Comparison of two statistical methods for handling missing values of quantitative data in Bayesian N-of-1 trials: a simulation study

Background:Missing data are frequently occurred in clinical studies.Due to the development of precision medicine,there is an increased interest in N-of-1 trial.Bayesian models are one of main statistical methods for analyzing the data of N-of-1 trials.This simulation study aimed to compare two statistical methods for handling missing values of quantitative data in Bayesian N-of-1 trials.Methods:The simulated data of N-of-1 trials with different coefficients of autocorrelation,effect sizes and missing ratios are obtained by SAS 9.1 system.The missing values are filled with mean filling and regression filling respectively in the condition of different coefficients of autocorrelation,effect sizes and missing ratios by SPSS 25.0 software.Bayesian models are built to estimate the posterior means by Winbugs 14 software.Results:When the missing ratio is relatively small,e.g.5%,missing values have relatively little effect on the results.Therapeutic effects may be underestimated when the coefficient of autocorrelation increases and no filling is used.However,it may be overestimated when mean or regression filling is used,and the results after mean filling are closer to the actual effect than regression filling.In the case of moderate missing ratio,the estimated effect after mean filling is closer to the actual effect compared to regression filling.When a large missing ratio(20%)occurs,data missing can lead to significantly underestimate the effect.In this case,the estimated effect after regression filling is closer to the actual effect compared to mean filling.Conclusion:Data missing can affect the estimated therapeutic effects using Bayesian models in N-of-1 trials.The present study suggests that mean filling can be used under situation of missing ratio≤10%.Otherwise,regression filling may be preferable.

Computer simulation studies of pulsed Doppler signals from vortices

A computer simulation method for pulsed Doppler signals from vortices was pro- posed to generate simulated vortex Doppler signals under various given circumstances. The relative waveforms, such as the maximum frequency waveform, the mean frequency waveform and the bandwidth waveform, were obtained using the short time Fourier analysis of those simulated signals. The relations were studied between several spectrum parameters obtained from these waveforms and given simulation conditions, such as the position and the size of the sample volume, the distance between two vortices, the free stream velocity and the maximum tangent velocity of the vortex. The sensitive parameters were found to detect vortices using the pulsed Doppler techniques.

Simulation study on factors influencing the entrainment behavior of liquid steel as bubbles pass through the steel/slag interface

In this study, a water/silicone oil interface was used to simulate the steel/slag interface in a converter. A high-speed camera was used to record the entrainment process of droplets when air bubbles were passed through the water/silicone oil interface. Motion parameters of the bubbles and droplets were obtained using particle kinematic analysis software, and the entrainment rate of the droplets was calculated. It was found that the entrainment rate decreased from 29.5% to 0 when the viscosity of the silicone oil was increased from 60 mPa.s to 820 mPa.s in the case of bubbles with a 5 mm equivalent diameter passing through the water/silicone oil interface. The results indicate that in- creasing the viscosity of the silicone oil is conducive to reducing the entrainment rate. The entrainment rate increased from 0 to 136.3% in the case of silicone oil with a viscosity of 60 mPa.s when the equivalent diameter of the bubbles was increased from 3 mm to 7 ram. We there- fore conclude that small bubbles are also conductive to reducing the entrainment rate. The force analysis results for the water colmnn indicate that the entrainment rate of droplets is affected by the velocity of the bubble passing through the water/silicone oil interface and that the en- trainment rate decreases with the bubble velocity.

Study on Numerical Simulation of Mold Filling and Heat Transfer in Die Casting Process

A 3-D mathematical model considering turbulence phenomena has been established based on a computational fluid dynamics technique, so called 3-D SOLA-VOF (Solution Algorithm-Volume of Fluid), to simulate the fluid flow of mold filling process of die casting. In addition, the mathematical model for simulating the heat transfer in die casting process has also been established. The computation program has been developed by the authors with the finite difference method (FDM) recently. As verification, the mold filling process of a S-shaped die casting has been simulated and the simulation results coincide with that of the benchmark test. Finally, as a practical application, the gating design of a motorcycle component was modified by the mold filling simulation and the dies design of another motorcycle component was optimized by the heat transfer simulation. All the optimized designs were verified by the production practice.

