Fuzzy hysteresis control method and circuit for high-power and high-stability remote plasma source

Abstract

The present invention belongs to the technical field of plasma sources, and provides a fuzzy hysteresis control method and circuit for a high-power and high-stability remote plasma source. When the reaction chamber is successfully ignited, the plasma source system is controlled to operate at a constant power output with a fixed frequency and variable duty cycle, including: respectively obtaining error control variables e and ec according to the sampled current value and duty cycle, and the predicted current value and duty cycle; using the error control variables e and ec as the input values of a fuzzy controller to obtain the actual control quantity hr output by the fuzzy controller; obtaining the compensation variables of current and duty cycle based on the error control variables e and ec; using the actual control quantity hr as the hysteresis width value of a hysteresis comparator, and compensating the current and duty cycle at the k-th moment with the output compensation variables according to a preset rule. The method provided by the present invention can ensure a relatively wide power window, improve the working stability of the remote plasma source, and enhance the uniformity of the semiconductor manufacturing process.

Description

Technical Field

[0001] The present invention relates to the technical field of plasma sources, and particularly relates to a fuzzy hysteresis control method and circuit for a high-power and high-stability remote plasma source. Background Art

[0002] As a new technology that uses high-energy plasma for material processing, plasma technology has been widely applied in fields such as industrial semiconductors, biomedicine, and food environment. Due to its characteristics of high efficiency, high flexibility, and sustainability, remote plasma sources are more widely used in processes such as surface cleaning, modification, and deposition of various materials, and their application prospects are becoming increasingly broad. However, current remote plasma source systems generally have problems such as low power output of the power supply, few types of ionizable gases that can be satisfied, a small gas flow range, and unstable plasma concentration, resulting in poor process uniformity and etching rate. In addition, when processing large samples or in large quantities, the methods of multi-source parallel connection and multi-stage series connection can increase the power supply power and processing efficiency, but they bring problems such as difficult synchronous control of remote plasma sources.

[0003] Currently, remote plasma source systems generally have problems such as low power output of the power supply, few types of ionizable gases that can be satisfied, a small gas flow range, and unstable plasma concentration. When processing large samples or in large quantities, the methods of multi-source parallel connection and multi-stage series connection can increase the power supply power and processing efficiency, but they bring problems such as resource waste, poor uniformity, and difficult synchronous control. Therefore, how to ensure the working stability of each system and the accuracy of control variables during the power increase of the plasma source has become an urgent problem for those skilled in the art. Summary of the Invention

[0004] In view of the deficiencies in the prior art, the present invention provides a fuzzy hysteresis control method and circuit for a high-power and high-stability remote plasma source to solve the problem of low output power of current remote plasma source systems.

[0005] In a first aspect, a fuzzy hysteresis control method for a high-power and high-stability remote plasma source provided by the present invention, when the reaction chamber is successfully ignited, controls the plasma source system to operate at a constant power output with a fixed frequency and variable duty cycle, including:

[0006] According to the sampled current value and duty cycle, and the predicted current value and duty cycle, error control variables e and ec are respectively obtained; where , , and are respectively the current sampling value and duty cycle at time k, and are respectively the predicted current sampling value and duty cycle at time k + 1;

[0007] Taking the error control variables e and ec as the input values of the fuzzy controller, the actual control quantity hr output by the fuzzy controller is obtained;

[0008] Based on the error control variables e and ec, the compensation variables for current and duty cycle are obtained and ;

[0009] Taking the actual control quantity hr as the hysteresis width value of the hysteresis comparator, the compensation variables are output according to the preset rules and respectively compensate the current and duty cycle at the k-th moment.

[0010] As can be seen from the above technical solutions, the method provided by the present invention combines the control of fuzzy control and variable hysteresis width hysteresis comparator, and feeds back and compensates to the output current and duty cycle D, which can improve the power regulation accuracy of the system, ensure a wide power window, improve the working stability of the remote plasma source, and enhance the uniformity of the semiconductor process technology.

[0011] Optionally, when compensating the current at the current moment with the output compensation variable , the preset rules include:

[0012] When the power error , subtract the variable from the current moment current and feedback and compensate it to the output current at the next moment;

[0013] When the power error , keep the at the k + 1 moment the same as the at the k moment;

[0014] When the power error , add the variable to the current moment current and feedback and compensate it to the output current at the next moment.

