Prediction System Research Articles

Droplet volume prediction methods in electrohydrodynamic jet printing based on multi-source data fusion

Electrohydrodynamic jet printing technology has garnered significant interest in additive manufacturing due to its advantages in higher resolution printing and greater viscosity material applicability. Nonetheless, the challenges associated with femtoliter-level droplets and the micrometer-scale printing space have rendered traditional optical observation methods for volume calculation impractical for integration into the printing systems. Obtaining the droplet volume accurately has become a formidable challenge in electrohydrodynamic printing. While existing research utilizes machine learning for the end-to-end prediction of process parameters to droplet volume, traditional prediction methods cannot achieve online prediction due to the complex spray states characteristic of electrohydrodynamic jet printing and pose integration difficulties within printing systems. This paper introduces a multi-source data fusion model suitable for electrohydrodynamic printing droplet volume prediction. The model employs the VGG network to extract features from Taylor cone images in the jetting state and synchronously fuse these with process parameter features. The fused features are then correlated with droplet volume labels through the MLP network for comprehensive model training. The performance of the proposed model has improved by 10% in prediction accuracy compared to single modal data. We integrated the prediction method into an electrohydrodynamic jet printing system and experimented with printing pixel-pit substrates. The results indicate that the prediction accuracy of the volume prediction system is over 92%. The printing efficiency has improved approximately 3 times compared to the traditional method, significantly enhancing overall performance.

Regional significant wave height forecast in the East China Sea based on the Self-Attention ConvLSTM with SWAN model

Accurate prediction of regional significant wave heights is crucial for marine resource development, especially in areas with limited wave observations. This paper proposes a method for predicting regional significant wave heights by integrating the numerical wave model Simulating Wave Nearshore (SWAN) with the spatio-temporal prediction network Self-Attention ConvLSTM. The SWAN model is used to simulate regional significant wave heights in the East China Sea. To improve simulation accuracy, calibration of wind input, whitecapping dissipation, quadruplet wave–wave interactions, and bottom friction parameterization schemes in the model source terms were conducted, resulting in a simulated correlation coefficient of 0.97 with measured significant wave heights. The regional significant wave height training dataset constructed using the SWAN model preserves the intrinsic relationships between waves, wind speed, and wind direction over a recent period. Subsequently, a regional prediction system is built based on the output of the SWAN model, integrating the self-attention mechanism into the spatio-temporal prediction model ConvLSTM to construct SA-ConvLSTM. This model extracts correlated features among regional waves from the training dataset to predict future significant wave heights at different time intervals. The prediction performance of four spatio-temporal forecasting networks was compared to verify the accuracy of the SA-ConvLSTM model. The average errors in predicting regional significant wave heights at 3, 6, and 12 h were 0.07 m, 0.09 m, and 0.09 m, respectively. The R-Square metrics for these three prediction intervals were 0.85, 0.78, and 0.7, respectively.

An effective IoT and deep learning-based patient healthcare monitoring system for chronic diseases prediction and classification

The state of the environment and human behavior today contributes to a wide range of diseases that affect people. Medical professionals often find it challenging to detect disorders by themselves appropriately; it is crucial to recognize and anticipate them early on. This paper aims to detect and predict people with more widespread chronic illnesses. To prevent such diseases from worsening, this research proposes a new deep learning-based technique to predict chronic diseases. Initially, patient data will be collected using Internet of Things (IoT) devices. Then, the missing values from input data are eliminated, and categorical data encoding, outlier detection, and data transformation are performed in the pre-processing stage. After that, the necessary attributes are selected to optimize the performance by eliminating unnecessary features using Binary Grasshopper Whale Optimization Algorithm (BGWOA), which combines the benefits of the Binary Grasshopper Optimization Algorithm (BGOA) and Binary Whale Optimization Algorithm (BWOA) algorithms. Then, the disease can be classified as chronic or not, utilizing a three-layer stacked bidirectional long short-term memory (TLSBLSTM) technique. The performance is evaluated on two chronic disease datasets that are publicly available. It successfully obtained good results by preparing the dataset on heart disease and comparing the findings using the most recent state-of-the-art approaches. According to the experimental findings, the proposed approach performs better in evaluating performance measures than the existing approaches. The observed accuracy of the proposed method is 99.87% and 99.84% for chronic kidney disease dataset and cardiovascular disease dataset, respectively.

