Hybrid model for BOF oxygen blowing time prediction based on oxygen balance mechanism and deep neural network

The amount of oxygen blown into the converter is one of the key parameters for the control of the converter blowing process,which directly affects the tap-to-tap time of converter. In this study, a hybrid model based on oxygen balance mechanism (OBM) and deep neural network (DNN) was established for predicting oxygen blowing time in converter. A three-step method was utilized in the hybrid model. First, the oxygen consumption volume was predicted by the OBM model and DNN model, respectively. Second, a more accurate oxygen consumption volume was obtained by integrating the OBM model and DNN model. Finally, the converter oxygen blowing time was calculated according to the oxygen consumption volume and the oxygen supply intensity of each heat. The proposed hybrid model was verified using the actual data collected from an integrated steel plant in China, and compared with multiple linear regression model, OBM model, and neural network model including extreme learning machine, back propagation neural network, and DNN. The test results indicate that the hybrid model with a network structure of 3 hidden layer layers, 32-16-8 neurons per hidden layer, and 0.1 learning rate has the best prediction accuracy and stronger generalization ability compared with other models. The predicted hit ratio of oxygen consumption volume within the error±300 m^(3)is 96.67%;determination coefficient (R^(2)) and root mean square error (RMSE) are0.6984 and 150.03 m^(3), respectively. The oxygen blow time prediction hit ratio within the error±0.6 min is 89.50%;R2and RMSE are0.9486 and 0.3592 min, respectively. As a result, the proposed model can effectively predict the oxygen consumption volume and oxygen blowing time in the converter.

Electrothermal Model Based Remaining Charging Time Prediction of Lithium-Ion Batteries against Wide Temperature Range

Battery remaining charging time(RCT)prediction can facilitate charging management and alleviate mileage anxiety for electric vehicles(EVs).Also,it is of great significance to improve EV users’experience.However,the RCT for a lithiumion battery pack in EVs changes with temperature and other battery parameters.This study proposes an electrothermal model-based method to accurately predict battery RCT.Firstly,a characteristic battery cell is adopted to represent the battery pack,thus an equivalent circuit model(ECM)of the characteristic battery cell is established to describe the electrical behaviors of a battery pack.Secondly,an equivalent thermal model(ETM)of the battery pack is developed by considering the influence of ambient temperature,thermal management,and battery connectors in the battery pack to calculate the temperature which is then fed back to the ECM to realize electrothermal coupling.Finally,the RCT prediction method is proposed based on the electrothermal model and validated in the wide temperature range from-20℃to 45℃.The experimental results show that the prediction error of the RCT in the whole temperature range is less than 1.5%.

A nonlocal dispersal and time delayed HIV infection model with general incidences

Biologically,because of the impact of reproduction period and nonlocal dispersal of HIV-infected cells,time delay and spatial heterogeneity should be considered.In this paper,we establish an HIV infection model with nonlocal dispersal and infection age.Moreover,applying the theory of Fourier transformation and von Foerster rule,we transform the model to an integrodifferential equation with nonlocal time delay and dispersal.The well-posedness,positivity,and boundedness of the solution for the model are studied.

A novel framework for predicting non-stationary production time series of shale gas based on BiLSTM-RF-MPA deep fusion model

