CN115979533A - Leak rate detection method and semiconductor process equipment - Google Patents
Leak rate detection method and semiconductor process equipment Download PDFInfo
- Publication number
- CN115979533A CN115979533A CN202211438988.0A CN202211438988A CN115979533A CN 115979533 A CN115979533 A CN 115979533A CN 202211438988 A CN202211438988 A CN 202211438988A CN 115979533 A CN115979533 A CN 115979533A
- Authority
- CN
- China
- Prior art keywords
- pressure
- value
- data
- leak rate
- data set
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P70/00—Climate change mitigation technologies in the production process for final industrial or consumer products
- Y02P70/50—Manufacturing or production processes characterised by the final manufactured product
Abstract
The invention provides a leak rate detection method and semiconductor process equipment, and relates to the technical field of leak rate detection, wherein the leak rate detection method comprises the following steps: acquiring a pressure data set of a reaction chamber in a closed state within a preset time length; the pressure data set is a pressure value set acquired under a preset frequency; filtering the pressure data set to obtain pressure filtering data; performing curve fitting on the pressure filtering data to obtain a fitting function; a pressure increase per unit time length is determined based on the fitting function, and a leak rate of the reaction chamber is determined based on the pressure increase per unit time length. The invention can filter abnormal data points with larger fluctuation range in the pressure data set, avoid the abnormal data points from causing larger leakage rate fluctuation, reduce the false alarm probability, ensure the production efficiency of equipment and reduce the personnel maintenance cost; meanwhile, the leakage rate detection of the same container in each process is relatively stable, and the accuracy of the leakage rate detection is improved.
Description
Technical Field
The invention relates to the technical field of leakage rate detection, in particular to a leakage rate detection method and semiconductor process equipment.
Background
The leak rate is a pressure increase value per minute in a closed environment, and the leak rate detection is generally applied to a process with strict requirements on pressure, such as the pressure requirement on a furnace tube in a plasma deposition process in the production process of a crystalline silicon solar cell module. PECVD (Plasma Enhanced Chemical Vapor Deposition) equipment is key equipment for a coating process in the production process of a photovoltaic cell, and the equipment controls a furnace tube to realize Plasma Deposition under stable pressure through a butterfly valve. In order to prevent impurities from entering the furnace tube during the plasma deposition process, it is necessary to ensure that the leak rate of the furnace tube in the bottom pressure state (in the vacuum state with the pressure of 0 Pa) meets specific requirements, so that the leak rate detection of the furnace tube before the ion deposition process is necessary.
The related leakage rate detection technology generally calculates the leakage rate by comparing the pressure change value of the starting time with the pressure change value of the current time, however, because the pressure sensor in the furnace tube is easy to generate detection deviation under the bottom pressure state, the acquired pressure data is easy to fluctuate, the calculated leakage rate of the same furnace tube in each process is large in fluctuation, the deviation exists from the actual leakage rate condition of the furnace tube, the false alarm condition of equipment is easy to occur, the production efficiency of the equipment is influenced, the maintenance cost of personnel is increased, and the accuracy rate of the leakage rate detection is reduced.
Disclosure of Invention
In view of the above, the present invention provides a leak rate detection method and a semiconductor process apparatus, which can avoid large leak rate fluctuation caused by abnormal data points, reduce false alarm probability, ensure production efficiency of the apparatus, reduce personnel maintenance cost, and simultaneously enable leak rate detection of the same container in each process to be relatively stable, thereby improving the accuracy of leak rate detection.
In order to achieve the above purpose, the embodiment of the present invention adopts the following technical solutions:
in a first aspect, an embodiment of the present invention provides a leak rate detection method, including: acquiring a pressure data set of the reaction chamber in a closed state within a preset time length; the pressure data set is a pressure value set acquired under a preset frequency; filtering the pressure data set to obtain pressure filtering data; performing curve fitting on the pressure filtering data to obtain a fitting function; determining a pressure increase per unit time based on the fit function, and determining a leak rate of the reaction chamber based on the pressure increase per unit time.
