An improved convolutional neural network with load range discretization for probabilistic load forecasting

[Departement_IRSTEA]Eaux [TR1_IRSTEA]ARCEAU

Nonparametric probabilistic load forecasting based on quantile combination in electrical power systems

Probabilistic load forecasting for buildings considering weather forecasting uncertainty and uncertain peak load

GEFCom2014 probabilistic electric load forecasting: An integrated solution with forecast combination and residual simulation

Objects and processes classification is a common experimental problem. Its solution, first of all, is needed in automatic diagnosis systems, for example, to determine the equipment operation state through the diagnostic signal or to identify abnormalities with medical images. With the development of convolutional neural networks, new prospects for solving such problems have risen up. However, the classification accuracy that can be achieved on these networks is not sufficient enough for all diagnosis issues. It is subject to, for example, timely diagnosis of the onset of transient phenomena. At the same time, another type of neural network, Kohonen self-organizing maps, has a conceptual property for training on unclassified set of classes, that is, giving the opportunity to solve such an issue. Therefore, the accuracy enhancement of the classification on the basis of Kohonen networks implementation in the architecture of convolutional networks is a relevant objective and has practical significance. The article analyzes the means of improving the accuracy of convolutional neural networks and arising problems solution. The ways of increasing the proportion of correct clustering on Kohonen networks due to the ‘growth’ of its grid shape in determining new classes in the learning process are also given. It is this property that makes it possible to recognize transient phenomena. It is determined that the existing solutions of the combination of Kohonen networks and convolutional networks are aimed at improving the efficiency of only self-organizing maps, with the purpose to improve the accuracy of classification by convolutional networks, it became necessary to develop a new architecture. The paper provides a description of this issue. Since the initial information of the Kohonen networks is the weighting matrix values of the grid shape neurons, it was necessary to associate it with the representation of the images in order to process the diagnostic images. The paper proposes the concept of a built-in associative array block based on Kohonen networks. According to the proposed method, a software implementation of a hybrid neural network is developed. The formulation and results of computational experiments are presented. The efficiency of the proposed method is experimentally proved. Keywords: convolutional neural networks, CNN, Kohonen self-organizing maps, SOM, signal classification, image classification.

ABSTRACTIn order to maintain the safe operation of various types of equipment, the health status of main components should be monitored in real-time, and the demand for intelligent fault diagnosis algorithm has increased sharply. However, the traditional intelligent diagnosis algorithm is based on the manual method for signal feature extraction, which has high requirements for expert experience and poor generality. A convolutional neural network, with big data as its engine, is the most effective pattern classification algorithm at present. In this paper, the convolutional neural network is applied to time-domain vibration signal fault diagnosis, taking the bearing as an example, and an intelligent diagnosis method of bearing based on convolution neural network is proposed. The proposed method does not need manual feature extraction, and can automatically complete feature extraction and automatic fault recognition. The convolutional neural network has three convolutional layers. We use data enhancement techniques for the input raw signal and convert the one-dimensional original time-domain vibration signal into a two-dimensional signal. The model shows good results on CWRU dataset and the recognition accuracy of the algorithm in the CWRU bearing database is more than 96%.

Identification of crops is an important topic in the agricultural domain. Hyperspectral remote sensing data are very useful for crop feature extraction and classification. Remote sensing data is an unstructured data and Convolutional Neural Network (CNN) can work on unstructured data efficiently. This paper presents an evaluation of CNN for crop classification using the Indian Pines standard dataset obtained from the AVIRIS sensor and the study area dataset obtained from the EO-1hyperion sensor. Optimized CNN has been tuned by training the model on different parameters. It has been compared with two classification algorithms: Deep Neural Network (DNN) and Convolutional Autoencoder. According to the test results, the proposed optimized CNN model provided better results as compared to the other two methods. CNN has given 97 ± 0.58% overall accuracy for the Indian Pines standard dataset and 78 ± 2.43% for our study area dataset.

With the expansion of scale of the grid-connected wind power, wind power forecasting plays an increasing important role in ensuring the security and steady operation and instructing the dispatch of power systems. In consideration of the randomness and intermittency of wind power, the probabilistic forecasting is required in quantifying the uncertainty of wind power. This study proposes a probabilistic wind power prediction method that combines variational modal decomposition (VMD), singular spectrum analysis (SSA), quantile regression (QR), convolutional neural network (CNN) and bidirectional gated neural network (BGRU). Firstly, a combination decomposition method VMDS combining VMD and SSA is proposed to decompose wind power sequence to reduce the complexity of the sequence. Next, a feature extractor based on CNN and BGRU (CBG) is used to extract complex dynamic features of NWP data and high-frequency components. Then, the QR is performed by the BGRU based on the high-order features to obtain the predicted values for different quantiles. Finally, the kernel density estimation (KDE) is employed to estimate the probability density curve of wind power. The proposed model can achieve reliable probabilistic prediction while achieving accurate deterministic prediction. According to comparisons with related prediction models, the effectiveness of the proposed method is verified with the example test using datasets from the wind farm in China.

