multi-layer perceptron(MLP)Neural Network is proposed to overcome the difficulty.Instead of complex flow field solving process,it provides reliable turbine aerodynamic performance and flow field distribution characteristics in a short solution time by“learning the measurement results”.The validation results illustrated that the predicted maximum relative errors of isentropic efficiency,corrected mass flow rate and corrected power are only 0.03%,0.22%and 0.26%respectively.The predicted flow distribution parameters in chamber,shroud cavity and outlet region of rotor are also basically consistent with the experimental results.In the chamber,it can be found that a pressure stagnation point is observed at circumferential angle of 270°when total pressure ratio is decreased.In the shroud cavity,obvious pressure variation is found near outlet of shroud cavity which although labyrinth seals exist.At outlet of rotor,obvious variations of velocity and pressure are found in the 0.0–0.4 and 0.6–0.8 of blade height.At the same time,obvious variations of velocity and pressure are found in the 0.0–0.4 and 0.6–0.8 of blade height and this is because the influence of upper passage vortex,lower passage vortex and end wall secondary flow.The present study can provide further reference for the dynamic performance evaluation of CAES radial inflow turbine.

Mapping salt diapirs and salt diapir-affected areas using MLP neural network model and ASTER data

This study employs visible-near infrared and short wave infrared datasets of Advanced Spaceborne Thermal Emission and Reflection Radiometer(ASTER)to map salt diapirs and salt diapir-affected areas using Multi-Layer Perceptron(MLP)in the Zagros Folded Belt,Iran,and introduces the role of earth observation technology and a type of digital earth processing in lithological mapping and geo-environmental impact assessment.MLP neural network model with several learning rates between 0.01 and 0.1 was carried out on ASTER L1B data,and the results were compared using confusion matrices.The most appropriate classification image for L1B input to MLP was produced by learning rate of 0.01 with Kappa coefficient of 0.90 and overall accuracy of 92.54%.The MLP result of input data set mapped lithological units of salt diapirs and demonstrated affected areas at the southern and western parts of the Konarsiah and Jahani diapirs,respectively.Field observations and X-ray diffraction analyses of field samples confirmed the dominant mineral phases identified remotely.It is concluded that MLP is an efficient approach for mapping salt diapirs and salt-affected areas.

SSA-MLP模型在岩质边坡稳定性预测中的应用

岩质边坡的力学参数量化及稳定性分析对岩质边坡灾害的防治具有重要意义。Hoek-Brown(H B)准则是一种用于确定岩体力学参数的经典方法,能反映出边坡岩体变形和位移的非线性破坏特征。在此基础上,首先,提出一种麻雀搜索算法(Sparrow Search Algorithm,SSA)改进多层感知器(Multi-Layer Perceptron,MLP)的神经网络模型,并用于边坡稳定性预测、指标敏感性分析及参数反演。其次,将收集的1085组岩质边坡的几何参数和H B准则参数等作为输入变量,极限平衡理论Bishop法求解的安全系数作为输出变量,对SSA MLP模型进行训练学习和性能评估。最后,将该模型运用于25个边坡实例,验证模型的有效性。结果显示,该模型收敛速度快、精度高,为边坡稳定性分析和参数量化提供了一种新思路。

GraphMLP-Mixer:基于图-多层感知机架构的高效多行为序列推荐方法

在多行为序列推荐领域,图神经网络(GNNs)虽被广泛应用,但存在局限性,如对序列间协同信号建模不足和处理长距离依赖性等问题.针对这些问题,提出了一种新的解决框架GraphMLP-Mixer.该框架首先构造全局物品图来增强模型对序列间协同信号的建模,然后将感知机-混合器架构与图神经网络结合,得到图-感知机混合器模型对用户兴趣进行充分挖掘.GraphMLP-Mixer具有2个显著优势:一是能够有效捕捉用户行为的全局依赖性,同时减轻信息过压缩问题;二是其时间与空间效率显著提高,其复杂度与用户交互行为的数量成线性关系,优于现有基于GNN多行为序列推荐模型.在3个真实的公开数据集上进行实验,大量的实验结果验证了GraphMLP-Mixer在处理多行为序列推荐问题时的有效性和高效性.