[0015] Optionally, when compensating the duty cycle at the current moment with the output compensation variable , the preset rules include:

[0016] When the current error , subtract the variable from the current moment duty cycle and feedback and compensate it to the output duty cycle at the next moment;

[0017] When the current error occurs, maintain the switching tube state at the previous moment to keep the duty cycle D relatively stable;

[0018] When the current error occurs, add the variable to the duty cycle at the current moment and feedback-compensate it to the output duty cycle at the next moment.

[0019] Optionally, the at the (k + 1)-th moment is predicted according to the following method:

[0020] The resonant frequency and the switching frequency of the resonant converter are respectively:

[0021] ; ; is the output frequency;

[0022] The equivalent resistance and the root mean square value of the fundamental component of the input square wave voltage are expressed as:

[0023] ; ;

[0024] Among them, the inductance ratio , characteristic impedance and quality factor of the equivalent circuit are respectively

[0025] ; ; ; , and are the resonant inductor, resonant capacitor and leakage inductor of the resonant converter respectively;

[0026] Then the primary winding current and voltage are respectively:

[0027] ,

[0028] ,

[0029] Predict the primary winding current at the (k + 1)-th moment.

[0030] Optionally, the duty cycle d(k + 1) at the (k + 1)-th moment is obtained by fitting with the arctangent function:

[0031] ,

[0032] Among them, is a sampling period time, A is a time-varying factor, and B is an influence factor of the duty cycle change rate.

[0033] Optionally, taking the error control variables e and ec as the input values of the fuzzy controller to obtain the actual control quantity hr output by the fuzzy controller includes:

[0034] Determining membership values E and EC according to the error control variables e, ec and the membership function; both membership values E and EC are divided into 5 fuzzy subspaces, and the linguistic values are respectively: much less than the lower limit value of the set range NB, slightly less than the lower limit value of the set range NS, within the set range ZO, slightly greater than the upper limit value of the set range PS, much greater than the upper limit value of the set range PB;

[0035] Determining the fuzzy output quantity H according to the membership values E and EC; the fuzzy output quantity H is divided into 5 fuzzy subspaces, and the linguistic values are respectively: negative large NB, negative small NS, zero ZO, positive small PS, positive large PB;

[0036] Defuzzifying according to the fuzzy output quantity H to obtain the actual control quantity hr.

[0037] Optionally, after receiving the ignition signal, controlling the plasma source system to work at a constant current output at the resonant frequency point to ignite the reaction chamber, including:

[0038] Controlling the introduction of the excitation gas;

[0039] At the first preset time before the ignition operation, instantaneously reducing the resonant output current reference value to the current reference value ;

[0040] When receiving the ignition signal, driving to close the relay S, otherwise the system continues to introduce the excitation gas;

[0041] Comparing the resonant output current reference value with the current value sampled by the power circuit and outputting the current error , after current PI modulation, starting the PWM drive unit;

[0042] Judging whether it satisfies and maintaining for more than the second preset time. If it is satisfied, then continue to judge whether the system ignites successfully, otherwise report an error for the ignition failure and disconnect the relay S;

[0043] If it is determined that the ignition is abnormal, adaptively adjust the plasma ignition duration and ignition voltage, amplify the ignition duration and ignition voltage by a ratio of (1 + K), and perform cyclic processing;

[0044] If the ignition is still not successful after exceeding the cyclic upper limit value, it is determined that the ignition fails, the system reports an error, and the relay is disconnected; when the ignition is successful, the relay is disconnected, the constant current stage ends, and the constant power stage is entered.

[0045] As can be seen from the above technical solution, by adopting the hybrid control method of constant current and constant power, various gases and a wider range of gas flow rates can be ionized, the process time can be shortened, and the applicability and etching rate of the remote plasma source can be improved.