Although many machine learning methods have been widely used to predict PM2.5 concentrations, these single or hybrid methods still have some shortcomings. This study integrated the advantages of convolutional neural network (CNN) feature extraction and the regression ability of random forest (RF) to propose a novel CNN-RF ensemble framework for PM2.5 concentration modeling. The observational data from 13 monitoring stations in Kaohsiung in 2021 were selected for model training and testing. First, CNN was implemented to extract key meteorological and pollution data. Subsequently, the RF algorithm was employed to train the model with five input factors, namely the extracted features from the CNN and spatiotemporal factors, including the day of the year, the hour of the day, latitude, and longitude. Independent observations from two stations were used to evaluate the models. The findings demonstrated that the proposed CNN-RF model had better modeling capability compared with the independent CNN and RF models: the average improvements in root mean square error (RMSE) and mean absolute error (MAE) ranged from 8.10% to 11.11%, respectively. In addition, the proposed CNN-RF hybrid model has fewer excess residuals at thresholds of 10 μg/m3, 20 μg/m3, and 30 μg/m3. The results revealed that the proposed CNN-RF ensemble framework is a stable, reliable, and accurate method that can generate superior results compared with the single CNN and RF methods. The proposed method could be a valuable reference for readers and may inspire researchers to develop even more effective methods for air pollution modeling. This research has important implications for air pollution research, data analysis, model estimation, and machine learning.

Self-mixing interferometry (SMI) is superior to other laser interferometry methods due to its simplicity and compactness. However, SMI signals are often complex and difficult to process due to interference in the form of variations in the effective reflectivity of the target, noisy signals, complex signal shapes, and other dependencies. Deep neural networks have been a very popular area of research in computer artificial intelligence in recent years, allowing more implicit features to be uncovered than traditional shallow machine learning. It has been shown that convolutional and back-propagation neural networks can be used for SMI signal processing. There are also studies that have used machine learning genetic algorithms for absolute distance measurement. Based on the above, this study used convolutional neural networks to form a deep neural network for absolute distance measurement based on SMI technology. We first trained the deep convolutional neural network at different feedback strengths. The results of the Convolutional Neural Network (CNN) model showed a coefficient of determination of 0.9987. which is consistent with the required model. The trained network was then used to estimate absolute distances with and without the addition of noise. The comparison proves that the proposed method is noise-proof and has high adaptability for measurements under different conditions.

Bus load forecasting plays an important role in terms of the safety of power grid dispatching and operation and the accuracy of online analysis and decision-making. In order to improve accuracy of bus load forecasting, this paper proposes an ultra-short-term load forecasting method based on multi-type data and hybrid model. Under the background of power big data, the multi-type data such as historical bus load data, date information and weather information are firstly taken as the input features of the prediction model, which is established based on the fusion of back propagation (BP-ANN) neural network and convolutional neural network(CNN). BP-ANN is used to extract the feature vectors of numeric type and category type data, which is concatenated with feature vectors extracted by CNN from image type data. The combined features are fed to a fully connected multilayer Neural Network to predict the bus load. In this paper, the historical data of active power load at the high voltage side of a 220kV substation in a certain area of China and the corresponding weather information are used for example analysis. The analysis of experimental results shows that the proposed method can make full use of the characteristics of multi-source data to conduct ultra-short-term bus load prediction, and it has higher load prediction accuracy compared to the single model prediction of BP-ANN and CNN.

Power quality gains more and more attentions because disturbances in power quality may damage equipment security, power availability and system reliability in power system. Detection and classification of the power quality disturbances is the first step before taking measures to lessen their harmful effects. Common methods to classify power quality disturbances includes signal processing methods, machine learning methods and deep learning methods. Signal processing methods are good at feature extraction, while machine learning methods and deep learning methods are expert in multi-classification tasks. Via combing their respective advantages, this paper proposes a combined method based on variational mode decomposition and convolutional neural networks, which needs a small quantity of samples but achieves high classification precision. The proposed method is proved to be a qualified and competitive scheme for the detection and classification of power quality disturbances.

