GRU Enabled Intrusion Detection System for IoT Environment with Swarm Optimization and Gaussian Random Forest Classification

In recent years,machine learning(ML)and deep learning(DL)have significantly advanced intrusion detection systems,effectively addressing potential malicious attacks across networks.This paper introduces a robust method for detecting and categorizing attacks within the Internet of Things(IoT)environment,leveraging the NSL-KDD dataset.To achieve high accuracy,the authors used the feature extraction technique in combination with an autoencoder,integrated with a gated recurrent unit(GRU).Therefore,the accurate features are selected by using the cuckoo search algorithm integrated particle swarm optimization(PSO),and PSO has been employed for training the features.The final classification of features has been carried out by using the proposed RF-GNB random forest with the Gaussian Naïve Bayes classifier.The proposed model has been evaluated and its performance is verified with some of the standard metrics such as precision,accuracy rate,recall F1-score,etc.,and has been compared with different existing models.The generated results that detected approximately 99.87%of intrusions within the IoT environments,demonstrated the high performance of the proposed method.These results affirmed the efficacy of the proposed method in increasing the accuracy of intrusion detection within IoT network systems.

Fault Detection Based on Hierarchical Cluster Analysis in Wide Area Backup Protection System

In wide area backup protection of electric power systems, the prerequisite of protection device's accurate, fast and reliable performance is its corresponding fault type and fault location can be discriminated quickly and defined exactly. In our study, global information will be introduced into the backup protection system. By analyzing and computing real-time PMU measurements, basing on cluster analysis theory, we are using mainly hierarchical cluster analysis to search after the statistical laws of electrical quantities' marked changes. Then we carry out fast and exact detection of fault components and fault sections, and finally accomplish fault isolation. The facts show that the fault detection of fault component (fault section) can be performed successfully by hierarchical cluster analysis and calculation. The results of hierarchical cluster analysis are accurate and reliable, and the dendrograms of hierarchical cluster analysis are in intuition.

Novel Approach for Fault Detection and Diagnosis of Sensors in Air Handling Units Using Wavelet Energy Entropy

A novel fault diagnosis method for sensors in air handling unit（AHU） using wavelet energy entropy was presented. Instead of directly comparing the numerous data under noise conditiom, the wavelet energy entropy residual was compared in the proposed method. Three.level wavelet analysis was used to decompose the measurement data under both fault-free and faulty operation conditions. The concept of Shannon entropy was referred to define wavelet energy entropy of the wavelet coefficients. The sensor faults were diagnosed by comparing the deviation of the wavelet energy entropy of the measured signal and the estimated one with the preset threshold. Testing results showed that the wavelet energy entropy was sensitive to diagnose the biased faults. The wavelet energy entropy residuals exceed the threshold significantly when faults occur. In addition, the severer the faults were, the larger the residuals would be. The results prove that the proposed method is feasible and effective for the fault detection and diagnosis of the sensors.

Micro-seismic Event Detection of Hot Dry Rock based on the Gated Recurrent Unit Model and a Support Vector Machine

Micro-seismic monitoring is one of the most critical technologies that guide hydraulic fracturing in hot dry rock resource development. Micro-seismic monitoring requires high precision detection of micro-seismic events with a low signal-to-noise ratio. Because of this requirement, we propose a recurrent neural network model named gated recurrent unit and support vector machine(GRU;VM). The proposed model ensures high accuracy while reducing the parameter number and hardware requirement in the training process. Since micro-seismic events in hot dry rock produce large wave amplitudes and strong vibrations, it is difficult to reverse the onset of each individual event. In this study, we utilize a support vector machine(SVM) as a classifier to improve the micro-seismic event detection accuracy. To validate the methodology, we compare the simulation results of the short-term-average to the long-term-average(STA/LTA) method with GRU;VM method by using hot dry rock micro-seismic event data in Qinghai Province, China. Our proposed method has an accuracy of about 95% for identifying micro-seismic events with low signal-to-noise ratios. By ignoring smaller micro-seismic events, the detection procedure can be processed more efficiently, which is able to provide a real-time observation on the types of hydraulic fracturing in the reservoirs.

