Filter Model Research Articles

Delineation of Shoreline and Associated Land Use/Land Cover Changes along the Coast of Chattogram, Bangladesh Based on Remote Sensing and GIS Techniques

This study aimed at delineating the shoreline by using the Digital Shoreline Analysis System (DSAS 5.0) tool and detected the changes of land use/land cover (LULC) by Google Earth Engine (GEE) platform. The shoreline is divided into two zones, whereas Zone I covered 87.12[Formula: see text]km and Zone II possessed 168.05[Formula: see text]km. According to End Point Rate (EPR), the mean shoreline change rate of Zone I is 3.55[Formula: see text]m/year and Zone II is [Formula: see text]6.84[Formula: see text]m/year. Likewise, based on Linear Regression Rate (LRR), the mean shoreline change rate of Zone I is 5.46[Formula: see text]m/year and Zone II is [Formula: see text]4.71[Formula: see text]m/year, respectively. Apart from that, the Net Shoreline Movement (NSM) recorded in Zone I is 109.42[Formula: see text]m as well as Zone II is [Formula: see text]213.25[Formula: see text]m, which also revealed how much the shoreline has changed during the last 32 years. This study also used the Kalman filter model to forecast the shoreline positions for 20 years. The most destructive signal is that more than 70% of the coastline is vulnerable due to erosion, whereas 6% is highly vulnerable. By contrast, the results of LULC changes demonstrated the increasing trend of water bodies, built up, and agricultural land while vegetation along with bare land is reduced continuously. The overall accuracy is recorded above 88%, and the kappa co-efficient is found above 0.87 for all three years. The outcome of this study will provide fruitful insight into coastal land use management and adaptation measures against the ongoing along with future threats of shoreline changes to coastal ecosystems and livelihoods.

Flow Dynamics through a High Swelling Nanofiber Membrane Processed at Different Relative Humidities: A Study on a FexOy/Polyvinyl Alcohol Composite.

Addressing the global problem of polluted water requires sustainable, efficient, and scalable remediation solutions, such as electrospun polyvinyl alcohol (PVA) membranes incorporating specific nanoadsorbents. The retention of contaminants depends on membrane swelling, morphology, and the adsorbent within the nanofiber. This study investigated the effect of relative humidity (RH) within the electrospinning chamber on the morphology of the resulting mats and how this affected the flow dynamics depending on whether or not the permeating liquid induced swelling in the membranes. An insolubilized PVA membrane was used as a hydrophilic filter model and a PVA membrane filled with iron oxide nanoparticles (IONPs) as a composite model (PVA + IONPs). The presence of IONPs increases the nanofiber diameter, which decreases when prepared under intermediate RH (IRH). Consequently, the nanofiber configuration, which is critical for filtration tortuosity, is influenced by RH. The initial swelling results in over 60% greater water flux through PVA + IONPs compared to PVA at an equivalent RH. This characterization helps to optimize membrane applications, highlighting that PVA + IONPs exhibit lower permeability values at IRH, indicating improved contaminant retention capabilities.

Enhancing Temporal Knowledge Graph Representation with Curriculum Learning

Temporal knowledge graph representation approaches encounter significant challenges in handling the complex dynamic relations among entities, relations, and time. These challenges include the high difficulty of training and poor generalization performance, particularly with large datasets. To address these issues, this paper introduces curriculum learning strategies from machine learning, aiming to improve learning efficiency through effective curriculum planning. The proposed framework constructs a high-dimensional filtering model based on graph-based high-order receptive fields and employs a scoring model that uses a curriculum temperature strategy to evaluate the difficulty of temporal knowledge graph data quadruples at each stage. By progressively expanding the receptive field and dynamically adjusting the difficulty of learning samples, the model can better understand and capture multi-level information within the graph structure, thereby improving its generalization capabilities. Additionally, a temperature factor is introduced during model training to optimize parameter gradients, alongside a gradually increasing training strategy to reduce training difficulty. Experiments on the benchmark datasets ICEWS14 and ICEWS05-15 demonstrate that this framework not only significantly enhances model performance on these datasets but also substantially reduces training convergence time.

