High Prediction Accuracy Research Articles

Uncertainty analysis for forest height inversion using L / P band PolInSAR datasets and RVoG model over kryclan forest site

Forest height, as a measure of the quantity and quality of forest resources, plays a significant role in the study of the ecological functions performed by forests. Although the polarimetric synthetic aperture radar interferometry (PolInSAR) technique has evolved as a potent method for forest height inversion, uncertainties still exist in the process of estimating forest height, and the uncertainties in predicted forest height directly lead into the uncertainty of terrestrial carbon stock calculation results. In this study, we took the Random Volume over Ground (RVoG) model as likelihood function and constructed a hierarchical Bayesian framework to calculate and reduce the uncertainty of forest height inversion using L / P band PolInSAR airborne data via RVoG model. Uncertainties resulted from five canopy types and three forest densities were analyzed, respectively. The results showed that among the five different canopy types, L band has the highest prediction accuracy in pure coniferous canopy with Acc. = 0.90. The uncertainty is extremely low for pure forest, with the ratio of uncertainty values of 0.09 for L band and 0.15 for the P band in pure coniferous canopy, and uncertainty values of 0.16 for L band and 0.11 for P band in pure broadleaf canopy, respectively. Furthermore, when the forest density is between 300 and 600 stems/ha, the ratio of uncertainties for the L band is 0.27, whereas the P band is 0.24. As forest density increases, the uncertainty in forest height estimates decreases for both bands. The changes in canopy types and forest density affect forest height estimation uncertainties obviously, the effects are different at each frequency. The forest height inversion accuracy of the L band in pure coniferous canopy surpasses that in other canopy types, with the lowest uncertainty. P band performed well in broadleaf canopy forest height inversion. The inversion uncertainties at both frequencies decrease with increase of forest densities.

Different data-driven prediction of global ionospheric TEC using deep learning methods

Ionospheric Total Electron Content (TEC) is a crucial parameter for monitoring the ionosphere and space weather disasters. Its accurate prediction is vital for precise applications of Interferometric Synthetic Aperture Radar (InSAR) and Global Navigation Satellite System (GNSS). This study proposes a novel method for ionospheric TEC prediction that considers multiple TEC-related factors. We present the random forest method and the Autoformer deep learning with multilayer perceptron (Autoformer-MLP) to predict the global TEC by incorporating the geomagnetic and solar activity parameters. Two schemes with different ionospheric data, i.e., spherical harmonic coefficients (SHC) and vertical TEC (VTEC), are performed by Autoformer-MLP. Ionospheric products from 2009 to 2019 (Cycle 24), obtained from the Center for Orbit Determination in Europe (CODE), are collected to test and evaluate the proposed method. Experimental results demonstrate that the root mean square errors (RMSEs) of predicted global ionospheric maps (GIMs) using the SHC scheme are 3.79 and 1.39 TECU in 2015 and 2019, respectively, and those are 3.55 and 1.28 TECU for the VTEC scheme. Moreover, the RMSEs of the prediction results are 0.5 and 0.2 TECU lower than that of CODE's 1-day predicted global ionospheric map (GIM) product (C1PG) during high and low solar activity years, respectively. These analyses indicate that the proposed random forest and Autoformer-MLP deep learning methods exhibit high accuracy and stability for data-driven prediction of global ionospheric TEC in various scenarios.

Enhanced Sequence-to-Sequence Deep Transfer Learning for Day-Ahead Electricity Load Forecasting