[0046] In a second aspect, a fuzzy hysteresis control circuit for a high-power and high-stability remote plasma source provided by the present invention uses the remote plasma source fuzzy hysteresis control method provided by any possible implementation manner of the first aspect, and includes a control unit, a PWM driving unit, a resonant converter, an ignition circuit, a sampling unit, a relay, and a reaction cavity;

[0047] The control unit is used to control the PWM driving unit to output a pulse signal, and to control the relay to be attracted or disconnected;

[0048] The resonant converter converts the pulse signal output by the PWM driving unit into an AC source, and is used to provide an input AC source for the ignition circuit;

[0049] The ignition circuit generates a high-voltage ignition signal to cause the plasma load in the reaction cavity to oscillate and ionize at a high frequency;

[0050] The sampling unit is used to collect the ignition current of the ignition circuit, and the output voltage and output current of the resonant converter;

[0051] The resonant converter includes a transformer T1, and the ignition circuit is connected to the secondary coil N3 of the transformer T1;

[0052] The relay is controlled by the control unit to be attracted or disconnected.

[0053] Optionally, it further includes a maintenance circuit. The maintenance circuit is connected in series with the secondary coil N2 of the transformer T1, and uses the reaction cavity as the secondary side of the transformer T1 to only provide , to meet the low-voltage energy input in the constant power stage;

[0054] The maintenance circuit is used to maintain the stability of the bus voltage after entering the constant power stage from the constant current stage when the relay is disconnected.

[0055] By adopting the above technical solution, the present application has the following beneficial effects:

[0056] The method provided by the present invention combines the control of fuzzy control and variable hysteresis width hysteresis comparator, and feeds back and compensates the output current and duty cycle D, which can improve the power regulation accuracy of the system, ensure a wide power window, improve the working stability of the remote plasma source, and enhance the uniformity of the semiconductor manufacturing process.

[0057] The present invention adopts a hybrid control method of constant current and constant power, which can ionize a variety of gases and a wider range of gas flow rates, shorten the process time, and improve the applicability and etching rate of the remote plasma source. Brief Description of the Drawings

[0058] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the specific embodiments or the prior art. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0059] Figure 1 Shows a schematic diagram of a high-power and high-stability remote plasma source fuzzy hysteresis control circuit provided by an embodiment of the present invention;

[0060] Figure 2 Shows a flowchart of the constant current stage in the remote plasma source fuzzy hysteresis control method provided by an embodiment of the present invention;

[0061] Figure 3 Shows one of the flowcharts of the constant power stage in the remote plasma source fuzzy hysteresis control method provided by an embodiment of the present invention;

[0062] Figure 4 Shows another flowchart of the constant power stage in the remote plasma source fuzzy hysteresis control method provided by an embodiment of the present invention;

[0063] Figure 5 Shows a schematic diagram of the fuzzy controller provided by an embodiment of the present invention;

[0064] Figure 6 Shows a schematic diagram of the membership function provided by an embodiment of the present invention. Detailed Embodiments

[0065] The following will describe in detail the embodiments of the technical solutions of the present invention in conjunction with the drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present invention, and therefore are only examples and cannot be used to limit the protection scope of the present invention.

[0066] It should be noted that unless otherwise specified, the technical terms or scientific terms used in this application should have the ordinary meaning understood by those skilled in the art to which the present invention belongs.

[0067] As Figure 1 shown, this embodiment provides a fuzzy hysteresis control circuit for a high-power remote plasma source, including a remote plasma source power circuit, a constant current stage unit, and a constant power stage unit;

[0068] The remote plasma source power circuit includes a control unit, a PWM drive unit, a DC / AC resonant converter, an ignition circuit, a sampling unit, and a reaction chamber;

[0069] The control unit is used to control the PWM drive unit to output a pulse signal;

[0070] The PWM drive unit controls the output current of the resonant converter;

[0071] The output current of the resonant converter affects the input current of the ignition circuit; the resonant converter converts the pulse signal output by the PWM drive unit into an AC source and is used to provide an input AC source for the ignition circuit;

[0072] The ignition circuit generates a high-voltage ignition signal to cause the plasma load to oscillate and ionize at high frequency, and the resonant converter is used to provide an input AC source;

[0073] The resonant converter includes a transformer T1, and the ignition circuit is connected to the secondary coil N3 of the transformer T1;

[0074] The sampling unit is used to collect the ignition current of the ignition circuit, as well as the output voltage and output current of the resonant converter;

[0075] The relay S is controlled by the control unit to be closed or opened;

[0076] The constant current stage unit, after receiving the ignition signal, outputs a current and a duty cycle that enable the plasma source system to operate at a constant current output within ±1% near the resonant frequency point, for igniting the reaction chamber;

[0077] The constant power stage unit, when the reaction chamber is successfully ignited, outputs a current and a duty cycle that enable the plasma source system to operate at a constant power output with a fixed frequency and variable duty cycle.

