A deep learning driven hybrid beamforming method for millimeter wave MIMO system

The hybrid beamforming is a promising technology for the millimeter wave MIMO system,which provides high spectrum efficiency,high data rate transmission,and a good balance between transmission performance and hardware complexity.The most existing beamforming systems transmit multiple streams by formulating multiple orthogonal beams.However,the Neural network Hybrid Beamforming(NHB)adopts a totally different strategy,which combines multiple streams into one and transmits by employing a high-order non-orthogonal modulation strategy.Driven by the Deep Learning(DL)hybrid beamforming,in this work,we propose a DL-driven nonorthogonal hybrid beamforming for the single-user multiple streams scenario.We first analyze the beamforming strategy of NHB and prove it with better Bit Error Rate(BER)performance than the orthogonal hybrid beamforming even with the optimal power allocation.Inspired by the NHB,we propose a new DL-driven beamforming scheme to simulate the NHB behavior,which avoids time-consuming neural network training and achieves better BERs than traditional hybrid beamforming.Moreover,our simulation results demonstrate that the DL-driven nonorthogonal beamforming outperforms its traditional orthogonal beamforming counterpart in the presence of subconnected schemes and imperfect Channel State Information(CSI).

A One-Dimensional Hybrid Simulation of DC/RF Combined Driven Capacitive Plasma

We developed a one-dimensional hybrid model to simulate the DC/RF combined driven capacitively coupled plasma for argon discharges. The numerical results are used to analyze the influence of the DC source on the plasma density distribution, ion energy distributions （IEDs） and ion angle distributions （IADs） on both the RF and DC electrodes. The increase in DC voltage drives more high-energy ions to the electrode applied to the DC source, which makes the IEDs at the DC electrode shift towards higher energy, and the peaks in the IADs shift towards small angle regions. At the same time, it also decreases the ion energy at the RF electrode and enlarges the incident angles of the ions, which strike the RF electrode.

Dynamic Modeling and Motion Simulation for A Winged Hybrid-Driven Underwater Glider

PETREL, a winged hybrid-driven underwater glider is a novel and practical marine survey platform which combines the features of legacy underwater glider and conventional AUV （autonomous underwater vehicle）. It can be treated as a multi-rigid-body system with a floating base and a particular hydrodynamic profile. In this paper, theorems on linear and angular momentum are used to establish the dynamic equations of motion of each rigid body and the effect of translational and rotational motion of internal masses on the attitude control are taken into consideration. In addition, due to the unique external shape with fixed wings and deflectable rudders and the dual-drive operation in thrust and glide modes, the approaches of building dynamic model of conventional AUV and hydrodynamic model of submarine are introduced, and the tailored dynamic equations of the hybrid glider are formulated. Moreover, the behaviors of motion in glide and thrust operation are analyzed based on the simulation and the feasibility of the dynamic model is validated by data from lake field trials.

Stability Analysis of Hybrid-Driven Underwater Glider

Hybrid-driven underwater glider is a new type of tmmanned underwater vehicle, which combines the advantages of autonomous underwater vehicles and traditional underwater gliders. The autonomous underwater vehicles have good maneuverability and can travel with a high speed, while the traditional underwater gliders are highlighted by low power consumption, long voyage, long endurance and good stealth characteristics. The hybrid-driven underwater gliders can realize variable motion profiles by their own buoyancy-driven and propeller propulsion systems. Stability of the mechanical system determines the performance of the system. In this paper, the Petrel-II hybrid-driven underwater glider developed by Tianjin University is selected as the research object and the stability of hybrid-driven underwater glider unitedly controlled by buoyancy and propeller has been targeted and evidenced. The dimensionless equations of the hybrid-driven underwater glider are obtained when the propeller is working. Then, the steady speed and steady glide path angle under steady-state motion have also been achieved. The steady-state operating conditions can be calculated when the hybrid-driven underwater glider reaches the desired steady-state motion. And the steady- state operating conditions are relatively conservative at the lower bound of the velocity range compared with the range of the velocity derived from the method of the composite Lyapunov function. By calculating the hydrodynamic coefficients of the Petrel-II hybrid-driven underwater glider, the simulation analysis has been conducted. In addition, the results of the field trials conducted in the South China Sea and the Danjiangkou Reservoir of China have been presented to illustrate the validity of the analysis and simulations.and to show the feasibility of the method of the composite Lyapunov function which verifies the stability of the Petrel-II hybrid-driven underwater glider.