Electricity load forecasting is a crucial undertaking within all the deregulated markets globally. Among the research challenges on a global scale, the investigation of deep transfer learning (DTL) in the field of electricity load forecasting represents a fundamental effort that can inform artificial intelligence applications in general. In this paper, a comprehensive study is reported regarding day-ahead electricity load forecasting. For this purpose, three sequence-to-sequence (Seq2seq) deep learning (DL) models are used, namely the multilayer perceptron (MLP), the convolutional neural network (CNN) and the ensemble learning model (ELM), which consists of the weighted combination of the outputs of MLP and CNN models. Also, the study focuses on the development of different forecasting strategies based on DTL, emphasizing the way the datasets are trained and fine-tuned for higher forecasting accuracy. In order to implement the forecasting strategies using deep learning models, load datasets from three Greek islands, Rhodes, Lesvos, and Chios, are used. The main purpose is to apply DTL for day-ahead predictions (1–24 h) for each month of the year for the Chios dataset after training and fine-tuning the models using the datasets of the three islands in various combinations. Four DTL strategies are illustrated. In the first strategy (DTL Case 1), each of the three DL models is trained using only the Lesvos dataset, while fine-tuning is performed on the dataset of Chios island, in order to create day-ahead predictions for the Chios load. In the second strategy (DTL Case 2), data from both Lesvos and Rhodes concurrently are used for the DL model training period, and fine-tuning is performed on the data from Chios. The third DTL strategy (DTL Case 3) involves the training of the DL models using the Lesvos dataset, and the testing period is performed directly on the Chios dataset without fine-tuning. The fourth strategy is a multi-task deep learning (MTDL) approach, which has been extensively studied in recent years. In MTDL, the three DL models are trained simultaneously on all three datasets and the final predictions are made on the unknown part of the dataset of Chios. The results obtained demonstrate that DTL can be applied with high efficiency for day-ahead load forecasting. Specifically, DTL Case 1 and 2 outperformed MTDL in terms of load prediction accuracy. Regarding the DL models, all three exhibit very high prediction accuracy, especially in the two cases with fine-tuning. The ELM excels compared to the single models. More specifically, for conducting day-ahead predictions, it is concluded that the MLP model presents the best monthly forecasts with MAPE values of 6.24% and 6.01% for the first two cases, the CNN model presents the best monthly forecasts with MAPE values of 5.57% and 5.60%, respectively, and the ELM model achieves the best monthly forecasts with MAPE values of 5.29% and 5.31%, respectively, indicating the very high accuracy it can achieve.

Gaussian processes modeling for the prediction of polymeric nanoparticle formulation design to enhance encapsulation efficiency and therapeutic efficacy.

Conventional drugs have been facing various drug delivery obstacles, including first-pass metabolism for oral medications, drug degradation by cellular enzymes, off-target effects, and cytotoxicity of healthy cells. Nanoparticles (NP) application in drug delivery can compensate for these drawbacks to a great extent. NPs can be fabricated using different materials and structures to achieve desired therapeutic effects. For each type of NP material, its physicochemical properties determine compatibility with specific drugs and other supplemental compositions. The optimized material selection becomes prominent in NP development to improve NP performances. Due to the nature of NP fabrication, the process is long and expensive. To accelerate NP composition optimization, machine learning (ML) techniques are among the most promising methods for efficient data predictions and optimizations.As a proof-of concept, we created Gaussian Process (GP) models to make predictions for drug encapsulation efficiency (EE%) and therapeutic efficacy of 32 poly (lactic-co-glycolic acid) (PLGA) NPs that are formed with materials with different physicochemical properties. Two model drugs, doxorubicin (DOX) and docetaxel (DTX) were loaded separately. The IC50 values for the various NPs formulations were evaluated using the OVCAR3 epithelial ovarian cancer cell line. EE% GP model has the highest prediction accuracy with the lowest normalized root-mean-squared-error (RMSE) of 0.187. The DOX and DTX IC50 GP models have normalized RMSEs of 0.296 and 0.206, respectively, which are higher than that of the EE% GP model.

Site Identification and Next Choice Protocol for Hit-to-Lead Optimization.

