A Kind of Second-Order Learning Algorithm Based on Generalized Cost Criteria in Multi-Layer Feed-Forward Neural Networks

A kind of second order algorithm--recursive approximate Newton algorithm was given by Karayiannis. The algorithm was simplified when it was formulated. Especially, the simplification to matrix Hessian was very reluctant, which led to the loss of valuable information and affected performance of the algorithm to certain extent. For multi layer feed forward neural networks, the second order back propagation recursive algorithm based generalized cost criteria was proposed. It is proved that it is equivalent to Newton recursive algorithm and has a second order convergent rate. The performance and application prospect are analyzed. Lots of simulation experiments indicate that the calculation of the new algorithm is almost equivalent to the recursive least square multiple algorithm. The algorithm and selection of networks parameters are significant and the performance is more excellent than BP algorithm and the second order learning algorithm that was given by Karayiannis.

Prediction Model of Drilling Costs for Ultra-Deep Wells Based on GA-BP Neural Network

Drilling costs of ultra-deepwell is the significant part of development investment,and accurate prediction of drilling costs plays an important role in reasonable budgeting and overall control of development cost.In order to improve the prediction accuracy of ultra-deep well drilling costs,the item and the dominant factors of drilling costs in Tarim oilfield are analyzed.Then,those factors of drilling costs are separated into categorical variables and numerous variables.Finally,a BP neural networkmodel with drilling costs as the output is established,and hyper-parameters(initial weights and bias)of the BP neural network is optimized by genetic algorithm(GA).Through training and validation of themodel,a reliable prediction model of ultra-deep well drilling costs is achieved.The average relative error between prediction and actual values is 3.26%.Compared with other models,the root mean square error is reduced by 25.38%.The prediction results of the proposed model are reliable,and the model is efficient,which can provide supporting for the drilling costs control and budget planning of ultra-deep wells.

Fault Diagnosis Based on BP Neural Network Optimized by Beetle Algorithm

In the process of Wavelet Analysis,only the low-frequency signals are re-decomposed,and the high-frequency signals are no longer decomposed,resulting in a decrease in frequency resolution with increasing frequency.Therefore,in this paper,firstly,Wavelet Packet Decomposition is used for feature extraction of vibration signals,which makes up for the shortcomings of Wavelet Analysis in extracting fault features of nonlinear vibration signals,and different energy values in different frequency bands are obtained by Wavelet Packet Decomposition.The features are visualized by the K-Means clustering method,and the results show that the extracted energy features can accurately distinguish the different states of the bearing.Then a fault diagnosis model based on BP Neural Network optimized by Beetle Algo-rithm is proposed to identify the bearing faults.Compared with the Particle Swarm Algorithm,Beetle Algorithm can quickly find the error extreme value,which greatly reduces the training time of the model.At last,two experiments are conducted,which show that the accuracy of the model can reach more than 95%,and the model has a certain anti-interference ability.

Intelligent direct analysis of physical and mechanical parameters of tunnel surrounding rock based on adaptive immunity algorithm and BP neural network

Because of complexity and non-predictability of the tunnel surrounding rock, the problem with the determination of the physical and mechanical parameters of the surrounding rock has become a main obstacle to theoretical research and numerical analysis in tunnel engineering. During design, it is a frequent practice, therefore, to give recommended values by analog based on experience. It is a key point in current research to make use of the displacement back analytic method to comparatively accurately determine the parameters of the surrounding rock whereas artificial intelligence possesses an exceptionally strong capability of identifying, expressing and coping with such complex non-linear relationships. The parameters can be verified by searching the optimal network structure, using back analysis on measured data to search optimal parameters and performing direct computation of the obtained results. In the current paper, the direct analysis is performed with the biological emulation system and the software of Fast Lagrangian Analysis of Continua (FLAC3D. The high non-linearity, network reasoning and coupling ability of the neural network are employed. The output vector required of the training of the neural network is obtained with the numerical analysis software. And the overall space search is conducted by employing the Adaptive Immunity Algorithm. As a result, we are able to avoid the shortcoming that multiple parameters and optimized parameters are easy to fall into a local extremum. At the same time, the computing speed and efficiency are increased as well. Further, in the paper satisfactory conclusions are arrived at through the intelligent direct-back analysis on the monitored and measured data at the Erdaoya tunneling project. The results show that the physical and mechanical parameters obtained by the intelligent direct-back analysis proposed in the current paper have effectively improved the recommended values in the original prospecting data. This is of practical significance to the appraisal of stability and informationization design of the surrounding rock.