Hybrid position/force control strategy and experiment of the six-cable driven parallel manipulator for the forty-meter aperture radio telescope

This paper mainly analyzes a hybrid position/force control strategy and experiment of a six-cable driven parallel manipulator for a forty-meter aperture radio telescope. Through the establishments of a kinematic model, a catenary model and a cable-force characteristics model, a basic method is presented based on preventing the pseudo-drag problem of a flexible cable and realizing the hybrid position/force control for a six-cable driven parallel manipulator, and a hybrid position/force control system is developed. Some specific experiments in two typical velocities for astronomical observations are carried out. Experimental results show that the tracking accuracy is related to the speed of the movement. The desired tracking accuracy of the support system is achieved through an effective hybrid position/force control strategy, the cable forces are controlled effectively, and the pseudo-drag problem of flexible cable is solved. This study establishes the foundation of achieving the harmonious movement of the six-cable driven parallel manipulator, A-B rotating platform and the Stewart fine tuninz olatform.

Direct driven wind energy conversion system based on hybrid excitation synchronous machine

A novel direct-drive type wind power generation system based on hybrid excitation synchronous machine（HESM）is introduced in this paper.The generator is connected to an uncontrollable rectifier,and a fully controlled voltage-sourceinverter is used to connect the system to utility grid.An intermediate DC bus exists between the rectifier and inverter.A new control strategy is proposed which achieves the maximum power point tracking（MPPT） with the control of excitation current of HESM and stabilizes the DC link voltage with the control of inverter output current simultaneously.Specially-designed buck circuit is used to control the excitation current of HESM,and grid voltage-oriented vector control strategy is employed to realize the decoupling of the inverter output power.Simulation results and experiment in 3 kW lab prototype show an excellent static and dynamic performance of the proposed system.

数模联动的多特征工件加工能耗预测方法研究

在实际切削加工过程中材料去除率是不断变化的,现有将其视为恒量的能耗建模方法难以实现能耗准确预测。为了提高切削过程能耗预测精度,提出了一种基于材料去除率的数模联动加工能耗预测方法。首先,基于切削过程刀具与工件的接触关系分析了切入、完全切入和切出阶段材料去除率变化规律,并对相应的加工能耗特性进行了分析;其次,提出了数据驱动的刀具切入,切出阶段加工能耗预测方法,以及模型驱动的完全切入阶段加工能耗预测方法,实现加工过程能耗准确预测;最后,利用实验案例验证了所提模型及方法的有效性,为今后研究能耗预测精度奠定了基础。

A New Design Method for Realizing Accurate Track of Hybrid-driven Five-bar Mechanism

The concept of virtual slider crank mechanism is proposed and decoupled to obtain parameters of controllable five bar mechanism without any principle error for any given trajectory. The model is simple and easy to solve. This method has no convergence,flexible workspace and singularity of the mechanism problem. Through this method,we don’ t need any curve to fit the trajectory point. Using MATLAB program to calculate,the computation time can be reduced to less than 3% of the original. Finally,an example is given to illustrate the method which is meanwhile compared with the traditional five bar design method.

机电系统网络控制课程的多维度混合式教学探索与实践

针对研究生课程机电系统网络控制的教学现状与发展需求,开展多维度混合式教学探索与实践。构建包含课堂讲授、资源库、专题研讨等在内的线上线下多维度教学体系;针对主题研讨实践环节引入翻转课堂,以项目驱动和研讨式教学引导学生分组完成课程相关的项目任务,掌握并运用所学知识分析解决问题,体验科学研究的过程与思维方法,形成理论知识到科研实践的闭环训练;最终建立形成性和终结性相结合的多元考核评价体系,旨在通过课程学习培养专业基础扎实、科研素养和工程实践能力突出的高端创新型人才。

高职“机械设计基础”课程混合式教学改革创新路径研究

基于“机械设计基础”这门课程的特点和高职学习者学情,探索“学做研创”专业基础课程改革路径,通过设计不同的产品结构设计主题,以机械产品设计为主线,有机融入“机械原理设计→机械结构设计→机械产品评价”教学方式,创新实施基于成果导向的“1+1+N”线上线下混合式教学模式,以宿舍为单位组建学习型组织,团队协作完成不同难易程度的机械作品以检验学习成果,实现“教与学、学与做、做与创”深度融合,让“学习者成为课堂主体”“教师成为课堂主导教练员”,打造高效优质金课堂。

基于数据驱动的混合式教学质量保障机制的研究——以民办高职院校为例

本研究的目的是将基于数据驱动的混合式教学质量保障机制融合到高等院校的教学质量保证体系中,试图对所确定的问题进行更深入的理解,以便提出干预和变革,加强高等院校混合式教学和学习过程,最重要的是,改善高等院校的教学质量和学习体验。为了实现这一目标,本研究首先明确基于数据驱动的混合式教学质量保障机制的作用,进而论述基于数据驱动的混合式教学质量保障机制的设计,最终提出基于数据驱动的混合式教学质量保障机制的实现。所做研究与高等教育混合式教学质量增强维度有关,是协助高等院校教务部门加深对混合式教学质量管理及保障的探索,具有一定的参考和借鉴意义。