A boundary perturbation approach for regional wave ensemble forecast

In this study, we developed and validated two wave ensemble prediction systems (WEPS) to forecast wave conditions along the southeastern coast of Australia. Using the SWAN model (GEN3 ST6), we integrated complex bathymetric features with an unstructured grid and validated model outputs against buoy observations from Sydney, Port Kembla, and Batemans Bay. The two WEPS, SWAN-WW3 and SWAN-Pert, utilize different methodologies: SWAN-WW3 derives boundary conditions from NCEP’s Global Wave Ensemble System, while SWAN-Pert employs Latin Hypercube Sampling for boundary perturbations based on historical data. Our results demonstrate that both systems effectively predict significant wave height (Hs), with SWAN-Pert showing improved forecast accuracy in certain metrics compared to SWAN-WW3. Despite underdispersion in spread-skill diagrams, both WEPS exhibited good agreement with observed data. Additionally, rank histograms revealed that SWAN-Pert is more reliable at shorter lead times. This study highlights the potential of integrating statistical sampling methods and ensemble systems for enhancing regional wave forecasting accuracy.

Comparative Analysis of Supervised Machine Learning Algorithms for Diabetes Prediction

Diabetes Mellitus is a chronic disease with far-reaching consequences, necessitating innovative diagnostic approaches. Current methods have limitations, highlighting the need for advanced techniques. This research leveraged machine learning algorithms to predict diabetes using the Pima Indian Diabetes Dataset. Six algorithms were developed and compared: Logistic Regression, Support Vector Machine, K-Nearest Neighbor, Naive Bayes, Random Forest, and Decision Tree. Standard evaluation metrics assessed their performance. The Random Forest Classifier emerged as the top-performing algorithm, achieving an impressive 91% accuracy and demonstrating exceptional reliability. This study showcases machine learning's potential to transform diabetes diagnosis and management. The developed web application provides a user-friendly interface for predicting diabetes, facilitating timely intervention and improved health outcomes. The findings contribute significantly to the development of intelligent disease prediction systems, promoting early detection and enhanced patient care. By harnessing machine learning, this research pioneers a new frontier in diabetes diagnosis, paving the way for improved patient outcomes and enhanced healthcare services.

Heatwaves in Vietnam: Characteristics and relationship with large‐scale climate drivers

AbstractThis study analyses the spatio‐temporal variability of heatwave characteristics and their association with large‐scale climate drivers across seven climatic sub‐regions in Vietnam, including the Northwest (R1), Northeast (R2), Red River Delta (R3), North Central (R4), South Central (R5), Central Highlands (R6) and the South (R7). The analysis is based on observed daily maximum temperatures from 102 meteorological stations, spanning the period 1980–2020. The obtained results reveal diverse heatwave patterns across the country. Amongst the seven climatic sub‐regions of Vietnam, the R3 and R4 sub‐regions experienced more frequent heatwaves and a higher number of heatwave days, but shorter durations. In contrast, other sub‐regions had fewer heatwave events and heatwave days but experienced longer‐lasting heatwaves. The intensity of heatwave events varies amongst sub‐regions, with the highest value in the R4 sub‐region, and the lowest in R7. Notably, the R1–R5 sub‐regions are affected by heatwaves over larger areas, compared to others. Additionally, the findings confirm that the lagged influence of El Niño–Southern Oscillation (ENSO) is the primary climatic driver of heatwave characteristics in Vietnam. Generally, heatwaves tend to occur more frequently in the years following El Niño events than after La Niña events. This observation provides opportunities for developing a system of seasonal predictions of heatwaves in Vietnam. The impact of ENSO on the number of heatwave events and heatwave days is evident in five out of seven sub‐regions, with less impact in the R2 and R7 sub‐regions. However, it does not significantly affect heatwave intensity.