A new hybrid fixed and adaptive gains control algorithm to reduce power losses on LLCL filter-based renewable energy conversion systems with systematic parametrization using Grey Wolf optimizer

Currently, there is a transition in the energy matrix around the world, where traditional sources of energy generation are continually being replaced by energy generation systems based on renewable sources to mitigate the climate crisis. In this bias, this work presents the mathematical modeling of an LLCL filter, used to connect power generation systems based on renewable energy sources to the electrical grid, and presents a novel hybrid fixed-and adaptive gains control strategy for current injection into the grid using this system. The developed hybrid controller is composed of a proportional–integral controller and a direct robust adaptive controller. The first term of the controller guarantees the reference current, while the second term of the controller is used for disturbance rejection. Furthermore, a systematic procedure for the controller’s parametrization based on Grey Wolf Optimizer is also provided. The control of the current injected into the grid is carried out considering the LLCL filter without passive damping resistors in the filter structure to avoid power losses due to the passive filter elements. Additionally, the LLCL filter model considers minimal parasitic resistances to evaluate the controller’s performance and optimize it for the application of interest, aiming to maximize the system performance by ensuring a short transient regime due to the fast closed-loop system response. Simulation results indicate high performance of this optimized control strategy with small tracking error even considering grid impedance variations.

Graph neural collaborative filtering with medical content-aware pre-training for treatment pattern recommendation

Recently, considering the advancement of information technology in healthcare, electronic medical records (EMRs) have become the repository of patients’ treatment processes in hospitals, including the patient’s treatment pattern (standard treatment process), the patient’s medical history, the patient’s admission diagnosis, etc. In particular, EMRs-based treatment recommendation systems have become critical for optimizing clinical decision-making. EMRs contain complex relationships between patients and treatment patterns. Recent studies have shown that graph neural collaborative filtering can effectively capture the complex relationships in EMRs. However, none of the existing methods take into account the impact of medical content such as the patient’s admission diagnosis, and medical history on treatment recommendations. In this work, we propose a graph neural collaborative filtering model with medical content-aware pre-training (CAPRec) for learning initial embeddings with medical content to improve recommendation performance. First the model constructs a patient-treatment pattern interaction graph from EMRs data. Then we attempt to use the medical content for pre-training learning and transfer the learned embeddings to a graph neural collaborative filtering model. Finally, the learned initial embedding can support the downstream task of graph collaborative filtering. Extensive experiments on real world datasets have consistently demonstrated the effectiveness of the medical content-aware training framework in improving treatment recommendations.

Online numerical simulation method based on adaptive unscented Kalman filter for shear wall structures

Hybrid test method is an effective approach to evaluate the seismic performance of engineering structures. Particularly, online numerical simulation method (ONSM) based on unscented Kalman filter (UKF) model updating significantly improves the accuracy of the hybrid test results. However, the sensitivity of UKF to the initial values of the state vector, state covariance, and observation noise covariance may lead to the distortion of hybrid test results. To address these issues, one novel ONSM based on adaptive UKF (ONSM-AUKF) is proposed in this paper. The AUKF algorithm incorporates an adaptive variance module to maintain equilibrium. This not only reduces the sensitivity of UKF to the initial parameter settings, prevents filter divergence, but also enhances the accuracy and stability of estimation. The proposed ONSM-AUKF accurately identifies the constitutive model parameters by utilizing measured data from specimens through the AUKF algorithm, which improves the accuracy and the stability of parameter estimation. The simulation results indicate that AUKF algorithm has the capability to decrease the sensitivity of initial parameter settings, thus enhancing the stability and robustness of parameter estimation in comparison to the UKF algorithm. The experimental results validate the feasibility and effectiveness of the proposed ONSM-AUKF, as well as the advantages of the ONSM-AUKF over the ONSM-UKF. The results indicate that the proposed ONSM-AUKF may have broad application prospect in evaluating the seismic performance of various of complex engineering structures.

Synthesis of hardware and software for adaptive discrete high-order control systems based on a model of the desired transient process with a guaranteed degree of stability

In this work, analytical expressions for the connection between optimization criteria andsettings parameters of the adaptive PI-PID controller with variable parameters of the object andadaptive filter have been removed. A method for synthesizing a model of discrete controllersbased on given dynamic characteristics has been developed. A method has been developed forsynthesizing a model of a discrete PID controller based on given dynamic characteristics basedon the criterion of the transient process of a guaranteed level of stability. 3 figures, 0 tables, 32formulas, 19 bibliographic sources..Analytical expressions have been removed to link the optimization criteria andadjustment parameters of the adaptive PI-PID controller with the variable parameters of theobject and adaptive filter.A method for synthesizing a model of discrete controllers from specified dynamiccharacteristics has been developed. The average warehouse of output pulses at the output of anonlinear channel, which is formed at the output of such a regulator and is seen as an inertialinduced uninterrupted part, changes according to the law, which is designated as a reversetransmission the function of the lanyard of the new molded filter model at the collar of the non-linear element.A method has been developed for synthesizing a model of a discrete PID controller fromspecified dynamic characteristics based on the criterion of the required transient process of aguaranteed stability level.The modeling showed that while the parameters of the object and the adaptive filter willchange during the ASC operation, then, obviously, the optimal parameters for adjusting theadaptive PID controller will immediately change.

