A multiscale adaptive framework based on convolutional neural network:Application to fluid catalytic cracking product yield prediction

Since chemical processes are highly non-linear and multiscale,it is vital to deeply mine the multiscale coupling relationships embedded in the massive process data for the prediction and anomaly tracing of crucial process parameters and production indicators.While the integrated method of adaptive signal decomposition combined with time series models could effectively predict process variables,it does have limitations in capturing the high-frequency detail of the operation state when applied to complex chemical processes.In light of this,a novel Multiscale Multi-radius Multi-step Convolutional Neural Network(Msrt Net)is proposed for mining spatiotemporal multiscale information.First,the industrial data from the Fluid Catalytic Cracking(FCC)process decomposition using Complete Ensemble Empirical Mode Decomposition with Adaptive Noise(CEEMDAN)extract the multi-energy scale information of the feature subset.Then,convolution kernels with varying stride and padding structures are established to decouple the long-period operation process information encapsulated within the multi-energy scale data.Finally,a reconciliation network is trained to reconstruct the multiscale prediction results and obtain the final output.Msrt Net is initially assessed for its capability to untangle the spatiotemporal multiscale relationships among variables in the Tennessee Eastman Process(TEP).Subsequently,the performance of Msrt Net is evaluated in predicting product yield for a 2.80×10^(6) t/a FCC unit,taking diesel and gasoline yield as examples.In conclusion,Msrt Net can decouple and effectively extract spatiotemporal multiscale information from chemical process data and achieve a approximately reduction of 30%in prediction error compared to other time-series models.Furthermore,its robustness and transferability underscore its promising potential for broader applications.

改进Transformer在产油量预测中的应用研究

产油量预测有利于制定合理的采油策略。本文提出一种包含卷积神经网络、门控循环单元和Transformer的组合模型CNN-GRU-Transformer,可用于产油量预测。该模型应用CNN提取部分深层空间特征,GRU提取产油量数据的时序特征,并根据油井数据的特点,改进了Transformer原有结构。通过改进的Transformer,将提取到的特征与预测相结合。实验的结果表明,CNN-GRU-Transformer模型在预测产油量各项指标中均为最优值,在适应产油量基本趋势方面表现最佳。

Matrix Inequalities for the Fan Product and the Hadamard Product of Matrices

A new inequality on the minimum eigenvalue for the Fan product of nonsingular M-matrices is given. In addition, a new inequality on the spectral radius of the Hadamard product of nonnegative matrices is also obtained. These inequalities can improve considerably some previous results.

Blind reconstruction of convolutional code based on segmented Walsh-Hadamard transform

Walsh-Hadamard transform （WriT） can solve linear error equations on Field F2, and the method can be used to recover the parameters of convolutional code. However, solving the equations with many unknowns needs enormous computer memory which limits the application of WriT. In order to solve this problem, a method based on segmented WriT is proposed in this paper. The coefficient vector of high dimension is reshaped and two vectors of lower dimension are obtained. Then the WriT is operated and the requirement for computer memory is much reduced. The code rate and the constraint length of convolutional code are detected from the Walsh spectrum. And the check vector is recovered from the peak position. The validity of the method is verified by the simulation result, and the performance is proved to be optimal.

Cross-CNN:基于CNN和Transformer混合模型的动画跨帧线稿着色算法

对长序列的动画线稿帧进行着色是计算机视觉中一项具有挑战性的任务.一方面,线稿中包含的信息较为稀疏,需要着色算法对缺失的信息进行推断;另一方面,连续帧之间的色彩需要保持一致,以确保整个视频的视觉质量.现有的着色算法多数只针对单张图片进行着色,这类算法只给出一个开放性的符合合理范围的色彩结果,无法适用于帧序列着色.另一些基于参考帧的着色算法,并没有将2帧之间的关系有机地联系起来,导致着色效果不够出色.在同一镜头序列中,同一对象的特征往往不会发生太大变化,因此,可以设计一个根据给定参考帧,即可给线稿自动着色的模型.为此,本文提出了基于CNN(Convolutional Neural Networks)和Transformer相结合的模型Cross-CNN,该模型能够从参考帧中寻找并匹配颜色,从而保证时间维度上的特征一致性.Cross-CNN模型参考帧和线稿帧在通道维度叠加,输入预训练的Resnet50网络提取局部融合特征,将融合特征图传给Transformer结构进行编码以提取全局特征.在Transformer结构中设计了交叉注意力机制更好地匹配远距离特征.最后使用带有跳层连接的卷积解码器完成着色图片输出.本文在数据集方面从8部电影中截取画面并经过严格筛选,最终制作了一个包含20000对二元组的数据集用于实验研究.Cross-CNN的SSIM(Structural SIMilarity)达到了0.932,高于SOTA算法0.014.本文算法代码链接:https://github.com/silenye/Cross-CNN.