Time efficiency and cost savings are major challenges in drug discovery and development. In this process, the hit-to-lead stage is expected to improve efficiency because it primarily exploits the trial-and-error approach of medicinal chemists. This study proposes a site identification and next choice (SINCHO) protocol to improve the hit-to-lead efficiency. This protocol selects an anchor atom and growth site pair, which is desirable for a hit-to-lead strategy starting from a 3D complex structure. We developed and fine-tuned the protocol using a training data set and assessed it using a test data set of the preceding hit-to-lead strategy. The protocol was tested for experimentally determined structures and molecular dynamics (MD) ensembles. The protocol had a high prediction accuracy for applying MD ensembles, owing to the consideration of protein flexibility. The SINCHO protocol enables medicinal chemists to visualize and modify functional groups in a hit-to-lead manner.

A Two-Layer SD-ANN-CA Model Framework for Multi-Typed Land Use and Land Cover Change Prediction under Constraints: Case Study of Ya’an City Area, Western China

Land use and land cover change (LUCC) prediction of cities in Western China requires higher accuracy in quantitative demand and spatial layout because of complex challenges in balancing relationships between urban constructions and ecological developments. Considering city-level areas and various types of land use and land cover, existing LUCC models without constraint or with only loose demand constraints were impractical in providing evidence of high accuracy and high-resolution predictions in areas facing fierce land competition. In this study, we proposed a two-layer SD-ANN-CA model to simulate and explore the LUCC trend and layout predictions for 2018, 2028, and 2038 in Ya’an City, Western China. The two-layer structure with an upper layer of the SD model and a lower layer of the ANN-CA model, as well as the advantages of all three methods of system dynamics (SD), artificial neural network (ANN), and cellular automata (CA), have allowed us to consider the macro-level demand constraints, meso-level driving factors constraints, and the micro-level spatial constraints into a unified model framework. The simulation results of the year 2018 have shown significant improvement in the accuracy of the ANN-CA model constructed in our earlier work, especially in types of forest land (error-accuracy: 0.08%), grassland (error-accuracy: 0.23%), and construction land (error-accuracy: 0.18%). The layout predictions of all six types of land use in 2028 and 2038 are then carried out to provide visual evidence support, which may improve the efficiency of planning and policy-making processes. Our work may also provide insights into new ways to combine quantitative methods into spatial methods in constructing city-level or even regional-level LUCC models with high resolution.

Asymmetric influence-based superposed random walk link prediction algorithm in complex networks

Random walk-based link prediction algorithms have achieved desirable results for complex network mining, but in these algorithms, the transition probability of particles usually only considers node degrees, resulting in particles being able to randomly select adjacent nodes for random walks in an equal probability manner, to solve this problem, the asymmetric influence-based superposed random walk link prediction algorithm is proposed in this paper. This algorithm encourages particles to choose the next node at each step of the random walk process based on the asymmetric influence between nodes. To this end, we fully consider the topological information around each node and propose the asymmetric influence between nodes. Then, an adjustable parameter is applied to normalize the degree of nodes and the asymmetric influence between nodes into transition probability. Based on this, the proposed new transition probability is applied to superposed random walk process to measure the similarity between all nodes in the network. Empirical experiments are conducted on 16 real-world network datasets such as social network, ecology network, and animal network. The experimental results show that the proposed algorithm has high prediction accuracy in most network, compared with 10 benchmark indices.

Development and validation of a novel nomogram for predicting good neoangiogenesis after encephaloduroarteriosynangiosis in patients with moyamoya disease and type 2 diabetes mellitus: a case-control study.