Combined wavelet transforms and neural network feed-forward model for ECG peak detection and classification

We have focused on development of a combined approach for electrocardiogram (ECG) signal filtering and various ECG peak detection. The filtering model is based on the combination of wavelet transform and neural network where after computing the wavelet coefficients the neural network feed-forward model is used to update the weights. The filtered signal is processed through the convolution layers and bidirectional long short-term memory (Bi-LSTM) architecture to perform the ECG peak detection. Further, we apply a combined feature extraction strategy where wavelet transform and morphological feature are extracted to classify the ECG beats as classify 5 different types of heartbeats, including premature ventricular contraction (PVC), left bundle branch block (LBBB), right bundle branch block (RBBB), PACE, and atrial premature contraction (APC) to examine the heart condition. The feature extraction phase uses wavelet transform, morphological features and high-order statistics to generate the robust features. The obtained feature vector is processed through the principal component analysis (PCA) module to reduce the dimension of feature vector. These features are trained by using support vector machine (SVM) and k-nearest neighbor (KNN) supervised model. The proposed approach is tested on publicly available MIT-BIH dataset where performance analysis shows that the proposed approach obtained average precision, sensitivity and error as 99.98%, 99.96%, and 0.101 which outperforms the existing filtering and peak detection schemes.

Comparing different segments in shut-in pressure signals: new insights into frequency range and energy distribution

Water hammer diagnostics is an important fracturing diagnosis technique to evaluate fracture locations and other downhole events in fracturing. The evaluation results are obtained by analyzing shut-in water hammer pressure signal. The field-sampled water hammer signal is often disturbed by noise interference. Noise interference exists in various pumping stages during water hammer diagnostics, with significantly different frequency range and energy distribution. Clarifying the differences in frequency range and energy distribution between effective water hammer signals and noise is the basis of setting specific filtering parameters, including filtering frequency range and energy thresholds. Filtering specifically could separate the effective signal and noise, which is the key to ensuring the accuracy of water hammer diagnosis. As an emerging technique, there is a lack of research on the frequency range and energy distribution of effective signals in water hammer diagnostics. In this paper, the frequency range and energy distribution characteristics of field-sampled water hammer signals were clarified quantitatively and qualitatively for the first time by a newly proposed comprehensive water hammer segmentation-energy analysis method. The water hammer signals were preprocessed and divided into three segments, including pre-shut-in, water hammer oscillation, and leak-off segment. Then, the three segments were analyzed by energy analysis and correlation analysis. The results indicated that, one aspect, the frequency range of water hammer oscillation spans from 0 to 0.65 Hz, considered as effective water hammer signal. The pre-shut-in and leak-off segment ranges from 0 to 0.35 Hz and 0 to 0.2 Hz respectively. Meanwhile, odd harmonics were manifested in water hammer oscillation segment, with the harmonic frequencies ranging approximately from 0.07 to 0.75 Hz. Whereas integer harmonics were observed in pre-shut-in segment, ranging from 6 to 40 Hz. The other aspect, the energy distribution of water hammer signals was analyzed in different frequency ranges. In 0 − 1 Hz, an exponential decay was observed in all three segments. In 1 − 100 Hz, a periodical energy distribution was observed in pre-shut-in segment, an exponential decay was observed in water hammer oscillation, and an even energy distribution was observed in leak-off segment. In 100 − 500 Hz, an even energy distribution was observed in those three segments, yet the highest magnitude was noted in leak-off segment. In this study, the effective frequency range and energy distribution characteristics of the field-sampled water hammer signals in different segments were sufficiently elucidated quantitatively and qualitatively for the first time, laying the groundwork for optimizing the filtering parameters of the field filtering models and advancing the accuracy of identifying downhole event locations.