基于MLP-Bagging集成分类模型的在线学习行为分析

针对教育者难以对学习者多样化的在线学习行为进行监测和研判等问题,提出一种带嵌入层的MLP-Bagging集成分类模型对学习者的在线学习行为进行分析与判别.考虑到学习者的在线学习行为以及个体特性,从学习准备行为、知识获取行为、交互学习行为、学习巩固行为和辅助特征5个方面构建在线学习行为模型,并采用MLP-Bagging集成分类模型对学习者进行分类判别.实验结果表明,所构建的学习模型可对在线学习者的学习行为进行符合实际的建模,加入辅助特征有利于对各类学习者的在线学习行为进行深入的分析与指导,并且在分类模型中加入嵌入层可以有效克服标签编码带来的数据冗余和误差缺陷,从而获得更好的分类效果.与其他分类模型相比,融合多个MLP分类器的Bagging集成模型可以减少单个MLP分类器的方差,其分类准确率达到98.72%,具有较好的实际应用价值.

湿度对PM_(2.5)/PM_(10)比值影响及MLP神经网络大气污染预测

颗粒物污染严重影响大气环境质量和公众身体健康.基于郑州、太原、成都2022—2023年秋冬季大气污染和气象数据,讨论了PM_(2.5)、PM_(10)污染特征,通过计算斯皮尔曼系数并进行偏相关分析,得出PM_(2.5)/PM_(10)与相对湿度之间的相关性,并利用MLP神经网络建立关于PM_(2.5)和PM_(2.5)/PM_(10)的预测模型.结果表明,郑州、太原、成都2022—2023年秋冬季PM_(2.5)均值分别为63.52、52.68、50.95μg/m^(3),PM_(2.5)/PM_(10)比值都超过0.5.郑州、太原PM_(2.5)/PM_(10)与相对湿度呈现较强的正相关性.利用MLP神经网络对PM_(2.5)和PM_(2.5)/PM_(10)建立预测模型,误差分别为6.75μg/m^(3)和0.076,模型精度较高.建议未来从秋冬供暖、推广工业水蒸气回收装置、合理开展道路洒水等方面巩固颗粒物污染治理成效,探索和优化机器学习模型用于PM_(2.5)/PM_(10)预测,为大气污染防治提供数据参考,以求发挥大数据在做到精准治污中的作用.

PCA-MLP神经网络模型在黄河宁夏段径流预测中的应用

为提高短时间尺度的月径流量预测效果并简化神经网络模型结构,将主成分分析(PCA)和多层感知器神经网络(MLP)相结合,构建PCA-MLP神经网络模型预测汛期月径流量。模型首先采用主成分分析法确定影响径流量的主要影响因子,再将主要影响因子数据输入MLP神经网络模型,预测月径流量数据。将宁夏青铜峡水文站2010—2019年汛期的月径流量和影响因子数据作为训练样本训练神经网络模型,以2020—2022年汛期月径流量和影响因子数据作为检验样本进行对比分析。预测结果表明:目前影响汛期径流量的因素主要是历史径流和气候特征,检验集预测结果确定性系数为0.851,模型可为宁夏汛期月径流量预测提供相应指导。

基于MLP和Transformer模型的大气温度预测

文章以运城市2015年1月1日至2020年12月21日期间监测的大气温度数据作为研究的基础资料,运用MLP模型和Transformer模型,预测了运城市大气温度。由于温度数据具有很强的时序性,对MLP模型与Transformer模型,各选取了两层、四层(MLP-2、MLP-4、Transformer-2、Transformer-4),进行了3天、5天、7天多组试验对比。结果显示:MLP-4模型7天的均方误差为3.2649,Transformer-4模型3天的均方误差为5.3767,预测精度都比较高,且MLP模型预测温度的精度高于Transformer模型预测温度的精度;MLP-2模型的均方误差分别为3.2662、3.2996、3.3579,MLP-4模型的均方误差分别为3.2674、3.2996、3.2649,均方误差有变化,但比较平稳;Transformer-2模型的均方误差分别为5.6225、5.9491、5.3892,Transformer-4模型的均方误差分别为5.3767、6.3787、6.1108,增加模型层数和参数量,均方误差增大,存在过拟合现象。运用Transformer模型进行预测,出现过拟合现象,原因是Transformer模型太过庞大(接近四百万个参数),而研究数据只有1531组,即使使用Weight decay和Dropout正则化的方法,仍然过拟合文章中提供的1531组研究数据,使其预测精度出现一定程度的下降。