Recurrent Autoencoder Ensembles for Brake Operating Unit Anomaly Detection on Metro Vehicles

The anomaly detection of the brake operating unit (BOU) in thebrake systems on metro vehicle is critical for the safety and reliability ofthe trains. On the other hand, current periodic inspection and maintenanceare unable to detect anomalies in an early stage. Also, building an accurateand stable system for detecting anomalies is extremely difficult. Therefore,we present an efficient model that use an ensemble of recurrent autoencodersto accurately detect the BOU abnormalities of metro trains. This is the firstproposal to employ an ensemble deep learning technique to detect BOUabnormalities in metro train braking systems. One of the anomalous caseson metro vehicles is the case when the air cylinder (AC) pressures are less thanthe brake cylinder (BC) pressures in certain parts where the brake pressuresincrease before coming to a halt. Hence, in this work, we first extract the dataof BC and AC pressures. Then, the extracted data of BC and AC pressuresare divided into multiple subsequences that are used as an input for bothbi-directional long short-term memory (biLSTM) and bi-directional gatedrecurrent unit (biGRU) autoencoders. The biLSTM and biGRU autoencodersare trained using training dataset that only contains normal subsequences. Fordetecting abnormalities from test dataset which consists of abnormal subsequences, the mean absolute errors (MAEs) between original subsequences andreconstructed subsequences from both biLSTM and biGRU autoencoders arecalculated. As an ensemble step, the total error is calculated by averaging twoMAEs from biLSTM and biGRU autoencoders. The subsequence with totalerror greater than a pre-defined threshold value is considered an abnormality.We carried out the experiments using the BOU dataset on metro vehiclesin South Korea. Experimental results demonstrate that the ensemble modelshows better performance than other autoencoder-based models, which showsthe effectiveness of our ensemble model for detecting BOU anomalies onmetro trains.

Clustering residential electricity load curve via community detection in network

Performing analytics on the load curve(LC)of customers is the foundation for demand response which requires a better understanding of customers'consumption pattern(CP)by analyzing the load curve.However,the performances of previous widely-used LC clustering methods are poor in two folds:larger number of clusters,huge variances within a cluster(a CP is extracted from a cluster),bringing huge difficulty to understand the electricity consumption pattern of customers.In this paper,to improve the performance of LC clustering,a clustering framework incorporated with community detection is proposed.The framework includes three parts:network construction,community detection,and CP extraction.According to the cluster validity index(CVI),the integrated approach outperforms the previous state-of-the-art method with the same amount of clusters.And the approach needs fewer clusters to achieve the same performance measured by CVI.

Online Fault Detection Configuration on Equipment Side of a Variable-Air-Volume Air Handling Unit

With the development of the technology of the Internet of Things,more and more operational data can be collected from air conditioning systems.Unfortunately,the most of existing air conditioning controllers mainly provide controlling functions more than storing,processing or computing the measured data.This study develops an online fault detection configuration on the equipment side of air conditioning systems to realize these functions.Modbus communication is served to collect real-time operational data.The calculating programs are embedded to identify whether the measured signals exceed their limits or not,and to detect if sensor reading is frozen and other faults in relation to the operational performance are generated or not.The online fault detection configuration is tested on an actual variable-air-volume(VAV)air handling unit(AHU).The results show that the time ratio of fault detection exceeds 95.00%,which means that the configuration exhibits an acceptable fault detection effect.

An Intrusion Detection Method for Advanced Metering Infrastructure System Based on Federated Learning

An advanced metering infrastructure(AMI)system plays a key role in the smart grid(SG),but it is vulnerable to cyberattacks.Current detection methods for AMI cyberattacks mainly focus on the data center or a distributed independent node.On one hand,it is difficult to train an excellent detection intrusion model on a self-learning independent node.On the other hand,large amounts of data are shared over the network and uploaded to a central node for training.These processes may compromise data privacy,cause communication delay,and incur high communication costs.With these limitations,we propose an intrusion detection method for AMI system based on federated learning(FL).The intrusion detection system is deployed in the data concentrators for training,and only its model parameters are communicated to the data center.Furthermore,the data center distributes the learning to each data concentrator through aggregation and weight assignments for collaborative learning.An optimized deep neural network(DNN)is exploited for this proposed method,and extensive experiments based on the NSL-KDD dataset are carried out.From the results,this proposed method improves detection performance and reduces computation costs,communication delays,and communication overheads while guaranteeing data privacy.

