Machine-learning-assisted efficient reconstruction of the quantum states generated from the Sagnac polarization-entangled photon source

Neural networks are becoming ubiquitous in various areas of physics as a successful machine learning(ML)technique for addressing different tasks.Based on ML technique,we propose and experimentally demonstrate an efficient method for state reconstruction of the widely used Sagnac polarization-entangled photon source.By properly modeling the target states,a multi-output fully connected neural network is well trained using only six of the sixteen measurement bases in standard tomography technique,and hence our method reduces the resource consumption without loss of accuracy.We demonstrate the ability of the neural network to predict state parameters with a high precision by using both simulated and experimental data.Explicitly,the mean absolute error for all the parameters is below 0.05 for the simulated data and a mean fidelity of 0.99 is achieved for experimentally generated states.Our method could be generalized to estimate other kinds of states,as well as other quantum information tasks.

Acoustic emission source identification based on harmonic wavelet packet and support vector machine

In order to solve the fatigue damage identification problem of helicopter moving components, a new approach for acoustic emission （AE） source type identification based on the harmonic wavelet packet （HWPT） feature extraction and the hierarchy support vector machine （H-SVM） classifier is proposed. After a four-level decomposition of the HWPT, the energy feature of AE signals in different frequency bands is extracted, which overcomes the shortcomings of the traditional wavelet packet including energy leakage, and inflexible frequency band selection and different frequency resolutions on different levels. The H-SVM classifier is trained with a subset of the experimental data for known AE source types and tested using the remaining set of data. The results of pressure-off experiments on the specimens of carbon fiber materials indicate that the proposed approach can effectively implement the AE source type identification, and has a better performance in terms of computational efficiency and identification accuracy than the wavelet packet （WPT） feature extraction.

Machine Learning Security Defense Algorithms Based on Metadata Correlation Features

With the popularization of the Internet and the development of technology,cyber threats are increasing day by day.Threats such as malware,hacking,and data breaches have had a serious impact on cybersecurity.The network security environment in the era of big data presents the characteristics of large amounts of data,high diversity,and high real-time requirements.Traditional security defense methods and tools have been unable to cope with the complex and changing network security threats.This paper proposes a machine-learning security defense algorithm based on metadata association features.Emphasize control over unauthorized users through privacy,integrity,and availability.The user model is established and the mapping between the user model and the metadata of the data source is generated.By analyzing the user model and its corresponding mapping relationship,the query of the user model can be decomposed into the query of various heterogeneous data sources,and the integration of heterogeneous data sources based on the metadata association characteristics can be realized.Define and classify customer information,automatically identify and perceive sensitive data,build a behavior audit and analysis platform,analyze user behavior trajectories,and complete the construction of a machine learning customer information security defense system.The experimental results show that when the data volume is 5×103 bit,the data storage integrity of the proposed method is 92%.The data accuracy is 98%,and the success rate of data intrusion is only 2.6%.It can be concluded that the data storage method in this paper is safe,the data accuracy is always at a high level,and the data disaster recovery performance is good.This method can effectively resist data intrusion and has high air traffic control security.It can not only detect all viruses in user data storage,but also realize integrated virus processing,and further optimize the security defense effect of user big data.

A Method of Identifying Electromagnetic Radiation Sources by Using Support Vector Machines

Electromagnetic Radiation Source Identification(ERSI) is a key technology that is widely used in military and radiation management and in electromagnetic interference diagnostics.The discriminative capability of machine learning methods has recently been used for facilitating ERSI.This paper presents a new approach to improve ERSI by adopting support vector machines,which are proven to be effective tools in pattern classification and regression,on the basis of the spatial distribution of electromagnetic radiation sources.Spatial information is converted from 3D cubes to 1D vectors with subscripts as inputs in order to simplify the model.The model is trained with 187 500 data sets in order to enable it to identify the types of radiation source types with an accuracy of up to 99.9%.The influence of parameters(e.g.,penalty parameter,reflection and noise from the ambient environment,and the scaling method for the input data) are discussed.The proposed method has good performance in noisy and reverberant environment.It has an identification accuracy of 82.15% when the signal-to-noise ratio is 20 dB.The proposed method has better accuracy in a noisy environment than artificial neural networks.Given that each Electromagnetic(EM) source has unique spatial characteristics,this method can be used for EM source identification and EM interference diagnostics.

