CN110809355B

CN110809355B - Langmuir probe multi-Mmewwell electron distribution automatic analysis method

Info

Publication number: CN110809355B
Application number: CN201911066633.1A
Authority: CN
Inventors: 叶孜崇; 张炜; 徐国盛
Original assignee: Hefei Institutes of Physical Science of CAS
Current assignee: Hefei Institutes of Physical Science of CAS
Priority date: 2019-11-04
Filing date: 2019-11-04
Publication date: 2022-02-11
Anticipated expiration: 2039-11-04
Also published as: CN110809355A

Abstract

The invention discloses an automatic analysis method for multi-MmWigner electron distribution of a Langmuir probe, which is an analysis method for automatically fitting data of a probe voltammetry curve according to multi-MmWigner electron energy distribution and automatically judging which distribution of single, double and triple MmWigner can better accord with plasma electron distribution represented by the analyzed voltammetry curve, and a software system built according to the method. Through the automation of the analysis of the multiple Maxwell distributions of the Langmuir probe voltammograms, the labor time required for realizing related experimental methods can be greatly reduced in related experiments, so that the realization of some research works involving hundreds or even thousands of curves in each experiment becomes possible.

Description

Langmuir probe multi-Mmewwell electron distribution automatic analysis method

Technical Field

The invention relates to the technical field of diagnosis of local potential, electron temperature and electron density in plasma by using a Langmuir probe, in particular to a Langmuir probe data automatic processing method for Langmuir probe Dogmaiwell electron distribution automatic analysis.

Background

The langmuir probe is the earliest plasma parameter diagnostic system, and has the advantages of simple construction, low cost, large measurable physical quantity, and the like. Furthermore, the langmuir probe, especially the single probe scanning measurement, belongs to the direct measurement of plasma electron and ion distribution, and is a difficult-to-replace electron distribution measurement method for the plasma with unknown electron specific distribution mode. Langmuir probes are divided into single probes, double probes and triple probes, wherein the single probe obtains a voltammogram by potential scanning of the probe to the ground (device wall), and plasma parameters are analyzed from the voltammogram; the double probes scan through the potential between the two probe surfaces to obtain a voltammetry curve, and plasma parameters are analyzed from the voltammetry curve; the triple probe then obtains real-time plasma parameters through the current between one floating probe and two energized probes.

Although the use of the dual probe and the triple probe is relatively simple, the interference to the plasma is generally less due to the limitation of the total absorption current. However, the two probes have the problems that sheath layers may overlap between the probes and the accuracy is affected, and in plasmas with non-single-McWire electron distribution, the diagnostic accuracy of the two probes is greatly reduced, so that the reliability of the obtained parameters is affected. Therefore, the method has unique advantages in the parameter interval of the plasma to be diagnosed and the environment with unclear specific electron energy distribution by using a single probe to obtain the volt-ampere characteristic line and then carrying out detailed electron energy distribution analysis.

Most commercial single probes on the market have automatic analysis system software, but the related software is analyzed according to the assumption that electrons are in single-Maifaner distribution, and a volt-ampere characteristic line drawn on a semilogarithmic graph is fitted to obtain plasma parameters. The maxwell distribution is the classical way of energy distribution of electrons after sufficient thermalization, and single or multiple maxwell electron distributions are common in plasma devices. In some surface treatment plasmas, the electron distribution is usually depleted by high-energy electrons, so that it is assumed that a single-wheatweil distribution may obtain better plasma parameters, but in most plasma sources with electron confinement but not long confinement time, the electron distribution is usually double-wheatweil because the input power of the plasma source itself produces a batch of high-energy electrons, which in turn produce a batch of cold electrons through ionization by collision, and the confinement of the plasma device is not enough to allow the high-energy electrons to escape without much to cause depletion, but not enough to allow the electrons to reach the thermal equilibrium of the single-wheatweil. In some plasma sources that rely on an implanted electron beam to generate plasma, such as a tungsten filament or lanthanum hexaboride hot cathode discharge, four energy distributions of electrons are typically readily present: electron beams originating from a hot cathode, thermal electrons that decelerate after energy loss by impact ionization, thermal electrons resulting from secondary emission from the device wall, and cold electrons resulting from ionization. Even though the behavior of a thermionic electron beam in a voltammogram does not necessarily differ significantly in ion current, and therefore is easily removed when dealing with saturated flows of ions, these plasma sources tend to produce a trimaran electron energy distribution. Therefore, physical studies involving the above-described apparatus and other plasma sources producing multimaxwell distributions are particularly important by analyzing the electron temperature and density of each of these multimaxwell distributions and calculating the relevant effective temperatures as needed for the physical study.