<p>One key strategy to fight climate change worldwide is to invest in renewable energy sources (RES) and increase their integration into the power system. In recent years, we observed how extreme weather conditions, together with growing penetration levels of RES, are increasingly affecting the power system operation and planning, as well as electricity markets. The inherent uncertainty of such events and the associated uncertainty in the power generation from RES can no longer be ignored by the energy industry. In other words, current deterministic methods have reached their limit due to the inherent inability to model and convey forecast uncertainties.</p><p>Probabilistic information and forecasts have been shown to improve decision-making in many weather-related processes. By dealing with uncertainties, the end-user takes responsibility, but also gets the possibility to harvest the benefits of knowing and being able to calculate what is at stake. Last but not least, knowing the uncertainty of an event in advance opens the possibility to act upon such uncertainty rather than acting on the event itself and thereby mitigating costly side effects or being able to secure safety.</p><p>In 2020, the IEA Wind Task 36 “Wind Energy Forecasting” has for this reason started an initiative “Probabilistic Forecasting Games and Experiments” in collaboration with the Max-Planck Institute for Human Development. The main goal of this initiative is to empirically investigate the psychological barriers to the adoption of probabilistic forecasts and to enable stakeholders to understand and explore their benefit and use.  With the initiative, the IEA Wind Task 36 wants to establish interdisciplinary teams to promote testing and playing with forecast games and experiments to give end-users a “feel” of where the hidden possibilities are to improve decisions and developers a platform to:</p><ul><li>Discuss</li> <li>Educate</li> <li>Inspire</li> </ul><p>the energy and meteorology community for the development, deployment and communication of uncertainties of weather and energy forecasts to end-users for better decision making.</p><p>The task leaders have started to setup a platform with a list of forecasting games and experiments  developed by the task, in cooperation or by cooperating institutions, researchers or companies as well as invite others outside the tasks community to share links or data to games and experiments.</p><p>The initiative will be presented and the first experience with the task’s own games and experiments briefly discussed. The many open questions and considerations when looking forward towards the establishment of training and educational tools for probabilistic forecasts will be formulated and posed to the meteorological and psychological/behaviorism research community to enhance the collaboration and establish a stronger link for this interdiciplinary work. </p>

The difficulty of feature extraction and the small sample size are two challenges in the field of mechanical fault diagnosis for a long time. Here we propose an intelligent mechanical fault diagnosis method for scenario with small sample datasets. This method can not only diagnose bearing faults but also gear faults, and has strong generalization performance. We use convolutional neural network to realize automatic feature extraction. Through sliding window scanning, one sample set is expanded to three sub-sample sets with different scales to meet the needs of deep learning training. Three convolutional networks are used to extract the features of the subsets respectively to ensure that their useful features are fully extracted. After feature extraction, the feature is reconstructed through feature splicing. Because of the unique advantages of SVM in dealing with small sample sets, we use SVM to classify the reconstructed features. We use the bearing data set collected by Case Western Reserve University in the United States, the bearing fault data set collected by Xi'an Jiaotong University in China, and the gearbox fault data collected by the University of Connecticut in the United States to conduct experiments. The experimental results show that the accuracy of training, validation and testing of the proposed method on the three data sets all reach 100%. This proves that our method can not only tackle the two challenges, but also has high fault diagnosis accuracy and strong generalization performance. It is hoped that our proposed method can contribute to the development of mechanical fault diagnosis.

Convolutional Neural Networks (CNNs) have demonstrated impressive performance in complex machine learning tasks such as classification and regression problems. A reliable neural network structure plays a decisive role in CNN studies. Through comparing and analyzing the structure of neural networks, a model structure for better visualization performance has been discovered, and such a method supports the development of deep learning research. These studies are of particular importance in end-to-end systems for autonomous driving to imitate human driving, where the interpretability of the system is limited. Because of the uncertainty of the ground truth, for the determination of human steering in an image, it is difficult to accurately compare the visualization performance of different CNN models or different visualization methods. For practical applications, however, an objective and quantitative measure for assessing visualization performance is necessary. Therefore, a method to evaluate the visualization performance of CNN models using a driver model instead of human drivers is proposed, to generate a data set which can be used to determine the decisional point (ground truth) in the input image. Then, an exclusive method is also put forth, to quantitatively calculate the relationship between the decisional point (ground truth) and the visualization results produced by CNN models. In this paper, five CNN models as an autonomous steering controller are designed based on PilotNet, and the visualization abilities of each CNN models is compared by three evaluation indicators. By comparing the visualization performance of five different CNN models, it is shown that the proposed method can successfully assess the visualization level of the CNN model.