基于MLP-Bagging的PCB电热耦合建模方法

随着三维集成电路性能的提高和复杂程度的增加,印制电路板(PCB)的散热问题日益突出。研究了PCB在电热多物理场相互作用下各部件的发热情况,提出了基于混合激活函数的多层感知机(MLP)-Bagging多物理参数算法。通过使用ReLU和Sigmoid两个激活函数进行学习和训练,建立了精度更高的MLP模型。之后,结合Bagging算法构建多个并行的MLP模型。所提出的神经网络多物理模型可以快速准确地预测PCB的电热响应。实验结果表明,此方法与有限元法相比,可以节省约97%的计算内存和99%的计算时间,与传统神经网络如随机森林(RF)、长短时记忆(LSTM)网络、MLP相比,该方法表现优良且泛化能力较好,为提高PCB设计效率提供了一种可行方法,为PCB热分析提供了更高效的解决方法。

Simple model based on artificial neural network for early prediction and simulation winter rapeseed yield

The aim of the research was to create a prediction model for winter rapeseed yield.The constructed model enabled to perform simulation on 30 June,in the current year,immediately before harvesting.An artificial neural network with multilayer perceptron(MLP) topology was used to build the predictive model.The model was created on the basis of meteorological data(air temperature and atmospheric precipitation) and mineral fertilization data.The data were collected in the period 2008–2017 from 291 productive fields located in Poland,in the southern part of the Opole region.The assessment of the forecast quality created on the basis of the neural model has been verified by defining forecast errors using relative approximation error(RAE),root mean square error(RMS),mean absolute error(MAE),and mean absolute percentage error(MAPE) metrics.An important feature of the created predictive model is the ability to forecast the current agrotechnical year based on current weather and fertilizing data.The lowest value of the MAPE error was obtained for a neural network model based on the MLP network of 21:21-13-6-1:1 structure,which was 9.43%.The performed sensitivity analysis of the network examined the factors that have the greatest impact on the yield of winter rape.The highest rank 1 was obtained by an independent variable with the average air temperature from 1 January to 15 April of 2017(designation by the T1-4_CY model).

A Neural Network Approach for Misuse and Anomaly Intrusion Detection

An MI.P(Multi-Layer Perception)/Elman neural network is proposed in thispaper, which realizes classification with memory of past events using the real-time classificationof MI.P and the memorial functionality of Elman. The system's sensitivity for the memory of pastevents ean be easily reconfigured without retraining the whole network. This approach can he usedfor both misuse and anomaly detection system. The intrusion detection systems(TDSs) using the hybridMLP/Elman neural network are evaluated by the intrusion detection evaluation data sponsored by U.S.Defense Advanced Research Projects Agency CDARPA) Ihc results of experiment are presented inReceiver Operating Characteristic CROC) curves. Thc capabilites of these IDSs to identify DenyofService(DOS) and probing attacks are enhanced.