Using agent-based simulation to assess diseaseprevention measures during pandemics

Despite the growing interest in macroscopic epidemiological models to deal with threats posed by pandemics such as COVID-19,little has been done regarding the assessment of disease spread in day-to-day life,especially within buildings such as supermarkets where people must obtain necessities at the risk of exposure to disease.Here,we propose an integrated customer shopping simulator including both shopper movement and choice behavior,using a force-based and discrete choice model,respectively.By a simple extension to the force-based model,we implement the following preventive measures currently taken by supermarkets;social distancing and one-way systems,and different customer habits,assessing them based on the average individual disease exposure and the time taken to complete shopping(shopping efficiency).Results show that maintaining social distance is an effective way to reduce exposure,but at the cost of shopping efficiency.We find that the one-way system is the optimal strategy for reducing exposure while minimizing the impact on shopping efficiency.Customers should also visit supermarkets less frequently,but buy more when they do,if they wish to minimize their exposure.We hope that this work demonstrates the potential of pedestrian dynamics simulations in assessing preventative measures during pandemics,particularly if it is validated using empirical data.

Study and Application of Mold Filling Simulation of Shaped Castings

In this paper, an algorithm for simulating fluid flow and heat transfer for mold filling of shaped castings is presented. The main features of the algorithm include: 1) a simple but practical technique based on VOF method to determine free surface, 2) an explicit scheme of enthalpy to solve the energy equation more efficiently, and 3) an effective treatment to modify the flux deviation due to pressure iteration. In order to verify these methods, well controlled experiments have been repeatedly done with both water analog and gray iron pouring experiments to record the flow patterns and temperature variations. The calculated results are in accordance with the experimental ones. For the applications, the simulated initial temperature distribution right after mold filling was used to analyse subsequent solidification and to predict shrinkage defects. Actual castings were poured and tested in a foundry plant. The reuslts show that the defects predication with considering fluid flow effects is more precise than that without considering the effects.

Dynamic Simulation and Valve Structure Optimization of an Electro-Hydraulic Clutch Shift Control System in an Automatic Transmission

The electro-hydraulic clutch control system controls the transferred torque of gear-shifting clutches in clutch-to-clutch transmissions. A nonlinear dynamic model of an electro-hydraulic clutch shift control system is presented. The mechanical and fluid subsystems of all valves are investigated, including their interactions. Model validation of the electro-hydraulic valve system is performed by comparing the simulated and measured pressure curves. The dynamic characteristics of the electro-hydraulic clutch shift control system with different supply pressures and different fluid temperatures are simulated and evaluated. It is found that pipes which are often ignored between the electro-hydraulic valve system and the clutch piston,have strong influence on clutch piston chamber pressures. In order to satisfy the required time and reduce the fluctuation of the clutch piston chamber pressures,the orifices' diameters and valve structure are optimized.

A Study on Numerical Simulation of Core-Shooting Process

In order to evaluate the main factors influencing the core-shooting process and to optimize the design of core boxes, the fluid-particle (air-sand) model has been built based on the two-phase flow theory. The fluid phase, air, and the particulate phase, sand granules, have been treated as a continuum. By using this model, it is possible to simulate the flow and compaction behavior of sand particles during the core-shooting process. To benchmark the calculated results, the shooting process has been recorded with a digital high speed camera, and the inlet condition of sand particles has also been achieved by using the camera. The preliminary results have showed that the calculation is in agreement with the testing results.

Impact of Temperature Variation on Performance of Carbon Nanotube Field-Effect Transistor-Based on Chaotic Oscillator:A Quantum Simulation Study

We evaluate the impact of temperature on the output behavior of a carbon nanotube field effect transistor （CNFET） based chaotic generator. The sources cause the variations in both current-voltage characteristics of the CNFET device and an overall chaotic circuit is pointed out. To verify the effect of temperature variation on the output dynamics of the chaotic circuit, a simulation is performed by employing the CNFET compact model of Wong et al. in HSPICE with a temperature range from -100℃ to 100℃. The obtained results with time series, frequency spectra, and bifurcation diagram from the simulation demonstrate that temperature plays a significant role in the output dynamics of the CNFET-based chaotic circuit. Thus, temperature-related issues should be taken into account while designing a high-quality chaotic generator with high stability.

Simulation Study on the ITB Formation during LHCD in JT-60U

A transport simulation has been done by using a i.5D time dependent transportcode to reproduce a formation of the ITB on electron temperature Profile during the long pu1seLHCD in JT-60U tokamak. The transport coefficients were assumed to reduce with a reversedmagnetic shear and the LH driven current profile was evaluated by fitting dynamic change inthe measured current profile. The gradual increase in the central electron temperature could beexplained by the change in the current profile during LHCD in the present simulation model. Theestimated LH-driven current profile by the transport code analysis shows a finite current densityat the plasma center. Analysis of transport simulation suggests some mechanisms for broadeningthe LH-driven current profile at the central region of the plasma.

