A deep reinforcement learning approach to gasoline blending real-time optimization under uncertainty

The gasoline inline blending process has widely used real-time optimization techniques to achieve optimization objectives,such as minimizing the cost of production.However,the effectiveness of real-time optimization in gasoline blending relies on accurate blending models and is challenged by stochastic disturbances.Thus,we propose a real-time optimization algorithm based on the soft actor-critic(SAC)deep reinforcement learning strategy to optimize gasoline blending without relying on a single blending model and to be robust against disturbances.Our approach constructs the environment using nonlinear blending models and feedstocks with disturbances.The algorithm incorporates the Lagrange multiplier and path constraints in reward design to manage sparse product constraints.Carefully abstracted states facilitate algorithm convergence,and the normalized action vector in each optimization period allows the agent to generalize to some extent across different target production scenarios.Through these well-designed components,the algorithm based on the SAC outperforms real-time optimization methods based on either nonlinear or linear programming.It even demonstrates comparable performance with the time-horizon based real-time optimization method,which requires knowledge of uncertainty models,confirming its capability to handle uncertainty without accurate models.Our simulation illustrates a promising approach to free real-time optimization of the gasoline blending process from uncertainty models that are difficult to acquire in practice.

A virtual 3D interactive painting method for Chinese calligraphy and painting based on real-time force feedback technology

A novel 3D interactive painting method for Chinese calligraphy and painting based on force feedback technology is proposed. The relationship between the force exerted on the brush and the resulting brush deformation is analyzed and a spring-mass model is used to build a model of the 3D Chinese brush. The 2D brush footprint between the brush and the plane of the paper or object is calculated according to the deformation of the 3D brush when force is exerted on the 3D brush. Then the 3D brush footprint is obtained by projecting the 2D brush footprint onto the surface of the 3D object in real time, and a complete 3D brushstroke is obtained by superimposing 3D brush footprints along the painting direction. The proposed method has been suc- cessfully applied in a virtual 3D interactive drawing system based on force feedback technology. In this system, users can paint 3D brushstrokes in real time with a Phantom Desktop haptic device, which can effectively serve as a virtual reality interface to the simulated painting environment for users.

COMPENSATION CONTROL OF REAL-TIME UNBALANCE FORCE FOR ACTIVE MAGNETIC BEARING SYSTEM

Aiming at the stability and others properties of active magnetic bearing （AMB） system influenced by the periodic unbalance stimulation synchronous with rotor rotational speed, a new real-time adaptive feed-forward unbalance force compensation scheme is proposed based on variable step-size least mean square（LMS） algorithm as the feed-forward compensation controller. The controller can provide some suitable sinusoidal signals to com- pensate the feedback unbalance response signals synchronous with the rotary frequency, then reduce the fluctua- tion of the control currents and weaken the active control of AMB system. The variable step-size proportional to the rotational frequency is deduced by analyzing the principle of normal LMS algorithm and its deficiency in the application of real-time filtering of AMB system. Experimental results show that the new method can implement real-time unbalance force compensation in a wide frequency band, reduce the effect of unbalance stimulant force on the housing of AMB system, and provide convenience to improve rotational speed.

Force and impulse multi-sensor based on flexible gate dielectric field effect transistor

Nowadays,force sensors play an important role in industrial production,electronic information,medical health,and many other fields.Two-dimensional material-based filed effect transistor(2D-FET)sensors are competitive with nano-level size,lower power consumption,and accurate response.However,few of them has the capability of impulse detection which is a path function,expressing the cumulative effect of the force on the particle over a period of time.Herein we fabricated the flexible polymethyl methacrylate(PMMA)gate dielectric MoS_(2)-FET for force and impulse sensor application.We systematically investigated the responses of the sensor to constant force and varying forces,and achieved the conversion factors of the drain current signals(I_(ds))to the detected impulse(I).The applied force was detected and recorded by I_(ds)with a low power consumption of~30 nW.The sensitivity of the device can reach~8000%and the 4×1 sensor array is able to detect and locate the normal force applied on it.Moreover,there was almost no performance loss for the device as left in the air for two months.

