The evolution of telecommunications has allowed the development of broadband services based mainly on fiber optic backbone networks. The operation and maintenance of these optical networks is made possible by using su...The evolution of telecommunications has allowed the development of broadband services based mainly on fiber optic backbone networks. The operation and maintenance of these optical networks is made possible by using supervision platforms that generate alarms that can be archived in the form of log files. But analyzing the alarms in the log files is a laborious and difficult task for the engineers who need a degree of expertise. Identifying failures and their root cause can be time consuming and impact the quality of service, network availability and service level agreements signed between the operator and its customers. Therefore, it is more than important to study the different possibilities of alarms classification and to use machine learning algorithms for alarms correlation in order to quickly determine the root causes of problems faster. We conducted a research case study on one of the operators in Cameroon who held an optical backbone based on SDH and WDM technologies with data collected from 2016-03-28 to “2022-09-01” with 7201 rows and 18. In this paper, we will classify alarms according to different criteria and use 02 unsupervised learning algorithms namely the K-Means algorithm and the DBSCAN to establish correlations between alarms in order to identify root causes of problems and reduce the time to troubleshoot. To achieve this objective, log files were exploited in order to obtain the root causes of the alarms, and then K-Means algorithm and the DBSCAN were used firstly to evaluate their performance and their capability to identify the root cause of alarms in optical network.展开更多
随着电子战技术的快速发展,低截获概率(Low Probability of Interception,LPI)雷达得到了越来越广泛的应用,基于深度学习的智能检测方法在低信噪比(Signal-to-Noise Ratio,SNR)条件下对低截获概率雷达信号展现出不俗的检测性能。由于噪...随着电子战技术的快速发展,低截获概率(Low Probability of Interception,LPI)雷达得到了越来越广泛的应用,基于深度学习的智能检测方法在低信噪比(Signal-to-Noise Ratio,SNR)条件下对低截获概率雷达信号展现出不俗的检测性能。由于噪声对神经网络拟合会造成很大的影响,智能检测方法在低信噪比条件下表现为虚警概率(Probability of False Alarm,PFA)随着信噪比的降低而增高,虚警概率不恒定。针对此问题,提出了一种注意力增强的分类智能检测方法Channel-DeepIQ。首先,利用数字信道化(Digital Channelization,DC)技术得到网络输入数据,然后利用子信道分类网络(Subchannel Classification Network,SCN)对子信道进行分类,最后采用脉冲边沿检测网络(Pulse Edge Detection Network,PEDN)完成对应子信道的最终检测。Channel-DeepIQ在处理过程中引入注意力机制,具有更强的表示能力和特征细化能力,能够在不同信噪比条件下保持稳定的虚警概率。仿真和实验结果表明,当信噪比为-10 dB时,信号检测概率达90%以上,且虚警概率维持在1×10^(-3)的水平。与其他方法相比,所提方法的检测性能更佳,并且在不同信噪比条件下具有更强的泛化能力。展开更多
水电站计算机监控系统长期运行于强电磁干扰、多源异构数据并发及工况瞬变的复杂工业环境中,导致传统监控方法虚警频发。为此,提出基于径向基函数神经网络(radial basis function neural network,RBFNN)的水电站监控系统智能告警方法。...水电站计算机监控系统长期运行于强电磁干扰、多源异构数据并发及工况瞬变的复杂工业环境中,导致传统监控方法虚警频发。为此,提出基于径向基函数神经网络(radial basis function neural network,RBFNN)的水电站监控系统智能告警方法。该方法首先采集了水电站监控系统中的设备状态数据、日志数据和配置数据,构建了包含输入层、隐含层和输出层的RBFNN故障诊断模型。其中输入层设置3个节点对应三类监控数据,输出层设置4个节点分别对应电气、机械、环境及其他监控部分的故障分类。通过参数寻优确定隐含层结构,并采用聚类方法确定基函数中心,使用最小二乘法优化网络权重,完成模型训练。实验依托某实际水电站监控系统展开,设计涵盖紧急至信息级的七级故障分类体系,并部署分层告警框架进行验证。结果表明,所提方法能有效抑制环境干扰引致的虚警,精准识别并分级呈现故障,仅在非关键性预警与信息级故障中各出现一次误判,显著优于对照方法。研究表明,所提方法能显著提升复杂工业背景下的故障辨识鲁棒性与告警直观性,为水电站监控系统的智能化运维提供了一种有效解决方案。展开更多
In this work a review of existing fire-detector types has been carried out along with the development of a low cost, portable, and reliable microcontroller based automated fire alarm system for remotely alerting any f...In this work a review of existing fire-detector types has been carried out along with the development of a low cost, portable, and reliable microcontroller based automated fire alarm system for remotely alerting any fire incidents in household or industrial premises. The aim of the system designed is to alert the distant property-owner efficiently and quickly by sending short message (SMS) via GSM network. A Linear integrated temperature sensor detects temperature beyond preset value whereas semiconductor type sensor detects presence of smoke or gas from fire hazards. The sensor units are connected via common data line to ATMega8L AVR microcontroller. A SIM300CZ GSM kit based network module, capable of operating in standard GSM bands, has been used to send alert messages. The system is implemented on printed circuit board (PCB) and tested under different experimental conditions to evaluate its performances.展开更多