Shale gas, as an environmentally friendly fossil energy resource, has gained significant commercial development and shows immense potential. However, accurately predicting shale gas production faces substantial challenges due to the complex law of decline, nonlinear and non-stationary features in production data, which greatly repair the robustness of current models in predicting shale gas production time series. To address these challenges and improve accuracy in production forecasting, this paper introduces a novel and innovative approach: a hybrid proxy model that combines the bidirectional long short-term memory(BiLSTM) neural network and random forest(RF) through deep learning. The BiLSTM neural network is adept at capturing long-term dependencies, making it suitable for understanding the intricate relationships between input and output variables in shale gas production.On the other hand, RF serves a dual purpose: reducing model variance and addressing the concept drift problem that arises in non-stationary time series predictions made by BiLSTM. By integrating these two models, the hybrid approach effectively captures the inherent dependencies present in long and nonstationary production time series, thereby reducing model uncertainty. Furthermore, the combination of BiLSTM and RF is optimized using the recently-proposed marine predators algorithm(MPA) to fine-tune hyperparameters and enhance the overall performance of the proxy model. The results demonstrate that the proposed BiLSTM-RF-MPA model achieves higher prediction accuracy and demonstrates stronger generalization capabilities by effectively handling the complex nonlinear and non-stationary characteristics of shale gas production time series. Compared to other models such as LSTM, BiLSTM, and RF, the proposed model exhibits superior fitting and prediction performance, with an average improvement in performance indicators exceeding 20%. This innovative framework provides valuable insights for forecasting the complex production performance of unconventional oil and gas reservoirs, which sheds light on the development of data-driven proxy models in the field of subsurface energy utilization.

Space station short-term mission planning using ontology modelling and time iteration

This paper studies the problem of the space station short-term mission planning, which aims to allocate the executing time of missions effectively, schedule the corresponding resources reasonably and arrange the time of the astronauts properly. A domain model is developed by using the ontology theory to describe the concepts, constraints and relations of the planning domain formally, abstractly and normatively. A method based on time iteration is adopted to solve the short-term planning problem. Meanwhile, the resolving strategies are proposed to resolve different kinds of conflicts induced by the constraints of power, heat, resource, astronaut and relationship. The proposed approach is evaluated in a test case with fifteen missions, thirteen resources and three astronauts. The results show that the developed domain ontology model is reasonable, and the time iteration method using the proposed resolving strategies can successfully obtain the plan satisfying all considered constraints.

Multi-Time Scale Operation and Simulation Strategy of the Park Based on Model Predictive Control

Due to the impact of source-load prediction power errors and uncertainties,the actual operation of the park will have a wide range of fluctuations compared with the expected state,resulting in its inability to achieve the expected economy.This paper constructs an operating simulation model of the park power grid operation considering demand response and proposes a multi-time scale operating simulation method that combines day-ahead optimization and model predictive control(MPC).In the day-ahead stage,an operating simulation plan that comprehensively considers the user’s side comfort and operating costs is proposed with a long-term time scale of 15 min.In order to cope with power fluctuations of photovoltaic,wind turbine and conventional load,MPC is used to track and roll correct the day-ahead operating simulation plan in the intra-day stage to meet the actual operating operation status of the park.Finally,the validity and economy of the operating simulation strategy are verified through the analysis of arithmetic examples.

Modelling and Analysis on Noisy Financial Time Series

Building the prediction model(s) from the historical time series has attracted many researchers in last few decades. For example, the traders of hedge funds and experts in agriculture are demanding the precise models to make the prediction of the possible trends and cycles. Even though many statistical or machine learning (ML) models have been proposed, however, there are no universal solutions available to resolve such particular problem. In this paper, the powerful forward-backward non-linear filter and wavelet-based denoising method are introduced to remove the high level of noise embedded in financial time series. With the filtered time series, the statistical model known as autoregression is utilized to model the historical times aeries and make the prediction. The proposed models and approaches have been evaluated using the sample time series, and the experimental results have proved that the proposed approaches are able to make the precise prediction very efficiently and effectively.

Bayesian Joint Modelling of Survival Time and Longitudinal CD4 Cell Counts Using Accelerated Failure Time and Generalized Error Distributions