The Impact of Buckling Restrained Braces in Strengthening Deficient Reinforced Concrete Structures

Seismic strengthening for existing structures is a sustainable solution that is utilized to enhance building safety, reduce damages, and prevent failure in a future earthquake event. The choice of seismic strengthening techniques has to be accurate, efficient, and adjusted to make RC structures stronger in the building sector. Buckling-restrained brace (BRB) system is one of the successful strengthening strategies, that it is possible to utilize in both RC and steel structures. Therefore, this paper explores the possibility of employing buckling restrained braces in existing RC buildings and assesses the impact of different BRB bracing distributions and positions on seismic force resistance. In this work, a five-story RC building was considered, and to upgrade their performance seismic was modeled using four types of BRB systems, consisting of two types of bracing configurations with two arrangements: diagonal in the central bay, diagonal in the corner bays, chevron in the central bay, and chevron in the corner bays. To assess the efficiency of the four proposed BRB systems, firstly, the nonlinear static pushover method was conducted to investigate the lateral strength of structures. Secondly, a parametric study was undertaken using dynamic time history analysis to study various factors such as roof displacement, shear force, and roof acceleration of the original and strengthened models. The numerical study was executed using the Seismostruct software. The results and different performance levels were examined and compared. The obtained results indicate that the BRB and concrete structures can successfully work together to resist the reliability of strengthening RC structures. It was observed that the four prediction systems of the BRB models were excessively effective at upgrading the seismic resistance of the existing structure and provided significantly less damage, especially when using the chevron BRBs with the corner arrangement compared to the other models.

Evaluation of Seasonal Prediction of Extreme Wind Resource Potential over China Based on a Dynamic Prediction System SIDRI-ESS V1.0

Wind resources play a pivotal role in building sustainable energy systems, crucial for mitigating and adapting to climate change. With the increasing frequency of extreme events under global warming, effective prediction of extreme wind resource potential can improve the safety of wind farms and other infrastructure, while optimizing resource allocation and emergency response plans. In this study, we evaluate the seasonal prediction skill for summer extreme wind events over China using a 20-year hindcast dataset generated by a dynamical seamless prediction system designed by Shanghai Investigation, Design and Research Institute Co., Ltd. (Shanghai, China) (SIDRI-ESS V1.0). Firstly, the hindcast effectively simulates the spatial distribution of summer extreme wind speed thresholds, even though it tends to overestimate the thresholds in most regions. Secondly, high prediction skills, measured by temporal correlation coefficient (TCC) and normalized root mean square error (nRMSE), are observed in northeast China, central east China, southeast China, and the Tibetan Plateau (TCC is about 0.5 and the nRMSE is below 0.9 in these regions). The highest skills emerge in southeast China with a maximum TCC greater than 0.7, and effective prediction skill can extend up to a 5-month lead time. Ensemble prediction significantly enhances predictive skill and reduces uncertainty, with 24 ensemble members being sufficient to saturate TCC and 12–16 members for nRMSE in most key regions and lead times. Furthermore, we show that the prediction skill for extreme wind counts is strongly related to the prediction skill for summer mean wind speeds, particularly in southeast China. Overall, SIDRI-ESS V1.0 shows promising performance in predicting extreme winds and has great potential to provide services to the wind industry. It can effectively help to optimize wind farm operating strategies and improve power generation efficiency. However, further improvements are needed, particularly in areas where prediction skills for extreme winds are influenced by smaller-scale weather phenomena and areas with complex underlying surfaces and climate characteristics.

Internet of things and optimized knn based intelligent transportation system for traffic flow prediction in smart cities

The rapid expansion of urban areas and the increasing number of vehicles on the road have resulted in accidents, traffic congestion, economic repercussions, environmental deterioration, and excessive fuel consumption. A dependable traffic management system is necessary to anticipate and regulate urban traffic patterns. Traffic forecast aids in the prevention of traffic issues. Urban traffic predictions often utilise historical and current traffic flow data to forecast road conditions. This article presents a traffic flow prediction system that utilises the Internet of Things (IoT), machine learning, and feature selection. Internet of Things (IoT) devices located on highways or within cars gather sensor data in real-time. The input data set comprises both real-time Internet of Things (IoT) data and historical traffic statistics. The input data is stored in a centralized cloud. The data is subjected to preprocessing in order to eliminate any unwanted interference and identify any exceptional values. The accuracy and root mean square error are contingent upon the process of feature selection. Particle swarm optimization identifies and extracts crucial features from input data. The classification model is constructed using K Nearest Neighbor, Multi layer Perceptron, and Bayes network approaches. The UCI traffic data is used for conducting experiments. The dataset has 47 attributes and 2102 occurrences. The accuracy of traffic flow prediction using PSO KNN is 96 %. The PSO KNN algorithm achieved a Mean Square Error (MSE) of 0.00289 and a Root Mean Square Error (RMSE) of 0.0595.