基于MLP神经网络的水泥分别粉磨强度预测模型与分析

为研究分别粉磨配制水泥中多组分下各原料对强度的影响,提出利用多层感知器(multi-layer perceptron,MLP)神经网络建立模型的方法。选用矿粉、粉煤灰和石灰石粉作为混合材,进行单因素试验得到活性数据,设计分别粉磨配制水泥配比组成试验样本,通过训练数据构建MLP神经网络模型,归纳出相关因素与强度之间的表达式。结果表明:活性越高的混合材多掺或适当磨细更有助于提高强度,增加石灰石粉含量有助于提升3 d抗压强度;所建立模型对3 d和28 d龄期抗压强度的预测结果精准,平均相对误差小于2%。

Determination of penetration depth at high velocity impact using finite element method and artificial neural network tools

Determination of ballistic performance of an armor solution is a complicated task and evolved significantly with the application of finite element methods(FEM) in this research field.The traditional armor design studies performed with FEM requires sophisticated procedures and intensive computational effort,therefore simpler and accurate numerical approaches are always worthwhile to decrease armor development time.This study aims to apply a hybrid method using FEM simulation and artificial neural network(ANN) analysis to approximate ballistic limit thickness for armor steels.To achieve this objective,a predictive model based on the artificial neural networks is developed to determine ballistic resistance of high hardness armor steels against 7.62 mm armor piercing ammunition.In this methodology,the FEM simulations are used to create training cases for Multilayer Perceptron(MLP) three layer networks.In order to validate FE simulation methodology,ballistic shot tests on 20 mm thickness target were performed according to standard Stanag 4569.Afterwards,the successfully trained ANN(s) is used to predict the ballistic limit thickness of 500 HB high hardness steel armor.Results show that even with limited number of data,FEM-ANN approach can be used to predict ballistic penetration depth with adequate accuracy.

Performance Comparison of Neural Networks for HRTFs Approximation

In order to approach to head related transfer functions (HRTFs), this paper employs and compares three kinds of one input neural network models, namely, multi layer perceptron (MLP) networks, radial basis function (RBF) networks and wavelet neural networks (WNN) so as to select the best network model for further HRTFs approximation. Experimental results demonstrate that wavelet neural networks are more efficient and useful.

线性回归方法与MLP相结合预测李家河水库的综合水质指数

为了提高水库水质的监测和管理水平,提升对城市供水的质量,采用李家河水库2021全年实测的水质指标数据(每周测试一次),通过研究对比选择了七个重要指标(分别为:总氮、氨氮、总磷、高锰酸盐系数、总铁、总锰和溶解氧)作为自变量,将反映水库水质水平高低的综合水质指数(WQI)作为因变量,分别构建、训练得到了多元线性回归模型和MLP多层人工神经网络模型;并利用所建模型对2022年上半年水库的WQI进行预测,经与实测数据对比发现,将多元线性回归模型和MLP模型相结合预测的综合水质指数明显优于两个单独模型的预测结果,WQI变化趋势的预测准确率高达95.24%;本研究所获得的这一综合水质指数复合预测模型对于李家河水库后续水质监测和生产管理会起到有效的辅助作用,对于同类型中小水库也具有一定的推广借鉴价值。

Performance comparison of three artificial neural network methods for classification of electroencephalograph signals of five mental tasks

In this paper, performance of three classifiers for classification of five mental tasks were investigated. Wavelet Packet Transform (WPT) was used for feature extraction of the relevant frequency bands from raw Electroencephalograph (EEG) signal. The three classifiers namely used were Multilayer Back propagation Neural Network, Support Vector Machine and Radial Basis Function Neural Network. In MLP-BP NN five training methods used were a) Gradient Descent Back Propagation b) Levenberg-Marquardt c) Resilient Back Propagation d) Conjugate Learning Gradient Back Propagation and e) Gradient Descent Back Propagation with movementum.

An Improved SPSA Algorithm for System Identification Using Fuzzy Rules for Training Neural Networks

Simultaneous perturbation stochastic approximation （SPSA） belongs to the class of gradient-free optimization methods that extract gradient information from successive objective function evaluation. This paper describes an improved SPSA algorithm, which entails fuzzy adaptive gain sequences, gradient smoothing, and a step rejection procedure to enhance convergence and stability. The proposed fuzzy adaptive simultaneous perturbation approximation （FASPA） algorithm is particularly well suited to problems involving a large number of parameters such as those encountered in nonlinear system identification using neural networks （NNs）. Accordingly, a multilayer perceptron （MLP） network with popular training algorithms was used to predicate the system response. We found that an MLP trained by FASPSA had the desired accuracy that was comparable to results obtained by traditional system identification algorithms. Simulation results for typical nonlinear systems demonstrate that the proposed NN architecture trained with FASPSA yields improved system identification as measured by reduced time of convergence and a smaller identification error.