An Image Encryption Algorithm Based on BP Neural Network and Hyperchaotic System

To reduce the bandwidth and storage resources of image information in communication transmission, and improve the secure communication of information. In this paper, an image compression and encryption algorithm based on fractional-order memristive hyperchaotic system and BP neural network is proposed. In this algorithm, the image pixel values are compressed by BP neural network, the chaotic sequences of the fractional-order memristive hyperchaotic system are used to diffuse the pixel values. The experimental simulation results indicate that the proposed algorithm not only can effectively compress and encrypt image, but also have better security features. Therefore, this work provides theoretical guidance and experimental basis for the safe transmission and storage of image information in practical communication.

Mechanical Properties Prediction of the Mechanical Clinching Joints Based on Genetic Algorithm and BP Neural Network

For optimal design of mechanical clinching steel-aluminum joints, the back propagation(BP) neural network is used to research the mapping relationship between joining technique parameters including sheet thickness, sheet hardness, joint bottom diameter etc., and mechanical properties of shearing and peeling in order to investigate joining technology between various material plates in the steel-aluminum hybrid structure car body.Genetic algorithm(GA) is adopted to optimize the back-propagation neural network connection weights.The training and validating samples are made by the BTM? Tog-L-Loc system with different technologic parameters.The training samples' parameters and the corresponding joints' mechanical properties are supplied to the artificial neural network(ANN) for training.The validating samples' experimental data is used for checking up the prediction outputs.The calculation results show that GA can improve the model's prediction precision and generalization ability of BP neural network.The comparative analysis between the experimental data and the prediction outputs shows that ANN prediction models after training can effectively predict the mechanical properties of mechanical clinching joints and prove the feasibility and reliability of the intelligent neural networks system when used in the mechanical properties prediction of mechanical clinching joints.The prediction results can be used for a reference in the design of mechanical clinching steel-aluminum joints.

CNC Thermal Compensation Based on Mind Evolutionary Algorithm Optimized BP Neural Network

Thermal deformation error is one of the most important factors affecting the CNCs’ accuracy, so research is conducted on the temperature errors affecting CNCs’ machining accuracy;on the basis of analyzing the unpredictability and pre-maturing of the results of the genetic algorithm, as well as the slow speed of the training speed of the particle algorithm, a kind of Mind Evolutionary Algorithm optimized BP neural network featuring extremely strong global search capacity was proposed;type KVC850MA/2 five-axis CNC of Changzheng Lathe Factory was used as the research subject, and the Mind Evolutionary Algorithm optimized BP neural network algorithm was used for the establishment of the compensation model between temperature changes and the CNCs’ thermal deformation errors, as well as the realization method on hardware. The simulation results indicated that this method featured extremely high practical value.

Neural Network Based on GA-BP Algorithm and its Application in the Protein Secondary Structure Prediction

The advantages and disadvantages of genetic algorithm and BP algorithm are introduced. A neural network based on GA-BP algorithm is proposed and applied in the prediction of protein secondary structure, which combines the advantages of BP and GA. The prediction and training on the neural network are made respectively based on 4 structure classifications of protein so as to get higher rate of predication---the highest prediction rate 75.65%,the average prediction rate 65.04%.

Parameters Optimization of the Heating Furnace Control Systems Based on BP Neural Network Improved by Genetic Algorithm

The heating technological requirement of the conventional PID control is difficult to guarantee which based on the precise mathematical model,because the heating furnace for heating treatment with the big inertia,the pure time delay and nonlinear time-varying.Proposed one kind optimized variable method of PID controller based on the genetic algorithm with improved BP network that better realized the completely automatic intelligent control of the entire thermal process than the classics critical purporting(Z-N)method.A heating furnace for the object was simulated with MATLAB,simulation results show that the control system has the quicker response characteristic,the better dynamic characteristic and the quite stronger robustness,which has some promotional value for the control of industrial furnace.

Circle BP Algorithm for MLP Neural Network

１ＩｎｔｒｏｄｕｃｔｉｏｎＩｎｔｈｅｐａｓｔｓｅｖｅｒａｌｄｅｃａｄｅｓ，ｔｈｅｎｅｒｖｏｕｓｓｙｓｔｅｍｈａｓｂｅｎｓｔｕｄｉｅｄ，ａｎａｌｙｚｅｄａｎｄｍｏｄｅｌｅｄｉｎｔｈｅｈｏｐｅｏｆａｃｈｉｅｖｉｎｇｈｕｍａｎ－ｌｉｋｅｉｎｔｅｌｉｇｅｎｃ...

