In order to solve the problems of oil‐paper insulation deterioration of oil‐immersed power transformers, including the instability of performance degradation quantity at the early stage, non‐linear deterioration process and uncertainty of life prediction, an improved Wiener stochastic process based on strong tracking filter (STF) is proposed to predict the remaining life of transformer oil‐paper insulation in this paper. Firstly, in order to grasp the dynamic process from healthy state to current deterioration state, a deterioration model is established based on Wiener random process. Secondly, in order to improve the accuracy, the initial values of model parameters are estimated based on expectation maximization (EM) algorithm, then with the increase of monitoring data, model parameters are updated by using STF algorithm, and the probability density function of remaining life is derived to predict the remaining life of oil‐paper insulation. Finally, the accuracy of the proposed method is verified by using furfural content as performance degradation index and accelerated thermal aging experimental data. Compared with the Bayesian–Wiener algorithm, the MSE of the proposed algorithm is reduced to 0.2138 and its MAPE is reduced more than 4%. Besides, the proposed method has the advantages of strong robustness and low uncertainty of residual life prediction.

To solve the oxygen supply problem of proton exchange membrane fuel cell (PEMFC), this paper presents a reduced-order active disturbance rejection control method for PEMFC air intake system based on the estimation of oxygen excess ratio(EOER-RADRC). First, a control-oriented third-order nonlinear model was established based on PEMFC air intake system. Then, based on the control-oriented model, the disturbance extended state observer was designed to solve the problem of unmeasured cathode pressure parameters in the calculation of oxygen excess ratio (OER), and realize the real-time estimation of OER. Next, aiming at the large phase lag of active disturbance rejection control (ADRC), a reduced-order ADRC method based on reduced-order extended state observer was proposed by a variable substitution technique. So, the proposed method reduces the observation phase lag and improves the dynamic response performance under good antidisturbance ability. Finally, the simulation and hardware-in-the-loop experience results showed that the proposed method has an excellent control effect compared with other methods.

This study is for the case where the available data of power transformer oil-paper insulation is limited to a small amount furfural data, to solve the problems in oil-paper insulation degradation modelling, such as few samples available, unknown function form of the degradation process, differences of individual transformers among degradation processes, and commonality of degradation trends. A power transformer oil-paper insulation degradation modelling and prediction method based on functional principal component analysis (FPCA) is proposed. First, discrete furfural data of oil-paper insulation degradation are converted into continuous functional data, and the common degradation information of transformers is extracted based on functional time warping technology. Second, the principal components of insulation degradation are extracted based on FPCA method, and the difference of degradation information of individual transformers is obtained by analysing the differential of principal component scores. Subsequently, power transformer oil-paper insulation degradation model is constructed, and finally, the degradation model is updated based on Bayesian theory and the oil-paper insulation degradation is predicted. The example results show that compared with traditional transformer oil-paper insulation degradation modelling method, the proposed method has obvious superiority in model accuracy.

The irregularities in the collection and transmission of user power data in the low-voltage power distribution station area have led to errors in the subsequent application analysis of the station area. In order to ensure the integrity of power data in low-voltage stations, a multi-user power missing data complement method based on improved deep convolutional self-encoding is proposed. First, according to the characteristics of the lack of multi-user power data in the low-voltage station area, the power data is formed into a spatio-temporal tensor data format that can be used for one-dimensional convolution operations. Then use the encoding and decoding capabilities of the improved deep convolutional selfencoding network to realize the reconstruction of missing data, and optimize the network structure by introducing residual learning and batch normalization (BN). Finally, based on the proposed method, two cases of random and continuous loss of user power data in a certain area are complemented. The results show that the method can accurately complete 40% of randomly missing data and 2 consecutive days of missing data. The proposed method has improved completion accuracy compared with traditional methods to varying degrees. INDEX TERMSIntelligent distribution network, low-voltage power distribution station area, deep convolutional autoencoder, residual learning, missing data completion.

No abstract