scholarly journals Real Time Security Assessment of the Power System Using a Hybrid Support Vector Machine and Multilayer Perceptron Neural Network Algorithms

2019 ◽  
Vol 11 (13) ◽  
pp. 3586 ◽  
Author(s):  
Oyeniyi Akeem Alimi ◽  
Khmaies Ouahada ◽  
Adnan M. Abu-Mahfouz
Keyword(s):  
Neural Network ◽  
Support Vector Machine ◽  
Power System ◽  
Real Time ◽  
Multilayer Perceptron ◽  
Hybrid Algorithm ◽  
Test System ◽  
Support Vector ◽  
Detection Accuracy ◽  
Bus System

In today’s grid, the technological based cyber-physical systems have continued to be plagued with cyberattacks and intrusions. Any intrusive action on the power system’s Optimal Power Flow (OPF) modules can cause a series of operational instabilities, failures, and financial losses. Real time intrusion detection has become a major challenge for the power community and energy stakeholders. Current conventional methods have continued to exhibit shortfalls in tackling these security issues. In order to address this security issue, this paper proposes a hybrid Support Vector Machine and Multilayer Perceptron Neural Network (SVMNN) algorithm that involves the combination of Support Vector Machine (SVM) and multilayer perceptron neural network (MPLNN) algorithms for predicting and detecting cyber intrusion attacks into power system networks. In this paper, a modified version of the IEEE Garver 6-bus test system and a 24-bus system were used as case studies. The IEEE Garver 6-bus test system was used to describe the attack scenarios, whereas load flow analysis was conducted on real time data of a modified Nigerian 24-bus system to generate the bus voltage dataset that considered several cyberattack events for the hybrid algorithm. Sising various performance metricion and load/generator injections, en included in the manuscriptmulation results showed the relevant influences of cyberattacks on power systems in terms of voltage, power, and current flows. To demonstrate the performance of the proposed hybrid SVMNN algorithm, the results are compared with other models in related studies. The results demonstrated that the hybrid algorithm achieved a detection accuracy of 99.6%, which is better than recently proposed schemes.

scholarly journals Random Forest Regressor-Based Approach for Detecting Fault Location and Duration in Power Systems

Sensors ◽  
2022 ◽  
Vol 22 (2) ◽  
pp. 458
Author(s):  
Zakaria El Mrabet ◽  
Niroop Sugunaraj ◽  
Prakash Ranganathan ◽  
Shrirang Abhyankar
Keyword(s):  
Neural Network ◽  
Random Forest ◽  
Power Systems ◽  
Real Time ◽  
Fault Location ◽  
State Of The Art ◽  
Economic Consequences ◽  
Support Vector ◽  
Detection Accuracy ◽  
Data Driven Approach

Power system failures or outages due to short-circuits or “faults” can result in long service interruptions leading to significant socio-economic consequences. It is critical for electrical utilities to quickly ascertain fault characteristics, including location, type, and duration, to reduce the service time of an outage. Existing fault detection mechanisms (relays and digital fault recorders) are slow to communicate the fault characteristics upstream to the substations and control centers for action to be taken quickly. Fortunately, due to availability of high-resolution phasor measurement units (PMUs), more event-driven solutions can be captured in real time. In this paper, we propose a data-driven approach for determining fault characteristics using samples of fault trajectories. A random forest regressor (RFR)-based model is used to detect real-time fault location and its duration simultaneously. This model is based on combining multiple uncorrelated trees with state-of-the-art boosting and aggregating techniques in order to obtain robust generalizations and greater accuracy without overfitting or underfitting. Four cases were studied to evaluate the performance of RFR: 1. Detecting fault location (case 1), 2. Predicting fault duration (case 2), 3. Handling missing data (case 3), and 4. Identifying fault location and length in a real-time streaming environment (case 4). A comparative analysis was conducted between the RFR algorithm and state-of-the-art models, including deep neural network, Hoeffding tree, neural network, support vector machine, decision tree, naive Bayesian, and K-nearest neighborhood. Experiments revealed that RFR consistently outperformed the other models in detection accuracy, prediction error, and processing time.

scholarly journals A Novel Hybrid Particle Swarm- Multiverse Optimization based Voltage Stability Improvement in IEEE 57 Bus System

2018 ◽  
Vol 7 (2.24) ◽  
pp. 381
Author(s):  
P K.Dhal ◽  
K Ramash Kumar
Keyword(s):  
Power System ◽  
Hybrid Algorithm ◽  
Voltage Stability ◽  
System Analysis ◽  
Test System ◽  
Analysis Tool ◽  
Power System Analysis ◽  
Sudden Loading ◽  
And Control ◽  
Bus System

The major role of power system is voltage stability. It is required to plan properly and smooth operation and control. It presents a new approach of voltage stability improvement in IEEE 57 bus system using hybrid algorithm. The hybrid algorithm (PSO-MVO) is combination of PSO which is used for exploitation and MVO used for exploration. It is used in an uncertain environment. The FACTS device as STATCOM is connected in IEEE 57 test system to check for event of voltage stability improvement through power system analysis tool (PSAT) software. Once the ability of system goes through sudden loading, its stability gets affected. It desires compensation to boost voltage from disturbances. The varied operative condition while not used STATCOM in the system, used with STATCOM tuned by PSO-MVO algorithm are measured judge the performance of the projected system. The hybrid PSO-MVO technique is implemented in this paper to solve the proposed problem. The simulation results are obtained by PSAT software for 57 IEEE bus systems. The hybrid algorithm validates its effectiveness compare to individual PSO and MVO algorithm.     