Simulation Study on the Controllable Dielectrophoresis Parameters of Graphene

The method of using dielectrophoresis （DEP） to assemble graphene between micro-electrodes has been proven to be simple and efficient. We present an optimization method for the kinetic formula of graphene DEP, and discuss the simulation of the graphene assembly process based on the finite element method. The simulated results illustrate that the accelerated motion of graphene is in agreement with the distribution of the electric field squared gradient. We also conduct research on the controllable parameters of the DEP assembly such as the alternating current （AC） frequency, the shape of micro-electrodes, and the ratio of the gap between electrodes to the characteristic/geometric length of graphene （λ）. The simulations based on the Clausius-Mossotti factor reveal that both graphene velocity and direction are influenced by the AC frequency. When graphene is close to the electrodes, the shape of micro-electrodes will exert great influence on the velocity of graphene. Also, λ has a great influence on the velocity of graphene. Generally, the velocity of graphene would be greater when λ is in the range of 0.4 0.6. The study is of a theoretical guiding significance in improving the precision and efficiency of the graphene DEP assembly.

A simulation study on CO_2 assimilation and crop growth in agroforest ecosystems in the East China Plain

ＡｓｉｍｕｌａｔｉｏｎｓｔｕｄｙｏｎＣＯ２ａｓｉｍｉｌａｔｉｏｎａｎｄｃｒｏｐｇｒｏｗｔｈｉｎａｇｒｏｆｏｒｅｓｔｅｃｏｓｙｓｔｅｍｓｉｎｔｈｅＥａｓｔＣｈｉｎａＰｌａｉｎＺｈｏｕＬｉｎ，ＷａｎｇＨａｎｊｉｅＥｃｏＥｎｖｉｒｏｎｍｅｎｔａｌＲｅｓｅ...

A Simulation Study on Boiler Fault Diagnosis Expert System

The boiler is one of the key parts of power station. Monitoring the operation of the boiler automatically, continuously and accurately is very important. In case of malfunctions, an expert system might be helpful to find the sources of faults, and thereafter will assist the operator to act correctly and to prevent the fault from further development.In this paper, a simulation study of an expert system, which can diagnose the running faults of boiler system, is conducted. Shallow and deep two-layer knowledge base structure is used in this two layers of knowledge base. So. this kind of expert system can overcome the disadvantages of the old types of expert systems, which are only able to deal with experience rules base.A simulator of power boiler is used for the research on fault diagnose expert system. This boiler simulator gives all necessary information for the diagnosis. The knowledge base of expert system is built in hierarchies, so the speed of reasoning may be invreased. After the reasoning the

Vertical Profile Comparison of Aerosol and Cloud Optical Properties in Dominated Dust and Smoke Regions over Africa Based on Space-Based Lidar

This study evaluates the vertical profiles of aerosol and cloud optical properties in 40 dominated dust and smoke regions in Western-Northern Africa (WNA) and Central-Southern Africa (CSA), respectively, from the surface to 10km and from 2008 to 2011 based on LIVAS (LIdar climatology of Vertical Aerosol Structure for space-based lidar simulation studies). Aerosol extinction (AE), aerosol backscatter (AB), and aerosol depolarization (AD) generally increase from the surface to 1.2 km and decrease from 1.2 km to the upper layers in both WNA and CSA. AE and AB in CSA (maximum of 0.13 km-1, 0.14 km-1, 0.0021 km-1‧sr-1, 0.0033 km-1‧sr-1) are higher than in WNA (maximum of 0.07 km-1, 0.08 km-1, 0.0017 km-1‧sr-1, 0.0015 km-1‧sr-1) at 532 and 1064 nm respectively. AD in WNA (maximum of 0.25) is significantly higher than in CSA (maximum of 0.05). There is a smooth change with the height of cloud extinction and backscatter in WNA and CSA, while there is a remarkable increase of cloud depolarization with height, whereby it is high in CSA and low in WNA due to high and low fraction of cirrus respectively. Altocumulus has the highest extinction in NA (0.0139 km-1), CA (0.058 km-1), WA (0.013 km-1), while low overcast transparent (0.76 km-1) below 1 km in SA. The major findings of this study may contribute to the improvement of our understanding of aerosol-cloud interaction studies in dominated dust and smoke aerosol regions.