Restoring force correction based on online discrete tangent stiffness estimation method for real-time hybrid simulation

In real-time hybrid simulation（RTHS）, it is difficult if not impossible to completely erase the error in restoring force due to actuator response delay using existing displacement-based compensation methods. This paper proposes a new force correction method based on online discrete tangent stiffness estimation（online DTSE） to provide accurate online estimation of the instantaneous stiffness of the physical substructure. Following the discrete curve parameter recognition theory, the online DTSE method estimates the instantaneous stiffness mainly through adaptively building a fuzzy segment with the latest measurements, constructing several strict bounding lines of the segment and calculating the slope of the strict bounding lines, which significantly improves the calculation efficiency and accuracy for the instantaneous stiffness estimation. The results of both computational simulation and real-time hybrid simulation show that：（1） the online DTSE method has high calculation efficiency, of which the relatively short computation time will not interrupt RTHS; and（2） the online DTSE method provides better estimation for the instantaneous stiffness, compared with other existing estimation methods. Due to the quick and accurate estimation of instantaneous stiffness, the online DTSE method therefore provides a promising technique to correct restoring forces in RTHS.

Evolutionary-assisted reinforcement learning for reservoir real-time production optimization under uncertainty

Production optimization has gained increasing attention from the smart oilfield community because it can increase economic benefits and oil recovery substantially.While existing methods could produce high-optimality results,they cannot be applied to real-time optimization for large-scale reservoirs due to high computational demands.In addition,most methods generally assume that the reservoir model is deterministic and ignore the uncertainty of the subsurface environment,making the obtained scheme unreliable for practical deployment.In this work,an efficient and robust method,namely evolutionaryassisted reinforcement learning(EARL),is proposed to achieve real-time production optimization under uncertainty.Specifically,the production optimization problem is modeled as a Markov decision process in which a reinforcement learning agent interacts with the reservoir simulator to train a control policy that maximizes the specified goals.To deal with the problems of brittle convergence properties and lack of efficient exploration strategies of reinforcement learning approaches,a population-based evolutionary algorithm is introduced to assist the training of agents,which provides diverse exploration experiences and promotes stability and robustness due to its inherent redundancy.Compared with prior methods that only optimize a solution for a particular scenario,the proposed approach trains a policy that can adapt to uncertain environments and make real-time decisions to cope with unknown changes.The trained policy,represented by a deep convolutional neural network,can adaptively adjust the well controls based on different reservoir states.Simulation results on two reservoir models show that the proposed approach not only outperforms the RL and EA methods in terms of optimization efficiency but also has strong robustness and real-time decision capacity.

Velocity plus displacement equivalent force control for real-time substructure testing

This paper employs a velocity plus displacement（V＋D）-based equivalent force control（EFC） method to solve the velocity/displacement difference equation in a real-time substructure test. This method uses type 2 feedback control loops to replace mathematical iteration to solve the nonlinear dynamic equation. A spectral radius analysis of the amplification matrix shows that the type 2 EFC-explicit, Newmark-β method has beneficial numerical characteristics for this method. Its stability limit of Ω = 2 remains unchanged regardless of the system damping because the velocity is achieved with very high accuracy during simulation. In contrast, the stability limits of the central difference method using direct velocity prediction and the EFC-average acceleration method with linear interpolation are shown to decrease with an increase in system damping. In fact, the EFC-average acceleration method is shown to change from unconditionally stable to conditionally stable. We also show that if an over-damped system with a damping ratio of 1.05 is considered, the stability limit is reduced to Ω =1.45. Finally, the results from an experiment with a single-degree-of-freedom structure installed with a magneto-rheological（MR） damper are presented. The results demonstrate that the proposed method is able to follow both displacement and velocity commands with moderate accuracy, resulting in improved test performance and accuracy for structures that are sensitive to both velocity and displacement inputs. Although the findings of the study are promising, additional test data and several further improvements will be required to draw general conclusions.

Force Sensitive Resistors-Based Real-Time Posture Detection System Using Machine Learning Algorithms

To detect the improper sitting posture of a person sitting on a chair,a posture detection system using machine learning classification has been proposed in this work.The addressed problem correlates to the third Sustainable Development Goal(SDG),ensuring healthy lives and promoting well-being for all ages,as specified by the World Health Organization(WHO).An improper sitting position can be fatal if one sits for a long time in the wrong position,and it can be dangerous for ulcers and lower spine discomfort.This novel study includes a practical implementation of a cushion consisting of a grid of 3×3 force-sensitive resistors(FSR)embedded to read the pressure of the person sitting on it.Additionally,the Body Mass Index(BMI)has been included to increase the resilience of the system across individual physical variances and to identify the incorrect postures(backward,front,left,and right-leaning)based on the five machine learning algorithms:ensemble boosted trees,ensemble bagged trees,ensemble subspace K-Nearest Neighbors(KNN),ensemble subspace discriminant,and ensemble RUSBoosted trees.The proposed arrangement is novel as existing works have only provided simulations without practical implementation,whereas we have implemented the proposed design in Simulink.The results validate the proposed sensor placements,and the machine learning(ML)model reaches a maximum accuracy of 99.99%,which considerably outperforms the existing works.The proposed concept is valuable as it makes it easier for people in workplaces or even at individual household levels to work for long periods without suffering from severe harmful effects from poor posture.