A production prediction method of single well in water flooding oilfield based on integrated temporal convolutional network model

Since the oil production of single well in water flooding reservoir varies greatly and is hard to predict, an oil production prediction method of single well based on temporal convolutional network(TCN) is proposed and verified. This method is started from data processing, the correspondence between water injectors and oil producers is determined according to the influence radius of the water injectors, the influence degree of a water injector on an oil producer in the month concerned is added as a model feature, and a Random Forest(RF) model is built to fill the dynamic data of water flooding. The single well history is divided into 4 stages according to its water cut, that is, low water cut, middle water cut, high water cut and extra-high water cut stages. In each stage, a TCN based prediction model is established, hyperparameters of the model are optimized by the Sparrow Search Algorithm(SSA). Finally, the models of the 4 stages are integrated into one whole-life model of the well for production prediction. The application of this method in Daqing Oilfield, NE China shows that:(1) Compared with conventional data processing methods, the data obtained by this processing method are more close to the actual production, and the data set obtained is more authentic and complete.(2) The TCN model has higher prediction accuracy than other 11 models such as Long Short Term Memory(LSTM).(3) Compared with the conventional full-life-cycle models, the model of integrated stages can significantly reduce the error of production prediction.

Fine-Grained Classification of Product Images Based on Convolutional Neural Networks

With the rapid development of the Internet of things and e-commerce, feature-based image retrieval and classification have become a serious challenge for shoppers searching websites for relevant product information. The last decade has witnessed great interest in research on content-based feature extraction techniques. Moreover, semantic attributes cannot fully express the rich image information. This paper designs and trains a deep convolutional neural network that the convolution kernel size and the order of network connection are based on the high efficiency of the filter capacity and coverage. To solve the problem of long training time and high resource share of deep convolutional neural network, this paper designed a shallow convolutional neural network to achieve the similar classification accuracy. The deep and shallow convolutional neural networks have data pre-processing, feature extraction and softmax classification. To evaluate the classification performance of the network, experiments were conducted using a public database Caltech256 and a homemade product image database containing 15 species of garment and 5 species of shoes on a total of 20,000 color images from shopping websites. Compared with the classification accuracy of combining content-based feature extraction techniques with traditional support vector machine techniques from 76.3% to 86.2%, the deep convolutional neural network obtains an impressive state-of-the-art classification accuracy of 92.1%, and the shallow convolutional neural network reached a classification accuracy of 90.6%. Moreover, the proposed convolutional neural networks can be integrated and implemented in other colour image database.

Identifying Brand Consistency by Product Differentiation Using CNN

This paper presents a new method of using a convolutional neural network(CNN)in machine learning to identify brand consistency by product appearance variation.In Experiment 1,we collected fifty mouse devices from the past thirty-five years from a renowned company to build a dataset consisting of product pictures with pre-defined design features of their appearance and functions.Results show that it is a challenge to distinguish periods for the subtle evolution of themouse devices with such traditionalmethods as time series analysis and principal component analysis(PCA).In Experiment 2,we applied deep learning to predict the extent to which the product appearance variation ofmouse devices of various brands.The investigation collected 6,042 images ofmouse devices and divided theminto the Early Stage and the Late Stage.Results show the highest accuracy of 81.4%with the CNNmodel,and the evaluation score of brand style consistency is 0.36,implying that the brand consistency score converted by the CNN accuracy rate is not always perfect in the real world.The relationship between product appearance variation,brand style consistency,and evaluation score is beneficial for predicting new product styles and future product style roadmaps.In addition,the CNN heat maps highlight the critical areas of design features of different styles,providing alternative clues related to the blurred boundary.The study provides insights into practical problems for designers,manufacturers,and marketers in product design.It not only contributes to the scientific understanding of design development,but also provides industry professionals with practical tools and methods to improve the design process and maintain brand consistency.Designers can use these techniques to find features that influence brand style.Then,capture these features as innovative design elements and maintain core brand values.

A Hybrid Deep Learning Approach for Green Energy Forecasting in Asian Countries

Electricity is essential for keeping power networks balanced between supply and demand,especially since it costs a lot to store.The article talks about different deep learning methods that are used to guess how much green energy different Asian countries will produce.The main goal is to make reliable and accurate predictions that can help with the planning of new power plants to meet rising demand.There is a new deep learning model called the Green-electrical Production Ensemble(GP-Ensemble).It combines three types of neural networks:convolutional neural networks(CNNs),gated recurrent units(GRUs),and feedforward neural networks(FNNs).The model promises to improve prediction accuracy.The 1965–2023 dataset covers green energy generation statistics from ten Asian countries.Due to the rising energy supply-demand mismatch,the primary goal is to develop the best model for predicting future power production.The GP-Ensemble deep learning model outperforms individual models(GRU,FNN,and CNN)and alternative approaches such as fully convolutional networks(FCN)and other ensemble models in mean squared error(MSE),mean absolute error(MAE)and root mean squared error(RMSE)metrics.This study enhances our ability to predict green electricity production over time,with MSE of 0.0631,MAE of 0.1754,and RMSE of 0.2383.It may influence laws and enhance energy management.