正交实验结合AHP和GA-BP神经网络优化益黄散醇提工艺

目的 优化益黄散的醇提工艺。方法 采用回流提取法,以乙醇体积分数、液料比、提取时间为考察因素设计正交实验,以橙皮苷、川陈皮素、橘皮素、没食子酸、诃黎勒酸、诃子酸、甘草苷、甘草酸、丁香酚含量和干浸膏得率为指标,采用层次分析法(AHP)进行赋权并计算综合评分。通过验证正交实验和遗传算法(GA)-反向传播神经网络(BP神经网络)所预测的结果确定益黄散最佳醇提工艺参数。结果 正交实验优选的最佳醇提工艺参数为乙醇体积分数60%、液料比14∶1(mL/g)、提取时间90 min、提取2次,验证所得综合评分为79.19分;GA-BP神经网络优选的最佳醇提工艺参数为乙醇体积分数65%、液料比14∶1(mL/g)、提取时间60 min、提取2次,验证所得综合评分为85.30分,高于正交实验所得结果。结论 采用正交实验结合GA-BP神经网络的寻优方法较传统的正交实验寻优方法效果更佳,其优选出的益黄散最佳醇提工艺稳定可靠。

基于FA-BP神经网络的生姜干燥含水率预测

为探索生姜的干燥特性,并实现生姜干燥的含水率预测,研究了不同干燥温度(50、55、60℃)、干燥风速(1.0、2.0、3.0m/s)、切片长度(30、35、40mm)对生姜干燥时间和干燥速率的影响。结合BP神经网络自适应能力、泛化能力、学习能力强和萤火虫算法(FA)参数少、寻优能力强、收敛速度快等特点,将干燥温度、干燥风速、切片长度和干燥时间作为输入层,隐藏层个数为10,输出层为生姜的含水率,搭建一个拓扑结构为“4-10-1”的FA-BP神经网络模型。研究结果表明:干燥温度、干燥风速、切片长度都是影响生姜含水率的关键因素,增加干燥风速、提高干燥温度和减少切片长度能有效缩短生姜的干燥时间,提高干燥效率。选用萤火虫算法优化BP神经网络的权值和阈值,减少了神经网络的训练时间,提高了精准度,其含水率预测值与试验值之间的决定系数R2=0.999 02,均方根误差RMSE为0.002 99,含水率预测结果准确且迅速,能够为生姜干燥过程中的含水率在线预测提供科学依据。

基于CSSA-BPNN模型的胶结充填体动态抗压强度预测

充填采矿法二步骤回采时胶结充填体稳定性受爆破扰动而降低。为快速准确地获得充填体动态抗压强度,利用分离式霍普金森压杆(SHPB)进行了40组不同应变率的单轴冲击实验,以灰砂比、充填体密度、养护龄期和平均应变率作为输入参数,充填体动态抗压强度作为输出参数,建立了一种基于Logistic混沌麻雀搜索算法(CSSA)优化BP神经网络(BPNN)的预测模型,并与传统BPNN和麻雀搜索算法优化的BPNN进行了对比分析。结果表明:CSSA-BPNN模型的平均相对误差为4.11%,预测值与实测值之间拟合的相关系数均在0.96以上,模型预测精度高。CSSA-BPNN模型的均方根误差为0.395 0 MPa,平均绝对误差为0.359 2 MPa,决定系数为0.995 2,均优于另外两种预测模型。实现了对充填体动态抗压强度的准确预测,可大幅减小物理实验量,为矿山胶结充填体的强度设计提供了一种新方法。

小波包与遗传算法优化BP神经网络相结合的井架钢结构损伤识别

井架钢结构损伤影响其承载安全性,为快速、准确对损伤位置进行识别,提出小波包与遗传算法优化BP神经网络相结合的井架钢结构损伤识别方法。首先利用小波包处理非平稳振动信号的优良性能对原始振动信号进行特征提取,获得表征井架钢结构损伤的信息;再通过特征参数建立数据集训练并测试井架钢结构损伤识别模型,该模型结合遗传算法自身特点改善了传统BP神经网络的不足。本文识别方法不需要损伤前的数据特征进行对比,便可对损伤位置进行确定。经过对石油井架钢结构模型实验验证:该方法对井架钢结构损伤识别准确率超过90%,相对于BP网络识别准确率以及识别速度均有所提高。

基于BP神经网络算法的异步电机故障诊断系统研究

为了确保电机安全可靠地运行,研究了BP神经网络算法对异步电动机进行故障诊断。通过MATLAB平台,分别使用附加动量因子和自适应学习率两种梯度下降法进行网络训练,搭建故障诊断BP网络模型。以MSE值为指标优化最佳隐含层节点数、动量因子与学习率,并通过遗传算法来优化BP网络的初始权值,对故障测试样本进行仿真测试。结果表明,GA-BP网络模型比MF-BP和AG-BP的MSE值更低,仅为0.009163,优化后的诊断预测结果与目标值几乎没有差别。基于遗传算法改进的故障诊断系统模型能够满足异步电动机故障诊断的应用需求。

