TIME SERIES NEURAL NETWORK MODEL FOR HYDROLOGIC FORECASTING

Time series analysis plays an important role in hydrologic forecasting,while the key to this analysis is to establish a proper model.This paper presents a time series neural network model with back propagation procedure for hydrologic forecasting.Free from the disadvantages of previous models,the model can be parallel to operate information flexibly and rapidly.It excels in the ability of nonlinear mapping and can learn and adjust by itself,which gives the model a possibility to describe the complex nonlinear hydrologic process.By using directly a training process based on a set of previous data, the model can forecast the time series of stream flow.Moreover,two practical examples were used to test the performance of the time series neural network model.Results confirm that the model is efficient and feasible.

Research on Plant Species Identification Based on Improved Convolutional Neural Network

Plant species recognition is an important research area in image recognition in recent years.However,the existing plant species recognition methods have low recognition accuracy and do not meet professional requirements in terms of recognition accuracy.Therefore,ShuffleNetV2 was improved by combining the current hot concern mechanism,convolution kernel size adjustment,convolution tailoring,and CSP technology to improve the accuracy and reduce the amount of computation in this study.Six convolutional neural network models with sufficient trainable parameters were designed for differentiation learning.The SGD algorithm is used to optimize the training process to avoid overfitting or falling into the local optimum.In this paper,a conventional plant image dataset TJAU10 collected by cell phones in a natural context was constructed,containing 3000 images of 10 plant species on the campus of Tianjin Agricultural University.Finally,the improved model is compared with the baseline version of the model,which achieves better results in terms of improving accuracy and reducing the computational effort.The recognition accuracy tested on the TJAU10 dataset reaches up to 98.3%,and the recognition precision reaches up to 93.6%,which is 5.1%better than the original model and reduces the computational effort by about 31%compared with the original model.In addition,the experimental results were evaluated using metrics such as the confusion matrix,which can meet the requirements of professionals for the accurate identification of plant species.

A Hybrid Neural Network and Box-Jenkins Models for Time Series Forecasting

Time series forecasting plays a significant role in numerous applications,including but not limited to,industrial planning,water consumption,medical domains,exchange rates and consumer price index.The main problem is insufficient forecasting accuracy.The present study proposes a hybrid forecastingmethods to address this need.The proposed method includes three models.The first model is based on the autoregressive integrated moving average(ARIMA)statistical model;the second model is a back propagation neural network(BPNN)with adaptive slope and momentum parameters;and the thirdmodel is a hybridization between ARIMA and BPNN(ARIMA/BPNN)and artificial neural networks and ARIMA(ARIMA/ANN)to gain the benefits of linear and nonlinearmodeling.The forecasting models proposed in this study are used to predict the indices of the consumer price index(CPI),and predict the expected number of cancer patients in the Ibb Province in Yemen.Statistical standard measures used to evaluate the proposed method include(i)mean square error,(ii)mean absolute error,(iii)root mean square error,and(iv)mean absolute percentage error.Based on the computational results,the improvement rate of forecasting the CPI dataset was 5%,71%,and 4%for ARIMA/BPNN model,ARIMA/ANN model,and BPNN model respectively;while the result for cancer patients’dataset was 7%,200%,and 19%for ARIMA/BPNNmodel,ARIMA/ANN model,and BPNNmodel respectively.Therefore,it is obvious that the proposed method reduced the randomness degree,and the alterations affected the time series with data non-linearity.The ARIMA/ANN model outperformed each of its components when it was applied separately in terms of increasing the accuracy of forecasting and decreasing the overall errors of forecasting.

Proton exchange membrane fuel cells modeling based on artificial neural networks

To understand the complexity of the mathematical models of a proton exchange membrane fuel cell (PEMFC) and their shortage of practical PEMFC control, the PEMFC complex mechanism and the existing PEMFC models are analyzed, and artificial neural networks based PEMFC modeling is advanced. The structure, algorithm, training and simulation of PEMFC modeling based on improved BP networks are given out in detail. The computer simulation and conducted experiment verify that this model is fast and accurate, and can be used as a suitable operational model for PEMFC real-time control.

