Selection of antioxidants for capacitor grade polypropylene film: Insights into electrical performance of the oil-film system.

High voltage power capacitors employ the oil-impregnated polypropylene film as the insulation. The swelling phenomenon might drive the antioxidants and small molecules within the film to migrate into the oil. It is necessary to comprehensively investigate the physical migration mechanism of antioxidants and their impact on the electrical performance of the oil-film combination insulation system and, consequently, formulate the proper selective prescription of antioxidants. Theoretical elucidation of the competitive interaction mechanism between the film and the oil in attracting antioxidant molecules was achieved through the calculation of inter-molecular binding energy, and the migration coefficient ηm was introduced to quantify the migration characteristics of antioxidants. Experimentally, the effects of antioxidants on the space charge distribution of the film, the dielectric properties of the oil, and the breakdown characteristics of both the film and oil were investigated. The experimental conclusions are consistent with theoretical analysis. The lamellar structure antioxidant molecules with ηm > 1 tend to migrate from the film to the oil, which results in increased dielectric loss and decreased breakdown strength of the insulating oil. In addition, the presence of phosphorus atoms in phosphite antioxidants contributes to a reduction in the breakdown strength of the film. For capacitor grade polypropylene film, in addition to the synergistic effect between different types of antioxidants on the thermo-oxidative stability, the structure of the antioxidant molecules and its influence on the electrical performance of the oil-film systems should also be taken into account.

Large-capacity AC superimposed DC power supply for the degradation evaluation of DC-link capacitors.

The DC-link capacitor, whose operating voltage is a periodic irregular waveform, is a key device in the converter. A large-capacity DC/AC superimposed experimental power supply above 100 kVA is an important piece of equipment that must be used in the aging research of DC-link capacitors. The irregular periodic operating voltage of the DC support capacitor is equivalent in the form of DC voltage superimposed with multiple harmonics based on the Fast Fourier transform analysis. The power supply is mainly divided into four parts: an AC module, a DC module, a protection module, and an isolation component. The composited harmonics are output by the IGBT (Insulated Gate Bipolar Transistor) inverter of the AC module. The soft start method is applied to the AC module in order to ensure the long-term reliable operation of the power supply. The kV-level high-voltage is output by four inverter units connected in series. The amplitude and phase control of harmonic decomposition is adopted, and the inverter is controlled by Sinusoidal Pulse Width Modulation. The output voltage of the power supply is close to the operating voltage of the capacitor. Based on the designed power supply, the degradation experiment was carried out. The temperature rise and aging characteristics of DC-link capacitors under different AC/DC ratios were studied.

Development of a Wideband Precision Electric Field Measuring Sensor.

High-voltage electric field measurement technology has certain applications in electric field measurement of power systems, but due to the limitation of its measurement accuracy and bandwidth, it cannot be used for the measurement of lightning-impulse voltage. In order to calibrate the nonlinearity of the MV-level lightning-impulse voltage measurement system, this paper proposes the design and implementation of a high-precision inductive wideband electric field measuring sensor (EFMS). The influence of the metal shell on the electric field distribution was simulated, and the influence of the electric field non-uniformity coefficient was studied. The characteristics of the EFMS were tested, and the results showed that the EFMS can accurately reproduce the waveform of lightning-impulse voltage and power-frequency voltage, with a proportionality coefficient of 0.05664 V/(kV/m). In mostly uniform and extremely non-uniform fields, the nonlinearity of the EFMS for impulse voltage is less than ±0.25%, and the nonlinearity of the EFMS for power-frequency voltage is less than 0.1%. It is shown that the EFMS can be used for the nonlinearity calibration of ultra-high voltage impulse measurement devices.

Analytical model for ionic current dominated corrosion of nanoelectrodes in metalized films: Frequency and electric stress.

Metalized film capacitors in a.c. applications suffer high frequency and high voltage, which will induce electrode corrosion, leading to capacitance degradation. The intrinsic mechanism of the corrosion is oxidation caused by ionic migration in the oxide film formed on the electrode surface. In this work, a D-M-O illustration structure for the nanoelectrode corrosion process is established, and thereby, an analytical model is derived to study the influences of frequency and electric stress on corrosion speed in a quantitative approach. The analytical results well conform to the experimental facts. It is found the corrosion rate rises with frequency and finally tends to reach a saturation value. The electric field in oxide has an exponential-like contribution to the corrosion rate. In the case of aluminum metalized films, the saturation frequency and minimum field required for corrosion initiation are, respectively, 3434 Hz and 0.35 V/nm calculated by the proposed equations.