Diabetes is often linked to poorer outcomes in patients with moyamoya disease (MMD). However, experience has shown that certain individuals with diabetes have favorable outcomes after encephaloduroarteriosynangiosis (EDAS). The authors aimed to develop a nomogram to predict good neoangiogenesis in patients with MMD and type 2 diabetes mellitus (T2DM) to aid neurosurgeons in the identification of suitable candidates for EDAS. Adults with MMD and T2DM who underwent EDAS between June 2004 and December 2018 were included in the analysis. In total, 126 patients (213 hemispheres) with MMD and T2DM from the Fifth Medical Centre of the Chinese PLA General Hospital were included and randomly divided into training (152 hemispheres) and internal validation (61 hemispheres) cohorts at a ratio of 7:3. Univariate logistic and least absolute shrinkage and selection operator regression analyses were used to identify the significant factors associated with good neoangiogenesis, which were used to develop a nomogram. The discrimination, calibration, and clinical utility were assessed. A total of 213 hemispheres in 126 patients were reviewed, including 152 (71.36%) hemispheres with good postoperative collateral formation and 61 (28.64%) with poor postoperative collateral formation. The authors selected 4 predictors (FGD5 rs11128722, VEGFA rs9472135, Suzuki stage, and internal carotid artery [ICA] moyamoya vessels) for nomogram development. The C-indices of the nomogram in the training and internal validation cohorts were 0.873 and 0.841, respectively. The nomogram exhibited a sensitivity of 84.5% and specificity of 81.0%. The positive and negative predictive values were 92.1% and 66.7%, respectively. The calibration curves indicated high predictive accuracy, and receiver operating characteristic curve analysis showed the superiority of the nomogram. The decision-making analysis validated the fitness and clinical application value of this nomogram. Then a web-based calculator to facilitate clinical application was generated. The nomogram developed in this study accurately predicted neoangiogenesis in patients with MMD and T2DM after EDAS and may assist neurosurgeons in identifying suitable candidates for indirect revascularization surgery.

Improving the Estimation of Structural Parameters of a Mixed Conifer–Broadleaf Forest Using Structural, Textural, and Spectral Metrics Derived from Unmanned Aerial Vehicle Red Green Blue (RGB) Imagery

Forest structural parameters are crucial for assessing ecological functions and forest quality. To improve the accuracy of estimating these parameters, various approaches based on remote sensing platforms have been employed. Although remote sensing yields high prediction accuracy in uniform, even-aged, simply structured forests, it struggles in complex structures, where accurately predicting forest structural parameters remains a significant challenge. Recent advancements in unmanned aerial vehicle (UAV) photogrammetry have opened new avenues for the accurate estimation of forest structural parameters. However, many studies have relied on a limited set of remote sensing metrics, despite the fact that selecting appropriate metrics as powerful explanatory variables and applying diverse models are essential for achieving high estimation accuracy. In this study, high-resolution RGB imagery from DJI Matrice 300 real-time kinematics was utilized to estimate forest structural parameters in a mixed conifer–broadleaf forest at the University of Tokyo Hokkaido Forest (Hokkaido, Japan). Structural and textual metrics were extracted from canopy height models, and spectral metrics were extracted from orthomosaics. Using random forest and multiple linear regression models, we achieved relatively high estimation accuracy for dominant tree height, mean tree diameter at breast height, basal area, mean stand volume, stem density, and broadleaf ratio. Including a large number of explanatory variables proved advantageous in this complex forest, as its structure is influenced by numerous factors. Our results will aid foresters in predicting forest structural parameters using UAV photogrammetry, thereby contributing to sustainable forest management.

Exploring rearing factors to predict potential sensory quality of heifer meat throughout the farm-to-fork continuum

Sensory beef qualities could be impacted throughout the farm-to-fork continuum. The aim of this work was to predict the sensory quality classes of Longissimus muscle established from three sensory descriptors: tenderness, juiciness and flavour intensity. The extreme classes of meat quality were: Q + class including the highest scores for tenderness and juiciness, and an intermediate score for flavour intensity and Q − class including low scores for tenderness, juiciness and flavour intensity. To predict the extreme quality classes, seven decision trees were performed using the individual data related to rearing factors (p = 50), carcase traits (p = 13) and/or aged meat traits (p = 9) of 100 Charolais heifers. The decision trees established from rearing factors and carcass trait data (RF-CARCA-Tree) allowed the highest accuracy of prediction (79.7%) with 90.7% and 66.7% of correctly classified individuals, respectively. Our results showed that different combinations of factors could produce Q + class. Three rearing factors (i.e. the calculated average of concentrates’ net energy content in the diet during the pasture period of pre-weaning period (PWP); the number of days concentrates were offered in calf diet during PWP; and the calculated average of concentrates’ crude protein content in the fattening diet) and the conformation score could be considered as action levers to improve meat quality. These three rearing factors were related to the pre-weaning and fattening periods of the heifer, slowing a possible management of the potential beef quality from rearing factors throughout the life of the heifer. Highlights Beef quality can be improved throughout the heifer’s life by rearing factors related to concentrate in the diet. Sensory beef quality classes could be managed from rearing factors and carcass traits. Raw aged meat data did not allow to improve the prediction of sensory beef quality classes.

