Big Data in Power Generation

Moleda, Marek; Mrozek, Dariusz

doi:10.1007/978-3-030-19093-4_2

Cited by 8 publications

(5 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…𝑥 𝑑 0 (6) In the context of online equipment condition monitoring, the Mahalanobis Distance technique as the equipment's healthy score can merely detect the presence of an equipment anomaly without elucidating which feature is the most significant contributor to the anomalous condition. One purpose of this research was to determine the parameters that contributed to the source of the anomaly using a reconstructed error signal from each normalized individual variable.…”

Section: 𝛼 = 𝑃(𝑥 < 𝑥 𝑑 ) = ∫ 𝑓 ̂(𝑥)𝑑𝑥mentioning

confidence: 99%

“…The threshold value of MD for the anomalous condition is determined numerically using the CDF with PDF integration of the MD distribution with the training dataset, as indicated in Eqs. ( 4) to (6). The confidence level is set to 99.9%.…”

Section: Online Anomaly Detectionmentioning

confidence: 99%

“…Big data and artificial intelligence technologies are developing rapidly and have been widely researched to apply to the broad sector [2][3][4]. Previous studies have developed many methods to utilize big data in the electricity sector to increase the reliability and effectiveness of maintenance [4][5][6][7][8]. Big data analysis applications were developed in the electricity sectors, among others, to improve demand response, high-accuracy forecasting, failures/ downtimes detection, high-accuracy operation planning model, equipment monitoring, and lifetime extension with failure detection as well as feature extraction and advanced visualization [2,9].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

RNN-Autoencoder Approach for Anomaly Detection in Power Plant Predictive Maintenance Systems

2022

IJIES

View full text Add to dashboard Cite

Induced Draft Fans (IDF) and Primary Air Fans (PAF) are critical equipment in steam power plants. Anomaly detection based on machine learning models is an approach that is currently being developed for optimization and increasing the effectiveness of predictive maintenance (PDM) as well as increasing the reliability of thermal power plants. The research aims to develop a data-driven model for diagnostic and prognostic equipment, produce accurate predictions using many sensor data taken in real-time from the SCADA system, and design a PDM management framework using an anomaly detection system. This research proposes to use a combined recurrent neural network (RNN)-autoencoder approach as a "normal" behavior model (NBM) with the Mahalanobis Distance (MD) statistical method for the detection of anomalies in power plant equipment. Based on time-series input sensor data, the RNNautoencoder is utilized to predict the behavior of the equipment in health circumstances. In contrast, the MD is used to determine the distance between the actual parameters of the equipment and its "normal" behavior prediction to determine the anomaly condition. This study examined the performance of the long short-term memory (LSTM) and gated recurrent unit (GRU) models in modeling normal behavior with hyperparameter optimization. The LSTM with the best hyperparameters had a validation loss of 5,690 x 10-4 and a validation accuracy of 93.36 percent, whereas the GRU with the best hyperparameters had a validation loss of 4.484 x 10-4 and a validation accuracy of 93.47 percent. GRU can outperform the LSTM model. The proposed framework can detect anomaly conditions in various cases of IDF and PAF equipment disturbances, both in early warning of equipment failure and during downtime conditions.

show abstract

Section: 𝛼 = 𝑃(𝑥 < 𝑥 𝑑 ) = ∫ 𝑓 ̂(𝑥)𝑑𝑥mentioning

confidence: 99%

Section: Online Anomaly Detectionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

RNN-Autoencoder Approach for Anomaly Detection in Power Plant Predictive Maintenance Systems

2022

IJIES

View full text Add to dashboard Cite

show abstract

“…The patterns detected are called anomalies [3]. The CMSs store a large amount of operation data in power plants [4][5][6], and the use of this data represents an important field for future smart power plants (SPPs) [7,8]. Therefore, anomaly detection using data-driven approaches is a popular research topic for SPPs.…”

Section: Introductionmentioning

confidence: 99%

Anomaly Detection of Power Plant Equipment Using Long Short-Term Memory Based Autoencoder Neural Network

Zhang

Yang

et al. 2020

Sensors

View full text Add to dashboard Cite

Anomaly detection is of great significance in condition-based maintenance of power plant equipment. The conventional fixed threshold detection method is not able to perform early detection of equipment abnormalities. In this study, a general anomaly detection framework based on a long short-term memory-based autoencoder (LSTM-AE) network is proposed. A normal behavior model (NBM) is established to learn the normal behavior patterns of the operating variables of the equipment in space and time. Based on the similarity analysis between the NBM output distribution and the corresponding measurement distribution, the Mahalanobis distance (MD) is used to describe the overall residual (OR) of the model. The reasonable range is obtained using kernel density estimation (KDE) with a 99% confidence interval, and the OR is monitored to detect abnormalities in real-time. An induced draft fan is chosen as a case study. Results show that the established NBM has excellent accuracy and generalizability, with average root mean square errors of 0.026 and 0.035 for the training and test data, respectively, and average mean absolute percentage errors of 0.027%. Moreover, the abnormal operation case shows that the proposed framework can be effectively used for the early detection of abnormalities.

show abstract

“…A significant problem with analytical work in the production environment is to determine the veracity of the data [8][9][10]. In the case of manually entered data, such as data entered in failure logs, the data rarely contain the specific time of the event, and the content requires interpretation by an expert.…”

Section: Introductionmentioning

confidence: 99%

Predictive Maintenance of Boiler Feed Water Pumps Using SCADA Data

Moleda

Momot

Mrozek

2020

Sensors

Self Cite

View full text Add to dashboard Cite

IoT enabled predictive maintenance allows companies in the energy sector to identify potential problems in the production devices far before the failure occurs. In this paper, we propose a method for early detection of faults in boiler feed pumps using existing measurements currently captured by control devices. In the experimental part, we work on real measurement data and events from a coal fired power plant. The main research objective is to implement a model that detects deviations from the normal operation state based on regression and to check which events or failures can be detected by it. The presented technique allows the creation of a predictive system working on the basis of the available data with a minimal requirement of expert knowledge, in particular the knowledge related to the categorization of failures and the exact time of their occurrence, which is sometimes difficult to identify. The paper shows that with modern technologies, such as the Internet of Things, big data, and cloud computing, it is possible to integrate automation systems, designed in the past only to control the production process, with IT systems that make all processes more efficient through the use of advanced analytic tools.

show abstract

Big Data in Power Generation

Cited by 8 publications

References 13 publications

RNN-Autoencoder Approach for Anomaly Detection in Power Plant Predictive Maintenance Systems

RNN-Autoencoder Approach for Anomaly Detection in Power Plant Predictive Maintenance Systems

Anomaly Detection of Power Plant Equipment Using Long Short-Term Memory Based Autoencoder Neural Network

Predictive Maintenance of Boiler Feed Water Pumps Using SCADA Data

Contact Info

Product

Resources

About