Neutron-gamma discrimination method based on blind source separation and machine learning

The discrimination of neutrons from gamma rays in a mixed radiation field is crucial in neutron detection tasks.Several approaches have been proposed to enhance the performance and accuracy of neutron-gamma discrimination.However,their performances are often associated with certain factors,such as experimental requirements and resulting mixed signals.The main purpose of this study is to achieve fast and accurate neutron-gamma discrimination without a priori information on the signal to be analyzed,as well as the experimental setup.Here,a novel method is proposed based on two concepts.The first method exploits the power of nonnegative tensor factorization(NTF)as a blind source separation method to extract the original components from the mixture signals recorded at the output of the stilbene scintillator detector.The second one is based on the principles of support vector machine(SVM)to identify and discriminate these components.In addition to these two main methods,we adopted the Mexican-hat function as a continuous wavelet transform to characterize the components extracted using the NTF model.The resulting scalograms are processed as colored images,which are segmented into two distinct classes using the Otsu thresholding method to extract the features of interest of the neutrons and gamma-ray components from the background noise.We subsequently used principal component analysis to select the most significant of these features wich are used in the training and testing datasets for SVM.Bias-variance analysis is used to optimize the SVM model by finding the optimal level of model complexity with the highest possible generalization performance.In this framework,the obtained results have verified a suitable bias–variance trade-off value.We achieved an operational SVM prediction model for neutron-gamma classification with a high true-positive rate.The accuracy and performance of the SVM based on the NTF was evaluated and validated by comparing it to the charge comparison method via figure of merit.The results indicate that the proposed approach has a superior discrimination quality(figure of merit of 2.20).

A Resource Management Algorithm for Virtual Machine Migration in Vehicular Cloud Computing

In recent years,vehicular cloud computing(VCC)has gained vast attention for providing a variety of services by creating virtual machines(VMs).These VMs use the resources that are present in modern smart vehicles.Many studies reported that some of these VMs hosted on the vehicles are overloaded,whereas others are underloaded.As a circumstance,the energy consumption of overloaded vehicles is drastically increased.On the other hand,underloaded vehicles are also drawing considerable energy in the underutilized situation.Therefore,minimizing the energy consumption of the VMs that are hosted by both overloaded and underloaded is a challenging issue in the VCC environment.The proper and efcient utilization of the vehicle’s resources can reduce energy consumption signicantly.One of the solutions is to improve the resource utilization of underloaded vehicles by migrating the over-utilized VMs of overloaded vehicles.On the other hand,a large number of VM migrations can lead to wastage of energy and time,which ultimately degrades the performance of the VMs.This paper addresses the issues mentioned above by introducing a resource management algorithm,called resource utilization-aware VM migration(RU-VMM)algorithm,to distribute the loads among the overloaded and underloaded vehicles,such that energy consumption is minimized.RU-VMM monitors the trend of resource utilization to select the source and destination vehicles within a predetermined threshold for the process of VM migration.It ensures that any vehicles’resource utilization should not exceed the threshold before or after the migration.RU-VMM also tries to avoid unnecessary VM migrations between the vehicles.RU-VMM is extensively simulated and tested using nine datasets.The results are carried out using three performance metrics,namely number of nal source vehicles(nfsv),percentage of successful VM migrations(psvmm)and percentage of dropped VM migrations(pdvmm),and compared with threshold-based algorithm(i.e.,threshold)and cumulative sum(CUSUM)algorithm.The comparisons show that the RU-VMM algorithm performs better than the existing algorithms.RU-VMM algorithm improves 16.91%than the CUSUM algorithm and 71.59%than the threshold algorithm in terms of nfsv,and 20.62%and 275.34%than the CUSUM and threshold algorithms in terms of psvmm.

Research on Blind Source Separation for Machine Vibrations

Blind source separation is a signal processing method based on independent component analysis, its aim is to separate the source signals from a set of observations (output of sensors) by assuming the source signals independently. This paper reviews the general concept of BSS firstly;especially the theory for convolutive mixtures, the model of convolutive mixture and two deconvolution structures, then adopts a BSS algorithm for convolutive mixtures based on residual cross-talking error threshold control criteria, the simulation testing points out good performance for simulated mixtures.