A Novel Data-Driven Approach with a Long Short-Term Memory Autoencoder Model with a Multihead Self-Attention Deep Learning Model for Wind Turbine Converter Fault Detection

The imminent depletion of oil resources and increasing environmental pollution have driven the use of clean energy, particularly wind energy. However, wind turbines (WTs) face significant challenges, such as critical component failures, which can cause unexpected shutdowns and affect energy production. To address this challenge, we analyzed the Supervisory Control and Data Acquisition (SCADA) data to identify significant differences between the relationship of variables based on data reconstruction errors between actual and predicted values. This study proposes a hybrid short- and long-term memory autoencoder model with multihead self-attention (LSTM-MA-AE) for WT converter fault detection. The proposed model identifies anomalies in the data by comparing the reconstruction errors of the variables involved. However, more is needed. To address this model limitation, we developed a fault prediction system that employs an adaptive threshold with an Exponentially Weighted Moving Average (EWMA) and a fixed threshold. This system analyzes the anomalies of several variables and generates fault warnings in advance time. Thus, we propose an outlier detection method through data preprocessing and unsupervised learning, using SCADA data collected from a wind farm located in complex terrain, including real faults in the converter. The LSTM-MA-AE is shown to be able to predict the converter failure 3.3 months in advance, and with an F1 greater than 90% in the tests performed. The results provide evidence of the potential of the proposed model to improve converter fault diagnosis with SCADA data in complex environments, highlighting its ability to increase the reliability and efficiency of WTs.

Exploring the impact of word prediction assistive features on smartphone keyboards for blind users

Assistive technologies have been developed to enhance blind users’ typing performance, focusing on speed, accuracy, and effort reduction. One such technology is word prediction software, designed to minimize keystrokes required for text input. This study investigates the impact of word prediction on typing performance among blind users using an on-screen QWERTY keyboard. We conducted a comparative study involving eleven blind participants, evaluating both standard QWERTY input and word prediction-assisted typing. Our findings reveal that while word prediction slightly improves typing speed, it does not enhance typing accuracy and increases both physical and temporal workload compared to the default keyboard. We conclude with recommendations for improving word prediction systems, including more efficient editing methods and the integration of voice pitch variations to aid error recognition.

Analyzing the implementation of predictive control systems and application of stored data in non-residential buildings

In non-residential buildings, building energy management systems (BEMS) and the application of data hold significant promise in reducing energy consumption. Nevertheless, BEMS have different levels of complexity, benefit, and limitation. Despite the advanced technologies and improvements in building operation, there is a clear gap in the actual performance of buildings that has been attributed to the adoption of advanced technologies. Consequently, there is an increasing need for researchers and practitioners to study current practices in order to identify and address the challenges that compromise the core objectives of BEMS. For this reason, this paper aims to validate three research questions: (i) to examine the current state of BEMS and its functionalities; (ii) to analyze the type of control used; (iii) and to determine the availability of historical data compiled by BEMS and its application in non-residential buildings. A survey of 676 buildings and interviews with building professionals were conducted. The findings confirmed that most of the buildings applied BEMS with scheduled control. In addition, a lack of digitized data for analysis and predictions was detected. Indeed, only 0.60% of the investigated buildings implemented predictive control. Finally, using hierarchical clustering analysis, responses were grouped to analyze similarities between them. The study findings help to develop targeted actions for implementing predictive control in non-residential buildings.

Real-time identification of epistatic interactions in SARS-CoV-2 from large genome collections