Further, an embodiment of the present invention provides a first possible implementation manner of the first aspect, where the step of performing filtering processing on the pressure data set to obtain pressure filtering data includes: and sequentially determining the optimal pressure estimation value corresponding to each pressure value in the pressure data set based on a Kalman filtering algorithm, and taking the optimal pressure estimation value corresponding to each pressure value in the pressure data set as the pressure filtering data so as to optimize and remove pressure fluctuation noise data.
Further, an embodiment of the present invention provides a second possible implementation manner of the first aspect, where the step of sequentially determining an optimal pressure estimation value corresponding to each pressure value in the pressure data set based on a kalman filter algorithm includes: sequentially acquiring each pressure value in the pressure data set based on the sequence of the pressure value acquisition time, predicting a pressure predicted value corresponding to a pressure value acquired at the later time based on an optimal pressure estimated value corresponding to a pressure value acquired at the previous time, and predicting an error covariance predicted value corresponding to a pressure value acquired at the later time based on a covariance corresponding to a pressure value acquired at the previous time; determining a Kalman gain corresponding to the pressure value acquired at the later moment based on the error covariance predicted value corresponding to the pressure value acquired at the later moment; and determining an optimal pressure estimation value corresponding to the pressure value acquired at the later moment based on the pressure value acquired at the later moment, the corresponding pressure predicted value and the Kalman gain, and determining a covariance corresponding to the pressure value acquired at the later moment based on the Kalman gain corresponding to the pressure value acquired at the later moment and the error covariance predicted value until obtaining the optimal pressure estimation value corresponding to all the pressure values in the pressure data set.
Further, an embodiment of the present invention provides a third possible implementation manner of the first aspect, wherein the step of performing curve fitting on the pressure filtering data to obtain a fitting function includes: and establishing a regression model corresponding to the pressure filtering data based on a least square method, and solving a fitting function matched with the pressure filtering data based on the regression model.
Further, an embodiment of the present invention provides a fourth possible implementation manner of the first aspect, where the fitting function is a linear equation, and horizontal and vertical coordinates of the fitting function are the data sorting number and the pressure value corresponding to the data sorting number, respectively.
Further, an embodiment of the present invention provides a fifth possible implementation manner of the first aspect, wherein the step of determining a leak rate of the reaction chamber based on the fitting function and the preset frequency includes: obtaining the slope of the fitting function; determining the pressure value collection quantity per minute based on the preset frequency; and calculating the product of the slope of the fitting function and the pressure value collection quantity per minute to obtain the leakage rate of the reaction chamber.
Further, an embodiment of the present invention provides a sixth possible implementation manner of the first aspect, wherein the step of determining a leak rate of the reaction chamber based on the fitting function and the preset frequency includes: acquiring coordinates of any two points from the fitting function, and calculating a pressure difference value and a data sorting number difference value corresponding to the coordinates of any two points; determining the acquisition time difference corresponding to the coordinates of any two points based on the data sequence number difference and the preset frequency; determining a leak rate of the reaction chamber based on the pressure difference value and the acquisition time difference.
Further, an embodiment of the present invention provides a seventh possible implementation manner of the first aspect, wherein the preset frequency is 4Hz.
In a second aspect, embodiments of the present invention further provide a semiconductor processing apparatus, including a reaction chamber and a control unit, the control unit being configured to perform the method according to any one of the first aspect, so as to detect a leak rate of the reaction chamber.
Further, an embodiment of the present invention provides a first possible implementation manner of the second aspect, wherein the control unit communicates with a lower computer through a serial port, so that the lower computer collects a pressure value of the reaction chamber at a preset frequency; the serial port attribute is blocking.