The development of Internet brings a great challenge to the survivability of the supporting distributed intelligent optical networks. The emergence of Active-Fault-Alarm (AFA) technology makes it possible for the syst...The development of Internet brings a great challenge to the survivability of the supporting distributed intelligent optical networks. The emergence of Active-Fault-Alarm (AFA) technology makes it possible for the system to be aware of the incoming or possible fault in advance and to provide possibility to develop a more active restoration mechanism. On this base, an Active Segment Pre-Restoration (ASPR) mechanism for distributed optical network is proposed. ASPR allows establishing a Segment Pre-Restoration Path (SPR-Path) as a work path, which is initiated by the local node, in advance of potential fault occuring and keeps the SPR-Path only during the low-quality or fault period. Simulation results show that the ASPR mechanism has better restoration performance compared with that of Active Restoration (AR) scheme.展开更多
Sensors for fire alarms require a high level of predictive variables to ensure accurate detection, injury prevention, and loss prevention. Bayesian networks can aid in enhancing early fire detection capabilities and r...Sensors for fire alarms require a high level of predictive variables to ensure accurate detection, injury prevention, and loss prevention. Bayesian networks can aid in enhancing early fire detection capabilities and reducing the frequency of erroneous fire alerts, thereby enhancing the effectiveness of numerous safety monitoring systems. This research explores the development of optimized probabilistic graphic models for the discretization thresholds of alarm system predictor variables. The study presents a statistical model framework that increases the efficacy of fire detection by predicting the discretization thresholds of alarm system predictor variable fluctuations used to detect the onset of fire. The work applies the Bayesian networks and probabilistic visual models to reveal the specific characteristics required to cope with fire detection strategies and patterns. The adopted methodology utilizes a combination of prior knowledge and statistical data to draw conclusions from observations. Utilizing domain knowledge to compute conditional dependencies between network variables enabled predictions to be made through the application of specialized analytical and simulation techniques.展开更多
Quantized fault detection for sensor/actuator faults of networked control systems （NCSs） with time delays both in the sensor-to-controller channel and controller-to-actuator channel is concerned in this paper. A fau...Quantized fault detection for sensor/actuator faults of networked control systems （NCSs） with time delays both in the sensor-to-controller channel and controller-to-actuator channel is concerned in this paper. A fault model is set up based on the possible cases of sensor/atuator faults. Then, the model predictive control is used to compensate the time delay. When the sensors and actuators are healthy, an H stability criterion of the state predictive observer is obtained in terms of linear matrix inequality. A new threshold computational method that conforms to the actual situation is proposed. Then, the thresholds of the false alarm rate （FAR） and miss detection rate （MDR） are presented by using our proposed method, which are also compared with the ones given in the existin~ literatures. Finally, some numerical simulations are shown to demonstrate the effectiveness of the proposed method.展开更多
