Transformer fault diagnosis based on adversarial generative networks and deep stacked autoencoder.

Establishing a deep learning model for transformer fault diagnosis using transformer oil chromatogram data requires a large number of fault samples. The lack and imbalance of oil chromatogram data can lead to overfitting, lack of representativeness of the model, and unsatisfactory prediction results on test set data, making it difficult to accurately diagnose transformer faults. A conditional Wasserstein generative adversarial network with gradient penalty optimization (CWGAN-GP) is adopted in this paper, which based on gradient penalty optimization and expand the oil chromatography fault samples of 500 sets of transformer oil chromatography data with 5 types of faults. The proposed method is used to classify transformer faults using a deep autoencoder, and the sample quality of the neural network model proposed in this paper is compared with several other variants of generative adversarial neural network models. The research results show that after using the method proposed in this paper for sample expansion, the overall accuracy of fault diagnosis can reach 93.2 %, which is 4.98 % higher than the original imbalanced samples. Compared with other sample expansion methods, the accuracy of fault diagnosis of the algorithm in this paper is improved by 1.70 %-3.05 %.

Improving methane productivity of waste activated sludge by ultrasound and alkali pretreatment in microbial electrolysis cell and anaerobic digestion coupled system.

In order to enhance the productivity of methane from the waste activated sludge (WAS), a coupled system of microbial electrolysis cell (MEC) and anaerobic digestion (AD) was proposed. In this study, alkali, ultrasound-alkali, high-temperature coupled microaeration (TM) were applied as pretreatment methods to disintegrate the WAS flocs and break bacterial cell. After ultrasound-alkali pretreatment, the maximum accumulated concentration of VFAs and SCOD increased by 6.4 and 13.8 times compared with the initial concentration, which were 2.8 and 2.6 times of alkali pretreatment, and 2.1 and 2.1 times of TM pretreatment. Then, the pretreated sludge was transferred into MEC-AD coupled reactors and control group of AD reactors. The results showed that, methane production rate was enhanced to 0.15 m3 CH4/m3 reactor/d in the coupled reactors, which was improved by 3 times compared with control AD (0.05 m3 CH4/m3 reactor/d). The methane yield of MEC-AD coupled reactors achieved 808 ±â€¯8 mL, which were increased by 97.0% ±â€¯1.85% compared to control AD (410 mL). Using MEC can promote the rate of organics degradation and methane yield. The MEC-AD coupled system realized a good performance on the treatment of WAS and improved the efficiency of methane production.

Study on the hydrogen production ability of high-efficiency bacteria and synergistic fermentation of maize straw by a combination of strains.

Based on the principle of reciprocal symbiosis and co-metabolism of mixed culture microorganisms, a group of high-efficiency maize straw-degrading hydrogen-producing complex bacteria X9 + B2 was developed by a strain matching optimization experiment. Systematic research and optimization experiments were carried out on the mechanism of the main controlling factors affecting the hydrogen production of the complex bacteria. The results showed that the optimum conditions for the acid blasting pre-treatment of maize straw as a substrate were as follows: when the inoculation amount was 6% and the inoculum ratio was 1 : 1, at which point, we needed to simultaneously inoculate, the initial pH was 6, the substrate concentration was 12 g L-1, and the culture time was 40 h. The complex bacteria adopted the variable temperature and speed regulation hydrogen production operational mode; after the initial temperature of 37 °C for 8 hours, the temperature was gradually increased to 40 °C for 3 hours. The initial shaker speed was 90 rpm for 20 hours, and the speed was gradually increased to 130 rpm. The maximum hydrogen production rate obtained by the complex bacteria under these conditions was 12.6 mmol g-1, which was 1.6 times that of the single strain X9 with a maximum hydrogen production rate of 5.7 mmol g-1. Through continuous subculturing and the 10th, 20th, 40th, 60th, 80th, 100th and 120th generation fermentation hydrogen production stability test analysis, no significant difference was observed between generations; the maximum difference was not more than 5%, indicating better functional properties and stability.