Development and validation of a nomogram to predict lymph node metastasis in patients with progressive muscle-invasive bladder cancer

PurposeTo develop and validate a nomogram for preoperative prediction of lymph node metastasis in patients with progressive muscle-invasive bladder cancer.Materials and methodsWe retrospectively recruited patients, divided them into training and validation cohorts, and gathered patient demographics, pathology data of transurethral bladder tumor resection specimens, imaging findings, and laboratory information. We performed logistic regression analyses, both single-variable and multi-variable, to investigate independent preoperative risk variables and develop a nomogram. Both internal and external validations were conducted to evaluate the predictive performance of this nomogram.ResultsThe training cohort consisted of 144 patients with advanced muscle-invasive bladder cancer, while the validation cohort included 62 individuals. The independent preoperative risk factors identified were tumor pathology grade, platelet count, tumor size on imaging, and lymph node size, which were utilized to develop the nomogram. The model demonstrated high predictive accuracy, as evidenced by the area under the receiver operating characteristic curve values of 0.898 and 0.843 for the primary and external validation cohorts, respectively. Calibration curves and decision curve analysis showed a good performance of the nomogram in both cohorts, indicating its high clinical applicability.ConclusionA nomogram for preoperative prediction of lymph node metastasis in patients with advanced muscle-invasive bladder cancer was successfully developed; its accuracy, reliability, and clinical value were demonstrated. This new tool would facilitate better clinical decisions regarding whether to perform complete lymph node dissection in cases of radical cystectomy.

Residual network improves the prediction accuracy of genomic selection.

Genetic improvement of complex traits in animal and plant breeding depends on the efficient and accurate estimation of breeding values. Deep learning methods have been shown to be not superior over traditional genomic selection (GS) methods, partially due to the degradation problem (i.e. with the increase of the model depth, the performance of the deeper model deteriorates). Since the deep learning method residual network (ResNet) is designed to solve gradient degradation, we examined its performance and factors related to its prediction accuracy in GS. Here we compared the prediction accuracy of conventional genomic best linear unbiased prediction, Bayesian methods (BayesA, BayesB, BayesC, and Bayesian Lasso), and two deep learning methods, convolutional neural network and ResNet, on three datasets (wheat, simulated and real pig data). ResNet outperformed other methods in both Pearson's correlation coefficient (PCC) and mean squared error (MSE) on the wheat and simulated data. For the pig backfat depth trait, ResNet still had the lowest MSE, whereas Bayesian Lasso had the highest PCC. We further clustered the pig data into four groups and, on one separated group, ResNet had the highest prediction accuracy (both PCC and MSE). Transfer learning was adopted and capable of enhancing the performance of both convolutional neural network and ResNet. Taken together, our findings indicate that ResNet could improve GS prediction accuracy, affected potentially by factors such as the genetic architecture of complex traits, data volume, and heterogeneity.

Genomic loci associated with grain yield under well-watered and water-stressed conditions in multiple bi-parental maize populations