In the past, the multimacweil distribution has often used the human hand to select a fitted interval of voltammetric signatures from hot to cold to fit the temperature to obtain the temperature and density of each electron distribution because as the probe voltage decreases, the cold electron distribution is more thoroughly reflected by the sheath of the probe because of the bosmann relationship than hot electrons, resulting in only hot electrons entering the probe at lower voltages to form a current. By manually determining a fitting interval in the semi-logarithmic graph that is relatively flat and far from the plasma potential, a relatively reliable temperature of the thermions can be obtained. However, the manual fitting is always a slow manual process, and if the experiment itself needs to fit a large number of voltammetry characteristic curves, the requirement of human time for the experiment depending on the manual fitting will be greatly increased, which is not beneficial to the research of a small team. For example, some time analysis diagnostics involving plasma global instability often require analysis of hundreds or even thousands of voltammograms taken during an instability period in each experiment. Therefore, a set of relatively reliable probe volt-ampere curve analysis method and corresponding software for automatically fitting double-Maxwell and triple-Maxwell electron energy distribution can also greatly improve the manpower efficiency of related experiments. In the past, this automated analysis technique did not exist because of the lack of a method and digital judgment that automatically resolves the voltammetric curve interval of the electrons analyzed for a certain temperature. The invention provides a judgment mode which is one of the important creatives of the invention.

Disclosure of Invention

The invention aims to overcome the defects of the prior art and provide an automatic analysis method for the Domaxwell electron distribution of a Langmuir probe.

The invention is realized by the following technical scheme:

an automatic analysis method for the multi-McWiler electron distribution of Langmuir probe features that the voltammogram is fitted to the single McWiler distribution, after the circular feedback reaches the optimal result, the parameters are output as seed parameters of the fitting range of the double-McWiler distribution to carry out the fitting of the double-McWiler distribution, after the optimal result is achieved through cyclic feedback and repeated fitting, the parameters are further output as seed parameters of fitting ranges of double-Maxwell distribution and triple-Maxwell distribution to carry out fitting of the distribution modes, in each layer delivery process, whether the double or triple McWeir distribution analysis is suitable or not is determined according to the parameter result, after proper electron energy distribution is screened, parameters such as electron density, plasma potential and the like are fitted through the recombined voltammetry curve, and finally, the recombined voltammetry curve and data are subjected to overlapping drawing for a user to check the accuracy.

The fitting mode of the single Mceway is as follows: firstly, a user inputs a voltammogram file number to be analyzed, firstly, an ion saturated flow is subtracted through linear fitting of the lowest 10V part of the voltammogram, and then, the voltammogram is differentiated to obtain an inflection point potential V_infThen, the voltammogram is plotted in a logarithmic graph, and a voltammogram V is taken_inf>V>V_infFitting the-1V segmented seed fitting interval to obtain the single-wheat-well distributed temperature T_e1And, further, the system again changes to V_inf>V>V_inf-T_e1Fitting T again in the interval of/e_e1Until T is obtained_e1And T obtained last time_e1Within 5%, the convergence T is obtained_e1Then, by using the fitting T_e1Recombining a voltammetry curve and then utilizing linear fitting V_infLast several T_e1With its cross-over point as the plasma potential V_pAnd obtaining the electron density n by cross point current calculation_e。

