An improved domain adaption method for roughness prediction of milling surfaces under variable processes

DNA encodes the genetic information of most living beings, except viruses that use RNA. Unlike other types of molecules, DNA is not usually described by its atomic structure being instead usually described by its base-pair sequence, i.e., the textual sequence of its subsidiary molecules known as nucleotides ( adenine (A), cytosine (C), guanine (G), and thymine (T)). The three-dimensional assembling of DNA molecules based on its base-pair sequence has been, for decades, a topic of interest for many research groups all over the world. In this paper, we survey the major methods found in the literature to assemble and visualize DNA molecules from their base-pair sequences. We divided these methods into three categories: predictive methods, adaptive methods, and thermodynamic methods . Predictive methods aim to predict a conformation of the DNA from its base pair sequence, while the goal of adaptive methods is to assemble DNA base-pairs sequences along previously known conformations, as needed in scenarios such as DNA Monte Carlo simulations. Unlike these two geometric methods, thermodynamic methods are energy-based and aim to predict secondary structural motifs of DNA in cases where hydrogen bonds between base pairs might be broken because of temperature changes. We also present the major software tools that implements predictive, adaptive, and thermodynamic methods.

The construction of an underwater sensor network system necessitates the swift and cost-effective acquisition of precise sound velocity profile data, which is crucial for enhancing underwater positioning accuracy and guaranteeing secure underwater wireless communication. This letter introduces an adaptive sound velocity profile prediction method premised on deep reinforcement learning (DRL-ASP). The DRL-ASP algorithm dynamically interacts with the environment and adjusts its behavior to maximize environmental rewards. During this interactive process, it continuously updates the sound velocity profile, leading to improved accuracy in predicting the sound velocity profile. The efficacy of the DRL-ASP algorithm was evaluated through experiments conducted on datasets from the South China Sea, Arctic Ocean, and Southern Ocean. Its performance in terms of sound velocity error, mean square error, fitting accuracy, and relative error was compared against empirical formula sound velocity prediction methods, <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$Q$</tex-math></inline-formula> -learning-based sound velocity prediction methods, and convolutional-neural-network-based sound velocity prediction methods. The results affirmed that the DRL-ASP algorithm exhibits superior prediction accuracy, enhanced search efficiency, and robust stability.

Aiming at the challenges brought by the strong fluctuation and uncertainty of power output of distributed photovoltaic (DPV) system to power grid operation, and the limitations of existing capacity evaluation and prediction methods in complex scenarios, this paper proposes a multi-model adaptive DPV capacity evaluation and prediction method. Firstly, the method constructs a multi-model library including physical mechanism model, time series statistical model and machine learning model to realize complementary advantages under different working conditions. Then, a dynamic weight fusion algorithm based on entropy weight method is adopted to dynamically allocate weights according to the prediction errors of each model in the sliding window, so as to avoid the subjectivity of traditional methods. At the same time, the online updating mechanism driven by sliding window is designed, and the model parameters and fusion weights are continuously optimized through incremental training or periodic retraining to adapt to the dynamic changes of system state and environment. Experimental results show that the proposed method outperforms traditional single models and static weight fusion methods in terms of metrics such as Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and Normalized Mean Absolute Error (NMAE). It can effectively improve the prediction accuracy and robustness under different meteorological conditions. Moreover, it can achieve theoretical capacity assessment, actual performance evaluation, and dynamic capacity assessment with early fault warning for DPV systems, providing technical support for the efficient utilization of DPV and the coordinated control of source-grid-load-storage.

Predicting accurate remaining useful life (RUL) is quite necessary in many industrial cases. Recently, data-driven method has been widely used for RUL prediction because of its promising performance and ease of use. However, one practical challenging issue of the data-driven method is lack of labeled data. One way for dealing with this issue is domain adaptation. With domain adaptation technique, a RUL prediction model is built using labeled data of source domain and then the trained model is used to predict RUL of target domain where labeled data is unavailable. In this paper, we propose our method for RUL prediction through domain adaptation. Another practical issue of RUL prediction is that a single RUL prediction value may not be enough for successful maintenance of target systems. In other words, we need to know how uncertain RUL prediction is. In this paper, we study Monte Carlo (MC) dropout-based uncertainty metric for realizing better RUL prediction result. Through experiment, we first show that our domain adaptation technique shows reasonable performance. We also show that MC dropout-based approach generates the uncertainty metric only meaningful for source domain.