Research on high-precision pulse power supply.

With the development of pulse power technology, higher technical requirements have been put forward for the pulse power supply system. How to improve the precision of power supply is the hotspot of current research. In this paper, a 20 kJ high-precision pulse power supply is established for study, consisting of a high energy storage pulse power capacitor (6420 µF/2.5 kV), a pulse forming inductor (2.73 µH/0.945 mΩ), a 4 in. thyristor, and 2 4 in. diodes. Three characteristic parameters (current peak Imax, rise time Tr, and pulse width Tw) are proposed to define the precision of the pulse power supply: Imax = 54 kA, Tr = 205 µs, and Tw = 1470 µs. According to the electrical parameters, the numerical expressions of the power supply precision are theoretically solved. This can guide the device screening and active temperature control scheme. Then, the active temperature control experiment of the high-precision pulse power supply was carried out. The results show that ΔImax is less than 1.2‰, ΔTr is less than 1.8‰, and ΔTw is less than 2.1‰, within the ambient temperature of 10-30 °C. The rated Imax is different under different ambient temperatures. The extreme difference is 0.7 kA, and the error is 6.5‰. In addition, the life test of a high-precision pulsed power supply is carried out, which shows that the capacitance and inductance will decrease with the increase of rated discharge times. The change of capacitance is 1.19‰, and the change of inductance is 3.73‰. Finally, this high-precision pulse power supply has proved that the precision can reach a thousandth under the condition of active temperature control, and it is proposed that the precision of the power supply can be maintained by controlling parameters such as device temperature and discharge voltage.

Simulation of the magnetization process of a gapped magnetic core based on the Jiles-Atherton model.

The Jiles-Atherton (J-A) model is used to simulate the magnetization process in a gapped magnetic core. A test platform for investigating the magnetic properties of a gapped core is designed and constructed. The hysteresis curves of an amorphous alloy core are measured for different gap lengths. Simulated hysteresis curves based on the original and inverse J-A models are obtained and are compared with the measured curves. The simulation result based on the original J-A model is better than that based on the inverse J-A model. The applicability of a modified J-A model to gapped cores is analyzed. The influence of gap length on inductance is analyzed from the perspective of the magnetic circuit. The inductance is calculated based on the J-A model, and the results indicate that the J-A model has high accuracy in this application.

Application of Jiles-Atherton model in description of temperature characteristics of magnetic core.

The Jiles-Atherton (J-A) model is widely used in magnetic characteristic modeling in alternating field. A dynamic J-A model, which is applied under high-frequency condition, is established. A pointwise iterative calculation method is used, and an approximate calculation method of the anhysteretic magnetization intensity and its derivative is proposed. The solving procedure of the J-A model is optimized. The platform for the magnetization and temperature characteristic test is designed and constructed. The hysteresis curves of ferrite cores at different temperatures are measured. The variation of parameters of magnetic properties with temperature is analyzed, including maximum magnetic flux density, remanence, coercivity, and core loss. Based on the particle swarm optimization combined with natural selection, optimized parameters of the J-A model are obtained. Their variation is analyzed, and their fitted curves with temperature are obtained. It is found that J-A model's modification in temperature can be used to guide the modeling analysis of electrical equipment containing magnetic cores working in high-temperature environment and to predict magnetic properties at high temperatures.

Modeling analysis of pulsed magnetization process of magnetic core based on inverse Jiles-Atherton model.

The J-A (Jiles-Atherton) model is widely used to describe the magnetization characteristics of magnetic cores in a low-frequency alternating field. However, this model is deficient in the quantitative analysis of the eddy current loss and residual loss in a high-frequency magnetic field. Based on the decomposition of magnetization intensity, an inverse J-A model is established which uses magnetic flux density B as an input variable. Static and dynamic core losses under high frequency excitation are separated based on the inverse J-A model. Optimized parameters of the inverse J-A model are obtained based on particle swarm optimization. The platform for the pulsed magnetization characteristic test is designed and constructed. The hysteresis curves of ferrite and Fe-based nanocrystalline cores at high magnetization rates are measured. The simulated and measured hysteresis curves are presented and compared. It is found that the inverse J-A model can be used to describe the magnetization characteristics at high magnetization rates and to separate the static loss and dynamic loss accurately.