The double-Maxwell distribution fitting process is as follows: firstly, by utilizing a volt-ampere curve obtained by a monoMcewell distribution fitting process after deionized saturated flow, after semilogarithmization, derivation is carried out, the absolute value is taken again, and the lowest point V is searched_hfitAs hot electron temperature T_h2Reference point of fitting range, in V_hfit>V>V_hfit10V (section 10V can be customized as desired) fitting the Hot Electron temperature T_h2Then, one voltammogram is reconstructed from the fitting results and subtracted from the data, and then, the V of the remaining curve is_inf>V>V_inf-T_e1Fitting Cold Electron temperature T in the/e Range_c2After fitting, will be at V_hfit>V>V_hfit-T_h2E and V_inf>V>V_inf-T_c2Respectively as the fitted Hot Electron temperature T_h2And cold electron temperature T_c2The above fitting procedure is cycled through until the hot electron temperature T_h2And cold electron temperature T_c2Converge to 5%, at which time if any electron temperature is below zero, either T_c2>T_h21.7, automatically judging that the voltammogram is not suitable for double-Maifaner distribution analysis, and outputting parameters obtained by single-Maifaner distribution to a user, otherwise, fitting the plasma potential V by the system_pAnd electron density n_eAnd the density of each temperature distribution, and outputting parameters obtained by double-Maxwell distribution for TriMaxwell distribution analysis.

Flow of trimaran distribution analysisThe process is as follows: first obtaining the thermal electron temperature T in the double-McWigner distribution analysis_h2And cold electron temperature T_c2And re-fitting the hot electron temperature T in the trimaran electron energy distribution according to the analysis procedure of the double-trimaran distribution_h3Then subtracting the hot electron temperature T_h3Part of the voltammogram is again logarithmized and derived, and the lowest point V in the middle is found again_mfitAnd is in V_mfit+T_c2/e>V>V_mfit-T_c2Fitting intermediate electron temperature T in the/e range_m3And using the intermediate electron temperature T_m3Reconstituting a voltammogram to subtract from the data, and then V in the remaining data_inf>V>V_inf-T_c2E refitting the Cold Electron temperature T_c3And further by V_mfit+T _m3/2e>V>V_mfit-T_m3[ 2e ] and [ V ]_inf>V>V_inf-T_c3E instead of fitting range to fit the intermediate electron temperature T cyclically_m3And cold electron temperature T_c3Up to the intermediate electron temperature T_m3And cold electron temperature T_c3Converge to 5%, at which time if any temperature presents a negative number, or T_c3>T_m31.7, or T_m3>T_h31.7, judging that the curve is not suitable for the trimaran distribution analysis by the program, outputting double-trimaran distribution parameters, and otherwise, utilizing T_h3Intermediate electron temperature T_m3And cold electron temperature T_c3Recombining voltammetric curve, and fitting plasma potential V_pElectron density n_eThe density parameter for each temperature profile is provided to the user.

A Langmuir probe data automatic analysis method aiming at low-temperature plasma multi-MmWir electron distribution defines a relatively reliable searching condition of an electron temperature fitting interval from hot to cold for Langmuir probe volt-ampere characteristic lines caused by various MmWir electron energy distributions. And the correlation analysis result is used for defining a specific method for automatically screening whether the voltammogram is suitable for being analyzed according to the multimaxwell electron energy distribution.

The software constructed according to the method can automatically analyze the voltammetry curve of the Langmuir probe according to single, double and triple Maxwell distributions through the multiple Maxwell distribution analysis, and obtain corresponding plasma potential, Debye diameter and corresponding electron temperature and density of each Maxwell distribution. Then, the comparison of the temperature of each obtained Maxwell distribution automatically analyzes whether the voltammogram is suitable for the analysis of double or triple Maxwell distribution, and finally generates a voltammogram recombined by the analysis result, so that the user can check the accuracy of the analysis result.

The analysis software selects a file number to be processed by a user, then the software automatically fits and subtracts ion saturation flow in a conventional mode, further obtains a reference point of plasma potential by using an inflection point method, then draws a volt-ampere curve on a semilog graph, automatically fits electron temperature according to single, double and trimigraine distributions from hot to cold, and in each fitting, the software automatically ensures the most accurate result through the feedback circulation of the fitting result, and after obtaining the electron temperature of each distribution, the software further calculates the average of the inverse values of the electron temperature according to the fitted electron temperature and density, namely the effective electron temperature affecting the plasma mechanics of the Bohm criterion. Finally, software recombines a voltammetry characteristic line through each distributed electron temperature, so that the potential of the plasma is fitted, and data such as total electron density, Debye diameter, electron density composition and the like are obtained.