An innovative three-dimensional computational fluid dynamics-iterative ensemble Kalman filter model for the prediction of heavy gas leakage and dispersion in enclosed workplaces: Case study for hydrogen sulfide leakage

An innovative three-dimensional computational fluid dynamics-iterative ensemble Kalman filter model for the prediction of heavy gas leakage and dispersion in enclosed workplaces: Case study for hydrogen sulfide leakage

Heterogeneous orientation tuning in the primary visual cortex of mice diverges from Gabor-like receptive fields in primates

A key feature of neurons in the primary visual cortex (V1) of primates is their orientation selectivity. Recent studies using deep neural network models showed that the most exciting input (MEI) for mouse V1 neurons exhibit complex spatial structures that predict non-uniform orientation selectivity across the receptive field (RF), in contrast to the classical Gabor filter model. Using local patches of drifting gratings, we identified heterogeneous orientation tuning in mouse V1 that varied up to 90° across sub-regions of the RF. This heterogeneity correlated with deviations from optimal Gabor filters and was consistent across cortical layers and recording modalities (calcium vs. spikes). In contrast, model-synthesized MEIs for macaque V1 neurons were predominantly Gabor like, consistent with previous studies. These findings suggest that complex spatial feature selectivity emerges earlier in the visual pathway in mice than in primates. This may provide a faster, though less general, method of extracting task-relevant information.

The Application of Multi-View Collaborative Filtering Model for Education Platform

The Application of Multi-View Collaborative Filtering Model for Education Platform

Statistical Learning of Incidental Perceptual Regularities Induces Sensory Conditioned Cortical Responses.

Statistical learning of sensory patterns can lead to predictive neural processes enhancing stimulus perception and enabling fast deviancy detection. Predictive processes have been extensively demonstrated when environmental statistical regularities are relevant to task execution. Preliminary evidence indicates that statistical learning can even occur independently of task relevance and top-down attention, although the temporal profile and neural mechanisms underlying sensory predictions and error signals induced by statistical learning of incidental sensory regularities remain unclear. In our study, we adopted an implicit sensory conditioning paradigm that elicited the generation of specific perceptual priors in relation to task-irrelevant audio-visual associations, while recording Electroencephalography (EEG). Our results showed that learning task-irrelevant associations between audio-visual stimuli resulted in anticipatory neural responses to predictive auditory stimuli conveying anticipatory signals of expected visual stimulus presence or absence. Moreover, we observed specific modulation of cortical responses to probabilistic visual stimulus presentation or omission. Pattern similarity analysis indicated that predictive auditory stimuli tended to resemble the response to expected visual stimulus presence or absence. Remarkably, Hierarchical Gaussian filter modeling estimating dynamic changes of prediction error signals in relation to differential probabilistic occurrences of audio-visual stimuli further demonstrated instantiation of predictive neural signals by showing distinct neural processing of prediction error in relation to violation of expected visual stimulus presence or absence. Overall, our findings indicated that statistical learning of non-salient and task-irrelevant perceptual regularities could induce the generation of neural priors at the time of predictive stimulus presentation, possibly conveying sensory-specific information about the predicted consecutive stimulus.

Mid-infrared methane standoff sensor using a frequency channel attention based convolutional neural network filter

Highly sensitive standoff methane detection is vital for atmospheric science, environmental protection, and production safety. We develop a mid-infrared methane standoff sensor using a cooperative target based on tunable diode laser absorption spectroscopy (TDLAS). To enhance sensor sensitivity, we propose a one-dimensional frequency channel attention-based convolutional neural network (1D-FCACNN) filter, which can effectively denoise the methane absorbance signals and its second harmonic signals. The filter model is adequately trained on a simulated spectrum dataset, which we constructed based on data augmentation. In comparison with reported filtering algorithms, the proposed filter shows the best performance in both measuring modes and evaluation metrics. Real-time measurements show that the measuring accuracy and limit of detection (LOD) of the proposed sensor reach a minimum of 30.40 ppb and 5.04 ppb over a 10-meter optical range, a significant improvement compared to previous reports of methane standoff sensors. The proposed methane standoff sensor proves the feasibility of enhancing the performance of TDLAS gas sensors with the attention mechanism, bringing a new option for high-sensitivity measurements of methane and other atmospheric trace gases.

Efficient trust region filter modeling strategies for computationally expensive black-box optimization

The study and application of contemporary optimization techniques considerably enhance the efficiency of chemical research and manufacturing. With the dynamic progression of modern manufacturing technologies, the emergence of numerous black-box models characterized by inaccessible mathematical formulations and high evaluation costs poses new challenges to traditional optimization methods, leading to difficulty in programming and solving. Hence, based on the trust region filter (TRF) method, we define a new framework to elevate optimization efficiency in this study for chemical systems involving computationally expensive black-box functions. Sampling size per iteration is reduced, and sampling efficiency is improved by incorporating the known data beyond the trust region to assist in constructing reduced models, which is achieved by the application of the Gaussian process. Through comparison and validation of benchmark tests and case studies, this approach demonstrates that using the Gaussian process as the reduced model can lower the number of calls to black-box functions by more than half compared to common linear and quadratic models, and convergence to first-order critical points can still be guaranteed.