Survival of HIV/AIDS patients is crucially dependent on comprehensive and targeted medical interventions such as supply of antiretroviral therapy and monitoring disease progression with CD4 T-cell counts. Statistical modelling approaches are helpful towards this goal. This study aims at developing Bayesian joint models with assumed generalized error distribution (GED) for the longitudinal CD4 data and two accelerated failure time distributions, Lognormal and loglogistic, for the survival time of HIV/AIDS patients. Data are obtained from patients under antiretroviral therapy follow-up at Shashemene referral hospital during January 2006-January 2012 and at Bale Robe general hospital during January 2008-March 2015. The Bayesian joint models are defined through latent variables and association parameters and with specified non-informative prior distributions for the model parameters. Simulations are conducted using Gibbs sampler algorithm implemented in the WinBUGS software. The results of the analyses of the two different data sets show that distributions of measurement errors of the longitudinal CD4 variable follow the generalized error distribution with fatter tails than the normal distribution. The Bayesian joint GED loglogistic models fit better to the data sets compared to the lognormal cases. Findings reveal that patients’ health can be improved over time. Compared to the males, female patients gain more CD4 counts. Survival time of a patient is negatively affected by TB infection. Moreover, increase in number of opportunistic infection implies decline of CD4 counts. Patients’ age negatively affects the disease marker with no effects on survival time. Improving weight may improve survival time of patients. Bayesian joint models with GED and AFT distributions are found to be useful in modelling the longitudinal and survival processes. Thus we recommend the generalized error distributions for measurement errors of the longitudinal data under the Bayesian joint modelling. Further studies may investigate the models with various types of shared random effects and more covariates with predictions.

Refinement of the Proposed Gamma-Ray Burst Time Delay Model

This paper is the second instalment in our study of the observed time delay in the arrival times of radio photons emanating from Gamma Ray Bursts (GRBs). The mundane assumption in contemporary physics as to the cause of these pondersome time delays is that they are a result of the photon being endowed with a non-zero mass. While we do not rule out the possibility of a non-zero mass for the photon, our working assumption is that the major cause of these time delays may very well be that these photons are travelling in a rarefied cosmic plasma in which the medium’s electrons interact with the electric component of the Photon, thus generating tiny currents that lead to dispersion, hence, a frequency-dependent speed of Light (FDSL). In the present instalment, we “improve” on the model presented in the first instalment by dropping the assumption that the resultant pairs of these radio photons leave the shock front simultaneously. The new assumption of a non-simultaneous— albeit systematic—emission of these photon pairs allows us to obtain a much more convincing and stronger correlation in the time delay. This new correlation allows us to build a unified model for the four GRBs in our sample using a relative distance correction mechanism. The new unified model allows us to obtain as our most significant result a value for the frequency equivalence of the interstellar medium (ISM)’s conductance ν* ~ 1.500 ± 0.009 Hzand also an independent distance measure to the GRBs where we obtain for our four GRB samples an average distance of: ~69.40 ± 0.10, 40.00 ± 0.00, 58.40 ± 0.40, and 86.00 ± 1.00 Mpc, for GRB 030329, 980425, 000418 and 021004 respectively.

Application of a 2 Parameter Weibull Distribution in Modeling of State Holding Time in HIV/AIDS Transition Dynamics

This study investigates the application of the two-parameter Weibull distribution in modeling state holding times within HIV/AIDS progression dynamics. By comparing the performance of the Weibull-based Accelerated Failure Time (AFT) model, Cox Proportional Hazards model, and Survival model, we assess the effectiveness of these models in capturing survival rates across varying gender, age groups, and treatment categories. Simulated data was used to fit the models, with model identification criteria (AIC, BIC, and R2) applied for evaluation. Results indicate that the AFT model is particularly sensitive to interaction terms, showing significant effects for older age groups (50 - 60 years) and treatment interaction, while the Cox model provides a more stable fit across all age groups. The Survival model displayed variability, with its performance diminishing when interaction terms were introduced, particularly in older age groups. Overall, while the AFT model captures the complexities of interactions in the data, the Cox model’s stability suggests it may be better suited for general analyses without strong interaction effects. The findings highlight the importance of model selection in survival analysis, especially in complex disease progression scenarios like HIV/AIDS.