Smallholder maize farming systems in sub-Saharan Africa (SSA) are vulnerable to drought-induced yield losses, which significantly impact food security and livelihoods within these communities. Mapping and characterizing genomic regions associated with water stress tolerance in tropical maize is essential for future breeding initiatives targeting this region. In this study, three biparental F3 populations composed of 753 families were evaluated in Kenya and Zimbabwe and genotyped with high-density single nucleotide polymorphism (SNP) markers. Quantitative trait loci maping was performed on these genotypes to dissect the genetic architecture for grain yield (GY), plant height (PH), ear height (EH) and anthesis-silking interval (ASI) under well-watered (WW) and water-stressed (WS) conditions. Across the studied maize populations, mean GY exhibited a range of 4.55–8.55 t/ha under WW and 1.29–5.59 t/ha under WS, reflecting a 31–59% reduction range under WS conditions. Genotypic and genotype-by-environment (G × E) variances were significant for all traits except ASI. Overall broad sense heritabilities for GY were low to high (0.25–0.60). For GY, these genetic parameters were decreased under WS conditions. Linkage mapping revealed a significant difference in the number of QTLs detected, with 93 identified under WW conditions and 41 under WS conditions. These QTLs were distributed across all maize chromosomes. For GY, eight and two major effect QTLs (>10% phenotypic variation explained) were detected under WW and WS conditions, respectively. Under WS conditions, Joint Linkage Association Mapping (JLAM) identified several QTLs with minor effects for GY and revealed genomic region overlaps in the studied populations. Across the studied water regimes, five-fold cross-validation showed moderate to high prediction accuracies (−0.15–0.90) for GY and other agronomic traits. Our findings demonstrate the polygenic nature of WS tolerance and highlights the immense potential of using genomic selection in improving genetic gain in maize breeding.

Lap-slip model of rebar-to-concrete in RC/ECC/UHPC based on GA-BP neural network

Rebar lap splicing is a widely used method of reinforcement connection in precast concrete structures, and the lap-slip model serves as an important prerequisite for conducting detailed mechanical calculations of reinforced concrete lap splice members. Most traditional bond-slip mechanical models are semi-empirical and semi-theoretical formulas derived from experimental data; they do not consider the effect of reinforcement lap spacing, which prevents them from accurately reflecting the force transfer characteristics of lap reinforcement in concrete. In this paper, a Back Propagation (BP) neural network based on a Genetic Algorithm (GA) is used to establish a lap-slip model of reinforced concrete considering rebar spacing, which improves on the current bond-slip mechanical model. Several sets of rebar lap tests based on RC/ECC/UHPC were designed, and full-sample parameter sensitivity analyses were carried out using the lap length, the thickness of the concrete cover, the rebar spacing and the stirrup ratio as test parameters. The results showed that rebar lap spacing was an important influencing factor in the proposed lap-slip model, with a sensitivity coefficient reaching 0.471. The lap-slip model, based on the GA-BP neural network, was developed and compared with existing models; the predictive ability of the model was verified by the ten-fold crossover method. The generalizability of the GA-BP model was verified by randomly selecting non-training test data as well as lap test data of rebar in different types of concrete. The results indicated that the model exhibits high predictive and generalization abilities, with most correlation coefficients exceeding 0.95. The proposed model was compared with a mechanical model established using the traditional fitting formula method, and the results showed that the prediction accuracy of the lap-slip model established based on the GA-BP neural network is slightly lower than that of the mechanical model obtained using the fitting formula method, but it still offers high prediction accuracy, and the prediction results are more objective and realistic, which is of guiding significance for practical applications and related research.

Modelling and optimization of electrodeposited amorphous Fe-P alloys using central composite design and response surface method