[0078] Optionally, the remote plasma source power circuit further includes a maintenance circuit. The maintenance circuit is connected in series with the secondary coil N2 of the transformer T. Regarding the vacuum reaction chamber as the secondary side of the high-frequency transformer, it can only provide to meet the low-voltage energy input during the ionization maintenance stage; the maintenance circuit is used to maintain the stability of the bus voltage when the relay is disconnected and after entering the constant power stage (ionization stage) from the constant current stage (ignition stage);

[0079] As Figure 2As shown in the figure, a remote plasma source fuzzy hysteresis control method provided by this embodiment controls the plasma source system to output a constant current of ±1% near the resonant frequency point after receiving the ignition signal, so as to ignite the reaction chamber. At this time, the system is in the constant current stage, that is, the ignition working stage, including:

[0080] S110. After the remote plasma source is powered on and running, first the system sets the gas current excitation threshold , and controls the introduction of the excitation gas;

[0081] S120. Through the stage adjustment unit, 50 ms before the ignition operation, the resonant output current reference value is instantaneously reduced to the current reference value . This step can effectively avoid inrush current and play a role in protecting the relay S.

[0082] S130. Determine whether the ignition signal is received. When the ignition signal is received, drive the relay S to close, otherwise the system continues to introduce the excitation gas;

[0083] S140. Compare the resonant output current reference value with the current value sampled by the power circuit and output the current error . After current PI modulation, start the PWM drive unit;

[0084] S150. Determine whether is satisfied and maintained for more than 500 ms. If satisfied, then continue to determine whether the system ignites successfully, otherwise report an error for ignition failure and disconnect the relay S;

[0085] S160. If it is determined that the ignition is abnormal, adaptively adjust the plasma ignition duration and ignition voltage, amplify the ignition duration and ignition voltage by a ratio of (1 + K) and perform cyclic processing; the upper limit of the number of cycles is 10 times to improve the ignition success rate;

[0086] S170. If the ignition is still not successful after exceeding the cycle upper limit, it is determined that the ignition fails and the system reports an error and disconnects the relay; when the ignition is successful, disconnect the relay, end the constant current stage and enter the constant power stage.

[0087] As Figures 3 - 4 shown, a remote plasma source fuzzy hysteresis control method provided by this embodiment controls the plasma source system to output a constant power with a fixed frequency and variable duty cycle when the reaction chamber is successfully ignited, including:

[0088] S210. According to the sampled current value and duty cycle, and the predicted current value and duty cycle, respectively obtain the error control variables e and ec.

[0089] Among them, , , and are the current sampling value and duty cycle at the k-th moment respectively, and are the predicted current sampling value and duty cycle at the (k + 1)-th moment respectively.

[0090] Specifically, at the (k + 1)-th moment is predicted according to the following method:

[0091] In this embodiment, the resonant converter is an LCL resonant converter. Due to the resonant effect of the converter, the normalized values of the resonant frequency and the switching frequency are respectively: ;

[0092] ; ; is the output frequency;

[0093] The equivalent resistance and the root mean square value of the fundamental component of the input square wave voltage are expressed as:

[0094] ; ;

[0095] Among them, the inductance ratio , characteristic impedance and quality factor are respectively

[0096] ; ; ; , and are the resonant inductor, resonant capacitor and leakage inductor of the resonant converter respectively;

[0097] It can be known that the primary winding current and voltage are respectively:

[0098] ,

[0099] ,

[0100] Predict the primary winding current .

[0101] According to the characteristics of the LCL resonant circuit, when the frequency is fixed, the resonant current remains relatively constant;

[0102] Then the actual power at time k can be sampled by the plasma voltage / current and is expressed as:

[0103]

[0104] where is the current and the voltage vector angle, with a value range of ;

[0105] The duty cycle at the current time k can be obtained by passing the current error e1 in the current loop through current PI modulation. According to the duty cycle at the current time k of the object under test, the duty cycle at time k + 1 is predicted by fitting

[0106] The duty cycle at time k + 1 is obtained by fitting with the arctangent function:

[0107] ,

[0108] where is the time of one sampling period, A is the time-varying factor, and B is the influence factor of the duty cycle change rate.