The invention has the advantages that: the invention provides a selection condition of a high-temperature electron fitting section with physical significance aiming at the single-probe volt-ampere characteristic line diagnosis of probes of various Mais Weier electron energy distribution plasmas, and defines a fitting result feedback with physical significance to realize a feedback circulation condition of the best fitting result. By setting these conditions, automatic fitting of a voltammetric characteristic line composed of a plurality of meval electron distributions becomes possible. Through the automation of the analysis of the multiple Maxwell distributions of the Langmuir probe voltammograms, the labor time required for realizing related experimental methods can be greatly reduced in related experiments, so that the realization of some research works involving hundreds or even thousands of curves in each experiment becomes possible.

Drawings

FIG. 1(a) is a flow chart of a voltammetric profile unimaswell automated analysis method;

FIG. 1(b) is a schematic diagram of the system finding Vinf;

FIG. 1(c) is a comparison graph of the voltammogram of the system to be recombined and the acquired voltammogram, and the difference between two lines near Vp is noted;

FIG. 2(a) is a flow chart of a voltammetric signature line double Maxwell automated analysis method;

FIG. 2(b) shows the system finding the hot electron fitting point V_hfitAnd examples of their fitted hot electron partial voltammograms;

FIG. 3(a) is a flow chart of the voltammetric profile trimmiers automated analysis software;

FIG. 3(b) shows the system finding the intermediate question-pair electronic fitting point V_mfitAnd fitting an instance of its corresponding portion of the voltammogram;

FIG. 3(c) shows the selection of the fitting points for the cold, medium and hot electron temperatures on a voltammetric curve, and the comparison of the reconstructed curve of the fitting parameters to the collected data.

Detailed Description

The technical scheme of the invention is explained in detail in the following with the accompanying drawings.

The invention provides a Langmuir probe data automatic analysis method aiming at low-temperature plasma multi-Maxwell electron distribution, which is suitable for analysis work of Langmuir single-probe voltammetry curves in a low-temperature plasma device with most non-single-Maxwell electron distribution, can automatically analyze the voltammetry curves of Langmuir probes according to single, double and trimmiwell distribution through multi-Maxwell distribution analysis, and obtains corresponding plasma potential, Debye diameter and electron temperature and density corresponding to each Maxwell distribution.

The automation of the analysis of the probe data based on the multimausweil electron distribution is achieved according to the above method and the adoption of single, double or trimaral distribution data is automatically suggested according to the voltammetry line analysis results. The processing time for analyzing a plurality of voltammograms of the multi-Maxwell distribution is greatly reduced, and the voltammograms reconstructed according to the multi-Maxwell distribution data are automatically drawn and output for user calibration and use.

According to one embodiment of the invention, the method is implemented by first fitting the voltammetry curves once in a single-Maxwell distribution mode, outputting the parameters as seed parameters of a double-Maxwell distribution fitting range after the most accurate result is achieved by cyclic feedback, and further outputting the parameters as seed fitting range parameters for fitting the trimodal distribution after the optimal result is achieved by cyclic feedback and repeated fitting. In each layer delivery process, software automatically determines whether the double or triple Maxwell distribution analysis is suitable according to parameter results, and after the proper electron energy distribution is screened, parameters such as electron density, plasma potential and the like are fitted through a regrouping voltammetry curve. Finally, the software can automatically overlay and draw the recombined voltammetry curve and the data for the user to check the accuracy.

The fitting method of the monowheatwell in the present invention is shown in fig. 1 (a):

step 1: the user inputs the voltammetry profile number to be analyzed, and the system reads the data and draws the data into a voltammetry curve with the voltage V as the horizontal axis and the current I as the vertical axis. Then, the system automatically searches the lowest voltage point V of the range of the horizontal axis of the curve_minTo V_minAnd a +10V portion, fitting a straight line to the curve and subtracting the fitting result from the curve to exclude the ion saturated flow.

Step 2: deriving the voltammetry curve to obtain an inflection point V_inf. This step is schematically illustrated in fig. 1(b), wherein the ordinate is the current I and the abscissa is the voltage V.