Is every microorganism detected in the intensive care unit a nosocomial infection?Isn’t prevention more important than detection?

The present letter to the editor is related to the study entitled“Multidrug-resistant organisms in intensive care units and logistic analysis of risk factors.”Not every microorganism grown in samples taken from critically ill patients can be considered as an infectious agent.Accurate and adequate information about nosocomial infections is essential in introducing effective prevention programs in hospitals.Therefore,the development and implementation of care bundles for frequently used medical devices and invasive treatment devices(e.g.,intravenous catheters and invasive ventilation),adequate staffing not only for physicians,nurses,and other medical staff but also for housekeeping staff,and infection surveillance and motivational feedback are key points of infection prevention in the intensive care unit.

High Impedance Fault Detection of Distribution Network by Phasor Measurement Units

This paper proposes a new algorithm for High Impedance Fault (HIF) detection using Phasor Measurement Unit (PMU). This type of faults is difficult to detect by over current protection relays because of low fault current. In this paper, an index based on phasors change is proposed for HIF detection. The phasors are measured by PMU to obtain the square summation of errors. Two types of data are used for error calculation. The first one is sampled data and the second one is estimated data. But this index is not enough to declare presence of a HIF. Therefore another index introduces in order to distinguish the load switching from HIF. Second index utilizes 3rd harmonic current angle because this number of harmonic has a special behaviour during HIF. The verification of the proposed method is done by different simulation cases in EMTP/MATLAB.

Semantic Units Based Event Detection in Soccer Videos

A semantic unit based event detection scheme in soccer videos is proposed in this paper.The scheme can be characterized as a three-layer framework. At the lowest layer, low-level featuresincluding color, texture, edge, shape, and motion are extracted. High-level semantic events aredefined at the highest layer. In order to connect low-level features and high-level semantics, wedesign and define some semantic units at the intermediate layer. A semantic unit is composed of asequence of consecutives frames with the same cue that is deduced from low-level features. Based onsemantic units, a Bayesian network is used to reason the probabilities of events. The experiments forshoot and card event detection in soccer videos show that the proposed method has an encouragingperformance.

Development of systems for detection,early warning,and control of pipeline leakage in drinking water distribution:A case study

Water leakage in drinking water distribution systems is a serious problem for many cities and a huge challenge for water utilities.An integrated system for the detection,early warning,and control of pipeline leakage has been developed and successfully used to manage the pipeline networks in selected areas of Beijing.A method based on the geographic information system has been proposed to quickly and automatically optimize the layout of the instruments which detect leaks.Methods are also proposed to estimate the probability of each pipe segment leaking （on the basis of historic leakage data）,and to assist in locating the leakage points （based on leakage signals）.The district metering area （DMA） strategy is used.Guidelines and a flowchart for establishing a DMA to manage the large-scale looped networks in Beijing are proposed.These different functions have been implemented into a central software system to simplify the day-to-day use of the system.In 2007 the system detected 102 non-obvious leakages （i.e.,14.2% of the total detected in Beijing） in the selected areas,which was estimated to save a total volume of 2,385,000 m 3 of water.These results indicate the feasibility,efficiency and wider applicability of this system.

Ensemble Voting-Based Anomaly Detection for a Smart Grid Communication Infrastructure

Advanced Metering Infrastructure(AMI)is the metering network of the smart grid that enables bidirectional communications between each consumer’s premises and the provider’s control center.The massive amount of data collected supports the real-time decision-making required for diverse applications.The communication infrastructure relies on different network types,including the Internet.This makes the infrastructure vulnerable to various attacks,which could compromise security or have devastating effects.However,traditional machine learning solutions cannot adapt to the increasing complexity and diversity of attacks.The objective of this paper is to develop an Anomaly Detection System(ADS)based on deep learning using the CIC-IDS2017 dataset.However,this dataset is highly imbalanced;thus,a two-step sampling technique:random under-sampling and the Synthetic Minority Oversampling Technique(SMOTE),is proposed to balance the dataset.The proposed system utilizes a multiple hidden layer Auto-encoder(AE)for feature extraction and dimensional reduction.In addition,an ensemble voting based on both Random Forest(RF)and Convolu-tional Neural Network(CNN)is developed to classify the multiclass attack cate-gories.The proposed system is evaluated and compared with six different state-of-the-art machine learning and deep learning algorithms:Random Forest(RF),Light Gradient Boosting Machine(LightGBM),eXtreme Gradient Boosting(XGboost),Convolutional Neural Network(CNN),Long Short-Term Memory(LSTM),and bidirectional LSTM(biLSTM).Experimental results show that the proposed model enhances the detection for each attack class compared with the other machine learning and deep learning models with overall accuracy(98.29%),precision(99%),recall(98%),F_(1) score(98%),and the UNDetection rate(UND)(8%).