scholarly journals Research on Fault Detection for Three Types of Wind Turbine Subsystems Using Machine Learning

Energies ◽  
2020 ◽  
Vol 13 (2) ◽  
pp. 460 ◽  
Author(s):  
Zuojun Liu ◽  
Cheng Xiao ◽  
Tieling Zhang ◽  
Xu Zhang
Keyword(s):  
Neural Network ◽  
Support Vector Machine ◽  
Fault Detection ◽  
Convolutional Neural Network ◽  
Wind Turbine ◽  
Cost Effective ◽  
Support Vector ◽  
Detection Accuracy ◽  
Machine Method ◽  
Scada System

In wind power generation, one aim of wind turbine control is to maintain it in a safe operational status while achieving cost-effective operation. The purpose of this paper is to investigate new techniques for wind turbine fault detection based on supervisory control and data acquisition (SCADA) system data in order to avoid unscheduled shutdowns. The proposed method starts with analyzing and determining the fault indicators corresponding to a failure mode. Three main system failures including generator failure, converter failure and pitch system failure are studied. First, the indicators data corresponding to each of the three key failures are extracted from the SCADA system, and the radar charts are generated. Secondly, the convolutional neural network with ResNet50 as the backbone network is selected, and the fault model is trained using the radar charts to detect the fault and calculate the detection evaluation indices. Thirdly, the support vector machine classifier is trained using the support vector machine method to achieve fault detection. In order to show the effectiveness of the proposed radar chart-based methods, support vector regression analysis is also employed to build the fault detection model. By analyzing and comparing the fault detection accuracy among these three methods, it is found that the fault detection accuracy by the models developed using the convolutional neural network is obviously higher than the other two methods applied given the same data condition. Therefore, the newly proposed method for wind turbine fault detection is proved to be more effective.

scholarly journals Intracranial Hemorrhage Detection in Head CT Using Double-Branch Convolutional Neural Network, Support Vector Machine, and Random Forest

2020 ◽  
Vol 10 (21) ◽  
pp. 7577
Author(s):  
Agata Sage ◽  
Pawel Badura
Keyword(s):  
Neural Network ◽  
Support Vector Machine ◽  
Random Forest ◽  
Convolutional Neural Network ◽  
Intracranial Hemorrhage ◽  
Support Vector ◽  
Detection Accuracy ◽  
Final Decision ◽  
Brain Hemorrhage ◽  
Computed Tomography Images

Brain hemorrhage is a severe threat to human life, and its timely and correct diagnosis and treatment are of great importance. Multiple types of brain hemorrhage are distinguished depending on the location and character of bleeding. The main division covers five subtypes: subdural, epidural, intraventricular, intraparenchymal, and subarachnoid hemorrhage. This paper presents an approach to detect these intracranial hemorrhage types in computed tomography images of the head. The model trained for each hemorrhage subtype is based on a double-branch convolutional neural network of ResNet-50 architecture. It extracts features from two chromatic representations of the input data: a concatenation of the image normalized in different intensity windows and a stack of three consecutive slices creating a 3D spatial context. The joint feature vector is passed to the classifier to produce the final decision. We tested two tools: the support vector machine and the random forest. The experiments involved 372,556 images from 11,454 CT series of 9997 patients, with each image annotated with labels related to the hemorrhage subtypes. We validated deep networks from both branches of our framework and the model with either of two classifiers under consideration. The obtained results justify the use of a combination of double-source features with the random forest classifier. The system outperforms state-of-the-art methods in terms of F1 score. The highest detection accuracy was obtained in intraventricular (96.7%) and intraparenchymal hemorrhages (93.3%).

scholarly journals Predictive risk mapping of human leptospirosis using support vector machine classification and multilayer perceptron neural network

2019 ◽  
Vol 14 (1) ◽  
Author(s):  
Mehrdad Ahangarcani ◽  
Mahdi Farnaghi ◽  
Mohammad Reza Shirzadi ◽  
Petter Pilesjö ◽  
Ali Mansourian
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Support Vector Machine ◽  
Multilayer Perceptron ◽  
Disease Incidence ◽  
Characteristic Curve ◽  
Support Vector ◽  
Support Vector Machine Classification ◽  
Risk Maps ◽  
Predictive Risk

Leptospirosis is a zoonotic disease found wherever human is in direct or indirect contact with contaminated water and environment. Considering the increasing number of cases of this disease in the northern part of Iran, identifying areas characterized by high disease incidence risk can help policy-makers develop strategies to prevent its further spread. This study presents an approach for generating predictive risk maps of leptospirosis using spatial statistics, environmental variables and machine learning. Moran's I demonstrated that the distribution of leptospirosis cases in the study area in Iran was highly clustered. Pearson’s correlation analysis was conducted to examine the type and strength of relationships between climate and topographical factors and incidence of the disease. To handle the complex and nonlinear problems involved, machine learning based on the support vector machine classification algorithm and multilayer perceptron neural network was exploited to generate annual and monthly predictive risk maps of leptospirosis distribution. Performance of both models was evaluated using receiver operating characteristic curve and Kappa coefficient. The output results demonstrated that both models are adequate for the prediction of the probability of leptospirosis incidence.

Sign in / Sign up

Export Citation Format

Share Document