And step 3: the system draws a semilogarithmic graph with a voltammogram taking current (vertical coordinate) as logarithm and voltage as linear number, and takes V_inf>V>V_infCarrying out straight line fitting (on a semi-logarithmic graph) in the seed fitting interval of-1V, and taking the slope inverse value of the fitting result to obtain T_e1；

And 4, step 4: the system returns to V_inf>V>V_inf-T_e1E, (e is the electron charge) i.e. from the inflection point to one in the curveFitting T again in the interval of the electron temperature_e1Until T is obtained_e1And T obtained last time_e1The concentration is within 5 percent.

This is used to ensure that any fit that yields an e-order decay range of electrons is involved in the fit, providing a physical consistency that is obtained in terms of the amount of electrons that participate in the fit. Especially in the case of a monomysterious distribution, the credibility of the fitting result is in fact questionable in itself if the fitting parameters cannot be made consistent in one e order (63% electrons).

And 5: obtaining convergence T_e1The system then fits T by using_e1Recombining a voltammetry curve, and then utilizing linear fitting to be more than V_infA plurality of T_eThe voltammogram (plotted as a line) represents the trend line of increasing electron saturation flow through sheath expansion, and the linear saturation flow is represented by T_e1The intersection of the recombined curves being the plasma potential V_p. Probe electron saturation current is the electron heat entry velocity multiplied by the electron density multiplied by the probe area. Using this relationship, the system automatically takes the horizontal axis point V from the curve_pThe electron density n is calculated from the current point (electron saturation current) of_e。

Here not at V_infAs V_pThe reason is that the part of the volt-ampere curve close to the plasma potential can cause the current to be pseudo-low due to the polluted probe head and the virtual cathode effect, so that V is caused_pThe first few curves are flattened. This is particularly evident in plate-type probes where the system is particularly suited for use. This effect is illustrated in fig. 1 (c).

As shown in fig. 2(a), after the single-maxwell fitting is completed, the parameters are output to the double-maxwell fitting program as seed parameters. The double-maxwell distribution fitting process is as follows:

step 1: using the obtained result from step 1 of the fitting process of the Mono-Mai Weir distribution, subtracting the voltammetry curve of the ion saturated flow, obtaining the absolute value of the whole curve again after derivation after semilogarithmization, and searching the lowest point V of the whole curve_hfitAs hot electron temperature T_h2Fitting a range reference point.

The fitting reference point is selected in this way in consideration of the superposition of two probe principle factors: firstly, in a voltammetry curve caused by a plurality of Maifaner electron energy distributions at different temperatures, due to the Bowman relationship, as the potential is reduced, a probe sheath layer reflects electrons with low energy to the outside of a probe, so that the electrons do not form a probe current, and the proportion of electrons of the Maifaner electron energy distribution of the curve at higher temperature is higher in a lower voltage interval; on the other hand, as the total electron current of the voltammogram becomes weaker with the decrease of voltage due to the Bolsman relationship, the voltage of the very low-voltage part of the voltammogram gradually becomes indistinguishable from the background noise except for subtracting the electron saturation current, which causes the negative infinite singularity of the part of the voltammogram to appear due to the zero crossing of the noise after the semilogarithmization. V is therefore selected according to the method described above_hfitThe part of the voltammogram representing the hottest electron distribution is automatically selected while avoiding the part of the voltammogram where the signal-to-noise ratio is too low. The fitting method of this step and the following step 2 is shown in fig. 2 b.

Step 2: procedure is as follows V_hfit>V>V_hfit10V fitting the Hot Electron temperature T_h2Then using T_h2A recombined voltammogram is subtracted from the data because, according to the bosmann relationship, cold electrons will enter the probe as the probe potential decreases, while hot electrons will enter the probe with the cold electrons as the potential increases, thus requiring the hot electron fraction to be subtracted from the curve to accurately fit the cold electron temperature.

And step 3: v at the remaining curve_inf>V>V_inf-T_e1Fitting Cold Electron temperature T in the/e Range_c2。

And 4, step 4: after fitting, the program will be at V_hfit>V>V_hfit-T_h2E and V_inf>V>V_inf-T_c2Respectively as fitting T_h2And T_c2Range of (3) the above fitting procedure is cycled through until T_h2And T_c2Converge to 5%.