基于改进的SSA-BP神经网络的矿井突水水源识别模型研究

机器学习与寻优算法的结合在矿井突水水源识别上得到广泛应用,但突水水样数据具有随机性且寻优算法易陷入局部最优,提高模型泛化能力和跳出局部最优需进一步研究。针对上述问题,提出了一种改进的麻雀搜索算法(SSA)优化BP神经网络模型,用于对矿井突水水源进行定量辨识。以鲁能煤电股份有限公司阳城煤矿为研究对象,通过常规离子浓度分析、Piper三线图对该煤矿水样的水化学特征进行分析,初步判断矿井水来源于奥灰含水层和三灰含水层,并确定Na^(+)+K^(+)浓度、Ca^(2+)浓度、Mg^(2+)浓度、HCO_(3)^(-)浓度、SO_(4)^(2-)浓度、Cl^(-)浓度、矿化度、总硬度、pH值作为突水水源识别指标;建立基于改进SSA-BP神经网络的矿井突水水源识别模型:首先进行SSA参数设置,引入Sine混沌映射使麻雀种群均匀分布,然后通过计算适应度值进行麻雀种群的更新,引入随机游走策略扰动当前最优个体,如果满足终止条件,则获得最优BP神经网络权重和阈值,最后基于构建的BP神经网络,输出识别结果。研究结果表明:①改进的SSA-BP模型在训练集上的识别准确率达95.6%,在测试集上的识别准确率达100%。②改进的SSA-BP神经网络模型与BP神经网络模型、SSA-BP神经网络模型对比结果:BP神经网络模型误判率为5/18,SSA-BP神经网络模型的误判率为2/18,改进的SSA-BP神经网络模型误判率为0,迭代10次后趋于稳定,且与设定的目标误差相差最小,初始适应度值最优,识别结果可信度高。③将阳城煤矿5组矿井水水样数据作为输入层数据输入到训练好的模型中,矿井水水样的主要来源为奥灰含水层、三灰含水层和山西组含水层,模型识别结果与水化学特征分析的结论相互印证,实现了精准区分。

基于K-means聚类和BP神经网络的电梯能耗实时监测方法

针对现有方法在对电梯能耗进行监测时,存在监测精度低、用时长、监测结果不理想的问题,该文提出一种基于K-means聚类算法和BP神经网络相结合的电梯能耗实时监测方法 .在经过清洗的能耗数据中提取影响建筑能耗实时监测的主要因素特征值,利用相似系数法进行相似度计算,获取相似系数.对相似电梯能耗数据进行小波分解获取高低频序列,分别采用LSSVM-GSA检测方法和均方加权处理方法对低频和高频部分进行处理,将两个结果进行重构,得到最终的实时监测结果 .仿真实验结果表明:所提方法能够获取高精度、低耗时、高稳定性的监测结果 .

BP神经网络算法在求解数学建模最优化问题中的应用

为了解决目标函数较为复杂、无法用初等函数表示的最优化问题,文章采用了结合BP神经网络与遗传算法的方法进行求解。求解过程分为两个模块:第一,利用BP神经网络算法确定目标函数的解析式;第二,利用遗传算法寻找目标函数的最优解。为验证该方法的可行性,文章对单变量和多变量两种情况进行了验证。

基于改进PSO-BP神经网络的热采管柱应力预测

稠油热采过程中,油套管柱由于在温度、地层等多重载荷作用下发生塑性形变进而导致断裂或失效。文中根据热采管柱高温服役工况,引入异步变化学习因子和自适应权重建立输入参数为注汽温度、井深、非均匀系数和水泥环温度,输出参数为套管应力的改进PSO-BP模型。文中以N80热采套管为例,选取260、280、300、320、340℃5种温度工况下有限元模拟结果作为训练数据,对比BP模型、GA-BP模型、MEA-BP模型、PSO-BP模型和改进PSO-BP模型在300℃工况温度下井深200、300、400、500、600、700 m处套管应力的预测值和试验值、有限元计算值。结果表明:改进PSO-BP模型预测的应力与试验值最接近,最大和最小误差分别为2.69%和0.06%。最后从训练数据、预测误差、计算时间等方面对建立的改进PSO-BP模型进行了评价,为热采管柱服役过程中的强度安全分析提供智能高效的模型。