A complete defect detection task aims to achieve the specific class and precise location of each in an image, which makes it still challenging for applying this practice. The is a composite classification location, leading related methods often hard take into account accuracy both. implementation depends on special data set that contains expensive manual annotations. In paper, we proposed novel system based deep learning focused practical industrial application: steel plate inspection. order...

Surface defect detection is a critical task in industrial production process. Nowadays, there are lots of methods based on computer vision and have been successfully applied industry, they also achieved good results. However, achieving full automation surface remains challenge, due to the complexity defect, intraclass. While defects between interclass contain similar parts, large differences appearance defects. To address these issues, this article proposes pyramid feature fusion global...

Breast cancer is the most common female in world, and it poses a huge threat to women's health. There currently promising research concerning its early diagnosis using deep learning methodologies. However, some commonly used Convolutional Neural Network (CNN) their variations, such as AlexNet, VGGNet, GoogleNet so on, are prone overfitting breast classification, due both small-scale pathology image datasets overconfident softmax-cross-entropy loss. To alleviate issue for better...

Steel surface defect recognition is an important part of industrial product detection, which has attracted more and attention in recent years. In the development steel technology, there been a process from manual detection to automatic based on traditional machine learning algorithm, subsequently deep algorithm. this paper, we discuss key hardware systems offer suggestions for related options; second, present literature review algorithms recognition, includes texture features shape as well...

Improving the accuracy of power system load forecasting is important for economic dispatch. However, a sequence highly nonstationary and hence makes accurate difficult. In this paper, method based on wavelet decomposition (WD) second-order gray neural network combined with an augmented Dickey-Fuller (ADF) test proposed to improve forecasting. First, decomposed by WD reduce sequence. Then, ADF adopted as stationary each component after in which results determine best level. Finally, because...

Selection of the kernel function by support vector regression (SVR), for purposes load forecasting, is affected power characteristics. The non-ideal SVR with a has low forecasting accuracy and poor generalization ability. A novel method combining stacking proposed in this paper. Base models are constructed based on SVRs different functions, then multiple base merged to obtain model layer via algorithm. Finally, an connected as meta-model layer. fusion composed This trained k-fold cross...

A new prediction framework is proposed to improve short-term power load forecasting accuracy. The based on particle swarm optimization (PSO)-variational mode decomposition (VMD) combined with a time convolution network (TCN) embedded attention mechanism (Attention). follows two-step process. In the first step, PSO applied optimize VMD method. original electricity sequence decomposed, and fitness function uses sample entropy describe complexity of series. decomposed sub-sequences are relevant...

This paper proposes a new algorithm to integrate image registration into super-resolution (SR). Image SR is process reconstruct high-resolution (HR) by fusing multiple low-resolution (LR) images. A critical step in accurate of the LR images or, other words, effective estimation motion parameters. Conventional algorithms assume either estimated parameters existing methods be error-free or are known priori. assumption, however, impractical many applications, as most still experience various...

Rail surface defect (RSD) inspection is an essential routine maintenance task. Computer vision testing suitable for RSD with its intuitiveness and rapidity. Deep learning techniques, which can extract deep semantic features, have been applied to inspect RSDs in recent years. However, these methods demand thousands of samples. And sample collection requires hard-working costs high. To address the issue, a novel scheme presented limited samples line-level label, regards images as sequence data...

Surface defect segmentation of no-service rail is important for its quality assessment. There are several challenges uneven illumination, complex background, and difficulty sample collection surface defects (NRSDs). In this article, we propose an acquisition scheme with two lamp light color scan line charge-coupled device (CCD) to alleviate illumination. Then, a multiple context information network proposed improve NRSD segmentation. The makes full use based on dense block, pyramid pooling...

An optimized microgrid scheduling model is established considering demand responses, forecast errors, and the effects of uncertainties in different stages. A day-ahead, intraday, multi-time scale economic method based on light robust optimization predictive control (MPC) also proposed. In long-time-scale stage, robustness used to cope with low-frequency components prediction errors uncertainties, mitigates deviations between day-ahead plan actual outcomes under source load uncertainties. At...

Aiming at the problem that power load data are stochastic and it is difficult to obtain accurate forecasting results by a single algorithm, in this paper, combined method for short-term was proposed based on Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN)-sample entropy (SE), BP neural network (BPNN), Transformer model. Firstly, were decomposed into several subsequences obvious complexity differences using CEEMDAN-SE. Then, BPNN model used forecast low high...

Extracting accurate values for relevant unknown parameters of solar cell models is vital and necessary performance analysis a photovoltaic (PV) system. This paper presents an effective application young, yet efficient metaheuristic, named the symbiotic organisms search (SOS) algorithm, parameter extraction models. SOS, inspired by interaction ways employed to improve their overall competitiveness in ecosystem, possesses some noticeable merits such as being free from tuning algorithm-specific...

The automated classification of breast cancer histopathological images is one the important tasks in computer-aided diagnosis systems (CADs). Due to characteristics small inter-class and large intra-class variances images, extracting features for difficult. To address this problem, an improved autoencoder (AE) network using a Siamese framework that can learn effective from CAD was designed. First, inputted image processed at multiple scales Gaussian pyramid obtain multi-scale features....

The capacity configuration of the energy storage system plays a crucial role in enhancing reliability power supply, quality, and renewable utilization microgrids. Based on variational mode decomposition (VMD), optimization model for hybrid (HESS) consisting batteries supercapacitors is established to achieve optimal wind–solar complementary islanded Firstly, based mapping relationship between frequency domain time domain, number K VMD determined principle minimum total aliasing energy. Then,...