[0109] S220. Take the error control variables e and ec as the input values of the fuzzy controller, and obtain the actual control quantity hr output by the fuzzy controller.

[0110] The fuzzy controller quantifies and calculates the input quantity and the feedback quantity to obtain the system errors e and ec as the input of the fuzzy controller; performs fuzzy processing on the control variables to determine the fuzzy domain and the corresponding membership function to obtain E and EC; then makes a decision judgment according to the pre-established fuzzy control rules to obtain the fuzzy output quantity H; performs defuzzification processing on the fuzzy output quantity H to obtain the actual control quantity hr, and sends it to the control mechanism to complete the entire control process;

[0111] Suppose the input fuzzy sets e and ec of the fuzzy controller and the fuzzy set of its output quantity H are as follows:

[0112] In this embodiment, the difference between the resonant output current reference value iref and the sampled current value ipri is used to control the bandwidth hr of the hysteresis controller. The input fuzzy sets e and ec are divided into 5 fuzzy subspaces, and the linguistic values are: much less than the lower limit of the set range (NB), slightly less than the lower limit of the set range (NS), within the set range (ZO), slightly greater than the upper limit of the set range (PS), much greater than the upper limit of the set range (PB). The hysteresis width hr is divided into 5 fuzzy subspaces, and the linguistic values are: negative large (NB), negative small (NS), zero (ZO), positive small (PS), positive large (PB). Then there is:

[0113] ,

[0114] ,

[0115] The basic universes of discourse of the input fuzzy sets e and ec are [-1, 1], and the basic universe of discourse of the hysteresis width hr is [-1, 1]. To ensure that the fuzzy controller has a high sensitivity, at the boundaries of the basic universe of discourse, the input membership functions of each variable are selected as triangular functions with higher sensitivity and are evenly distributed.

[0116] Step S220 specifically includes:

[0117] S221. According to the error control variables e and ec and the membership functions, determine the membership values E and EC; the membership values E and EC are both divided into 5 fuzzy subspaces, and the linguistic values are: much less than the lower limit of the set range NB, slightly less than the lower limit of the set range NS, within the set range ZO, slightly greater than the upper limit of the set range PS, much greater than the upper limit of the set range PB;

[0118] S222. According to the membership values E and EC, determine the fuzzy output quantity H; the fuzzy output quantity H is divided into 5 fuzzy subspaces, and the linguistic values are: negative large NB, negative small NS, zero ZO, positive small PS, positive large PB;

[0119] S223. According to the fuzzy output quantity H, defuzzify to obtain the actual control quantity hr.

[0120] As Figure 6 shown, (a)-(c) are respectively the membership function diagrams of the output quantities E, EC, and H in the fuzzy rule base of the fuzzy controller. Table 1 also shows the fuzzy rule base of the fuzzy sets E, EC, and the fuzzy output quantity H.

[0121] Table 1

[0122]

[0123] In this embodiment, the Mamdani inference method is adopted for the fuzzy inference method. Mamdani is a commonly used method in fuzzy control. Its essence is a compositional inference method. The i-th rule in the rule base is expressed as:

[0124] Ri “If e is E and ec is EC, Then hr is H”

[0125] where e, ec, and hr are the language variables corresponding to E, EC, and H respectively;

[0126] where hr is the fuzzy output quantity, and the inherent fuzzy relationship is:

[0127]

[0128] The specific steps of the fuzzy inference are as follows: perform fuzzy inference on the fuzzy input quantities E and EC according to the set rule R, and output the fuzzy output quantity H; where ;

[0129] Then the inherent fuzzy relationship of the entire rule base is:

[0130]

[0131] The specific steps of the defuzzification process are as follows: adopt the centroid method to find the centroid value of each element in the fuzzy output quantity H and its corresponding membership degree to obtain the output quantity hr.

[0132]

[0133] where hr is the output quantity of the fuzzy controller, is the central value of the membership function interval corresponding to the output quantity, is the corresponding membership degree;

[0134] The steps to obtain the output quantity hr include: first, perform fuzzy processing on the input signal, then perform fuzzy inference according to the set data set and rule set to obtain the fuzzy output quantity H; finally, perform defuzzification processing on the fuzzy output quantity H using the centroid method to finally obtain the output quantity hr.

[0135] S230. Obtain the compensation variables of the current and duty cycle based on the error control variables e and ec and .