A multi-constrained domain adaptation network for remaining useful life prediction of bearings

Infrared thermography provides abundant spatiotemporal degradation information, facilitating non-contact condition monitoring. Reducing domain shift between simulated and industrial infrared images is significantly desired for leveraging labeled simulated data to tackle practical insufficiency of run-to-failure samples. Recently, adversarial-based domain adaptation (DA) techniques have aroused broad concern in solving domain shifts. However, simultaneously aligning marginal and conditional distributions in cross-domain remaining useful life (RUL) prediction is rarely researched in adversarial-based DA. In this article, an adversarial regressive domain adaptation (ARDA) approach is, thus, put forward to address this challenge. First, a regressive disparity discrepancy is designed to describe the dissimilarity between distributions and derive the generalization bound for cross-domain prognostics. Guided by this bound, the ARDA effectively aligns marginal and conditional distributions by learning indistinguishable features and considering the relationship between samples and prediction tasks. Simulated and experimental infrared degradation image datasets are used to demonstrate the effectiveness and superiority of the proposed approach over existing methods for cross-domain RUL prediction.

Dispersion-based surface wave prediction and suppression methods have obtained more and more attention in recent years. Currently the main problem of this type of method is that the dispersion spectrum accuracy is not high enough leading to inaccurate surface wave prediction, or the dispersion spectrum measurement is precise but the calculation is too time-consuming. We introduce an improved data-derived surface wave prediction and adaptive attenuation method to overcome those problems. With the help of adaptive surface wave data extraction technology, offset-varying dispersive wave reconstruction method, and noise model matching subtraction, we can accurately predict the surface wave noise, and make this procedure effectively apply in the processing of massive data. Through the test of actual data, this method can effectively remove the surface wave, protect the effective signal to the maximum extent, and has a wide application prospect in the area with severe surface wave noise.

An adaptive predictive control for steam-reforming plant which consist of a steam-gas reformer and a waste heat steam-boiler was studied by using MIMO bilinear model. ARMA model was used in the process identification and the adaptive predictive control. The simulation experiments of the process identification were performed by using linear and bilinear models. From the simulation results it was found that the bilinear model represented the dynamic behavior of a steam-reforming plant very well. To verify the performance and effectiveness of the adaptive predictive control method proposed in this study the simulation results of the adaptive predictive control based on bilinear model for steam-reforming plant were compared to those of linear model. The simulation results showed that the adaptive predictive control method based on bilinear model provides better performance than those of linear model.

In this paper, For 4th generation wireless communication systems, we propose a method to predict FH patterns in FH-OFDMA systems. OFDM is recognized as a promising modulation technique. Multi-user allocation in OFDM system can use FH that provides the spectrum-spread techniques. If one can generate more predictable FH sequences, then performance of the system can be easily improved. Current random FH and simple adaptive FH methods, however, are not considering predicting FH sequences. In this paper we show that the sampling of the wireless faded signal is not realized as a certain probability nature. With this regard, the proposed predictive FH allocation method is designed to embed the unknown probability models. Simulation study shows that the predictive FH method is more accurately predict FH sequences than the random or simple adaptive FH methods. We will further improve this proposed method to apply QoS control and MAC function development in OFDMA based wireless physical structures, especially maritime wireless data communications.

Among various methods in protein function prediction, rough set has recently been applied to prediction of protein structural classes. However, this was a blind application on a single but small data set of high homology, which did not consider investigation of various parameters in the rough set. The aim of this paper is therefore to study rough set in the area through comprehensive and consistent analysis and then to present a practical strategy in the rough set-based protein function prediction. To achieve this aim, three different data sets were considered: the first data set for prediction of six main enzyme classes, and other two for prediction of structural classes. Boolean reasoning, Entropy scaling and Equal frequency binning were used for discretization along with two methods for producing reducts and rules, genetic and Johnson's algorithms. It can be seen that the predictive accuracies were poor for the enzyme dataset whereas it performed better at prediction of the protein structural classes. It is also observed that the dataset with low homology produced poor accuracies than the dataset with high homology. Furthermore, various parameters and methods used in the rough set were sensitive to the problems in the area, as well as the data sets of low and high homology and different number of the features. The results appear to indicate that the equal frequency-based approach combined with genetic algorithm yields higher prediction. However, other methods such as Boolean reasoning with the genetic algorithm are also found to be promising. Further investigation will provide a practical strategy that can be used in the rough set-based protein function prediction as well as other areas of Bioinformatics.

