Multilayer perceptron neural networks training through charged system search and its Application for non-technical losses detection

Pereira, Luiz Haroldo; Afonso, Luis Cláudio Súgi; Papa, João P.; Vale, Zita; Ramos, Caio C. O.; Gastaldello, Danillo S.; Souza, Ana Cristina de

doi:10.1109/isgt-la.2013.6554383

Cited by 36 publications

(18 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The probability of a user bus having a non-zero measurement bias on any of its phases, i.e., the probability of a user deciding to steal energy, is set to 0.3. Each energy thief's measurement bias magnitude is uniformly chosen from the interval [3,10]A and has the same angle as its corresponding phase. The substation measurements have zero biases with probability 1.…”

Section: Simulation Resultsmentioning

confidence: 99%

See 1 more Smart Citation

State estimation for energy theft detection in microgrids

Salinas

Luo

Liao

et al. 2014

9th International Conference on Communications and Networking in China

View full text Add to dashboard Cite

In traditional power networks, energy theft is a significant problem that causes severe financial losses to utility companies and legitimate users, jeopardizes system stability, and enables other illegal activities. Recently, governments and utility companies propose the Smart Grid as the next generation electric network to improve the current grid's efficiency, reliability, and security. In the Smart Grid, smart meters are deployed at users' premises to facilitate data collection, system control, etc. However, smart meters are vulnerable to cyber attacks, thus enabling easier energy pilfering. In this paper, we model the amount of energy stolen by a smart meter as a measurement bias, and propose an energy theft detection algorithm based on state estimation. In particular, our algorithm employs weighted least squares (WLS) state estimation, and can identify all the energy thieves in the system. We conduct extensive simulations in IEEE 13-bus and 123-bus test systems to validate our algorithm.

show abstract

Section: Simulation Resultsmentioning

confidence: 99%

“…Pereira et. al [10] find energy thieves by analyzing fine-grained load profiles from users' smart meters with a neural network classifier called charged system search. Huang et.…”

Section: Related Workmentioning

confidence: 99%

State estimation for energy theft detection in microgrids

Salinas

Luo

Liao

et al. 2014

9th International Conference on Communications and Networking in China

View full text Add to dashboard Cite

show abstract

“…AI has numerous applications on knowledge representation, information retrieval, speech recognition, understanding natural language, computer vision, bioinformatics, expert systems, robotics, game playing, and cyber defense with the help of various algorithms like artificial neural networks, genetic algorithms, artificial immune systems, particle-swarm intelligence, stochastic algorithms and fuzzy logic [19,20].…”

Section: B Artifical Intelligencementioning

confidence: 99%

“…Input layer, hidden layers and output layer respectively. The hidden layers' neurons are connected with all neurons in previous and next layers, and their connections are properly weighted [20]. The number of neurons in the input and output layers rely on the application, whereas neurons in the hidden layers are usually decided by trials [21].…”

Section: B Artifical Intelligencementioning

confidence: 99%

An Intelligent Technique for Detecting Malicious Users on Mobile Stores

Terzı

Yavanoğlu

Sinanç

et al. 2014

2014 13th International Conference on Machine Learning and Applications

View full text Add to dashboard Cite

In this study, malicious users who cause to resource exhausting are tried to detect in a telecommunication company network. Non-Legitimate users could cause lack of information availability and need countermeasures to prevent threat or limit permissions on the system. For this purpose, ANN based intelligent system is proposed and compared to SVM which is well known classification technique. According to results, proposed technique has achieved approximately 70% general success rate, 33% false positive rate and 27% false negative rate in controlled environment. Also ANN has high ability to work compare to SVM for our dataset. As a result proposed technique and developed application shows sufficient and acceptable defense mechanism in huge company networks. We discussed about this is initial study and ongoing research which is compared to the current literature. By the way, this study also shows that non security information such as users mobile experiences could be potential usage to prevent resource exhausting also known as DoS related attacks. Keywords-resource exhausting; artificial neural network(ANN); DoS attack, mobile store security 2014 13th International Conference on Machine Learning and Applications 978-1-4799-7415-3/14 $31.00

show abstract

“…Nayak et al proposed a firefly based higher order neural network for data classification for maintaining fast learning and avoids the exponential increase of processing units [34]. Many other metaheuristic algorithms, like ant colony optimization (ACO) [35, 36], Cuckoo Search (CS) [37], Artificial Bee Colony (ABC) [38, 39], Charged System Search (CSS) [40], Grey Wolf Optimizer (GWO) [41], Invasive Weed Optimization (IWO) [42], and Biogeography-Based Optimizer (BBO) [43] have been adopted for the research of neural network.…”

Section: Introductionmentioning

confidence: 99%

Training Feedforward Neural Networks Using Symbiotic Organisms Search Algorithm

Zhou

Luo

et al. 2016

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

Symbiotic organisms search (SOS) is a new robust and powerful metaheuristic algorithm, which stimulates the symbiotic interaction strategies adopted by organisms to survive and propagate in the ecosystem. In the supervised learning area, it is a challenging task to present a satisfactory and efficient training algorithm for feedforward neural networks (FNNs). In this paper, SOS is employed as a new method for training FNNs. To investigate the performance of the aforementioned method, eight different datasets selected from the UCI machine learning repository are employed for experiment and the results are compared among seven metaheuristic algorithms. The results show that SOS performs better than other algorithms for training FNNs in terms of converging speed. It is also proven that an FNN trained by the method of SOS has better accuracy than most algorithms compared.

show abstract

Multilayer perceptron neural networks training through charged system search and its Application for non-technical losses detection

Cited by 36 publications

References 14 publications

State estimation for energy theft detection in microgrids

State estimation for energy theft detection in microgrids

An Intelligent Technique for Detecting Malicious Users on Mobile Stores

Training Feedforward Neural Networks Using Symbiotic Organisms Search Algorithm

Contact Info

Product

Resources

About