基于混合驱动的分离罗拉运动轨迹规划研究

针对当前棉精梳机分离接合机构运动输出特性固定、振动大等问题,分析棉精梳机分离罗拉分离接合工艺,确定分离罗拉运动的位移关键点位置,根据差动轮系的运动规律提出交替驱动和分段驱动2种动力分配方案,并采用三次样条曲线作为交替驱动位移曲线、S型加减速曲线作为分段驱动位移曲线,通过对比分析得到最佳分离罗拉运动轨迹。指出:当设定棉纤维长度为30 mm时,2种驱动方案均能满足分离接合要求;采用混合驱动机构对分离罗拉高速正反转运动进行分解,实现伺服电机混合驱动分离罗拉,其运动性能更佳。

机理-数据混合驱动的气象敏感负荷需求响应潜力评估

为准确评估电网中气象敏感负荷需求响应潜力,提出了一种基于机理-数据驱动方法融合的气象敏感负荷需求响应潜力评估方法。该方法根据数据驱动的双分支神经网络模型估算电网层级的气象敏感负荷功率,基于热力学等值模型建立气象敏感负荷的机理聚合模型,并以聚合功率与估算功率误差最小为目标,采用控制变量法和粒子群优化算法优化确定机理聚合模型的等效参数(温控负荷设备数量、温度设定值等),同时综合考虑用户舒适度与意愿度等因素评估气象敏感负荷需求响应潜力。实例验证结果表明,该方法估算的气象敏感负荷功率与温度具有显著相关性,获得的机理聚合模型等效参数及需求响应潜力与实际情况相符且合理。

精梳机分离罗拉齿轮传动机构优化

混合驱动的精梳机分离罗拉齿轮传动机构的动力学性能直接影响车速的提高,为改善传动机构的动力学性能,首先,分析现有混合驱动的精梳机分离罗拉齿轮传动机构,采用集中质量法建立传动机构的固有频率模型,并计算传动机构在不同精梳机车速下的激励频率,发现当车速为600钳次/min时机构会出现共振;然后,采用导数法建立传动机构固有频率对机构中各个构件转动惯量的灵敏度模型,发现行星轮p的转动惯量对一阶固有频率的影响较大;最后,通过优化传动机构中WW型差动轮系的各个齿轮齿数以改变其转动惯量,使传动机构的一阶固有频率比原先提高了43.44%,避免了机构在600钳次/min时出现共振。

机理模型与集成学习混合驱动的机器人关节摩擦建模方法

机器人一体化关节广泛应用于医疗、协作机器人等领域,其摩擦特性是影响机器人性能的关键因素.为此,提出了一种机理模型与集成学习混合驱动的机器人关节摩擦建模方法,以提高模型精度.首先,综合考虑转速、负载等关节摩擦特性的影响因素及其周期波动特性,基于先验知识和物理分析分别建立了伺服电机与谐波减速器的参数化机理模型,描述摩擦特性的变化规律.然后,针对机理建模中因线性假设、忽略高阶项等产生的非线性残差,提出了基于eXtreme gradient boosting(XGBoost)的残差补偿模型建模方法,通过采用Boosting集成学习策略,提高残差补偿模型的泛化能力.同时,采用贝叶斯优化方法进行XGBoost模型的超参数寻优,以提高模型精度和训练效率.相比于传统的参数化机理模型,本文所提出的混合驱动模型具有更高精度.与反向传播神经网络、支持向量机、长短时记忆神经网络等多种典型方法的对比实验表明,本文所提出的基于XGBoost的残差补偿模型具有更强的特征提取能力,能够较好地预测强非线性的波动摩擦残差,有效地提高了整体模型的精度.

