A New Method for Transformer Fault Prediction Based on Multifeature Enhancement and Refined Long Short-Term Memory

Ma, Xin; Hu, Hao; Shang, Yizi

doi:10.1109/tim.2021.3098383

Cited by 37 publications

(10 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…By comparison, the method proposed in this paper can achieve higher accuracy, does not rely on a large number of data and samples and can achieve fault prediction. Based on the analysis of typical variables, details are given in literature [31] Method based on clustering and data 80-94% High diagnostic accuracy A large number of data and samples are required and failure can-not be predicted It belongs to the method of machine learning, details are given in literature [28] Sensitive variable feature fusion analysis method proposed in this paper…”

Section: Comparative Analysis With Typical Methodsmentioning

confidence: 99%

“…In the literature [30], an early defect prediction model based on wavelet packet decomposition and dynamic kernel principal component analysis (wpd-dkpca) is investigated using machine learning technology to fulfill the demands of engineering applications. In the literature [31], to maintain the objectivity of prediction model, the multi-objective particle swarm optimization technique and random walk strategy are employed to optimize long-term and short-term memory (LSTM) network. The literature [32] provided a detection index based on residual SMD (square of the Mahalanobis distance) that could diagnose but not predict faults.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Fault Critical Point Prediction Method of Nuclear Gate Valve with Small Samples Based on Characteristic Analysis of Operation

Liu

et al. 2022

Materials

View full text Add to dashboard Cite

The number of fault samples for the new nuclear valve is commonly rare; thus, the machine learning algorithm is not suitable for the fault prediction of this kind of equipment. In order to overcome this difficulty, this paper proposes a novel method for the fault critical point prediction of the gate valve based on the characteristic analysis of the operation process variables. The operation process of gate valve switch often contains various fault characteristics and information, and this method first adopts the Shannon entropy to describe the power spectrum of vibration signal relevant to the operation process of gate valve switch, and then employs the mean value of the power spectrum entropy as an indirect process variable and further investigates the differences between the indirect process variable under the healthy state and the fault state with a different fault degree. In addition, the power signal of the gate valve is also employed as the direct process variable and the features of the direct process variable under the healthy state and the fault state with different fault degrees are also investigated. Based on the previous indirect process variable and direct process variable, the prediction approach for the critical point of the gate valve fault is established by analyzing the change in the indirect process variable and direct process variable before and after faults. Finally, the data of a nuclear gate valve experiment are employed to demonstrate the feasibility of the proposed method and the results show that the proposed method can effectively predict the fault critical point of the mentioned nuclear gate valve. If the diagnostic threshold is set properly, the critical point prediction of a nuclear gate valve fault can be realized as 100% or close to 100%. Furthermore, the proposed method can be directly applied to the nuclear gate valve in a nuclear power plant to improve the operation reliability of the valve. At the same time, the method can be applied to the fault diagnosis and prediction of valves in other fields, such as the chemical industry.

show abstract

Section: Comparative Analysis With Typical Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Fault Critical Point Prediction Method of Nuclear Gate Valve with Small Samples Based on Characteristic Analysis of Operation

Liu

et al. 2022

Materials

View full text Add to dashboard Cite

show abstract

“…The proposed methods in [21] and [22] are useful to carry out preventive maintenance and inspection of power transformers but identification of the types of failures is not achieved. To address this problem, an LSTM model integrated with an improved deep residual shrinkage network, the multiobjective PSO and a random walk strategy was presented in [23]. The proposed model can identify fault signs in advance, but the forecasting accuracy is reduced beyond 2 weeks prediction period.…”

Section: Introductionmentioning

confidence: 99%

“…In addition, the fault classification method is based on fault codes in the relevant IEC standard. In comparison to work in [23], this paper has introduced intelligent classifiers to distinguish between normal state and specific fault classes using large-scale DGA field data [18].…”

Section: Introductionmentioning

confidence: 99%

Dynamic Incipient Fault Forecasting for Power Transformers Using an LSTM Model

Wang

Littler

Liu

2023

IEEE Trans. Dielect. Electr. Insul.

View full text Add to dashboard Cite

“…Reviewing existing research, most of the studies focused on the concentration prediction of dissolved gases in oil [ 15 ]. Lu et al proposed the calculation of the gray correlation coefficient of gas feature selection based on gray correlation analysis (GRA) and then used Gaussian process regression (GPR) to predict the dissolved gas value.…”

Section: Introductionmentioning

confidence: 99%

Concentration Prediction of Polymer Insulation Aging Indicator-Alcohols in Oil Based on Genetic Algorithm-Optimized Support Vector Machines

Zhang

Wang

et al. 2022

Polymers

View full text Add to dashboard Cite

The predictive model of aging indicator based on intelligent algorithms has become an auxiliary method for the aging condition of transformer polymer insulation. However, most of the current research on the concentration prediction of aging products focuses on dissolved gases in oil, and the concentration prediction of alcohols in oil is ignored. As new types of aging indicators, alcohols (methanol, ethanol) are becoming prevalent in the aging evaluation of transformer polymer insulation. To address this, this study proposes a prediction model for the concentration of alcohols based on a genetic-algorithm-optimized support vector machine (GA-SVM). Firstly, accelerated thermal aging experiments on oil-paper insulation are conducted, and the concentration of alcohols is measured. Then, the data of the past 4 days of aging are used as the input feature of SVM, and the GA algorithm is utilized to optimize the kernel function parameter and penalty factor of SVM. Moreover, the concentrations of methanol and ethanol are predicted, after which the prediction accuracy of other algorithms and GA-SVM are compared. Finally, an industrial software program for predicting the concentration of methanol and ethanol is established. The results show that the mean square errors (MSE) of methanol and ethanol concentration predictions of the model proposed in this paper are 0.008 and 0.003, respectively. The prediction model proposed in this paper can track changes in methanol and ethanol concentrations well, providing a theoretical basis for the field of alcohol concentration prediction in transformer oil.

show abstract

A New Method for Transformer Fault Prediction Based on Multifeature Enhancement and Refined Long Short-Term Memory

Cited by 37 publications

References 27 publications

Fault Critical Point Prediction Method of Nuclear Gate Valve with Small Samples Based on Characteristic Analysis of Operation

Fault Critical Point Prediction Method of Nuclear Gate Valve with Small Samples Based on Characteristic Analysis of Operation

Dynamic Incipient Fault Forecasting for Power Transformers Using an LSTM Model

Concentration Prediction of Polymer Insulation Aging Indicator-Alcohols in Oil Based on Genetic Algorithm-Optimized Support Vector Machines

Contact Info

Product

Resources

About