Application of Grey Model and Neural Network in Financial Revenue Forecast

There are many influencing factors of fiscal revenue,and traditional forecasting methods cannot handle the feature dimensions well,which leads to serious over-fitting of the forecast results and unable to make a good estimate of the true future trend.The grey neural network model fused with Lasso regression is a comprehensive prediction model that combines the grey prediction model and the BP neural network model after dimensionality reduction using Lasso.It can reduce the dimensionality of the original data,make separate predictions for each explanatory variable,and then use neural networks to make multivariate predictions,thereby making up for the shortcomings of traditional methods of insufficient prediction accuracy.In this paper,we took the financial revenue data of China’s Hunan Province from 2005 to 2019 as the object of analysis.Firstly,we used Lasso regression to reduce the dimensionality of the data.Because the grey prediction model has the excellent predictive performance for small data volumes,then we chose the grey prediction model to obtain the predicted values of all explanatory variables in 2020,2021 by using the data of 2005–2019.Finally,considering that fiscal revenue is affected by many factors,we applied the BP neural network,which has a good effect on multiple inputs,to make the final forecast of fiscal revenue.The experimental results show that the combined model has a good effect in financial revenue forecasting.

Implementation of Neural Networks in Time Series to Generate a Portfolio of Investment in Cryptocurrencies

The present work aims to implement two types of neural networks and an analysis of a multivariate time series model of VAR type to predict the price of cryptocurrencies like Bitcoin,Dash,Ethereum,Litecoin,and Ripple.This subject has been popular in recent years due to the rapid price fluctuations and the immense amount of money involved in the cryptocurrencies market.Several technologies have been developed around cryptocurrencies,with Blockchain rising as the most popular.Blockchain has been implementing other information technology projects which have helped to open a wide variety of job positions in some industries.A“New Economy”is emerging and it is important to study its basis in order to establish the pillars that help us to understand its behavior and be ready for a new era.

A New Searching Strategy for the Lost Plane Based on RBF Neural Network Model and Global Optimization Model

In this paper, we construct two models for the searching task for a lost plane. Model 1 determines the searching area. We predict the trajectory of floats generated after the disintegration of the plane by using RBF neural network model, and then determine the searching area according to the trajectory. With the pass of time, the searching area will also be constantly moving along the trajectory. Model 2 develops a maritime search plan to achieve the purpose of completing the search in the shortest time. We optimize the searching time and transform the problem into the 0-1 knapsack problem. Solving this problem by improved genetic algorithm, we can get the shortest searching time and the best choice for the search power.

Predicting formation lithology from log data by using a neural network

In order to increase drilling speed in deep complicated formations in Kela-2 gas field, Tarim Basin, Xinjiang, west China, it is important to predict the formation lithology for drilling bit optimization. Based on the conventional back propagation （BP） model, an improved BP model was proposed, with main modifications of back propagation of error, self-adapting algorithm, and activation function, also a prediction program was developed. The improved BP model was successfully applied to predicting the lithology of formations to be drilled in the Kela-2 gas field.

Chinese News Text Classification Based on Convolutional Neural Network

With the explosive growth of Internet text information,the task of text classification is more important.As a part of text classification,Chinese news text classification also plays an important role.In public security work,public opinion news classification is an important topic.Effective and accurate classification of public opinion news is a necessary prerequisite for relevant departments to grasp the situation of public opinion and control the trend of public opinion in time.This paper introduces a combinedconvolutional neural network text classification model based on word2vec and improved TF-IDF:firstly,the word vector is trained through word2vec model,then the weight of each word is calculated by using the improved TFIDF algorithm based on class frequency variance,and the word vector and weight are combined to construct the text vector representation.Finally,the combined-convolutional neural network is used to train and test the Thucnews data set.The results show that the classification effect of this model is better than the traditional Text-RNN model,the traditional Text-CNN model and word2vec-CNN model.The test accuracy is 97.56%,the accuracy rate is 97%,the recall rate is 97%,and the F1-score is 97%.

3D laser scanning strategy based on cascaded deep neural network

A 3D laser scanning strategy based on cascaded deep neural network is proposed for the scanning system converted from 2D Lidar with a pitching motion device. The strategy is aimed at moving target detection and monitoring. Combining the device characteristics, the strategy first proposes a cascaded deep neural network, which inputs 2D point cloud, color image and pitching angle. The outputs are target distance and speed classification. And the cross-entropy loss function of network is modified by using focal loss and uniform distribution to improve the recognition accuracy. Then a pitching range and speed model are proposed to determine pitching motion parameters. Finally, the adaptive scanning is realized by integral separate speed PID. The experimental results show that the accuracies of the improved network target detection box, distance and speed classification are 90.17%, 96.87% and 96.97%, respectively. The average speed error of the improved PID is 0.4239°/s, and the average strategy execution time is 0.1521 s.The range and speed model can effectively reduce the collection of useless information and the deformation of the target point cloud. Conclusively, the experimental of overall scanning strategy show that it can improve target point cloud integrity and density while ensuring the capture of target.