Recent Developments of Modulation and Control for High-Power Current-Source-Converters Fed Electric Machine Systems

The pulse-width-modulated(PWM)current-source converters(CSCs)fed electric machine systems can be considered as a type of high reliability energy conversion systems,since they work with the long-life DC-link inductor and offer high fault-tolerant capability for short-circuit faults.Besides,they provide motor friendly waveforms and four-quadrant operation ability.Therefore,they are suitable for high-power applications of fans,pumps,compressors and wind power generation.The purpose of this paper is to comprehensively review recent developments of key technologies on modulation and control of high-power(HP)PWM-CSC fed electric machines systems,including reduction of low-order current harmonics,suppression of inductor–capacitor(LC)resonance,mitigation of common-mode voltage(CMV)and control of modular PWM-CSC fed systems.In particular,recent work on the overlapping effects during commutation,LC resonance suppression under fault-tolerant operation and collaboration of modular PMW-CSCs are described.Both theoretical analysis and some results in simulations and experiments are presented.Finally,a brief discussion regarding the future trend of the HP CSC fed electric machines systems is presented.

Systematic analysis of error sources during ultra-precision machining

During ultra-precision machining, machining accuracy is determined by many factors and interaction of these factors. Error sources are systematically analyzed for ultra-precision machine tools, and the influencing degree of each factor is presented to provide orientation for error reduction and error compensation.

Groundwater contaminant source identification considering unknown boundary condition based on an automated machine learning surrogate

Groundwater contamination source identification(GCSI)is a prerequisite for contamination risk evaluation and efficient groundwater contamination remediation programs.The boundary condition generally is set as known variables in previous GCSI studies.However,in many practical cases,the boundary condition is complicated and cannot be estimated accurately in advance.Setting the boundary condition as known variables may seriously deviate from the actual situation and lead to distorted identification results.And the results of GCSI are affected by multiple factors,including contaminant source information,model parameters,boundary condition,etc.Therefore,if the boundary condition is not estimated accurately,other factors will also be estimated inaccurately.This study focuses on the unknown boundary condition and proposed to identify three types of unknown variables(contaminant source information,model parameters and boundary condition)innovatively.When simulation-optimization(S-O)method is applied to GCSI,the huge computational load is usually reduced by building surrogate models.However,when building surrogate models,the researchers need to select the models and optimize the hyperparameters to make the model powerful,which can be a lengthy process.The automated machine learning(AutoML)method was used to build surrogate model,which automates the model selection and hyperparameter optimization in machine learning engineering,largely reducing human operations and saving time.The accuracy of AutoML surrogate model is compared with the surrogate model used in eXtreme Gradient Boosting method(XGBoost),random forest method(RF),extra trees regressor method(ETR)and elasticnet method(EN)respectively,which are automatically selected in AutoML engineering.The results show that the surrogate model constructed by AutoML method has the best accuracy compared with the other four methods.This study provides reliable and strong support for GCSI.

New Individual Identification Method of Radiation Source Signal Based on Entropy Feature and SVM

In this paper,according to the defect of methods which have low identification rate in low SNR,a new individual identification method of radiation source based on information entropy feature and SVM is presented. Firstly,based on the theory of multi-resolution wavelet analysis,the wavelet power spectrum of noncooperative signal can be gotten. Secondly,according to the information entropy theory,the wavelet power spectrum entropy is defined in this paper. Therefore,the database of signal's wavelet power spectrum entropy can be built in different SNR and signal parameters. Finally,the sorting and identification model based on SVM is built for the individual identification of radiation source signal. The simulation result indicates that this method has a high individual's identification rate in low SNR,when the SNR is greater than 4 dB,the identification rate can reach 100%. Under unstable SNR conditions,when the range of SNR is between 0 dB and 24 dB,the average identification rate is more than 92. 67%. Therefore,this method has a great application value in the complex electromagnetic environment.