Time Predictable Modeling Method for GPU Architecture with SIMT and Cache Miss Awareness

Graphics Processing Units(GPUs)are used to accelerate computing-intensive tasks,such as neural networks,data analysis,high-performance computing,etc.In the past decade or so,researchers have done a lot of work on GPU architecture and proposed a variety of theories and methods to study the microarchitectural characteristics of various GPUs.In this study,the GPU serves as a co-processor and works together with the CPU in an embedded real-time system to handle computationally intensive tasks.It models the architecture of the GPU and further considers it based on some excellent work.The SIMT mechanism and Cache-miss situation provide a more detailed analysis of the GPU architecture.In order to verify the GPU architecture model proposed in this article,10 GPU kernel_task and an Nvidia GPU device were used to perform experiments.The experimental results showed that the minimum error between the kernel task execution time predicted by the GPU architecture model proposed in this article and the actual measured kernel task execution time was 3.80%,and the maximum error was 8.30%.

Study on time effect and prediction model of shear strength of root-soil complex under dry-wet cycle

Triaxial compression tests were conducted on the alfalfa root-loess complex at different growthperiods obtained through artificial planting.The research focused on analyzing the time variation law of the shear strength index and deformation index of the alfalfa root-loess complex under dry-wet cycles.Additionally,the time effect of the shear strength index of the alfalfa root-loess complex under dry-wet cycles was analyzed and its prediction model was proposed.The results show that the PG-DWC(dry-wet cycle caused by plant water management during plant growth period)causes the peak strength of plain soil to change in a"V"shape with the increase of growth period,and the peak strength of alfalfa root-loess complex is higher than that of plain soil at the same growth period.The deterioration of the peak strength of alfalfa root-loess complex in the same growth period is aggravated with the increase of drying and wetting cycles.Compared with the 0 days growth period,the effective cohesion of alfalfa root-loess complex under different dry-wet cycles maximum increase rate is at the 180 days,which are 33.88%,46.05%,30.12%and 216.02%,respectively.When the number of dry-wet cycles is constant,the effective cohesion of the alfalfa root-loess complex overall increases with the growth period.However,it gradually decreases comparedwith the previous growth period,and the minimum increase rate are all at the 180 days.For the same growth period,the effective cohesion of the alfalfa root-loess complex decreases with the increase of the number of dry-wet cycles.This indicates that EC-DWC(the dry-wet cycles caused by extreme natural conditions such as continuous rain)have a detrimental effect on the time effect of the shear strength of the alfalfa root-loess complex.Finally,based on the formula of total deterioration,a prediction model for the shear strength of the alfalfa root-loess complex under dry-wet cycles was proposed,which exhibits high prediction accuracy.The research results provide useful guidance for the understanding of mechanical behavior and structural damage evolution of root-soil composite.

Stochastic Characteristics and Modelling of Monthly Rainfall Time Series of Ilorin, Nigeria

The analysis of time series is essential for building mathematical models to generate synthetic hydrologic records, to forecast hydrologic events, to detect intrinsic stochastic characteristics of hydrologic variables as well to fill missing and extend records. To this end, this paper examined the stochastic characteristics of the monthly rainfall series of Ilorin, Nigeria vis-à-vis modelling of same using four modelling schemes. The Decomposition, Square root transformation-deseasonalisation, Composite, and Periodic Autoregressive (T-F) modelling schemes were adopted. Results of basic analysis of the stochastic characteristics revealed that the monthly series does not show any discernible presence of long-term trend, though there is a seeming inter-decadal annual variation. The series exhibits strong seasonality throughout its length, both in the moments and autocorrelation and significantly intermittent. Based on assessment of the respective models, the performance of the different modelling schemes can be expressed in this order: T-F > Composite > Square root transformation-Deseasonalised > Decomposition. Considering the results obtained, modelling of monthly rainfall series in the presence of serial correlation between months should be based on the establishment of conditional probability framework. On the other hand, in view of the inadequacy of these modelling schemes, because of the autoregressive model components in the coupling protocol, nonlinear deterministic methods such as Artificial Neural Network, Wavelet models could be viable complements to the linear stochastic framework.