Central composite circumscribed (CCC) design and response surface methodology (RSM) were used to model and optimize the electrodeposition of amorphous Fe-P alloys. Based on our previous single-factor experimental results, the significance of influencing factors was ranked using analysis of variance (ANOVA) method. Three factors, significantly impacting the P content in deposits, hardness and corrosion current density, were identified, including bath temperature, pH values and H2PO2− concentration. The statistic relationships between process parameters and individual responses were established based on the CCC experimental data and RSM. The optimal parameters for each response were derived and the influences of interaction terms were investigated. The predicted maximum P content of 26.35 wt% was achieved at a bath temperature of 60℃, a pH of 1.2 and H2PO2− concentration of 60 g/L. The predicted highest microhardness was 534 HV0.1 under the optimal conditions of temperature of 60 ℃, pH value of 1.6 and H2PO2− concentration of 49 g/L. The lowest corrosion current density was predicted to be 1.34 μA·cm−2 under the conditions of a bath temperature of 60℃，a pH value of 1.2, and a H2PO2− concentration of 53 g/L. A desirability function was applied to explore the optimal comprehensive performance of both high hardness and low corrosion current density. The optimal parameter combination for comprehensive performance was bath temperature of 60℃, pH of 1.4, and H2PO2− concentration of 49 g/L, and the resulting hardness and corrosion current density were 527 HV0.1 and 2.40 μA·cm−2, respectively. Due to the complex electrodeposition mechanism of amorphous Fe-P alloys, the predicted P content in deposits largely deviated from the experimental result, but the hardness, corrosion current density and comprehensive performances show high prediction accuracy.

Feature engineering of environmental covariates improves plant genomic-enabled prediction

IntroductionBecause Genomic selection (GS) is a predictive methodology, it needs to guarantee high-prediction accuracies for practical implementations. However, since many factors affect the prediction performance of this methodology, its practical implementation still needs to be improved in many breeding programs. For this reason, many strategies have been explored to improve the prediction performance of this methodology.MethodsWhen environmental covariates are incorporated as inputs in the genomic prediction models, this information only sometimes helps increase prediction performance. For this reason, this investigation explores the use of feature engineering on the environmental covariates to enhance the prediction performance of genomic prediction models.Results and discussionWe found that across data sets, feature engineering helps reduce prediction error regarding only the inclusion of the environmental covariates without feature engineering by 761.625% across predictors. These results are very promising regarding the potential of feature engineering to enhance prediction accuracy. However, since a significant gain in prediction accuracy was observed in only some data sets, further research is required to guarantee a robust feature engineering strategy to incorporate the environmental covariates.

Benedict–Webb–Rubin–Starling Equation of State + Hydrate Thermodynamic Theories: An Enhanced Prediction Method for CO2 Solubility and CO2 Hydrate Phase Equilibrium in Pure Water/NaCl Aqueous Solution System

Accurately predicting the phase behavior and physical properties of carbon dioxide (CO2) in pure water/NaCl mixtures is crucial for the design and implementation of carbon capture, utilization, and storage (CCUS) technology. However, the prediction task is complicated by CO2 liquefaction, CO2 hydrate formation, multicomponent and multiphase coexistence, etc. In this study, an improved method that combines Benedict–Webb–Rubin–Starling equation of state (BWRS EOS) + hydrate thermodynamic theories was proposed to predict CO2 solubility and phase equilibrium conditions for a mixed system across various temperature and pressure conditions. By modifying the interaction coefficients in BWRS EOS and the Van der Waals–Platteeuw model, this new method is applicable to complex systems containing two liquid phases and a CO2 hydrate phase, and its high prediction accuracy was verified through a comparative evaluation with a large number of reported experimental data. Furthermore, based on the calculation results, the characteristics of CO2 solubility and the variation of phase equilibrium conditions of the mixture system were discussed. These findings highlight the influence of hydrates and NaCl on CO2 solubility characteristics and clearly demonstrate the hindrance of NaCl to the formation of CO2 hydrates. This study provides valuable insights and fundamental data for designing and implementing CCUS technology that contribute to addressing global climate change and environmental challenges.

A novel coupled rainfall prediction model based on stepwise decomposition technique