BLIND SPEECH SEPARATION FOR ROBOTS WITH INTELLIGENT HUMAN-MACHINE INTERACTION

Speech recognition rate will deteriorate greatly in human-machine interaction when the speaker's speech mixes with a bystander's voice. This paper proposes a time-frequency approach for Blind Source Seperation (BSS) for intelligent Human-Machine Interaction(HMI). Main idea of the algorithm is to simultaneously diagonalize the correlation matrix of the pre-whitened signals at different time delays for every frequency bins in time-frequency domain. The prososed method has two merits: (1) fast convergence speed; (2) high signal to interference ratio of the separated signals. Numerical evaluations are used to compare the performance of the proposed algorithm with two other deconvolution algorithms. An efficient algorithm to resolve permutation ambiguity is also proposed in this paper. The algorithm proposed saves more than 10% of computational time with properly selected parameters and achieves good performances for both simulated convolutive mixtures and real room recorded speeches.

HSP超前探测技术在煤矿TBM掘进巷道中的应用研究

随着全断面掘进机TBM(Tunnel Boring Machine)逐渐应用于煤矿岩巷掘进,对不良地质构造进行超前准确快速预测的需求日益迫切。通过对主动源地震波超前探测方法的特点和TBM破岩震源超前探测技术的适用性进行分析,结合煤矿巷道地质和生产条件,提出了适用于煤矿巷道TBM掘进的HSP超前探测方法。以河南平顶山首山一矿TBM掘进底板瓦斯治理巷道为工程背景,选用防爆硬件一体化设计的探测仪器在煤矿巷道中进行应用。构建了空间型观测方式对煤矿巷道近水平煤线进行探测,优化了双护盾TBM掘进巷道狭小空间检波器阵列式布置参数;基于时频分析、互相关干涉处理、反射与散射联合反演等方法处理原始信号并进行探测结果成像。研究表明:采用空间型观测方式可实现与巷道小角度斜交煤线的识别,设计震源与首检波器间距离为15 m时最优。通过时频分析提取有效信号,利用互相关干涉法获取虚拟震源道和反射特征曲线,并结合反射与散射联合反演成像得到探测区域地层反射能量分布图,能够较准确地推测得到围岩存在的不良地质构造。通过比较现场开挖揭露情况与探测结果发现两者吻合度较高,表明HSP超前探测方法可实现掘进工作面前方100 m范围内超前无损地质预测,有助于提高煤矿岩巷TBM掘进速度。

基于特征集重构与多标签分类模型的谐波源定位方法

传统基于谐波状态估计的谐波源定位方法需要专门的同步相量量测装置,工程应用受到限制。为此,基于电能质量监测装置所采集的非同步量测数据,提出了基于特征集重构与多标签分类模型的谐波源定位方法。利用监测数据的充分统计量来挖掘量测时段的谐波信息,同时利用标签特定特征学习算法重构特征集,从而消除冗余特征以及无关特征对于谐波源定位精度的影响;提出基于邻接矩阵以及灵敏度分析的测点配置方法,结合电路网络拓扑信息实现测点的优化配置;提出基于改进极限学习机的谐波源定位方法,该方法以重构特征集为输入,建立多标签分类模型,实现谐波源定位。通过仿真与算例分析,验证了所提方法的可行性及有效性。

基于可解释机器学习的黄河源区径流分期组合预报

黄河源区是黄河流域重要的产流区和我国重要的清洁能源基地,提高黄河源区径流预报准确率可为流域水资源科学调配和水风光清洁能源高效利用提供重要支撑。以黄河源区唐乃亥和玛曲水文站为研究对象,基于不同月份径流组分的差异,考虑积雪覆盖率及融雪水当量变化,构建了中长期径流分期组合机器学习预报模型及其可解释性分析框架。研究结果表明:1)年内的径流预报时段可划分为融雪影响期(3—6月)和非融雪主导(以降雨和地下水补给为主)期(7月—次年2月);2)与传统不分期模型相比,唐乃亥站和玛曲站分期组合预报模型的纳什效率系数分别达0.897、0.835,确定系数(R2)分别达0.897、0.839,均方根误差分别降低了10%、17%,提高了径流预报准确率,通过分位数映射校正,唐乃亥站和玛曲站预报模型的R2分别进一步提升至0.926和0.850;3)基于SHAP机器学习可解释性分析框架,辨识了预报因子对径流预报结果的贡献程度,由高到低依次为降水、前一个月流量、蒸发、气温、相对湿度、融雪水当量等,发现了不同预报因子之间交互作用散点分布具有拖尾式或阶跃式的特征。