[0136] In this step, the compensation variables and are positively correlated with the error control variables e and ec, and are specifically determined according to the following formula:

[0137]

[0138]

[0139] m and n are correlation coefficients, which can be selected by those skilled in the art according to the actual application situation.

[0140] S240. Take the actual control quantity hr as the hysteresis width value of the hysteresis comparator, and output the compensation variable according to the preset rules and Compensate the current and the duty cycle at time k respectively.

[0141] After fuzzy inference, hr is obtained as the hysteresis width value of the hysteresis comparator. The rules of the variable hysteresis width hysteresis comparator at this time are:

[0142]

[0143]

[0144] Specifically, it includes outputting the compensation variable to compensate the current at the current moment, and outputting the compensation variable to compensate the duty cycle at the current moment.

[0145] When compensating the current at the current moment with the output compensation variable , the preset rules specifically include:

[0146] When the power error , subtract the variable from the current moment current and feedback the compensation to the output current at the next moment;

[0147] When the power error , keep the at time k + 1 the same as at time k;

[0148] When the power error , add the variable to the current moment current and feedback the compensation to the output current at the next moment.

[0149] Specifically, calculate the actual output power according to the sampled output voltage and current, and compare the at time k + 1 with After comparison, the current error e is obtained, and the compensation variable of the power loop output is calculated from the current error e. ; When the power error output by the power loop is , subtract the variable from the current current and feedback-compensate it to the output current of the next moment ; When the power error is , keep the at the (k + 1)-th moment consistent with that at the k-th moment ; When the power error is , add the variable

[0150] to the current current and feedback-compensate it to the output current of the next moment .

[0151] When compensating the duty cycle at the current moment with the output compensation variable , the preset rules specifically include: When the current error is

[0152] , subtract the variable from the duty cycle

[0153] at the current moment and feedback-compensate it to the output duty cycle of the next moment ; When the current error is

[0154] , maintain the switching tube state of the previous moment to keep the duty cycle D relatively stable; When the current error is , subtract the variable from the duty cycle at the current moment and feedback-compensate it to the output duty cycle of the next moment , that is, turn off T1 and T3 and turn on T2 and T4 to reduce the duty cycle D; when the current error is , maintain the switching tube state of the previous moment to keep the duty cycle D relatively stable; when the current error is The feedback compensates for the output duty cycle at the next moment , that is, turning on T1 and T3 and turning off T2 and T4 to increase the duty cycle D.

[0155] The above embodiments are only used to introduce the technical solutions of the present application in detail. However, the description of the above embodiments is only used to help understand the method of the embodiments of the present invention and should not be construed as a limitation of the embodiments of the present invention. Any changes or substitutions that can be easily thought of by those skilled in the art should be covered within the protection scope of the embodiments of the present invention.

Claims

1. A high-power and high-stability remote plasma source fuzzy hysteresis control method, characterized in that: After receiving the ignition signal, the plasma source system is controlled to operate at a constant current output at a resonant frequency point to ignite the reaction chamber; When the reaction chamber is ignited successfully, the plasma source system is controlled to operate at a constant power output with a fixed frequency and variable duty cycle, including: According to the sampled current value and duty cycle, the predicted current value and duty cycle, the error control variables e and ec are obtained respectively; wherein, , , and are the current sampling value and duty cycle at time k, and are respectively the current sampling value and duty cycle predicted at time k+1; The error control variables e and ec are used as the input values ​​of the fuzzy controller to obtain the actual control quantity hr output by the fuzzy controller; Obtain compensation variables for current and duty cycle based on error control variables e and ec and ; The actual control variable hr is used as the loop width value of the hysteresis comparator, and the compensation variable is output according to the preset rules. and The current at time k is and duty cycle Make compensation.

2. The method according to claim 1, characterized in that: Output compensation variable The current at the current moment When making compensation, the preset rules include: When the power error When the current Subtracting variables Feedback compensation to the output current at the next moment ; When the power error When maintaining the k+1 moment and k moment Consistency; When the power error When the current Add variables Feedback compensation to the output current at the next moment .

3. The method according to claim 2, characterized in that Output compensation variable The duty cycle at the current moment When making compensation, the preset rules include: When the current error When the current duty cycle Subtracting variables Feedback compensation to the output duty cycle at the next moment ; When the current error When , the switch state of the previous moment is maintained to keep the duty cycle D relatively stable; When the current error When the current duty cycle Add variables Feedback compensation to the output duty cycle at the next moment .