The traditional decomposed ensemble prediction model decomposes the entire rainfall sequence into several sub-sequences, dividing them into training and testing periods for modeling. During sample construction, future information is erroneously mixed into the training data, making it challenging to apply in practical rainfall forecasting. This paper proposes a novel stepwise decomposed ensemble coupling model, realized through variational mode decomposition (VMD) and bidirectional long short-term memory neural network (BiLSTM) models. Model parameters are optimized using an improved particle swarm optimization (IPSO). The performance of the model was evaluated using rainfall data from the Southern Four Lakes basin. The results indicate that: (1) Compared to the PSO algorithm, the IPSO algorithm-coupled model shows a minimum decrease of 2.70% in MAE and at least 2.62% in RMSE across the four cities in the Southern Four Lakes basin; the IPSO algorithm results in a minimum decrease of 25.58% in MAE and at least 28.19% in RMSE for the VMD-BiLSTM model. (2) When compared to IPSO-BiLSTM, the VMD-IPSO-BiLSTM based on the stepwise decomposition technique exhibits a minimum decrease of 26.54% in MAE and at least 34.16% in RMSE. (3) The NSE for the testing period of the VMD-IPSO-BiLSTM model in each city surpasses 0.88, indicating higher prediction accuracy and providing new insights for optimizing rainfall forecasting.

A nomogram based on nutritional and inflammatory parameters to predict DMFS and identify beneficiaries of adjuvant chemotherapy in IVA-stage nasopharyngeal carcinoma

ObjectiveThis study aims to develop a nomogram integrating inflammation (NLR), Prognostic Nutritional Index (PNI), and EBV DNA (tumor burden) to achieve personalized treatment and prediction for stage IVA NPC. Furthermore, it endeavors to pinpoint specific subgroups that may derive significant benefits from S-1 adjuvant chemotherapy.MethodsA total of 834 patients diagnosed with stage IVA NPC were enrolled in this study and randomly allocated into training and validation cohorts. Multivariate Cox analyses were conducted to identify independent prognostic factors for constructing the nomogram. The predictive and clinical utility of the nomogram was assessed through measures including the AUC, calibration curve, DCA, and C-indexes. IPTW was employed to balance baseline characteristics across the population. Kaplan-Meier analysis and log-rank tests were utilized to evaluate the prognostic value.ResultsIn our study, we examined the clinical features of 557 individuals from the training cohort and 277 from the validation cohort. The median follow-up period was 50.1 and 49.7 months, respectively. For the overall cohort, the median follow-up duration was 53.8 months. The training and validation sets showed 3-year OS rates of 87.7% and 82.5%, respectively. Meanwhile, the 3-year DMFS rates were 95.9% and 84.3%, respectively. We created a nomogram that combined PNI, NRI, and EBV DNA, resulting in high prediction accuracy. Risk stratification demonstrated substantial variations in DMFS and OS between the high and low risk groups. Patients in the high-risk group benefited significantly from the IC + CCRT + S-1 treatment. In contrast, IC + CCRT demonstrated non-inferior 3-year DMFS and OS compared to IC + CCRT + S-1 in the low-risk population, indicating the possibility of reducing treatment intensity.ConclusionsIn conclusion, our nomogram integrating NLR, PNI, and EBV DNA offers precise prognostication for stage IVA NPC. S-1 adjuvant chemotherapy provides notable benefits for high-risk patients, while treatment intensity reduction may be feasible for low-risk individuals.

Joint prediction method for strip thickness and flatness in hot strip rolling process: A combined multi-indicator Transformer with embedded sliding window

Thickness and flatness are important quality indicators for strip. It is important that the rapid and accurate prediction of the exit thickness and flatness for the optimal control of the hot strip rolling process. Due to the fast and long rolling process, there are time delays, non-linearity and strong coupling among the variables, which cause difficulties in the establishment of prediction models. In this paper, the variables related to thickness and flatness are selected by analyzing the rolling process mechanism and data. Based on the data related to the rolling quality, a rolling exit thickness and flatness joint prediction model combined multi-indicator Transformer with embedded sliding window (SW-MTrans) is proposed. First, a sliding window is embedded into the input layer of the model in order to address the effect of the time delay among variables. Then a Transformer network is improved to achieve accurate prediction of thickness and flatness simultaneously. It is verified that the proposed method can predict the thickness and flatness at the same time with higher prediction accuracy and generalization ability compared with other methods through actual production data. The mean absolute error (MAE) for thickness prediction was reduced by 19.37% and MAE for flatness prediction was reduced by 14.03% compared to the existing prediction model.