基于强跟踪滤波器的水中高频振荡放电参数分析

Parameter Analysis of High Frequency Oscillation Discharge Underwater Based on Strong Tracking Filter

为探明水中放电高频振荡阶段参数及其变化特性,提出一种基于自适应噪声完备集合经验模态分解(CEEMDAN)和强跟踪滤波器的时变参数辨识方法。通过该方法分解水中放电实验平台采集的电压、电流信号得到不同频率特征的信号分量,对最适应原始波形的信号分量开展Hilbert变换并求得相应的瞬时幅值、频率,进而得到所需的电阻和电感。实验数据离散度分析结果表明,放电进程中参数变化具有随机性,故利用强跟踪滤波器进一步对实验数据进行辨识处理,可有效地降低随机放电造成的离散性,并获得具备普适性的电阻值和电感值。偏离度分析结果表明,辨识电阻与测量数据除在气泡崩塌阶段随机性过大外,前期偏离度集中在23.26%以下,降低了偏离度处于80%~110%内数据点的干扰,电感偏离度集中在2.35%以下。该方法能够有效地应用于水中高频振荡放电过程的时变参数处理研究中。

Underwater pulsed discharge is a complex physical process characterized.Based on the characteristics of plasma evolution,this process is divided into pre-breakdown discharge,main discharge,and high-frequency oscillation discharge stages.Existing research focuses on the impact of energy conversion during the pre-breakdown and main discharge stages,with the relatively limited investigation into parameters during the high-frequency oscillation discharge stage.The distribution of the electric field between electrodes is susceptible to environmental influences in the underwater pulsed discharge process,leading to a significant level of randomness in the development of arc channe.Therefore,the circuit parameters of single discharge data is not representative.This paper proposes a method for analyzing parameters of high-frequency oscillation discharge underwater by using a strong tracking filter(STF).Firstly,the variations of resistance and inductance were delineated based on the plasma dynamics in the high-frequency oscillation discharge stage,and the equivalent circuit model was established.Plasma density within the bubbles generated in the main discharge stage remains substantial and retains conductivity during this stage.Under the residual voltage of the capacitors,the equivalent capacitance and inductance induce attenuated oscillatory perturbations within the discharge circuit.At this stage,due to the loss of the arc channel,the bubble resembles a non-heat source structure,which means this bubble cannot continue to expand under the action of static water pressure,and the volume of it begins to pulsate regularly.The plasma density in the bubble is influenced by this phenomenon,resulting in oscillatory variations in the inter-electrode resistance and inductance.Constrained by the electrode configuration and its inherent morphological characteristics,the bubble undergo collapse,leading to a rapid increase in the inter-electrode resistance during this process.Next,the underwater pulsed discharge environment results in a non-uniform electric field within the fluid field between discharge electrodes,leading to stochastic characteristics in the morphology and spatial position of the plasma arc channel.This discharge characteristic makes variations in resistance and inductance values among different high-frequency oscillation discharge cycles,concurrently the variation trend of parameters in the time domain is also different.To obtain practical resistance values,the complete ensemble empirical mode decomposition with adaptive noise(CEEMDAN)was used to separate the inductance from the measured data.The obtained resistance and inductance data to filtering using the STF,thereby yielding representative identification values for resistance and inductance.Finally,the efficacy of the proposed method was validated followed by a subsequent analysis of the results'discreteness.To reflect the reliability of the discharge resistance and inductance values identified by the STF method,the discreteness coefficient is introduced as the quantization parameter.The research findings indicate that,during the expansion stage of bubble formation in the high-frequency oscillation discharge process,the deviations in both resistance and inductance are relatively minor,measuring less than 2.83%and 0.61%,respectively.In the bubble pulsation stage,the deviations in resistance and inductance are more significantly influenced by the morphology of the bubble,with the trend of deviation coefficient variations displaying a distinct pulsating pattern,the deviations are less than 23.26%and 1.59%for resistance and inductance,respectively.In the bubble collapse stage,the parameters are strongly affected by the stochastic nature of bubble deformation,resulting in the highest level of deviation in this phase,the maximum deviation coefficients are 110%and 24.32%,respectively.