Traffic Anomaly DetectionMethod Based on Improved GRU and EFMS-Kmeans Clustering

In recent years,with the continuous development of information technology and the rapid growth of network scale,network monitoring and management become more and more important.Network traffic is an important part of network state.In order to ensure the normal operation of the network,improve the availability of the network,find network faults in time and deal with network attacks;it is necessary to detect the abnormal traffic in the network.Abnormal traffic detection is of great significance in the actual network management.Therefore,in order to improve the accuracy and efficiency of network traffic anomaly detection,this paper proposes a comprehensive anomaly detection method based on improved GRU traffic prediction and improved K-means clustering,and cascade the traffic prediction and clustering to achieve the purpose of anomaly detection.Firstly,an improved highway-GRU algorithm HS-GRU(An improved Gate Recurrent Unit neural network based on Highway network and STL algorithm,HS-GRU)is proposed,which combines STL decomposition algorithm with highway GRU neural network and uses this improved algorithm to predict traffic.And then,we proposed the EFMS-Kmeans algorithm(An improved clustering algorithmthat combined Mean Shift algorithmbased on electrostatic force with K-means clustering)to solve the shortcoming of the traditional K-means clustering which cannot automatically determine the number of clustering.The sum of the squared errors(SSE)method and the contour coefficient method were used to double test the clustering effect.After determining the clustering center,the potential energy gradient was directly used for anomaly detection by using the threshold method,which considered the local characteristics of the data and ensured the accuracy of anomaly detection.The simulation results show that the anomaly detection algorithm based on HS-GRU and EFMS-Kmeans clustering proposed in this paper can effectively improve the accuracy of flow anomaly detection and has important application value.

Chiller faults detection and diagnosis with sensor network and adaptive 1D CNN

Computer-empowered detection of possible faults for Heating,Ventilation and Air-Conditioning(HVAC)subsystems,e.g.,chillers,is one of the most important applications in Artificial Intelligence(AI)integrated Internet of Things(IoT).The cyber-physical system greatly enhances the safety and security of the working facilities,reducing time,saving energy and protecting humans’health.Under the current trends of smart building design and energy management optimization,Automated Fault Detection and Diagnosis(AFDD)of chillers integrated with IoT is highly demanded.Recent studies show that standard machine learning techniques,such as Principal Component Analysis(PCA),Support Vector Machine(SVM)and tree-structure-based algorithms,are useful in capturing various chiller faults with high accuracy rates.With the fast development of deep learning technology,Convolutional Neural Networks(CNNs)have been widely and successfully applied to various fields.However,for chiller AFDD,few existing works are adopting CNN and its extensions in the feature extraction and classification processes.In this study,we propose to perform chiller FDD using a CNN-based approach.The proposed approach has two distinct advantages over existing machine learning-based chiller AFDD methods.First,the CNN-based approach does not require the feature selection/extraction process.Since CNN is reputable with its feature extraction capability,the feature extraction and classification processes are merged,leading to a more neat AFDD framework compared to traditional approaches.Second,the classification accuracy is significantly improved compared to traditional methods using the CNN-based approach.

GPS Spoofing Attack Detection in Smart Grids Based on Improved CapsNet

This paper analyzes the influence of the global positionong system(GPS)spoofing attack(GSA)on phasor measurement units(PMU)measurements.We propose a detection method based on improved Capsule Neural Network(CapsNet)to handle this attack.In the improved CapsNet,the gated recurrent unit(GRU)is added to the front of the full connection layer of the CapsNet.The improved CapsNet trains and updates the network parameters according to the historical measurements of the smart grid.The detection method uses different structures to extract the temporal and spatial features of the measurements simultaneously,which can accurately distinguish the attacked data from the normal data,to improve the detection accuracy.Finally,simulation experiments are carried out on IEEE 14-,IEEE 118-bus systems.The experimental results show that compared with other detection methods,our method is proved to be more efficient.