ForceCT双能量与低kV扫描技术在头颈CTA中的对比分析

探索并比较Force CT双能量与低kV扫描技术在头颈CTA中的作用。方法 以系统随机化法,将2022.01-2023.01在我院接受头颈CTA检查的40例患者均分为两组各20例。其中观察组患者采用低kV扫描技术,对照组则进行Force CT双能量扫描。比较两组的辐射剂量、图像质量。结果 观察组的辐射剂量指数低于对照组,P<0.05。观察组的大脑中动脉与颈内动脉图像质量均高于对照组,P<0.05。结论 运用低kV扫描技术,能较好地降低头颈CTA检查时的辐射剂量,与此同时还能得到更为高质量的图像,帮助病情的更快诊断,患者能尽快得到个性化的治疗,早日恢复健康。

Force CT高级建模迭代重建算法对门静脉图像质量的影响

目的:探讨Force CT高级建模迭代重建算法(ADMIRE)对门静脉图像质量影响。方法:回顾性分析2022年1—4月于中山大学附属第八医院行上腹部CT平扫及增强的40例患者影像资料,在syngo.via后处理工作站重建,采用门脉期原始数据重建出滤波反投影算法(FBP)、ADMIRE1、ADMIRE3、ADMIRE5 4组不同重建算法的图像,在4组不同算法图像上测量门静脉干中心位置、肝实质及同层竖脊肌的CT值和噪声值,计算门静脉和肝实质的信噪比(SNR)和对比噪声比(CNR)。此外,由两位具有5年以上诊断经验的医生对图像进行主观评分。结果:随着迭代等级的增高,图像的CNR、SNR呈现上升的趋势,图像的噪声值呈现下降的趋势,其中ADMIRE5组图像中门静脉和肝脏具有最高的CNR和SNR且具有最低的噪声值,以及最高的主观评分,采用ADMIRE5重建算法的门静脉与周围组织对比非常好,门静脉5级分支显示良好。结论:高级建模迭代重建ADMIRE算法相比于FBP重建算法,降噪效果更好,门静脉的显示更加锐利,能提升门静脉的图像质量。

基于Real-time PCR法检测乳粉中牛源性成分定量研究

基于Real-timePCR建立了乳粉中牛源性成分相对定量检测方法,并对牛的特异性引物与探针进行了特异性、灵敏度和稳定性测试。通过模拟不同浓度牛乳粉与马乳粉混合样本,根据其△Ct值的函数关系进行线性拟合进而绘制标准曲线,建立乳粉中牛源性成分的相对定量检测。结果显示,该方法的最低检测限为0.00001 mg/mL,回收率为91.11%~119.2%,组间变异系数≤0.58%、组内变异系数≤1.44%。说明该方法在特异性与稳定性上适用于乳粉中牛源性成分及含量的掺假检测。

A multifunctional shear apparatus for rocks subjected to true triaxial stress and high temperature in real-time

Deep engineering disasters,such as rockbursts and collapses,are more related to the shear slip of rock joints.A novel multifunctional device was developed to study the shear failure mechanism in rocks.Using this device,the complete shearedeformation process and long-term shear creep tests could be performed on rocks under constant normal stiffness(CNS)or constant normal loading(CNL)conditions in real-time at high temperature and true-triaxial stress.During the research and development process,five key technologies were successfully broken through:(1)the ability to perform true-triaxial compressioneshear loading tests on rock samples with high stiffness;(2)a shear box with ultra-low friction throughout the entire stress space of the rock sample during loading;(3)a control system capable of maintaining high stress for a long time and responding rapidly to the brittle fracture of a rock sample as well;(4)a refined ability to measure the volumetric deformation of rock samples subjected to true triaxial shearing;and(5)a heating system capable of maintaining uniform heating of the rock sample over a long time.By developing these technologies,loading under high true triaxial stress conditions was realized.The apparatus has a maximum normal stiffness of 1000 GPa/m and a maximum operating temperature of 300C.The differences in the surface temperature of the sample are constant to within5C.Five types of true triaxial shear tests were conducted on homogeneous sandstone to verify that the apparatus has good performance and reliability.The results show that temperature,lateral stress,normal stress and time influence the shear deformation,failure mode and strength of the sandstone.The novel apparatus can be reliably used to conduct true-triaxial shear tests on rocks subjected to high temperatures and stress.