Iterative Learning Control Algorithm with a Fixed Step

Iterative Learning Control （ILC） captures interests of many scholars because of its capability of high precision control implement without identifying plant mathematical models, and it is widely applied in control engineering. Presently, most ILC algorithms still follow the original ideas of ARIMOTO, in which the iterative-learning-rate is composed by the control error with its derivative and integral values. This kind of algorithms will result in inevitable problems such as huge computation, big storage capacity for algorithm data, and also weak robust. In order to resolve these problems, an improved iterative learning control algorithm with fixed step is proposed here which breaks the primary thought of ARIMOTO. In this algorithm, the control step is set only according to the value of the control error, which could enormously reduce the computation and storage size demanded, also improve the robust of the algorithm by not using the differential coefficient of the iterative learning error. In this paper, the convergence conditions of this proposed fixed step iterative learning algorithm is theoretically analyzed and testified. Then the algorithm is tested through simulation researches on a time-variant object with randomly set disturbance through calculation of step threshold value, algorithm robustness testing,and evaluation of the relation between convergence speed and step size. Finally the algorithm is validated on a valve-serving-cylinder system of a joint robot with time-variant parameters. Experiment results demonstrate the stability of the algorithm and also the relationship between step value and convergence rate. Both simulation and experiment testify the feasibility and validity of the new algorithm proposed here. And it is worth to noticing that this algorithm is simple but with strong robust after improvements, which provides new ideas to the research of iterative learning control algorithms.

Quantifying the effect of persistent dryer climates on forest productivity and implications for forest planning:a case study in northern Germany

Background: Forest management decisions are based on expectations of future developments. For sound decisions it is essential to accurately predict the expected values in future developments and to account for their inherent uncertainty,for example the impact of climate change on forests. Changing climatic conditions affect forest productivity and alter the risk profile of forests and forest enterprises. Intensifying drought stress is seen as one major risk factor threatening forest management in the north German lowlands. Drought stress reduces tree growth and vitality and might even trigger mortality. But so far, it is not possible to quantify effects of a persistent dryer climate on forest productivity at a level suitable for forest management.Methods: We apply a well-established single-tree forest growth simulator to quantify the effect of persistent dryer climates on future forest productivity. We analyse the growth of Scots pine(Pinus sylvestris L.), European beech(Fagus sylvatico L.) and oak(Quercus robur L. and Quercus petraea(Matt.) Liebl.) in two forest regions in the north German lowlands for a time interval of 60 years until 2070. The growth response under three different climate projections is compared to a baseline scenario.Results: The results show clear differences in volume increment to persistent dryer climates between tree species. The findings exhibit regional differences and temporal trends. While mean annual increment at biological rotation age of Scots pine and oak predominantly benefits from the projected climate conditions until 2070, beech might suffer losses of up to 3 m^3·ha^(-1)yr^(-1) depending on climate scenario and region. However, in the projection period2051 to 2070 the uncertainty ranges comprise positive as well as negative climatic effects for all species.Conclusions: The projected changes in forest growth serve as quantitative contributions to provide decision support in the evaluation of, for example, species future site suitability and timber supply assessments. The analysis of productivity changes under persistent dryer climate complements the drought vulnerability assessment which is applied in practical forestry in northwestern Germany today. The projected species' productivity has strong implications for forest management and the inherent uncertainty needs to be accounted for.

Numerical and Experimental Study of the 3D Effect on Connecting Arm of Vertical Axis Tidal Current Turbine

Vertical axis tidal current turbine is a promising device to extract energy from ocean current. One of the important components of the turbine is the connecting arm, which can bring about a significant effect on the pressure distribution along the span of the turbine blade, herein we call it 3D effect. However, so far the effect is rarely reported in the research, moreover, in numerical simulation. In the present study, a 3D numerical model of the turbine with the connecting arm was developed by using FLUENT software compiling the UDF（User Defined Function） command. The simulation results show that the pressure distribution along the span of blade with the connecting arm model is significantly different from those without the connecting arm. To facilitate the validation of numerical model, the laboratory experiment has been carried out by using three different types of NACA aerofoil connecting arm and circle section connecting arm. And results show that the turbine with NACA0012 connecting arm has the best start-up performance which is 0.346 m/s and the peak point of power conversion coefficient is around 0.33. A further study has been performed and a conclusion is drawn that the aerofoil and thickness of connecting arm are the most important factors on the power conversion coefficient of the vertical axis tidal current turbine.