Research on Hydrological Time Series Prediction Based on Combined Model

Water level prediction of river runoff is an important part of hydrological forecasting.The change of water level not only has the trend and seasonal characteristics,but also contains the noise factors.And the water level prediction ability of a single model is limited.Since the traditional ARIMA(Autoregressive Integrated Moving Average)model is not accurate enough to predict nonlinear time series,and the WNN(Wavelet Neural Network)model requires a large training set,we proposed a new combined neural network prediction model which combines the WNN model with the ARIMA model on the basis of wavelet decomposition.The combined model fit the wavelet transform sequences whose frequency are high with the WNN,and the scale transform sequence which has low frequency is fitted by the ARIMA model,and then the prediction results of the above are reconstructed by wavelet transform.The daily average water level data of the Liuhe hydrological station in the Chu River Basin of Nanjing are used to forecast the average water level of one day ahead.The combined model is compared with other single models with MATLAB,and the experimental results show that the accuracy of the combined model is improved by 7%compared with the traditional wavelet network under the appropriate wavelet decomposition function and the combined model parameters.

基于CNN-GRU-ISSA-XGBoost的短期光伏功率预测

针对光伏功率随机性及波动性大,单一预测模型往往难以准确分析历史数据波动规律,从而导致预测精度不高的问题,提出一种基于卷积神经网络-门控循环单元(CNN-GRU)和改进麻雀搜索算法(ISSA)优化的极限梯度提升(XGBoost)模型的短期光伏功率预测组合模型.首先去除历史数据中的异常值并对其进行归一化处理,利用主成分分析法(PCA)进行特征选取,以便更好地识别影响光伏功率的关键因素.然后采用CNN网络提取数据的空间特征,再经过GRU网络提取时间特征,针对XGBoost模型手动配置参数困难、随机性大的问题,利用ISSA对模型超参数寻优.最后对两种方法预测的结果用误差倒数法减小误差的同时对权重进行更新,得到新的预测值,从而完成对光伏功率的预测.实验结果表明,所提出的CNN-GRU-ISSA-XGBoost组合模型具有更强的适应性和更高的精度.

基于在线监测时间序列数据的水质预测模型研究进展

当前地表水突发性污染事件频发,已造成严重的环境和社会影响,对环境监管部门应急处置能力建设提出了新要求和新挑战。地表水水质在线监测数据具有高频率和高时效等特点,系统论述了基于在线监测时间序列数据的水质预测模型的研究现状和进展,包括数据软测量、预处理方法和水质预测模型等,分析了不同水质预测模型在应用过程中存在的问题,并对未来研究方向进行了展望,以期为水质预测预警和环境监管提供技术支持和方法参考。

基于卷积神经网络的入侵昆虫识别研究

现有昆虫相关识别算法识别种类较少,缺少针对数量庞大种类众多的入侵昆虫分类识别算法,难以为入侵昆虫综合系统的识别功能提供稳定高效的技术支持。该研究对31类入侵昆虫图像进行数据采集,并对图像数据进行处理与数据集划分,基于四种卷积神经网络模型DenseNet121、MobileNetV3、ResNet101和ShuffleNet对其进行训练测试分析讨论。结果表明,在入侵昆虫综合识别系统识别功能后台算法应用上,MobileNetV3表现出更好的综合性能。根据MobileNetV3模型现有缺陷和模型特性,对MobileNetV3模型指定瓶颈层的注意力机制和激活函数进行改进,改进后模型的准确率为92.8%,单张测试集图像的平均识别时间0.012 s,相较于原MobileNetV3模型分别提高0.5%、缩短15.2%,可以很好满足多昆虫识别分类需求。

基于低风速功率修正和损失函数改进的超短期风电功率预测

风电功率具有较强的波动性和随机性。为进一步提升风电功率的预测精度,提出一种基于低风速功率修正和损失函数改进的超短期风电功率预测模型。该模型采用卷积神经网络、自注意力机制和双向门控循环单元捕获风电功率序列的长期时序依赖关系。为了解决低风速下待风状态神经网络难以精确拟合的问题,模型通过预测风速并结合当前时段的风电功率对低风速段的预测功率进行修正。针对参数训练的稳定性问题,模型通过改进预测策略和共享权重,引入一种多元非线性的损失函数来提取序列间的关联性。结果表明,所提模型在多项误差指标中均优于对比模型,能够有效提升超短期风电功率的预测效果。