In this paper, the communication technology of seismic precursor network instrument is introduced, including instruction format and returned information format of instrument login, status information acquisition, and ...In this paper, the communication technology of seismic precursor network instrument is introduced, including instruction format and returned information format of instrument login, status information acquisition, and current measured data acquisition. The remote monitoring alarm software is based on this technology, and also introduced that the structure of monitoring information table, abnormal alarm index, and monitoring strategy. The application of the software raises instrument running rate and observation data quality.展开更多
文摘The evolution of telecommunications has allowed the development of broadband services based mainly on fiber optic backbone networks. The operation and maintenance of these optical networks is made possible by using supervision platforms that generate alarms that can be archived in the form of log files. But analyzing the alarms in the log files is a laborious and difficult task for the engineers who need a degree of expertise. Identifying failures and their root cause can be time consuming and impact the quality of service, network availability and service level agreements signed between the operator and its customers. Therefore, it is more than important to study the different possibilities of alarms classification and to use machine learning algorithms for alarms correlation in order to quickly determine the root causes of problems faster. We conducted a research case study on one of the operators in Cameroon who held an optical backbone based on SDH and WDM technologies with data collected from 2016-03-28 to “2022-09-01” with 7201 rows and 18. In this paper, we will classify alarms according to different criteria and use 02 unsupervised learning algorithms namely the K-Means algorithm and the DBSCAN to establish correlations between alarms in order to identify root causes of problems and reduce the time to troubleshoot. To achieve this objective, log files were exploited in order to obtain the root causes of the alarms, and then K-Means algorithm and the DBSCAN were used firstly to evaluate their performance and their capability to identify the root cause of alarms in optical network.
文摘随着电子战技术的快速发展,低截获概率(Low Probability of Interception,LPI)雷达得到了越来越广泛的应用,基于深度学习的智能检测方法在低信噪比(Signal-to-Noise Ratio,SNR)条件下对低截获概率雷达信号展现出不俗的检测性能。由于噪声对神经网络拟合会造成很大的影响,智能检测方法在低信噪比条件下表现为虚警概率(Probability of False Alarm,PFA)随着信噪比的降低而增高,虚警概率不恒定。针对此问题,提出了一种注意力增强的分类智能检测方法Channel-DeepIQ。首先,利用数字信道化(Digital Channelization,DC)技术得到网络输入数据,然后利用子信道分类网络(Subchannel Classification Network,SCN)对子信道进行分类,最后采用脉冲边沿检测网络(Pulse Edge Detection Network,PEDN)完成对应子信道的最终检测。Channel-DeepIQ在处理过程中引入注意力机制,具有更强的表示能力和特征细化能力,能够在不同信噪比条件下保持稳定的虚警概率。仿真和实验结果表明,当信噪比为-10 dB时,信号检测概率达90%以上,且虚警概率维持在1×10^(-3)的水平。与其他方法相比,所提方法的检测性能更佳,并且在不同信噪比条件下具有更强的泛化能力。
文摘水电站计算机监控系统长期运行于强电磁干扰、多源异构数据并发及工况瞬变的复杂工业环境中,导致传统监控方法虚警频发。为此,提出基于径向基函数神经网络(radial basis function neural network,RBFNN)的水电站监控系统智能告警方法。该方法首先采集了水电站监控系统中的设备状态数据、日志数据和配置数据,构建了包含输入层、隐含层和输出层的RBFNN故障诊断模型。其中输入层设置3个节点对应三类监控数据,输出层设置4个节点分别对应电气、机械、环境及其他监控部分的故障分类。通过参数寻优确定隐含层结构,并采用聚类方法确定基函数中心,使用最小二乘法优化网络权重,完成模型训练。实验依托某实际水电站监控系统展开,设计涵盖紧急至信息级的七级故障分类体系,并部署分层告警框架进行验证。结果表明,所提方法能有效抑制环境干扰引致的虚警,精准识别并分级呈现故障,仅在非关键性预警与信息级故障中各出现一次误判,显著优于对照方法。研究表明,所提方法能显著提升复杂工业背景下的故障辨识鲁棒性与告警直观性,为水电站监控系统的智能化运维提供了一种有效解决方案。
文摘In this work a review of existing fire-detector types has been carried out along with the development of a low cost, portable, and reliable microcontroller based automated fire alarm system for remotely alerting any fire incidents in household or industrial premises. The aim of the system designed is to alert the distant property-owner efficiently and quickly by sending short message (SMS) via GSM network. A Linear integrated temperature sensor detects temperature beyond preset value whereas semiconductor type sensor detects presence of smoke or gas from fire hazards. The sensor units are connected via common data line to ATMega8L AVR microcontroller. A SIM300CZ GSM kit based network module, capable of operating in standard GSM bands, has been used to send alert messages. The system is implemented on printed circuit board (PCB) and tested under different experimental conditions to evaluate its performances.