The embodiment of the invention provides a leak rate detection method and semiconductor process equipment, wherein the leak rate detection method comprises the following steps: acquiring a pressure data set of a reaction chamber in a closed state within a preset time length; the pressure data set is a pressure value set acquired under a preset frequency; filtering the pressure data set to obtain pressure filtering data; performing curve fitting on the pressure filtering data to obtain a fitting function; a pressure increase per unit time length is determined based on the fitting function, and a leak rate of the reaction chamber is determined based on the pressure increase per unit time length. According to the invention, the pressure data set of the reaction chamber within a certain time length is obtained, and the pressure data set is subjected to filtering treatment, so that abnormal data points with a large fluctuation range in the pressure data set can be filtered, the phenomenon that the leakage rate fluctuates greatly due to the abnormal data points is avoided, the false alarm probability is reduced, the production efficiency of equipment is ensured, and the personnel maintenance cost is reduced; the fitting function is obtained by fitting the pressure filtering data, and the leak rate is determined based on the pressure increase value in unit time length in the fitting function, so that the leak rate detection of the same container in each process is relatively stable, and the leak rate detection accuracy is improved.
Additional features and advantages of embodiments of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of embodiments of the invention as set forth above.
In order to make the aforementioned and other objects, features and advantages of the present invention comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
FIG. 1 is a schematic diagram showing a variation in pressure values in a furnace;
fig. 2 is a flow chart of a leak rate detection method according to an embodiment of the present invention;
FIG. 3 is a flow chart of a Kalman filtering algorithm provided by an embodiment of the present invention;
FIG. 4 illustrates a comparison graph of pressure data set filtering provided by an embodiment of the present invention;
FIG. 5 is a schematic diagram illustrating a pressure value fitting curve provided by an embodiment of the present invention;
FIG. 6 is a graph illustrating a leak rate detection experiment comparison provided by an embodiment of the present invention;
fig. 7 shows a flow chart of pressure value collection according to an embodiment of the present invention.
Detailed Description
To make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some, not all, embodiments of the present invention.
In the related leak rate detection technology, taking the leak rate detection of the PECVD equipment as an example, before the PECVD equipment performs a coating process, the furnace door of the furnace tube and all the valves of the gas inlet pipeline need to be closed, the air in the furnace is completely pumped out, the opening angle of the valves is controlled by a butterfly valve controller through an internal PID (proportional Integral Differential) algorithm to control the pressure in the furnace tube, and the evacuation valve is closed by the butterfly valve controller when the pressure in the furnace tube reaches a bottom pressure vacuum state of 0 Pa. And under a closed condition, recording the pressure value in the furnace in real time, and comparing the pressure change value at the starting moment with the pressure change value at the current moment (namely the 1min moment) after 1min, thereby obtaining the leakage rate of the equipment. Referring to the schematic diagram of the variation of the pressure value in the furnace shown in fig. 1, fig. 1 shows a variation curve of the pressure value in the furnace, where the abscissa of the schematic diagram is the sampling interval of the pressure value, the ordinate is the actual pressure value of the furnace tube, and the sampling frequency of the pressure value is 4Hz.
In the actual production process, the current value detected by the sensor is very small because the pressure in the furnace is very low. Therefore, the current output value is very small in the bottom pressure state, and the detection value fluctuates up and down due to insufficient sensor accuracy (the range of the sensor is large and the detection value is small) and insufficient filtering function, and as shown in fig. 1, although the pressure value remains rising during the detection, it is not stable but fluctuates with the up-down error. The fluctuation can cause the situation that the actual pressure values acquired at the starting time and the ending time of the leakage rate detection fluctuate up and down, the leakage rate obtained based on the pressure change values at the starting time and the current time can fluctuate greatly each time, and the leakage rate detection is inaccurate, so that the related leakage rate detection technology has the following defects:
1. the calculation deviation of the leakage rate is large, the fluctuation of the calculation result of the leakage rate of the same furnace tube in each process is large, the calculation result is not accurate, and the deviation of the calculation result and the actual condition of the furnace tube is large;
2. false alarm of equipment is easily caused, and the process operation is stopped when the acquisition of low values and peak values is started and ended, so that the production capacity of the equipment is influenced;
3. the maintenance cost of personnel is increased, the staff cannot accurately predict the real situation of the equipment, and the regular maintenance workload and the personnel cost of customers are increased.