汽车冷却风扇噪声主动控制仿真及声品质评价

以汽车冷却风扇噪声为研究对象,基于采集的噪声生成具有不同音调特性的模拟噪声,利用改进的音调度模型计算音调度,使用等级评分法对生成的模拟噪声进行主观评价实验。根据音调特性十分明显的旋转噪声频率与风扇转速密切相关的特点,提出以风扇转速信号来构造次级声源参考信号进行主动噪声控制。最后建立并联型灰色神经网络模型并检验降噪效果。结果表明,冷却风扇噪声的音调特性是影响感知烦恼的重要因素之一,通过有源噪声控制能有效降低窄带噪声峰值从而降低音调度。引入并联型灰色神经网络进行预测的平均相对误差为2.86%,降噪后烦恼度有较好的改善。

病毒入侵下无线网络覆盖漏洞感知模型设计

为了及时发现病毒入侵风险,保证无线网络安全,设计一种基于改进遗传神经网络的无线网络覆盖漏洞感知模型。考虑多径衰落效应,构建无线网络信号传播模型,计算信号传输路径损耗,采用地理信息系统获取可视化无线网络覆盖图;融合贝叶斯理论和属性攻击图,通过后验概率更新静态贝叶斯攻击图属性状态,实现覆盖漏洞动态风险评估;使用D-S证据理论融合多源数据,运用蚁群算法实施数据寻优,以过往与现阶段网络安全状态为基础,建立无线网络覆盖漏洞感知模型,凭借改进遗传神经网络优化模型感知准确度。仿真结果表明,所提方法在不同节点密度与感知半径下均具备良好的漏洞感知精度及效率,可广泛应用于现实场景。

基于动态神经网络NARX时间序列的双排桩基坑变形预测

针对目前基于含基本假设或经验公式的传统土力学计算方法,不能有效地反映具有多因素交叉性以及时空性的基坑变形规律,而监测数据时间序列能够真实地表现基坑土体变形的演变,以南宁市亭洪路72号河南水厂住宅小区危旧房改造项目双排桩基坑工程为依托,考虑开挖深度和土体暴露时间这2个因素对监测时间序列的影响,提出一种带有外部输入的非线性自回归(NARX)动态神经网络时间序列模型,多方位预测关键断面重要测点的竖向位移和水平位移。结果表明:预测值和实际监测数据的变化趋势具有较好的一致性,且竖向位移预测值与实际监测值的预测残差小于1.0 mm,水平位移预测残差小于0.3 mm。该模型预测效果良好,同时验证了此模型应用于双排桩基坑变形动态分析的可行性。

基于改进域对抗网络的新能源基地风光时序功率曲线生成方法

准确刻画风光时序功率曲线对于加快推动新能源大规模并网、指导联合发电系统规划运行具有重要意义。针对我国沙漠、戈壁、荒漠等地区新建大型风电光伏发电基地无历史功率数据可利用的现状,该文提出基于改进域对抗网络(improved domain adversarial neural networks,IDANN)的新能源基地风光时序功率曲线生成方法。以历史气象和功率数据充足的新能源场站作为源域,仅有气象数据的新建基地作为目标域。将源域上学习的输入气象信息到输出风光功率的非线性映射知识迁移到目标域,并添加最大均值差异(maximum mean discrepancy,MMD)作为度量域间特征分布相似性的损失函数以降低目标域泛化误差。最后采用实际风光场站算例验证所提方法的有效性,并进一步表明该方法的实用价值和意义。

基于BP-ANN与RBF-ANN的钢筋与混凝土黏结强度预测模型研究

为研究神经网络对钢筋与混凝土黏结强度的预测能力以及神经网络的输出性能,基于大量的试验数据,提出一种基于改进神经网络的变形钢筋与混凝土黏结强度预测模型,对混凝土结构的研究与实际工程应用均有着重要的意义。收集290组黏结锚固试验数据,引入基于反向传播人工神经网络(BP-ANN)与径向基函数神经网络(RBF-ANN)算法,揭示混凝土强度、保护层厚度、钢筋直径、锚固长度及配箍率对变形钢筋与混凝土黏结性能的影响规律,建立基于改进神经网络算法的钢筋与混凝土黏结强度预测模型。对比分析不同数据预处理方法和训练神经元个数对建议模型预测结果的影响,评估各经典模型与建议模型的预测精度和离散性,提出临界锚固长度计算公式。结果表明:BP-ANN预测值与试验值比值的均值、标准差及变异系数分别为1.009、0.188、0.86,其预测精度略高于RBF-ANN;建议模型能够更准确、更稳定地预测钢筋与混凝土的黏结强度,该方法为解决钢筋与混凝土黏结问题提供了新思路。