4. The method according to claim 1 or 3, characterized in that: At k+1 time The predictions are based on the following method: Resonant frequency of a resonant converter and switching frequency The normalized value of They are: ; ; is the output frequency; Equivalent resistance The RMS value of the fundamental component of the input square wave voltage is expressed as: ； ； Among them, the inductance ratio of the equivalent circuit is , characteristic impedance and quality factors They are ; ; ; , and They are the resonant inductance, resonant capacitance and leakage inductance of the resonant converter respectively; The primary winding current and voltage They are: ， ， Predict the primary winding current at time k+1 .

5. The method according to claim 4, characterized in that The duty cycle d (k+1) at time k+1 is obtained by fitting the inverse tangent function: ， in, is a sampling cycle time, A is the time-varying factor, and B is the factor affecting the duty cycle change rate.

6. The method according to claim 1, characterized in that The error control variables e and ec are used as input values ​​of the fuzzy controller to obtain the actual control quantity hr output by the fuzzy controller, including: According to the error control variables e, ec and the membership function, the membership values ​​E and EC are determined; the membership values ​​E and EC are divided into 5 fuzzy subspaces, and the language values ​​are: too less than the lower limit of the set range NB, slightly less than the lower limit of the set range NS, within the set range ZO, slightly greater than the upper limit of the set range PS, and too greater than the upper limit of the set range PB; According to the membership values ​​E and EC, the fuzzy output H is determined; the fuzzy output H is divided into five fuzzy subspaces, and the language values ​​are: negative large NB, negative small NS, zero ZO, positive small PS, and positive large PB; According to the fuzzy output H, the actual control quantity hr is obtained by defuzzification.

7. The method according to claim 1, characterized in that After receiving the ignition signal, the plasma source system is controlled to operate at a constant current output at the resonant frequency point to ignite the reaction chamber, including: Controlling the introduction of exciting gas; At the first preset time before the ignition operation, the resonant output current reference value Instantly reduce to the current reference value ; When receiving the ignition signal, the relay S is driven to close, otherwise the system continues to pass the excitation gas; The resonant output current reference value The current value sampled by the power circuit Output current error after comparison , after current PI modulation, the PWM drive unit is started; Determine whether it is satisfied And it is maintained for more than the second preset time. If it is satisfied, then it is determined whether the system is ignited successfully. Otherwise, it is ignition failure, the system reports an error and disconnects the relay S; If it is determined that the ignition is abnormal, the plasma ignition duration and ignition voltage are adaptively adjusted, and the ignition duration and ignition voltage are amplified by a ratio of (1+K) and cyclically processed; If the ignition fails after exceeding the cycle upper limit, it is judged as an ignition failure, the system reports an error and disconnects the relay; when the ignition is successful, the relay is disconnected, the constant current stage ends and the constant power stage begins.

8. A remote plasma source fuzzy hysteresis control circuit, characterized in that: A remote plasma source fuzzy hysteresis control method using any one of claims 1 to 7, comprising a control unit, a PWM drive unit, a resonant converter, an ignition circuit, a sampling unit, a relay and a reaction chamber; The control unit is used to control the PWM drive unit to output a pulse signal and control the relay to be closed or disconnected; The resonant converter converts the pulse signal output by the PWM drive unit into an AC source, and is used to provide an input AC source for the ignition circuit; The ignition circuit generates a high-voltage ignition signal to cause the plasma load in the reaction chamber to oscillate and ionize at a high frequency; The sampling unit is used to collect the ignition current of the ignition circuit, and the output voltage and output current of the resonant converter; The resonant converter includes a transformer T1, and the ignition circuit is connected to a secondary coil N3 of the transformer T1; The relay is controlled by the control unit to be closed or disconnected.

9. The circuit according to claim 8, characterized in that The maintenance circuit is also included. The maintenance circuit is connected in series with the secondary coil N2 of the transformer T1, and the reaction cavity is used as the secondary side of the transformer T1 to provide only , to meet the low-voltage energy input in the constant power stage; The maintaining circuit is used to maintain the bus voltage stable after entering the constant power stage from the constant current stage when the relay is disconnected.

Images