一种基于real-time PCR技术的TTV检测方法的建立及应用

目的:本研究旨在开发一种具有更高灵敏度和特异性的TTV检测技术,为揭示TTV在多种疾病过程中的作用提供重要的技术支持。方法:为了更精确、灵敏的检测TTV,本研究分析了目前公布的所有亚型的TTV基因序列,在此基础上建立了一种基于UTR区域的real-time PCR检测方法,并与文献报道应用较为广泛的PCR检测方法进行了对比。结果:本研究建立的方法在1×10^(7)~1×10^(1) copies/μL标准品浓度范围内具有良好的线性关系,相关系数为1.000,斜率为-3.446,检测下限为1×10^(1) copies/μL。重复性试验结果显示,组内变异系数为7.22%,表明本方法重复性、稳定性较强。针对30份临床样本,使用本研究建立的real-time PCR检测方法及目前被多个研究所使用的4套引物进行对比。结果表明,本研究所建立的方法灵敏度显著高于文献中报道的4种方法(P<0.01);Sanger测序结果表明,本方法检测出的30份阳性样本均为TTV,检测特异性为100%。结论:本研究采用基于TaqMan探针的real-time PCR检测方法,检测灵敏性高、覆盖基因型范围广,尤其对于TTV病毒载量较低的情况下能够进行定量检测,对于TTV病毒的致病性及作为免疫标志物的应用提供重要的技术支持。

Sensitivity impacts owing to the variations in the type of zero-range pairing forces on the fission properties using the density functional theory

Using the Skyrme density functional theory,potential energy surfaces of^(240)Pu with constraints on the axial quadrupole and octupole deformations(q_(20)and q_(30))were calculated.The volume-like and surface-like pairing forces,as well as a combination of these two forces,were used for the Hartree–Fock–Bogoliubov approximation.Variations in the least-energy fission path,fission barrier,pairing energy,total kinetic energy,scission line,and mass distribution of the fission fragments based on the different forms of the pairing forces were analyzed and discussed.The fission dynamics were studied based on the timedependent generator coordinate method plus the Gaussian overlap approximation.The results demonstrated a sensitivity of the mass and charge distributions of the fission fragments on the form of the pairing force.Based on the investigation of the neutron-induced fission of^(239)Pu,among the volume,mixed,and surface pairing forces,the mixed pairing force presented a good reproduction of the experimental data.

Comparisons of Wave Force Model Effects on the Structural Responses and Fatigue Loads of a Semi-Submersible Floating Wind Turbine

The selection of wave force models will significantly impact the structural responses of floating wind turbines.In this study,comparisons of wave force model effects on the structural responses and fatigue loads of a semi-submersible floating wind turbine(SFWT)were conducted.Simulations were performed by employing the Morison equation(ME)with linear or second-order wave kinematics and potential flow theory(PFT)with first-or second-order wave forces.A comparison of regular waves,irregular waves,and coupled wind/waves analyses with the experimental data showed that many of the simulation results and experimental data are relatively consistent.However,notable discrepancies are found in the response amplitude operators for platform heave,tower base bending moment,and tension in mooring lines.PFT models give more satisfactory results of heave but more significant discrepan-cies in tower base bending moment than the ME models.In irregular wave analyses,low-frequency resonances were captured by PFT models with second-order difference-frequency terms,and high-frequency resonances were captured by the ME models or PFT models with second-order sum-frequency terms.These force models capture the response frequencies but do not reasonably predict the response amplitudes.The coupled wind/waves analyses showed more satisfactory results than the wave-only analyses.However,an important detail to note is that this satisfactory result is based on the overprediction of wind-induced responses.

Real-time data processing method for CO_(2) dispersion interferometer on EAST

A real-time data processing system is designed for the carbon dioxide dispersion interferometer(CO_(2)-DI)on EAST.The system utilizes the parallel and pipelining capabilities of an fieldprogrammable gate array(FPGA)to digitize and process the intensity of signals from the detector.Finally,the real-time electron density signals are exported through a digital-to-analog converter(DAC)module in the form of analog signals.The system has been successfully applied in the CO_(2)-DI system to provide low-latency electron density input to the plasma control system on EAST.Experimental results of the latest campaign with long-pulse discharges on EAST(2022–2023)demonstrate that the system can respond effectively in the case of rapid density changes,proving its reliability and accuracy for future electron density calculation.