基于内嵌物理知识卷积神经网络的电力系统暂态稳定评估

针对现有数据驱动的电力系统暂态评估方法依赖大规模数据集且可解释性不足的问题,文中将物理知识嵌入传统数据驱动方法,提出一种基于内嵌物理知识卷积神经网络的电力系统暂态稳定评估方法。该方法考虑大规模风电并网的电力系统,将电力系统暂态稳定物理方程内嵌至神经网络损失函数,通过神经网络直接逼近物理过程,使输出结果满足物理规律,提高暂态稳定评估的可靠性与可解释性。通过数据与知识双驱动,所提方法不依赖大规模训练数据集,依然具有较好的鲁棒性与泛化能力。此外,所提方法通过卷积神经网络进行特征提取与降维,解决拓扑数据无法直接作为神经网络输入的难题。在含风机的IEEE 9节点和IEEE 39节点测试系统上的实验结果表明,所提方法在准确率、计算效率、泛化能力等方面相较现有方法有显著提升。

基于捕获点理论的混合驱动水下刀锋腿机器人稳定性判据

由8个推进器和6条刀锋腿混合驱动的水下机器人可在水底或水下结构物表面上行走.所提方法旨在研究这类机器人运动稳定性的评判准则,即稳定性判据.现有的稳定性判据多集中于同一机构(腿)驱动的陆地机器人,未涉及混合驱动的水下刀锋腿机器人.针对该问题,提出了基于捕获点理论的混合驱动水下刀锋腿机器人稳定性判据.首先,在建立混合驱动水下滚动倒立摆模型的基础上,利用机器人运动状态预测摆动腿和支撑腿切换瞬间机器人的动能;然后,根据推进器所能提供的推力范围,计算迫使机器人静止的捕获点变化范围,即获取捕获域;最后,根据捕获域与支撑域的空间关系,判断机器人是否稳定,并计算机器人的稳定裕度.水下实验结果表明,所提出的稳定性判据具有较好的充要性和普适性.

聚变能源研究态势及展望

聚变能源开发有望重塑未来能源发展格局,正处于突破工程能量“得失相当”的关键阶段,机遇与挑战并存。本文总结了聚变能源研究的主要进展,凝练了能量平衡尚未实现、氚自持尚未得到验证、实现高可利用率难度极高、耐辐照材料开发进展缓慢、使用经济性普遍较差等发展挑战。在梳理国际热核聚变实验堆(ITER)计划的共性基础技术突破、成员国配套研究的基础上,归纳了我国磁约束聚变研究的整体规划、自主项目部署、技术路线跟踪等方面的进展。进一步,围绕我国自主提出的Z箍缩聚变裂变混合堆(Z-FFR)概念,阐述了基本原理、应用优势、系列进展,提出了面向2040年实现商业化供能目标的发展规划,涵盖关键技术攻关、工程演示、商业发电推广等阶段的任务目标。为了全面推进我国聚变能源开发进程,建议在磁约束聚变方面深入参与ITER计划和相关国际合作,攻克商用聚变堆关键物理与工程技术,开展中国聚变工程实验堆(CFETR)主机关键部件研发并适时建设和运营CFETR;在Z-FFR方面,加快“电磁驱动大科学装置”建设,开展聚变能源关键技术攻关,推进Z-FFR工程演示和商业应用。

基于子网优先级驱动的交直流混合微网集群双向互联变流器分散式控制策略

随着可再生能源渗透率的不断提高,交直流混合微电网集群逐渐成为新型电力系统的弹性解决方案。考虑到实际运行中各子电网优先级和电能质量需求存在差异,该文提出一种基于子网优先级驱动的交直流混合微网集群双向互联变流器分散式控制策略,以实现系统的弹性功率交互。首先,通过将归一化的交流频率和直流电压值作为子电网的状态偏差因子,以最小化全局总偏差因子为目标构建了用于交直流混合微电网集群的最优电力交换模型,该模型充分考虑了子电网的优先级指数和供电容量;其次,基于最优电力交换目标,提出用于双向互联变流器的准下垂控制策略,由于该策略中双向互联变流器仅需检测本地交流频率/直流电压的状态偏差因子,因此可以实现对交直流混合微电网集群整体偏差因子最小化的分散式最优功率自主协同控制;最后,构造了两个仿真实例,并利用半实物仿真系统验证了所提策略的有效性。

Data-Model Hybrid Driven Topology Identification Framework for Distribution Networks

Extensive penetration of distribution energy resources(DERs)brings increasing uncertainties to distribution networks.Accurate topology identification is a critical basis to guarantee robust distribution network operation.Many algorithms that estimate distribution network topology have already been employed.Unfortunately,most are based on data-driven alone method and are hard to deal with ever-changing distribution network physical structures.Under these backgrounds,this paper proposes a data-model hybrid driven topology identification scheme for distribution networks.First,a data-driven method based on a deep belief network(DBN)and random forest(RF)algorithm is used to realize the distribution network topology rough identification.Then,the rough identification results in the previous step are used to make a model of distribution network topology.The model transforms the topology identification problem into a mixed integer programming problem to correct the rough topology further.Performance of the proposed method is verified in an IEEE 33-bus test system and modified 292-bus system.