We accurately established an air discharge simulation experiment system to obtain the variation of ozone concentration in air discharge process, which includes pin-to-plate discharge, air-gap discharge, creeping discharge as well as metal particle discharge model, consisting of voltage console, transformer, and discharge device, gas absorption cell, spectrometer, and computer. Through online monitoring, the ozone concentration in a relatively closed container to simulate internal partial discharges for an hour with the voltage increases from 3 to 0.2 kV increment, we obtained four basic models with discharge inception voltage grade of 4.0, 4.5, 4.7, and 4.1. Besides, we applied the method of Fourier transform-differential absorption spectroscopy to process and analyze the experimental data, and obtained the ozone concentration curve. The ozone concentration increases with the increasing voltage loading time, and it tends to be stable and reaches an equilibrium state under a certain voltage.

Air Quality Monitoring System (AQMS) is a real time system for environmental monitoring that has been developed in accordance with China EPA rules relevant to emission monitoring and control of pollutants from large combustion plants. AQMS is designed to provide collection, evaluation and storage of environmental data for the sake of air quality and environment safeguard and health protection inside and outside the factory or plant. It can be used in all kind of plants that may produce emissions to the atmosphere (e.g. fossil fired boilers, combustion turbines, kilns, etc.) by means of definition of specific limit and guideline values for all air pollutants with various sampling periods. This paper designs a set of AQMS for Industrial district.

Simple trigger logic is commonly used in actual wind farms to monitor unit conditions, which face problems such as a high false-alarm rate and overlapping alarms. In addition, the characteristics of SCADA data, such as large quantity, complexity, and variable correlation, lead to insufficient accuracy of fault diagnosis. To address this problem, an improved fault diagnosis method based on a Multi-Channel Attention Mechanism Convolutional Neural Network (MCAMCNN) is proposed. Firstly, feature analysis is performed after preprocessing SCADA data to fully explore the coupling characteristics between data, and a dataset is established. Then, the proposed fault diagnosis model is used for feature screening. Innovatively, a structure combining double-layer multi-scale convolution and multi-channel attention is adopted to extract multi-domain features and dynamically calibrate the weights of feature channels. Fault classification is realized after adaptive fusion of features by Efficient Channel Attention (ECA). Finally, experiments are designed based on real data from an onshore wind farm in China, which verify that the method is timely and accurate in fault diagnosis, with significantly improved accuracy and F1-score, and has obvious advantages over comparative methods.

An ozone generation and online monitoring system is built, which consists of a cylinder, ozone generator, flow control meter, ozone absorption pool and spectrometer and utilizes a xenon lamp as light source. Using differential optical absorption spectroscopy verified the feasibility of the system for monitoring ozone concentration. The method of Fourier transform and inverse Fourier transform is applied to the process and to analyze the experimental data. The Fourier transform of the ozone concentration amplitude curve is concluded, and the noise signal curves are obtained and the maximum detection limit of 157 ppm is determined. Comparing the least squares method and Fourier transform-differential optical absorption spectroscopy method, the result shows that the Fourier transform method has greater accuracy.