BackgroundThe emergence of the SARS-CoV-2 virus has highlighted the importance of genomic epidemiology in understanding the evolution of pathogens and guiding public health interventions. The Omicron variant in particular has underscored the role of epistasis in the evolution of lineages with both higher infectivity and immune escape, and therefore the necessity to update surveillance pipelines to detect them early on.ResultsIn this study, we apply a method based on mutual information between positions in a multiple sequence alignment, which is capable of scaling up to millions of samples. We show how it can reliably predict known experimentally validated epistatic interactions, even when using as little as 10,000 sequences, which opens the possibility of making it a near real-time prediction system. We test this possibility by modifying the method to account for the sample collection date and apply it retrospectively to multiple sequence alignments for each month between March 2020 and March 2023. We detected a cornerstone epistatic interaction in the Spike protein between codons 498 and 501 as soon as seven samples with a double mutation were present in the dataset, thus demonstrating the method’s sensitivity. We test the ability of the method to make inferences about emerging interactions by testing candidates predicted after March 2023, which we validate experimentally.ConclusionsWe show how known epistatic interaction in SARS-CoV-2 can be detected with high sensitivity, and how emerging ones can be quickly prioritized for experimental validation, an approach that could be implemented downstream of pandemic genome sequencing efforts.

Relational regression: a cognitively-inspired method for prediction system in cognitive IoT

Relational regression: a cognitively-inspired method for prediction system in cognitive IoT

Predictability of the early summer surface air temperature over Western South Asia

Variability of the Surface Air Temperature (SAT) over the Western South Asia (WSA) region leads to frequent heatwaves during the early summer (May-June) season. The present study uses the European Centre for Medium-Range Weather Forecast’s fifth-generation seasonal prediction system, SEAS5, from 1981 to 2022 based on April initial conditions (1-month lead) to assess the SAT predictability during early summer season. The goal is to evaluate the SEAS5’s ability to predict the El Niño-Southern Oscillation (ENSO) related interannual variability and predictability of the SAT over WSA, which is mediated through upper-level (200-hPa) geopotential height anomalies. This teleconnection leads to anomalously warm surface conditions over the region during the negative ENSO phase, as observed in the reanalysis and SEAS5. We evaluate SEAS5 prediction skill against two observations and three reanalyses datasets. The SEAS5 SAT prediction skill is higher with high spatial resolution observations and reanalysis datasets compared to the ones with low-resolution. Overall, SEAS5 shows reasonable skill in predicting SAT and its variability over the WSA region. Moreover, the predictability of SAT during La Niña is comparable to El Niño years over the WSA region.

Research and application of a novel graph convolutional RVFL and evolutionary equilibrium optimizer algorithm considering spatial factors in ultra-short-term solar power prediction

With the increasing proportion of photovoltaic power generation, solar power prediction systems have become more important. Accurate photovoltaic power prediction can assist the grid dispatch department in formulating effective power scheduling plans, improving grid stability and the capacity to accommodate solar energy. To address the challenges in solar power generation forecasting, the paper proposes a new neural network model called Graph Convolutional Random Vector Functional Link (GCRVFL). The output layer of this model includes both nonlinear transformed features from the hidden layer and the original input features, while also considering the spatial characteristics among photovoltaic sites. Furthermore, Singular Spectrum Analysis (SSA) is utilized to decompose the original solar power time series, extracting different component sequences from the photovoltaic power time series. Opposite based learning and cross-mutation are employed to improve the original Equilibrium Optimizer (EO) algorithm, resulting in an Evolutionary Equilibrium Optimizer (EEO) algorithm. The EEO algorithm is employed to optimize the weight threshold of GCRVFL, leading to a novel photovoltaic power prediction model called SSA-EEO-GCRVFL. Through four datasets and eight control experiments, it can be observed that the proposed model achieves the best predictive performance and is capable of fulfilling the task of photovoltaic power prediction.

Comparison of Regression Methods to Predict the First Spike Latency in Response to an External Stimulus in Intracellular Recordings for Cerebellar Cells.

The significance of intracellular recording in neurophysiology is emphasized in this article, with considering the functions of neurons, particularly the role of first spike latency in response to external stimuli. The study employs advanced machine learning techniques to predict first spike latency from whole cell patch recording data. Experiments were conducted on Control (Salin) and Experiment (Harmaline) groups, generating a dataset for developing predictive models. Because the dataset has a limited number of samples, we utilized models that are effective with small datasets. Among different groups of regression models (linear, ensemble, and tree models), the ensemble models, specifically the LGB method, can achieve better performance. The results demonstrate accurate prediction of first spike latency, with an average mean squared error of 0.0002 and mean absolute error of 0.01 in 10-fold cross-validation. The research suggests the potential of machine learning in forecasting the first spike latency, allowing reliable estimation without the need for extensive animal testing. This intelligent predictive system facilitates efficient analysis of first spike latency changes in both healthy and unhealthy brain cells, streamlining experimentation and providing more detailed insights into the captured signals.