And 5: if any electron temperature is below zero, or T_c2>T_h21.7, thenThe system automatically determines that the voltammograms are unsuitable for double-McWillebur distribution analysis and outputs parameters obtained from single-McWillebur distribution to a user. Otherwise, proceed to step 6.

Step 6: system fitting V_pAnd electron density n_eAnd the density of each temperature distribution, outputting parameters obtained by double-Maifaner distribution for analysis of the triple-Maifaner distribution.

As shown in fig. 3(a), the flow of trimaran distribution analysis is as follows:

step 1: the system first obtains T in double-Maiweier distribution analysis_h2，T_c2

Step 2: re-fitting T according to the analytical procedure of the double-McWigner distribution_h2. Note T_h3Is a thermoelectron fitting point V obtained when fitting electrons according to two Maxwell distributions in a volt-ampere curve with three Maxwell distributions of electrons_hfitRefitting as the hottest electron temperature in the trimaran distribution; it essentially follows T_h2Are not separate.

And step 3: after subtracting T_h2Part of the voltammograms are again semilogarithmic and derived;

and 4, step 4: find the lowest point V in the middle again_mfit. The fitting method of this step and the following step 5 is shown in fig. 3 (b).

And 5: at V_mfit+T_c2/e>V>V_mfit-T_c2Fitting intermediate electron temperature T in the/e range_m3(ii) a Temperature unit T_c2The temperature unit is already included in the Boersmann constant and only needs to be divided by e because the common temperature unit in the physical field of the plasma is 'volt-electron charge' eV.

Step 6: by T_m3Recombining a voltammetric curve, subtracting from the curve of step 5, excluding T_m3The effect of the relevant electrons on the voltammogram;

and 7: then V in the remaining data_inf>V>V_inf-T_c2E againFitting the Cold Electron temperature T_c3。

And 8: and then with V_mfit+T _m3/2e>V>V_mfit-T_m3Fitting range cycle fitting T is replaced by/2 e_m3With V_inf>V>V_inf-T_c3Cyclic fitting T with fitting range replaced by/e_c3Until the two temperatures approach each other.

And step 9: if any temperature exhibits a negative number, or T_c3>T_m31.7, or T_m3>T_h3And 1.7, judging that the curve is not suitable for trimaran distribution analysis, and outputting double-trimaran distribution parameters, otherwise, continuing to the step 10.

Step 10: by T_h3、T_m3And T_c3Recombining voltammetric curves, then fitting V_p、n_eAnd the density of each temperature and other parameters are provided for a user, and the recombined curve and the data are respectively overlapped in a logarithmic graph and a linear graph for the user to check the accuracy of the fitting result. FIG. 3(c) shows the selection of the fitting points of the cold, medium and hot electron temperatures on a voltammetric curve, and the comparison of the reconstructed curve of the fitting parameters to the collected data.

Although illustrative embodiments of the present invention have been described above to facilitate the understanding of the present invention by those skilled in the art, it should be understood that the present invention is not limited to the scope of the embodiments, but various changes may be apparent to those skilled in the art, and it is intended that all inventive concepts utilizing the inventive concepts set forth herein be protected without departing from the spirit and scope of the present invention as defined and limited by the appended claims.

Claims

1. A Langmuir probe Dogmaisville electron distribution automated analysis method, characterized in that:

step 1: fitting the volt-ampere curve in a single-Mcewell distribution mode, automatically redefining a fitting range by using the fitted electron temperature for fitting after fitting the parameters of the seeds used for one time, and repeatedly cycling to achieve the most reasonable self-convergence result in the fitting range defined by the electron temperature;

step 2: fitting in a double-Mcewell distribution mode by using the optimal result of the single-Mcewell fitting as a seed parameter, and achieving the optimal result through cyclic feedback and repeated fitting;

and step 3: further taking the parameter output as a seed parameter of a fitting range of the trimaran distribution to carry out fitting in the trimaran distribution mode, and determining whether the double or trimaran distribution analysis is suitable or not according to a parameter result in each hierarchical delivery process;

and 4, step 4: after determining that the parameter result is suitable for double or trimaran Weir distribution analysis, fitting electron density and plasma potential parameters by a reorganized volt-ampere curve;

and 5: overlapping and drawing the recombined voltammetry curve and the data for a user to check the accuracy;

the step 1 is specifically as follows:

step 1.1: a user inputs a voltammetry profile number to be analyzed, the system draws the voltammetry profile after reading data, and automatically fits and subtracts ion saturated flow;

step 1.2: deriving the voltammetry curve to obtain an inflection point potential V_inf；