Deep rolling is a widely applied mechanical surface and subsurface treatment method. It is typically used after conventional machining to improve the roughness, increase the surface hardness and to induce compressive residual stresses. The main influence parameters on the surface topography are the applied deep rolling pressure, the ball diameter and the feed. In general, low feeds, larger ball diameters and higher pressures result in an even surface finish. However, an exact prediction of the roughness is not possible. Therefore, it is the aim of the presented research to find a generally applicable method for surface roughness prediction after deep rolling for a variety of steel and aluminum materials. It is shown that the surface topography can be predicted by an analytical model with high accuracy.

As the number of vehicles increases, the traffic environment becomes more complicated. It is important to find a routing method for different scenarios in the vehicular ad hoc networks (VANETs). Although there are many routing methods, they rarely consider multiple road traffic states. In this paper, we propose a traffic state prediction method based on Hidden Markov Model (HMM), and then choose different routing methods according to different traffic states. Since we are aware that GPS may cause measurement errors, Kalman Filter is used to estimate the observation, which makes observation more accurate. For different road states, we can make appropriate methods to improve routing performance. When the road is in rush hour, we will use Extended Kalman Filter to predict vehicle information in a short time to reduce the number of broadcasts, which can alleviate channel load. The result show that our method is useful for reducing the number of packets and improving the delivery rate.

The existing adaptive single-pixel imaging methods suffer from a waste of sampling resources. The sampling resources are not used adequately for superior localization of significant coefficients and reconstruction. In this paper, an adaptive single-pixel imaging method via the guided coefficients in the Haar wavelet tree is proposed. The goal is to achieve high quality imaging with less sampling resources. The guided coefficients are selected from the unsampled coefficients by a proposed same-scale prediction method based on the sampled coefficients. These guided coefficients are used to localize the significant coefficients with higher resolution belonging to the sampled coefficients and the significant coefficients belonging to the guided coefficients by a proposed guided prediction method. The significant guided coefficients are then used in the composite reconstruction method to reconstruct the image. Performance analysis shows that the proposed method reduces waste of the sampling resources and localizes more significant coefficients. Simulation results demonstrate that the proposed method improves the imaging quality in terms of peak signal-to-noise ratio up to 29.7 dB for the images containing regular and chaotic textures in the noise-free environment. The sampling rate for the same imaging quality can be reduced up to 56%. Under the noisy condition, the proposed method also achieves better imaging quality at a lower sampling rate.

To achieve adaptive and accurate ultra-short-term power generation forecasting in power systems, this study proposes a novel prediction method combining Sparrow Search Algorithm (SSA) with Long Short-Term Memory (LSTM) networks. The methodology involves the following steps: (1) Collecting historical ultra-short-term power generation data from photovoltaic systems, where outlier detection and data cleaning are performed using horizontal processing methods; (2) Applying Pearson correlation analysis to identify key meteorological factors significantly influencing power output as feature inputs; (3) Developing an Adaptive Sparrow Search Algorithm (ASSA) by dynamically adjusting the quantities of discoverers and followers in traditional SSA; (4) Optimizing LSTM network parameters through ASSA to enhance prediction accuracy. The experimental results demonstrate superior performance with Root Mean Square Error (RMSE) values of 0.075, 0.088, and 0.089 for sunny, cloudy, and variable weather conditions respectively. The corresponding Mean Absolute Percentage Error (MAPE) values are 0.21 MW, 0.52 MW, and 0.13 MW, while Absolute Error (AE) values reach 0.17 MW, 0.46 MW, and 0.18 MW. These findings confirm the method’s effectiveness in achieving precise ultra-short-term power generation forecasting across diverse weather conditions.

In this paper, we will propose a novel pixelwise adaptive prediction method for lossless video compression. The proposed method adequately takes use of the temporal and spatial redundancy in the video sequences. Firstly, the pixel to be encoded is predicted in temporal domain and spatial domain respectively, and then the final prediction values are chosen by an adaptive mode selection. In addition, we introduce a forgetting factor α in the temporal prediction and the adaptive mode selection to improve the prediction accuracy. Experimental results show that our proposed method gains the minimum zero-entropy compared to other prediction methods.