A Value-at-Risk Based Approach for PMU Placement in Distribution Systems

With the application of phasor measurement units(PMU)in the distribution system,it is expected that the performance of the distribution system state estimation can be improved obviously with the PMU measurements into consideration.How to appropriately place the PMUs in the distribution is therefore become an important issue due to the economical consideration.According to the concept of efficient frontier,a value-at-risk based approach is proposed to make optimal placement of PMU taking account of the uncertainty of measure errors,statistical characteristics of the pseudo measurements,and reliability of the measurement instrument.The reasonability and feasibility of the proposed model is illustrated with 12-node system and IEEE-33 node system.Simulation results indicated that uncertainties of measurement error and instrument fault result in more PMU to be installed,and measurement uncertainty is the main affect factor unless the fault rate of PMU is quite high.

Personal glucose meters coupled with signal amplification technologies for quantitative detection of non-glucose targets:Recent progress and challenges in food safety hazards analysis

Ensuring food safety is paramount worldwide.Developing effective detection methods to ensure food safety can be challenging owing to trace hazards,long detection time,and resource-poor sites,in addition to the matrix effects of food.Personal glucose meter(PGM),a classic point-of-care testing device,possesses unique application advantages,demonstrating promise in food safety.Currently,many studies have used PGM-based biosensors and signal amplification technologies to achieve sensitive and specific detection of food hazards.Signal amplification technologies have the potential to greatly improve the analytical performance and integration of PGMs with biosensors,which is crucial for solving the challenges associated with the use of PGMs for food safety analysis.This review introduces the basic detection principle of a PGM-based sensing strategy,which consists of three key factors:target recognition,signal transduction,and signal output.Representative studies of existing PGM-based sensing strategies combined with various signal amplification technologies(nanomaterial-loaded multienzyme labeling,nucleic acid reaction,DNAzyme catalysis,responsive nanomaterial encapsulation,and others)in the field of food safety detection are reviewed.Future perspectives and potential opportunities and challenges associated with PGMs in the field of food safety are discussed.Despite the need for complex sample preparation and the lack of standardization in the field,using PGMs in combination with signal amplification technology shows promise as a rapid and cost-effective method for food safety hazard analysis.

基于Object Detection API的物流单元货架目标检测

随着人工智能的兴起,深度学习的方法已经被广泛地应用到各类图像目标的检测当中,并在复杂环境下取得了良好的效果。针对物流仓储环境,该文基于开源框架Tensorflow上的库Object Detection API,选择了Faster R-CNN算法和SSD-MobileNet算法,分别对物流单元货架上摆放的物流周转箱进行目标检测。实验结果表明,相比于Faster R-CNN算法,SSD-MobileNet算法能够同时满足实时性与准确率的要求。将训练所得的SSD-MobileNet模型移植到QT平台,设计了物流单元货架目标检测界面。

Loss-of-Main Monitoring and Detection for Distributed Generations Using Dynamic Principal Component Analysis

In this paper, our previous work on Principal Component Analysis (PCA) based fault detection method is extended to the dynamic monitoring and detection of loss-of-main in power systems using wide-area synchrophasor measurements. In the previous work, a static PCA model was built and verified to be capable of detecting and extracting system faulty events;however the false alarm rate is high. To address this problem, this paper uses a well-known ‘time lag shift’ method to include dynamic behavior of the PCA model based on the synchronized measurements from Phasor Measurement Units (PMU), which is named as the Dynamic Principal Component Analysis (DPCA). Compared with the static PCA approach as well as the traditional passive mechanisms of loss-of-main detection, the proposed DPCA procedure describes how the synchrophasors are linearly auto- and cross-correlated, based on conducting the singular value decomposition on the augmented time lagged synchrophasor matrix. Similar to the static PCA method, two statistics, namely T2 and Q with confidence limits are calculated to form intuitive charts for engineers or operators to monitor the loss-of-main situation in real time. The effectiveness of the proposed methodology is evaluated on the loss-of-main monitoring of a real system, where the historic data are recorded from PMUs installed in several locations in the UK/Ireland power system.