Deep learning is currently the mainstream method for ceramic defect detection, and it requires a large number of defect samples to train the network. However, collecting these defect samples is very time-consuming and...
详细信息
Deep learning is currently the mainstream method for ceramic defect detection, and it requires a large number of defect samples to train the network. However, collecting these defect samples is very time-consuming and deep learning suffers from few-shot learning problems. In this study, a StyleGAN3-based data augmentation method for ceramic defect detection was proposed which can generate ceramic defect samples and thus reduce the data collection work. Experiments show that our method uses less training time, has a more stable training process, and can improve the accuracy of the detection network.
Effective identification of faults or abnormal conditions can help operators make corrective decisions and plan equipment maintenance. Sequence matching and cluster analysis are important methods to distinguish differ...
Effective identification of faults or abnormal conditions can help operators make corrective decisions and plan equipment maintenance. Sequence matching and cluster analysis are important methods to distinguish different faults. Most existing sequence matching methods mainly focus on alarm event sequences, which reflect the amplitude change characteristics of process data. However, due to the complexity of the equipment and the coupling between variables, alarm event sequences caused by different faults may still assemble each other in a certain extent, which makes it difficult to distinguish faults based on alarms only. To solve this problem, this paper proposes a sequence similarity analysis method combining both alarm and trend events. A qualitative trend representation method is proposed to extract trend changes as trend events. A feature event fusion method is proposed to generate a hybrid sequence to distinguish different fault sequences. The proposed method is evaluated based on data generated by the Tennessee Eastman process model.
For hybrid energy storage systems in DC microgrids, a droop control consisting of virtual capacitors and virtual resistors can decompose power into high-frequency components and low-frequency components, then assign t...
For hybrid energy storage systems in DC microgrids, a droop control consisting of virtual capacitors and virtual resistors can decompose power into high-frequency components and low-frequency components, then assign them to batteries and supercapacitors to respond respectively. However, aiming at the service life of the energy storage system, this paper considers the characteristics and key parameters of the hybrid energy storage structure and proposes an adaptive drooping comprehensive control strategy considering the SOC of the energy storage unit given the shortcomings of power distribution within the current hybrid energy storage. According to the self-regulation capacity of each energy storage unit, it is sorted and constrained, and protected by using SOC, which ensures the economy and safety of the system while ensuring power distribution. The traditional droop control and adaptive droop control are simulated to verify the effectiveness of the proposed control strategy.
This paper explores the finite-time synchronization of a class of discrete-time nonlinear singularly perturbed complex networks using a dynamic event-triggered mechanism (DETM). The DETM is designed to optimize packet...
This paper explores the finite-time synchronization of a class of discrete-time nonlinear singularly perturbed complex networks using a dynamic event-triggered mechanism (DETM). The DETM is designed to optimize packet transmission, aiming to conserve network resources. By constructing a Lyapunov function considering singularly perturbed parameters (SPPs) and DETM information, a sufficient condition for the dynamics of synchronization error system to be finite-time stable is given. The parameters of the synchronization controller can be determined by solving a set of matrix inequalities. The effectiveness of the proposed controller is demonstrated through a numerical example.
In this paper, a template matching and trend feature analysis-based data pre-processing method for seismic wave detection is proposed with two stages. In the first stage, it involves extracting the rock physical param...
In this paper, a template matching and trend feature analysis-based data pre-processing method for seismic wave detection is proposed with two stages. In the first stage, it involves extracting the rock physical parameters from seismic wave detection results using OCR (Optical Character Recognition) method, and extracting the original rock physical parameters from the raw rock property table using keyword matching method. Using the rock physical parameters as a template, a template matching approach is employed to eliminate abnormal values from the original rock physical parameters. In the next stage, a technique is proposed to extract trend features of rock physical parameters for conducting advanced geological forecasting, which considered the expertise of experts in interpreting seismic wave detection data. Finally, the effectiveness of the proposed method is verified by the compared simulation results.
With the rapid development of sequencing technology, researchers can obtain a large number of single cell RNA sequencing (scRNA-seq) data which is useful for analysis of cell fate decision and growth process at indivi...
详细信息
Due to the significant time lag and under-regulation, predicting the blast furnace gas generation and formulating its scheduling strategy is complex. This paper proposes a blast furnace gas generation prediction metho...
详细信息
Due to the significant time lag and under-regulation, predicting the blast furnace gas generation and formulating its scheduling strategy is complex. This paper proposes a blast furnace gas generation prediction method based on time series feature extraction and designs a blast furnace gas scheduling strategy based on the prediction results. Firstly, Pearson correlation analysis is used to identify the parameters that have a significant correlation with the blast furnace gas generation, and the selected parameters are decomposed into several intrinsic mode components with different frequency characteristics using the complete ensemble empirical mode decomposition; Then, the principal component analysis method is used to extract the principal components of several intrinsic modal components, and these principal components are employed as the inputs of long short-term memory neural network to predict the blast furnace gas generation; Finally, according to the prediction results designs the scheduling strategy of blast furnace gas. The experiment and contrast experiments are carried out with the industrial field data, and experimental results illustrate that the proposed method is correct and effective.
This article investigates a fractional-order coupled Hindmarsh-Rose neural networks ***,the existence and stability of an equilibrium point in the system are ***,the periodic bifurcation behavior of the system on a tw...
This article investigates a fractional-order coupled Hindmarsh-Rose neural networks ***,the existence and stability of an equilibrium point in the system are ***,the periodic bifurcation behavior of the system on a twoparameter plane is studied,and numerical simulations show the existence of both non-chaotic and chaotic plus periodic bifurcation behavior on the two-parameter ***,a feedback controller was designed to stabilize the bifurcation point of the delayed system and increase the stable range of the system.
Ground Penetrating Radar (GPR) features are vital for accurately predicting the formation environment in the tunnel engineering. In this paper, a novel intelligent method for extracting the multi-domain (time domain, ...
Ground Penetrating Radar (GPR) features are vital for accurately predicting the formation environment in the tunnel engineering. In this paper, a novel intelligent method for extracting the multi-domain (time domain, time frequency domain, and spatial domain) features of ground penetrating radar data is proposed. Firstly, the GPR exploration data is read to obtain the relative amplitude matrix. Secondly, multi-domain features are obtained using the following methods. The relative amplitude matrix of GPR is averaged by rows to obtain the time domain feature called the average relative amplitude(ARA). The S-transform is used to extract the time frequency domain feature called the average maximum weight frequency(AMWF) of the electromagnetic waves. And the events of the GPR images are highlighted by using Gaussian filtering and edge detection, and the spatial domain feature called the maximum event length(M EL) is obtained through contour detection. Finally, the three extracted multi-domain features are stored in the GPR feature database. Compared simulation results verify the effectiveness of the proposed method.
A high-reliability constant current to constant voltage power supply system has the advantages of small volume of switching power supply, high power density, high efficiency was proposed. This paper use two controller...
A high-reliability constant current to constant voltage power supply system has the advantages of small volume of switching power supply, high power density, high efficiency was proposed. This paper use two controllers to control the shunt regulator(SR) circuit and single-end flyback converter part, and separate the two parts for small signal modeling and give the parameters to stabilize the closed loop. The state space average modeling idea was used to solve the state equations for the modes of the converters in a switching cycle. In order to ensure the stability of cascade system, this paper collaborative optimization of hardware filter parameters and the appropriate PI parameter design. The experimentals verify the correctness of our theory, and the system has good stability under closed-loop conditions.
暂无评论