Development of a Modelling Script of Time Series Suitable for Data Mining

Data Mining has become an important technique for the exploration and extraction of data in numerous and various research projects in different fields (technology, information technology, business, the environment, economics, etc.). In the context of the analysis and visualisation of large amounts of data extracted using Data Mining on a temporary basis (time-series), free software such as R has appeared in the international context as a perfect inexpensive and efficient tool of exploitation and visualisation of time series. This has allowed the development of models, which help to extract the most relevant information from large volumes of data. In this regard, a script has been developed with the goal of implementing ARIMA models, showing these as useful and quick mechanisms for the extraction, analysis and visualisation of large data volumes, in addition to presenting the great advantage of being applied in multiple branches of knowledge from economy, demography, physics, mathematics and fisheries among others. Therefore, ARIMA models appear as a Data Mining technique, offering reliable, robust and high-quality results, to help validate and sustain the research carried out.

Examine the Reliability of Econometrics Software: An Empirical Comparison of Time Series Modelling

Researchers must understand that naively relying on the reliability of statistical software packages may result in suboptimal, biased, or erroneous results, which affects applied economic theory and the conclusions and policy recommendations drawn from it. To create confidence in a result, several software packages should be applied to the same estimation problem. This study examines the results of three software packages (EViews, R, and Stata) in the analysis of time-series econometric data. The time-series data analysis which presents the determinants of macroeconomic growth of Sri Lanka from 1978 to 2020 has been used. The study focuses on testing for stationarity, cointegration, and significant relationships among the variables. The Augmented Dickey-Fuller and Phillips Perron tests were employed in this study to test for stationarity, while the Johansen cointegration test was utilized to test for cointegration. The study employs the vector error correction model to assess the short-run and long-term dynamics of the variables in an attempt to determine the relationship between them. Finally, the Granger Causality test is employed in order to examine the linear causation between the concerned variables. The study revealed that the results produced by three software packages for the same dataset and the same lag order vary significantly. This implies that time series econometrics results are sensitive to the software that is used by the researchers while providing different policy implications even for the same dataset. The present study highlights the necessity of further analysis to investigate the impact of software packages in time series analysis of economic scenarios.

Modelling time series properties of Australian lending interest rates

The purpose of this paper is to examine the time series properties of Australian residential mortgage interest rates, and in doing so, establish whether or not selected home loan rates （product-level monthly home loan interest rates for CBA） exhibit the expected cyclical and seasonal variations and whether seasonality, if present, is stochastic or deterministic. In particular, due to a well established presence of cyclicality in financial markets＇ interest rates and strong correlation between financial markets＇ interest rates and home loan interest rates, the paper presumes that cyclicality is also to be found in home loan interest rates. Furthermore, the paper tests the hypothesis that home loan interest rates, for selected products, exhibit the three identified （＂Spring＂, ＂Autumn＂ and ＂The end of the Financial Year＂） season-related interest rate reductions. The paper uses a structural time series modelling approach and product-level home loan interest rates data from one of the biggest banks in Australia, Commonwealth Bank of Australia （CBA）. As expected, the results overall confirm the existence of cyclicality in home loan interest rates. With respect to the seasonality of home loan interest rate, although most of the analysed variables show the presence of statistically significant seasonal factors, the majority of the statistically significant seasonal factors observed cannot be attributed to any of the three considered seasonal effects.

Observations and modelling of the travel time delay and leading negative phase of the 16 September 2015 Illapel,Chile tsunami