LIE-TROTTER FORMULA FOR THE HADAMARD PRODUCT

Suppose that A and B are two positive-definite matrices,then,the limit of(A^p/2B^pA^p/2)1/p as p tends to 0 can be obtained by the well known Lie-Trotter formula.In this article,we generalize the usual product of matrices to the Hadamard product denoted as*which is commutative,and obtain the explicit formula of the limit(A^p*B^p)^1/p as p tends to 0.Furthermore,the existence of the limit of(A^p*B^p)^1/p as p tends to+∞is proved.

Generalized Inversions of Hadamard and Tensor Products for Matrices

We shall give natural generalized solutions of Hadamard and tensor products equations for matrices by the concept of the Tikhonov regularization combined with the theory of reproducing kernels.

从三个角度考察Hadamard不等式

高等代数、数学分析以及解析几何是大学数学专业的三门基础课程,它们在理论基础、问题处理方法等方面存在差异,然而相互之间存在协同作用,相互渗透。Hadamard不等式是高等数学中一项重要的代数不等式。相较于以往单一代数视角的研究,本文旨在从高等代数、数学分析和解析几何这三门数学专业基础课程的不同视角探讨Hadamard不等式的成立,并分析在不同视角下证明该不等式的差异和联系。

A new model of the harmonic control based on Hadamard product

In terms of Hadamard product, a new model is proposed for the control of connection coefficients of the state variables of the systems. The control law to stabilize the systems via the regulations of connection coefficients is obtained via a Hadamard product involved bilinear matrix inequalities. This new control model may be of significant applications in many fields, especially may be of some special sense in the emergency control such as isolation and obstruction control.

A generalized Hadamard transformation from new entangled state

A new entangled state ｜η 0） is proposed by the technique of integral within an ordered product. A generalized Hadamaxd transformation is derived by virtue of η; θ）, which plays a role of Hadamard transformation for （a1 sinθ - a2 cosθ） and （a1 cosθ ＋ a2 sin θ）.

基于矩阵Hadamard乘积运算的空域冲突检测方法

针对现有的以经纬度为网格剖分的空域表征模型,在高纬度地区网格形变较大,且以空域边界坐标判定空域之间是否重合的冲突检测算法存在的计算速度慢的问题,提出以正二十面体球面菱形离散格网大圆弧剖分为基础,用全等菱形离散格网表征空域,结合空域优先级,利用多层级希尔伯特(Hilbert)空间填充曲线对空域进行统一编码。设计了基于矩阵的空域数字化表征方法,利用哈达玛积(Hadamard)乘积运算快速判定多个空域之间的用空属性是否存在冲突。仿真结果表明:该方法具有较高的网格精度,实现秒级冲突检测,与传统冲突检测算法相比,能够达到降低算法运算量,提高运算速度的目的。

A CONJECTURE CONCERNING THE HADAMARD PRODUCT OF INVERSE M-MATRICES

1 IntroductionFor an n×n matrix A which is an inverse M-matrix,M.Neumann in [1]conjecturedthat the Hadamard product A·A is an inverse of an M-matrix.They have checked hisconjecture without failure on Ultrametric matrices and inverse of MMA-matrices,Uni-pathicmatrices and the Willongby inverse M-matrices.Bo-Ying Wang et al.in[2]haveinvestigated Triangular inverse M-matrices which are closed under the Hadamard multipli-cation.Lu Linzheng,Sun Weiwei and Li Wen in[3]presented a more general

The Exact Solutions of Such Coupled Linear Matrix Fractional Differential Equations of Diagonal Unknown Matrices by Using Hadamard Product

In this paper, we present the general exact solutions of such coupled system of matrix fractional differential equations for diagonal unknown matrices in Caputo sense by using vector extraction operators and Hadamard product. Some illustrated examples are also given to show our new approach.

矩阵Hadamard积的最小特征值新下界

依据Gerschgorin定理,给出非奇异M-矩阵Hadamard积的最小特征值新的下界,新结果只与相关矩阵的元素有关,在计算上比现有结果更容易.新估计式改进了现有文献中的一些结果,是现有结果的有益补充.

Taylor-Hadamard乘积的q-级和q-型

通过构造Taylor-Hadamard乘积,讨论了泰勒级数的增长性,获得了此Taylor-Hadamard乘积与原泰勒级数关于q-级和q-型的几个关系定理,并给出相应的例子说明了结果的正确性.

A REMARK ON CERTAIN CONVOLUTION OPERATOR

A certain operator D^(a+p-1) defined by convolutions (or Hadamard products) is introduced. The object of this paper is to give an application of the convolution operator D^(a+p-1) to the differential inequalities.