基金supported in part by National High Technical Research and Development Program of China (863 Program)under Grant No.2009AA01z255, 2009AA01A345National Key Basic Research Program of China (973 Program) under Grant No.2007CB310705+1 种基金National Natural Science Foundation of China under Grant No. 60932004Specialized Research Fund for the Doctoral Program of Higher Education of China under Grant No.200800130001
文摘The development of Internet brings a great challenge to the survivability of the supporting distributed intelligent optical networks. The emergence of Active-Fault-Alarm (AFA) technology makes it possible for the system to be aware of the incoming or possible fault in advance and to provide possibility to develop a more active restoration mechanism. On this base, an Active Segment Pre-Restoration (ASPR) mechanism for distributed optical network is proposed. ASPR allows establishing a Segment Pre-Restoration Path (SPR-Path) as a work path, which is initiated by the local node, in advance of potential fault occuring and keeps the SPR-Path only during the low-quality or fault period. Simulation results show that the ASPR mechanism has better restoration performance compared with that of Active Restoration (AR) scheme.
文摘Sensors for fire alarms require a high level of predictive variables to ensure accurate detection, injury prevention, and loss prevention. Bayesian networks can aid in enhancing early fire detection capabilities and reducing the frequency of erroneous fire alerts, thereby enhancing the effectiveness of numerous safety monitoring systems. This research explores the development of optimized probabilistic graphic models for the discretization thresholds of alarm system predictor variables. The study presents a statistical model framework that increases the efficacy of fire detection by predicting the discretization thresholds of alarm system predictor variable fluctuations used to detect the onset of fire. The work applies the Bayesian networks and probabilistic visual models to reveal the specific characteristics required to cope with fire detection strategies and patterns. The adopted methodology utilizes a combination of prior knowledge and statistical data to draw conclusions from observations. Utilizing domain knowledge to compute conditional dependencies between network variables enabled predictions to be made through the application of specialized analytical and simulation techniques.
基金supported by National Natural Science Foundation of China(No.61074065)Natural Science Foundation of Hebei Province(No.F2012203184)Specialized Research Fund for the Doctoral Program of Higher Education of China(No.20111333120009)
文摘Quantized fault detection for sensor/actuator faults of networked control systems （NCSs） with time delays both in the sensor-to-controller channel and controller-to-actuator channel is concerned in this paper. A fault model is set up based on the possible cases of sensor/atuator faults. Then, the model predictive control is used to compensate the time delay. When the sensors and actuators are healthy, an H stability criterion of the state predictive observer is obtained in terms of linear matrix inequality. A new threshold computational method that conforms to the actual situation is proposed. Then, the thresholds of the false alarm rate （FAR） and miss detection rate （MDR） are presented by using our proposed method, which are also compared with the ones given in the existin~ literatures. Finally, some numerical simulations are shown to demonstrate the effectiveness of the proposed method.
文摘In this paper, the communication technology of seismic precursor network instrument is introduced, including instruction format and returned information format of instrument login, status information acquisition, and current measured data acquisition. The remote monitoring alarm software is based on this technology, and also introduced that the structure of monitoring information table, abnormal alarm index, and monitoring strategy. The application of the software raises instrument running rate and observation data quality.