Step 1.3: the system draws a volt-ampere curve by using a semi-logarithmic graph with current as logarithm and voltage as linear number, takes a seed fitting interval near an inflection point potential to perform straight line fitting, and obtains the single-McWille electron temperature T_e1；

Step 1.4: the system resumes with T_e1Defining a fitting interval to perform fitting, and circulating the fitting interval until the obtained T_e1And T obtained last time_e1Self-convergence;

step 1.5: obtaining convergence T_e1The system then fits T by using_e1Recombining a voltammetry curve, and then fitting a trend line of the electron saturated flow increased by the expansion of the sheath layer on the original data, so as to obtain a saturated flow straight line and use T_e1The intersection of the recombined curves being the plasma potential V_pAnd calculating the electron density n_e；

The step 2 is specifically as follows:

step 2.1: using the obtained result from step 1 of the fitting process of the Mono-Mai Weir distribution, subtracting the voltammetry curve of the ion saturated flow, obtaining the absolute value of the whole curve again after derivation after semilogarithmization, and searching the lowest point V of the whole curve_hfitAs hot electron temperature T_h2Fitting a range reference point;

step 2.2: according to V_hfit>V>V_hfit10V fitting the Hot Electron temperature T_h2Then using T_h2Reconstructing a voltammetric curve and subtracting the curve from the data;

step 2.3: v at the remaining curve_inf>V>V_inf-T_e1Fitting Cold Electron temperature T in the/e Range_c2；

Step 2.4: after fitting, will be at V_hfit>V>V_hfit-T_h2E and V_inf>V>V_inf-T_c2Respectively as fitting T_h2And T_c2Range of (3) the above fitting procedure is cycled through until T_h2And T_c2The two parameters converge on themselves;

step 2.5: if any electron temperature is below zero, or T_c2>T_h21.7, automatically judging that the volt-ampere curve is not suitable for double-Mcewell distribution analysis by the system, and outputting parameters obtained by single-Mcewell distribution to a user; otherwise, continuing to step 2.6;

step 2.6: system fitting V_pAnd electron density n_eAnd the density of each temperature distribution, outputting parameters obtained by double-Maxwell distribution for TriMaxwell distribution analysis;

the step 3 is specifically as follows:

step 3.1: the system first obtains T in double-Maiweier distribution analysis_h2，T_c2；

Step 3.2: re-fitting T according to the analytical procedure of the double-McWigner distribution_h2；

Step 3.3: after subtracting T_h2Part of the voltammograms are again semilogarithmic and derived;

step 3.4: find the lowest point V in the middle again_mfit；

Step 3.5: at V_mfit+T_c2/e>V>V_mfit-T_c2Fitting intermediate electron temperature T in the/e range_m3；

Step 3.6: by T_m3Recombining a voltammetric curve from the fit of (a) and subtracting from the curve of step 3.5;

step 3.7: then V in the remaining data_inf>V>V_inf-T_c2E refitting the Cold Electron temperature T_c3；

Step 3.8: and then with V_mfit+T_m3/2e>V>V_mfit-T_m3Fitting range cycle fitting T is replaced by/2 e_m3With V_inf>V>V_inf-T_c3Cyclic fitting T with fitting range replaced by/e_c3Until the two temperatures approach to the same temperature;

step 3.9: if any temperature exhibits a negative number, or T_c3>T_m31.7, or T_m3>T_h31.7, judging that the curve is not suitable for trimaran distribution analysis, and outputting double-trimaran distribution parameters, otherwise, continuing to the step 3.10;

step 3.10: by T_h3、T_m3And T_c3Recombining voltammetric curves, then fitting V_p、n_eAnd providing the density parameters of each temperature for a user, and overlapping the recombined curve and the data in a logarithmic graph and a linear graph respectively for the user to check the accuracy of the fitting result.