A 16 MJ compact pulsed power system for electromagnetic launch.

This paper has established a compact pulsed power system (PPS) of 16 MJ for electromagnetic rail gun. The PPS consists of pulsed forming network (PFN), chargers, monitoring system, and current junction. The PFN is composed of 156 pulse forming units (PFUs). Every PFU can be triggered simultaneously or sequentially in order to obtain different total current waveforms. The whole device except general control table is divided into two frameworks with size of 7.5 m × 2.2 m × 2.3 m. It is important to estimate the discharge current of PFU accurately for the design of the whole electromagnetic launch system. In this paper, the on-state characteristics of pulse thyristor have been researched to improve the estimation accuracy. The on-state characteristics of pulse thyristor are expressed as a logarithmic function based on experimental data. The circuit current waveform of the single PFU agrees with the simulating one. On the other hand, the coaxial discharge cable is a quick wear part in PFU because the discharge current will be up to dozens of kA even hundreds of kA. In this article, the electromagnetic field existing in the coaxial cable is calculated by finite element method. On basis of the calculation results, the structure of cable is optimized in order to improve the limit current value of the cable. At the end of the paper, the experiment current wave of the PPS with the load of rail gun is provided.

Modeling of stored charge in metallized biaxially oriented polypropylene film capacitors based on charging current measurement.

Metallized biaxially oriented polypropylene film (BOPP) capacitors are widely used in pulsed power systems. When the capacitor is used as the energy storage equipment under high electric field, more charges should be provided to maintain the voltage of the capacitor. This should be ascribed to the completion of the slow polarization which may take several hours or even longer. This paper focuses on the stored charge in metallized BOPP film capacitors. The modeling of the stored charge by the equivalent conversion of circuits is conducted to analyse the slow polarization in the BOPP film. The 3-RC network is proposed to represent the time-dependent charge stored in the capacitor. A charging current measurement system is established to investigate the charge storage property of the capacitor. The measurement system can measure the long time charging current with a sampling rate of 300 Hz. The total charge calculated by the charging current indicates that the stored charge in the capacitor under the electric field of 400 V/µm is 13.5% larger than the product of the voltage and the capacitance measured by the AC bridge. The nonlinear effect of the electric field on the slow polarization charge is also demonstrated. And the simulation of charge storage based on the 3-RC network can match well with the trend of the stored charge increasing with the time.

Study on erosion mechanism of graphite electrode in two-electrode spark gap switch.

In a high-powered single pulse system, the graphite electrode is better than other common metal electrodes for high energy transfer and pulse discharge. In this paper, the erosion mechanism of graphite electrode is investigated with the thermodynamics theory and the experimental results. Based on a simplified mathematical model, the graphite electrode erosion process of high-powered spark gap switch is also analyzed. The analysis results show that the relationship of the graphite electrode erosion and the charge transfer is linear, which is accordant with the experimental results.

Design and performance of a pulse transformer based on Fe-based nanocrystalline core.

A dry-type pulse transformer based on Fe-based nanocrystalline core with a load of 0.88 nF, output voltage of more than 65 kV, and winding ratio of 46 is designed and constructed. The dynamic characteristics of Fe-based nanocrystalline core under the impulse with the pulse width of several microseconds were studied. The pulse width and incremental flux density have an important effect on the pulse permeability, so the pulse permeability is measured under a certain pulse width and incremental flux density. The minimal volume of the toroidal pulse transformer core is determined by the coupling coefficient, the capacitors of the resonant charging circuit, incremental flux density, and pulse permeability. The factors of the charging time, ratio, and energy transmission efficiency in the resonant charging circuit based on magnetic core-type pulse transformer are analyzed. Experimental results of the pulse transformer are in good agreement with the theoretical calculation. When the primary capacitor is 3.17 µF and charge voltage is 1.8 kV, a voltage across the secondary capacitor of 0.88 nF with peak value of 68.5 kV, rise time (10%-90%) of 1.80 µs is obtained.