基于MLP和RBF神经网络模型预测在线开放课程学习者满意度

网络课程在近10年的发展中,在线学习者的满意度一直是一个重要的理论热点。依托某大规模在线教学与学习平台的虚拟学习环境,基于某校在网络学习空间开设的9门在线开放课程,结合LMS系统中的在线学习行为分析数据,研究开发、训练了具有不同激活函数的多层感知器(MLP)、径向基函数(RBF)以及分类树模型,以预测学生对于在线开放课程的满意度。Ⅰ型和Ⅱ型误差以及输入变量重要性用于模型比较和分类准确性判断。经过实验,模型的总体分类准确率高于50%。尽管t检验表明模型之间的差异在统计上并不显著,但是MLP模型为低课程满意度的学习者提供了最高的平均分类精度和最低的错误分类偏好。在所有模型中,在线论坛中学习者互动参与度被认为是预测学习者在线课程满意度高的重要指标。

串联网式过滤器拦截特性和过滤时间分析

针对网式过滤器在实际灌溉水源条件下易堵塞、使用寿命短等问题,该研究将泵前和泵后网式过滤器串联组合,通过室内原型试验分析各级过滤器对泥沙和木屑的拦截规律。结果表明:过滤器拦截泥沙和木屑的分布情况与各级滤网孔径有关,当水源杂质以泥沙为主时,建议选取0.32和0.20 mm孔径滤网的泵前和泵后过滤器进行过滤,当有机物杂质含量较高时,采用0.25~0.32 mm滤网孔径的泵前过滤器有利于延长过滤时间。对水头损失随过滤时间的变化规律进行分析,明确了水源条件和滤网孔径对过滤时间的影响,极差和方差分析结果表明:不同进水流量条件下,初始水头损失和水头损失峰值分别集中在2.43~5.87和13.92~28.92 m,杂质含量对水头损失影响较小;进水流量越大,过滤器过滤时间越短。建立水头损失计算公式和过滤时间拟合经验公式,拟合与试验结果误差小于7%。各因素对过滤时间影响程度由大到小依次是杂质含量、滤网孔径、含沙量、进水流量。基于多层感知机(multilayer perceptron,MLP)神经网络模型对过滤时间进行预测,预测值与实测值误差基本在10%以内,均方误差和平均相对误差分别为0.32%、5.85%,该模型适用于对复杂水源条件下过滤时间的预测。研究结果可为灌溉工程中串联网式过滤器各级滤网参数的合理配置提供参考。

基于锥形追踪和网络分解的NeRF三维重建方法

在计算机视觉领域,神经辐射场(NeRF)是以空间坐标或者时间、相机位姿等其他维度作为输入,通过多层感知机(MLP)网络模拟目标函数,生成颜色、深度等目标标量的过程。NeRF的应用包括对三维场景进行高质量的重建,而其在处理不同分辨率的场景时会产生过度模糊或者伪影的渲染效果,且存在训练耗时较长的问题。为了解决上述问题,提出基于锥形追踪和网络分解的NeRF三维重建方法。使用锥形追踪的方法,为每个像素投射一个圆锥体,并将投射的圆锥体切割成一系列的圆锥台,沿着该圆锥体进行特征化,通过高效渲染抗锯齿的圆锥台来降低模糊或者伪影效果。为了缩短训练时间,使用网络分解的方法,将原始NeRF接收5维数据的神经网络分解为两个网络,有效地缩短训练时间。实验结果表明,在NeRF_Synthetic、LLFF和Multiresolution数据集中,相比于NeRF、F 2-NeRF等方法,所提方法的峰值信噪比(PSNR)提升了14.4%~24.6%,能够重建出更丰富的细节特征,视觉效果更好,且训练时间大幅降低。