Study On Microcosmic Physical Properties And Propagation Law Of Corona Discharge In Corona Cage

With the development of China's economy, the society's demand for energy is increasing constantly, and the construction of AC/DC EHV/UHV transmission project is in full swing. The increase of the voltage level of transmission line makes the series of environmental problems brought by corona discharge become more and more prominent. The radio interference, as one of the environmental impact assessment factors, will interfere with the reception of the radio signal near the line, which becomes one of the key factors to be considered in the design of the UHV transmission line. The corona cage test is commonly utilized in the study of radio interference in transmission lines to obtain the excitation function of different conductor bundles configurations, which can be used to predict the radio interference level of the actual transmission line. The nature of radio interference formation is due to generation and propagation of corona discharge pulse current on the transmission line. Therefore, the study of the microphysical mechanism of the corona discharge pulse generation and the corona current propagation law of the corona discharge pulse along the test circuit are the basis for the reasonable prediction of the radio interference level of the transmission line, which is very important for the construction of the environment-friendly EHV/UHV transmission lines.At present, in the numerical simulation of the microphysical mechanism of the corona discharge pulse in the air, a single theoretical model is used to simulate the microscopic physical process, including no description of the complex micro-chemical reaction in the discharge process. The existing hybrid model considering the micro-chemical reaction process can not simulate the multiple pulses of corona discharge because of introducing too many parameters. In the study on propagation law of the corona discharge pulse along the test circuit, the domestic and foreign scholars introduced the concept of amplification factor. The corona current propagated on conductor bundles was obtained by the analytical solution to obtain the relationship between the corona current generated on the conductors and the corona current measured in the measuring circuit when the terminal of the conductors is open or matched with the characteristic impedance of the conductors. Because the measuring circuit equivalent impedance and the characteristic impedance of conductors do not match with each other, the amplification factor can not be obtained by simple analytical method in domestic experiments. There are neither reliable methods nor accepted conclusions for calibration analysis of corona current propagation characteristics in corona cage measuring circuit.In this paper, the corona discharge of tansmission lines is studied in two aspects: the microphysical mechanism of the corona discharge pulse formation and its propagation law along the corona cage measuring circuit. The main contents and conclusions are as follows:Firstly, based on the collision cross section of the existing 62 electron-air molecules collision reaction and the Boltzmann equation, the air is simplified as the composition of 21% O2 molecules and 79% N2 molecules, and two approximation theory is adopted to solve Electron Energy Distribution Function (EEDF) of Corona Discharge and the relation between electron average energy and reduced electric field intensity. The ionization coefficient of the electrons is calculated by the townsend coefficient of collision ionization reaction of 02 and N2 molecules, and the attachment coefficient is calculated by the townsend coefficient of the dissociation and non-dissociative attachment of electrons and 02 molecules. The corresponding corona discharge onset field intensity when effective ionization coefficient is equal to zero is calculated, and the results are compared with the experimental results in the literature. The relative errors are less than 3%, which verifies the accuracy of the Boltzmann equation solution. The electronic transport parameters such as electron mobility, electron diffusion coefficient, electron energy mobility, and electron energy diffusion coefficient are calculated according to EEDF.Secondly, when the electronic transport parameters are provided for particle continuity equation and the momentum conservation equation, and the electron energy transfer coefficient is provided for the energy conservation equation, a corona discharge hybrid numerical model based on the fluid-chemical reaction is proposed. The control equations consist of particle continuum equation, momentum conservation equation and energy conservation equation in fluid dynamics model, and the particle source terms in the control equations are modified according to the plasma chemical reaction model. The corona cage is reduced to a coaxial cylindrical electrode. The microphysical characteristics evolution of the single pulse and the time interval in the development process of the positive and negative corona pulses are calculated and analyzed. The secondary pulse characteristics and repetitive characteristics of the negative corona pulse are emphatically discussed according to the Tricer pulse sequence calculated in a certain period of time. The results show that, compared with the iorization collision, the elastic collision is dominant during the whole discharge process. The ionization reaction rate of electrons and N2 is larger than that of electrons and 02. The dissociation attachment rate of electrons and 02 is much higher than non-dissociation attachment. The calculated negative corona discharge pulses amplitudes are between 1-2mA, and pulse interval is in the order of 10-2s. The calculated positive corona discharge pulses amplitudes are in the order of 10-2A.Thirdly, in order to verify the accuracy of corona discharge hybrid numerical model and to study the statistical characteristics of corona pulses, based on small corona cage test, a high potential high frequency current collection system is connected directly into the high-voltage conductor to collect the corona current pulses. The corona current pulses at different voltages and different polarities are measured. The variation law of the corona pulse amplitude and time interval is discussed, and the results are compared with the simulation results. The mean amplitude of positive and negative corona pulse are about 20mA and between 1-2mA, which will not change significantly with the increase of conductor voltage. The mean interval time of the positive and negative corona pulse will decrease significantly with the increase of the conductor voltage, and interval time is stable in the order of 10-3s and 10-5s, respectively. According to the measured results, the characteristic quantities of the positive and negative corona current pulses are statistically analyzed. Negative corona current pulses amplitude distribution is similar to positive skew distribution, and negative corona current pulses interval time distribution obeys exponential distribution with significance level of 0.05. Positive corona current pulses amplitudes distribution is similar to negative skew distribution, and interval time distribution is similar to normal distribution.Finally, based on the actual radio interference measuring circuit on corona cage, an equivalent numerical simulation model considering distribution parameters of the measuring circuit is established. The response of the excitation signal in the measuring circuit is calculated by adopting three kinds of simulated single corona current pulse as the excitation signal, and the propagation characteristics of the corona current in the measuring circuit are analyzed. The experimental platform of corona current propagation characteristics is built. The single-pulse corona current waveform is simulated by arbitrary waveform generator in series of non-inductive resistor, and the amplification coefficient obtained by the experiment is compared with the simulation results, which shows that at the measuring frequency of 0.5MHz, and the maximum error between experimental and simulation results is less than 3.2%. Different excitation signal waveforms have little effect on the amplification factor. At the measuring frequency of 1MHz, the maximum error between experimental and simulation results is 17.0%. In the high voltage side measurement mode with coupling capacitor, the amplification factor decreases with the increase of the number of conductor bundles, and the experimental and numerical results are in good agreement.