Online prediction of joint mechanical properties of FSW thick AA2219-T8 based on multi-source information fusion using 1DCNN

High strength AA2219-T8, Al-Cu alloy of 18 mm plate is welding by friction stir welding (FSW) technology and used for heavy-lift launch vehicles tank. During FSW thick AA2219-T8 plate, joint tensile strength and microhardness are affected by multiple interconnected variables including welding force, temperature, and geometric quantity of the butt face. The joint mechanical properties are difficult to predict and ensure. This paper proposed a hybrid predicting approach based on multi-source information fusion, called Down-Sampling and One-Dimensional Convolutional Neural Network (DS-1DCNN). Initially, FSW experiments were conducted to collect the axial force, welding surface temperature at advancing and retreating side, gap, and mismatch signals, which were fused through DS technology. To extract temporal information and features from signals, a 1DCNN prediction model was established, with the fused multi-source signals as input and the tensile strength and Vickers hardness as output. Subsequently, the optimal model architecture and hyper-parameters of 1DCNN were determined through experiments and grid search combined with K-fold cross-validation, and an online prediction system was developed. Finally, effective validation experiments were conducted and the experimental results agreed with the DS-1DCNN prediction results with an acceptable error of approximately 5%. Parameter sensitivity analysis revealed that the number of convolutional layers and learning rate were identified as critical parameters affecting the model performance. Additionally, time analysis showed that the calculation time of the prediction system was less than 200 ms which can meet the time requirements of real-time thick-plate FSW application. The research lays a foundation on the real-time quality monitoring in the process of FSW thick-plate butt joint.

Research on energy consumption optimization of predictive cruise control considering the state of the leading vehicle

To comprehensively assess the economic performance of electric vehicles (EVs) cruise planning and mitigate conflicts with leading vehicles, a dynamic programming-based predictive cruise speed optimization control system was developed. This system comprises two layers: an upper V2X-based LSTM-DP cruise speed planning system and a lower adaptive cruise control system based on sliding mode control (SMC), and the mode switching strategy of the speed and distance tracking controller is designed. Initially, a segmented uniform variable motion algorithm processed the historical driving data of the leading vehicle, significantly enhancing the LSTM neural network's acceleration prediction capability, particularly on inclines, with an improvement in accuracy of 37.8 %. Subsequently, an adaptive cruise tracking controller was implemented to track the planned vehicle speed. Simulation tests on actual roads demonstrated that the LSTM-DP approach not only ensures driving safety but also markedly improves fuel efficiency over conventional constant speed cruising methods when planning is effective. Moreover, even in scenarios where planning fails, the LSTM-DP strategy can still achieve energy savings of up to 37 % without compromising the safety distance between vehicles. Finally, hardware-in-the-loop (HIL) testing confirmed the system's efficacy under real-time operational conditions.

Smart Predict: adding partner-suggested vocabulary to increase efficiency in a dual tablet AAC typing application

Smart Predict is a novel two-tablet application developed to improve conversational efficiency for people who use augmentative and alternative communication (AAC) devices. The Smart Predict system consists of two distinct applications that were built for use with two Android tablets and/or phones with Bluetooth® capability. One application is referred to as the AAC User app and the second application is called the Partner app. Smart Predict integrates vocabulary supplementation from communication partners in real-time while persons who rely on AAC generate text and maintain full control to choose words within a word prediction line. Using single-case research designs with adapted alternating treatments under two counterbalanced treatment conditions (AAC User app alone versus AAC User app + Partner app), we examined message efficiency with persons who rely on switch scanning for on-screen keyboards. User experience questionnaires were administered to determine satisfaction and workload, and to provide feedback for future development efforts. Three adults with complex communication needs and motor impairments conversed with research staff about fourteen movie trailers. Characters per minute and switch selections per character produced by the person with complex communication needs and motor impairments were measured during 15-minute interviews to represent communication speed and effort. Results indicate that message efficiency increased with the dual-tablet Smart Predict, indicating faster message generation with the vocabulary supplementation system. User satisfaction increased and perceived effort decreased under the Smart Predict conditions. Given these results within a research and development framework, the Smart Predict concept is a viable feature that could be considered within smart AAC technologies.