The systematic discrepancies in both tsunami arrival time and leading negative phase(LNP)were identified for the recent transoceanic tsunami on 16 September 2015 in Illapel,Chile by examining the wave characteristics from the tsunami records at 21 Deep-ocean Assessment and Reporting of Tsunami(DART)sites and 29 coastal tide gauge stations.The results revealed systematic travel time delay of as much as 22 min(approximately 1.7%of the total travel time)relative to the simulated long waves from the 2015 Chilean tsunami.The delay discrepancy was found to increase with travel time.It was difficult to identify the LNP from the near-shore observation system due to the strong background noise,but the initial negative phase feature became more obvious as the tsunami propagated away from the source area in the deep ocean.We determined that the LNP for the Chilean tsunami had an average duration of 33 min,which was close to the dominant period of the tsunami source.Most of the amplitude ratios to the first elevation phase were approximately 40%,with the largest equivalent to the first positive phase amplitude.We performed numerical analyses by applying the corrected long wave model,which accounted for the effects of seawater density stratification due to compressibility,self-attraction and loading(SAL)of the earth,and wave dispersion compared with observed tsunami waveforms.We attempted to accurately calculate the arrival time and LNP,and to understand how much of a role the physical mechanism played in the discrepancies for the moderate transoceanic tsunami event.The mainly focus of the study is to quantitatively evaluate the contribution of each secondary physical effect to the systematic discrepancies using the corrected shallow water model.Taking all of these effects into consideration,our results demonstrated good agreement between the observed and simulated waveforms.We can conclude that the corrected shallow water model can reduce the tsunami propagation speed and reproduce the LNP,which is observed for tsunamis that have propagated over long distances frequently.The travel time delay between the observed and corrected simulated waveforms is reduced to<8 min and the amplitude discrepancy between them was also markedly diminished.The incorporated effects amounted to approximately 78%of the travel time delay correction,with seawater density stratification,SAL,and Boussinesq dispersion contributing approximately 39%,21%,and 18%,respectively.The simulated results showed that the elastic loading and Boussinesq dispersion not only affected travel time but also changed the simulated waveforms for this event.In contrast,the seawater stratification only reduced the tsunami speed,whereas the earth’s elasticity loading was responsible for LNP due to the depression of the seafloor surrounding additional tsunami loading at far-field stations.This study revealed that the traditional shallow water model has inherent defects in estimating tsunami arrival,and the leading negative phase of a tsunami is a typical recognizable feature of a moderately strong transoceanic tsunami.These results also support previous theory and can help to explain the observed discrepancies.

DELAY-TIME MODEL BASED ON IMPERFECT INSPECTION OF AIRCRAFT STRUCTURE WITHIN FINITE TIME SPAN

According to the failure characteristics of aircraft structure, a delay-time model is an effective method to optimize maintenance for aircraft structure. To imitate the practical situation as much as possible, imperfect inspections, thresholds and repeated intervals are concerned in delay-time models. Since the suggestion by the existing delay-time models that the inspections are implemented in an infinite time span lacks practical value, a de- lay-time model with imperfect inspection within a finite time span is proposed. In the model, the nonhomogenous Poisson process is adopted to obtain the renewal probabilities between two different successive inspections on de- fects or failures. An algorithm is applied based on the Nelder-Mead downhill simplex method to solve the model. Finally, a numerical example proves the validity and effectiveness of the model.

TIME SERIES NEURAL NETWORK MODEL FOR HYDROLOGIC FORECASTING

Time series analysis plays an important role in hydrologic forecasting,while the key to this analysis is to establish a proper model.This paper presents a time series neural network model with back propagation procedure for hydrologic forecasting.Free from the disadvantages of previous models,the model can be parallel to operate information flexibly and rapidly.It excels in the ability of nonlinear mapping and can learn and adjust by itself,which gives the model a possibility to describe the complex nonlinear hydrologic process.By using directly a training process based on a set of previous data, the model can forecast the time series of stream flow.Moreover,two practical examples were used to test the performance of the time series neural network model.Results confirm that the model is efficient and feasible.

Mathematical model for precursor gas residence time in isothermal CVD process of C/C composites

In the chemical vapor deposition（CVD） process of C/C composites,the dynamics and mechanism of precursor gas flowing behavior were analyzed mathematically,in which the precursor gas was infiltrated by the pressure difference of the gas flowing through felt.Differential equations were educed which characterized the relations among the pressure inside the felt,the pressure outside the felt of the precursor gas and the porosity of the felt as a function of CVD duration.The gas residence time during the infiltration process through the felt was obtained from the differential equations.The numerical verification is in good agreement with the practical process,indicating the good reliability of the current mathematical model.