Study of the Lift Force Induced by An Interceptor on A High-Speed Mono-Hull:The Affecting Factors and Estimation Formula

To find a better way to estimate the lift force induced by an interceptor on a high-speed mono-hull ship,a series of high-speed mono-hull ship models are designed and investigated under different conditions.Different lift forces are obtained by numerical calculations and validated by a model test in a towing tank.The factors that influence the force are the interceptor height,velocity,draft,and deadrise angle.The relationship between each factor and the induced lift force is investigated and obtained.We found that the induced lift mainly depends on the interceptor height and advancing velocity,and is proportional to the square of the interceptor height and velocity.The results also showed that the effects of the draft and deadrise angle are relatively less important,and the relationship between the induced lift and these two factors is generally linear.Based on the results,a formula including the combined effect of all factors used to estimate the lift force induced by the interceptor is developed based on systematic analysis.The proposed formula could be used to estimate the lift force induced by interceptors,especially under high-speed condition.

Quantitative analysis on resistant forces at oil-surfactant-rock interfaces with dynamic wettability characterization

Adhesion of oil at rock surface plays an important role in the liberation of oil from micro-/nano-pores,especially for heavy oil that has extremely high viscosity.Although molecular dynamics simulation is widely used to study the interfacial interaction for some specific oil-water-rock systems,experimental measurements provide more realistic and reliable evidence.In this work,we propose a dynamic wettability characterization method to indirectly measure resistant forces at oil-surfactant-rock interfaces,including frictional force,wettability hysteresis force,and viscous force,which are parallel with the oil-solid interface.The adhesive force,which is normal to the oil-solid interface is calculated through measurement of work of adhesion.The results show that work of adhesion instead of contact angle can better describe the adhesion of oil at solid surface.The effect of surfactant concentration on work of adhesion is different for water-wet and oil-wet surfaces.Moreover,average viscous forces are calculated through force analysis on oil drops moving along solid surface in different surfactant environments.It is found that viscous force has a magnitude comparable to the frictional force during the movement,while the wettability hysteresis force is negligible.On the other hand,the adhesive force calculated from the work of adhesion is also comparable to the viscous force.Therefore,both the resistant forces parallel with and normal to the oil-solid interface should be minimized for the liberation of oil from rock surface.This work proposes a simple method to evaluate the wetting capability of different surfactants and measure the adhesive force between heavy oil and rock surfaces indirectly,which provides insight into the adhesion of heavy oil at rock surface and would be valuable for the development of surfactant-based oil recovery methods.

Real-Time Intelligent Diagnosis of Co-frequency Vibration Faults in Rotating Machinery Based on Lightweight-Convolutional Neural Networks

The co-frequency vibration fault is one of the common faults in the operation of rotating equipment,and realizing the real-time diagnosis of the co-frequency vibration fault is of great significance for monitoring the health state and carrying out vibration suppression of the equipment.In engineering scenarios,co-frequency vibration faults are highlighted by rotational frequency and are difficult to identify,and existing intelligent methods require more hardware conditions and are exclusively time-consuming.Therefore,Lightweight-convolutional neural networks(LW-CNN)algorithm is proposed in this paper to achieve real-time fault diagnosis.The critical parameters are discussed and verified by simulated and experimental signals for the sliding window data augmentation method.Based on LW-CNN and data augmentation,the real-time intelligent diagnosis of co-frequency is realized.Moreover,a real-time detection method of fault diagnosis algorithm is proposed for data acquisition to fault diagnosis.It is verified by experiments that the LW-CNN and sliding window methods are used with high accuracy and real-time performance.

Culturally competent care across borders: Implementing culturally responsive teaching for nurses in diverse workforces

The migration of healthcare professionals,including nurses,is a global phenomenon.It is driven by various factors,including the pursuit of better opportunities,living conditions,and professional development,as well as political instability or conflict in their home countries.The World Health Organization(WHO)has noted that high-income countries often rely on foreign-trained nurses to fill gaps in their healthcare systems[1].For instance,as of 2021,over 40%(52 million)of all nurses in the United States(US)were expatriates[2].In the United Kingdom(UK),the percentage of expatriate nurses was even higher,reaching approximately 18%in 2021[3].Owing to globalization and migration,healthcare providers must possess cultural competence to deliver improved care[4,5].Culturally responsive teaching(CRT)is rooted in the idea that culture plays a vital role in shaping people’s behaviors,beliefs,values,and communication styles[6].Furthermore,these cultural factors influence patients’perspectives on health,illness,healing,and their preferences for care and communication[7].By recognizing and embracing these cultural differences,nurses can provide more effective and compassionate care to their diverse patient population[8].