Evaluation of building deformation monitoring using stereophotogrammetry method and Kalman filter model

Evaluation of building deformation monitoring using stereophotogrammetry method and Kalman filter model

Color image hybrid noise filtering algorithm based on deep convolution neural network

To solve the problems of the classical color image hybrid noise filtering method, a deep convolutional neural network improved by evolutionary strategy and jump connection is proposed and applied to the filtering noise reduction of color images. First, the color information of the image is described quantitatively by digital means. The common method is to build color space model. According to the characteristics of color and the needs of human vision, mathematical algorithms are used to convert images into machine recognizable data. The distance between pixels is measured according to the difference of pixels in the color image determined above. Then, the probability density function and noise probability density function of Gaussian noise are calculated to determine the hybrid noise feature points of color image. The filtering algorithm structure designed this time is as follows: A color image hybrid noise filter is used to map the noise points in the mapped image to the feature space, and linear regression is performed on the noise point data. Relaxation variables are introduced in the network to improve the denoising ability. The experimental results show that the Peak Signal to Noise Ratio and structural similarity index values of the filtering algorithm designed in this study are higher than the two methods in the literature. The color image hybrid noise filtering model designed in this study has good filtering performance, good image cleanliness, and high filtering efficiency.

Numerical analysis of a time discretized method for nonlinear filtering problem with Lévy process observations

In this paper, we consider a nonlinear filtering model with observations driven by correlated Wiener processes and point processes. We first derive a Zakai equation whose solution is an unnormalized probability density function of the filter solution. Then, we apply a splitting-up technique to decompose the Zakai equation into three stochastic differential equations, based on which we construct a splitting-up approximate solution and prove its half-order convergence. Furthermore, we apply a finite difference method to construct a time semi-discrete approximate solution to the splitting-up system and prove its half-order convergence to the exact solution of the Zakai equation. Finally, we present some numerical experiments to demonstrate the theoretical analysis.

Optimized design of flexible sensors for snow and ice sports characterization and intelligent recognition of training technique movements

Skiing performance is affected by a variety of external factors, and real-time monitoring of skiers' skiing status is very important for improving athletes' skiing technique. There are problems in the training monitoring of snow sports athletes in a wide range of venues, such as difficulty in accurately obtaining the parameters of sports status, lack of quantitative assessment of the training effect, and inability to provide timely feedback. And traditional sensors cannot adapt to cold, humid and other weather conditions, which affects their stability. Traditional sensors can only provide limited motion data and cannot comprehensively evaluate the technical level of skiers. At the same time, it is difficult to provide sufficiently accurate data in complex snowy environments. To address this problem, this study proposes a flexible sensor that can contact with the human body and sense athletes' body postures and movements in real time. By combining the flexible sensor with ice and snow sports equipment, intelligent recognition of athletes' movements can be realized. By monitoring and analyzing athletes' movements in real time, problems can be detected in time and targeted adjustments and training can be carried out. The findings of the experiment demonstrate that the accuracy of the time synchronization model, the time-domain high-pass filtering model, the zero-speed correction model, and the improved zero-speed correction model are, respectively, 97, 94, 92, and 90 when the size of the training set is 800. At the distance of 2000 m, The CF-TENG, H-TENG and CS-TENG response times are 18 ms, 20 ms and 28 ms. The results show that the designed flexible sensors have good performance and can be used in the daily training of cross-country skiing and alpine skiing professional sports teams.

Marine management and monitoring system based on AHP cost-benefit analysis with integrated evaluation and genetic algorithms for recurrent searches

Marine management and monitoring systems need to develop a safety procedure for submersibles that predicts changes in the position of the submersible over time. This would effectively prevent the submersible from losing communication with the host and would allow it to react to possible deficiencies. In this paper, a random walk model is used to predict the position of the submersible at each moment on the seafloor. The probability of the highest point on each plane is 0.87-0.92. The error in the position of the submersible caused by the uncertainty on the seafloor is corrected by the Kalman filter model. In this paper, we use AHP cost-benefit analysis to comprehensively evaluate and select the best additional search equipment. In this paper, the genetic algorithm based on cyclic search is used to determine the optimal initial deployment point and search pattern. After a large number of iterations, the shortest search path length of 817.273 is obtained, and the shortest search path is taken as the optimal path for search and rescue. In this paper, the model is extended using particle swarm algorithm and environment modelling. The predictive model of the submersible is combined with the environmental model of different sea areas. Therefore, the actual motion model of the submersible is approximated. After simulation, it was found that the accuracy of the submersible probability increased by 23.72%.