含准Z源的双永磁电机驱动系统的研究及仿真

由于九开关逆变器通过共享模式,可减少电力电子器件数量,减小驱动装置的体积与重量,适用于双机驱动的应用场合。针对九开关逆变器双机驱动系统中单台电机电压利用率低的问题,研究了一种含准Z源的双永磁同步电机驱动系统。为了实现双机驱动系统中每台电机的独立可控,研究了基于SVPWM调制的双机分时控制原理。为满足不同的工况,研究了永磁同步电机的恒转矩-弱磁控制策略,利用间接电压法对Z源电压进行控制,以维持Z源电压的稳定。搭建了含准Z源的双永磁同步驱动系统仿真模型,仿真结果表明,永磁同步电机和准Z源的控制策略行之有效,分时控制可实现双机驱动系统的独立可控,可满足工程应用的需求。

基于多源数据与机器学习的降雨型滑坡灾害危险性动态预警

降雨型滑坡灾害是三峡库区发育的主要地质灾害之一,区域滑坡灾害危险性预警对滑坡风险减缓有重要意义。针对区域滑坡灾害危险性预警的时效性等问题,本文以重庆市云阳县为研究区,提出了基于多源数据与机器学习的滑坡灾害危险性动态预警方法(图1)。1滑坡易发性空间预测滑坡易发性预测可准确地预测出潜在滑坡空间分布规律(黄发明等,2020;林高聪等,2023)。

超声振动辅助电火花加工表面成形机理研究

以TC4钛合金作为工件材料,通过建立高斯热源在工件表面的分布模型,描述了单脉冲电火花加工凹坑的成形机理。为了验证模型的准确性,进行了仿真和实验验证。结果表明:通过分析仿真得到的温度场云图和实验测得的数据发现,高斯热源分布模型与两者一致。在引入超声振动后,表面粗糙度降低35%~46%,材料表面凹坑明显减少,表面形貌发生显著变化。随振幅增大,高斯热源分布曲线的最大值减小,温度场中心最高温度降低,深度变浅的表面凹坑数量增加,单个凹坑的面积增加,表面粗糙度变好。

基于多源地理数据的城市突发公共卫生事件风险评估研究

风险评估是突发公共卫生事件应急管理的关键环节,基于地理数据开展风险评估可有效提高精度.通过采集多源地理数据,结合随机森林算法和地理探测器、核密度分析、空间自相关等空间统计分析提出城市突发公共卫生事件风险评估方法,并通过实证分析验证模型可行性.结果表明:采用随机森林算法构建的风险评估模型表现良好;餐饮美食、公司企业和交通设施等场所是影响疫情的主要因素,疫情流行具有因子交互性,其中餐饮美食与其他因子的交互作用最强;空间传播上位于城市中心区域的社区风险等级较高并呈现向外围逐渐减弱的趋势,同时有明显的高值或低值聚集.

高比例新能源接入下的历史策略库辅助源网荷储协同实时电压控制研究

随着“双碳”目标的推进,储能、充电桩等用户侧新型负荷设备数量增多。在此背景下,由于量测条件不满足造成的不可观测区域将直接导致传统的电压控制方法难以完成对分布式电源的精准调控。为解决上述问题,提出一种基于蜣螂优化算法(DBO)的极限学习机(ELM)构建历史策略库,用以辅助源网荷储协同实时电压控制的方法,可实现对配电网电压实时精确控制。首先介绍了基于近似灵敏度计算的电压控制方法,然后介绍了DBO改进的极限学习机和历史策略库的概念及结合应用方法,构建了以基于近似灵敏度计算的电网节点有功及无功功率为输入,母线期望电压为输出的ELM模型。模型输出的母线电压作为控制依据,进一步转换为下发的用户侧可调设备调节指令。仿真算例的结果验证了所提方法的有效性和优越性。