In order to solve the above problems, the leakage rate detection method and the semiconductor processing equipment provided in the embodiments of the present invention can be applied to improve the accuracy of leakage rate detection, reduce the false alarm probability, and reduce the personnel maintenance cost. The following describes embodiments of the present invention in detail.
The present embodiment provides a leak rate detection method, which can be applied to semiconductor processing equipment to detect a leak rate of a reaction chamber in the semiconductor processing equipment, referring to a flow chart of the leak rate detection method shown in fig. 2, and the method mainly includes the following steps:
step S202, acquiring a pressure data set of the reaction chamber in a closed state within a preset time length.
The reaction chamber can be any container which needs to be subjected to leak rate detection, such as a furnace tube in PECVD equipment, before the leak rate detection is carried out, all air in the reaction chamber is pumped out, so that the reaction chamber reaches a bottom pressure vacuum state of 0Pa, the pressure value of the reaction chamber in a preset time period is acquired at a preset frequency based on a pressure sensor in a closed state, and the pressure value acquired in the preset time period is collected as a pressure data set.
The preset time duration can be set according to the requirement of the leakage rate detection time, and can be a time duration greater than or equal to 1 minute so as to collect more pressure value data and improve the leakage rate detection accuracy, such as 1 minute to 2 minutes; in the case of emergency leak rate detection time, the preset time period may also be a time period less than 1 minute to increase the leak rate detection speed, such as 30s to 50s. In order to facilitate the calculation of the leakage rate, the preferable value of the preset time is 1 minute.
The pressure data set is a pressure value set acquired under the preset frequency, the number of pressure values in the pressure data set can be controlled through the acquisition frequency of the set pressure values, the larger the preset frequency is, the smaller the pressure value acquisition interval is, and the more the number of pressure values in the pressure value data set is.
In the related leak rate detection technology, the pressure value is collected at a speed of 40 to 60 times per minute, and in a specific embodiment, in order to increase the pressure value collection speed, the value range of the preset frequency may be 3 to 6Hz, and the preferred value of the preset frequency may be 4Hz, that is, 4 pressure values are collected at each second and 240 pressure values are collected at each minute. In the embodiment, a large amount of process data in the leak detection process can be obtained by increasing the pressure value acquisition speed in the leak detection process, and data support is provided for improving the calculation accuracy of the leak rate.
And step S204, filtering the pressure data set to obtain pressure filtering data.
And performing data filtering processing on the pressure data set based on a filtering algorithm, filtering abnormal data points of noise and interference in the pressure data set, restoring real data, concentrating pressure values in the pressure data set more, reducing fluctuation deviation of the pressure values, and recording the filtered pressure data set as pressure filtering data. The filtering algorithm may be, for example, a kalman filtering algorithm, a bayesian filtering algorithm, a clipping filtering algorithm, or other filtering algorithms.
And step S206, performing curve fitting on the pressure filtering data to obtain a fitting function.
Because the pressure in the container is generally continuously increased in the leakage rate detection process and meets the linear condition (if the pressure increase value is stable and unchanged, the linear condition is met, and if the pressure increase value is continuously changed, the linear condition is met), a fitting curve can be set as a linear equation, the pressure filtering data obtained after filtering is subjected to curve fitting to obtain a fitting function matched with the pressure filtering data, and the fitting function is a relation curve of pressure acquisition time or the number of acquisition points and the pressure value.
And S208, determining a pressure increase value per unit time length based on the fitting function, and determining the leakage rate of the reaction chamber based on the pressure increase value per unit time length.
The unit time length can be any time length set by a user, and the leak rate of the reaction chamber is = the pressure increase value of the unit time length when the unit time length is 1 minute because the leak rate is the pressure increase value of each minute; when the above unit time length is 2 minutes, the leak rate of the reaction chamber = pressure increase value/2 per unit time length; when the above unit time length is 0.5 minutes, the leak rate of the reaction chamber = pressure increase value x2 per unit time length.
According to the leak rate detection method provided by the embodiment, the pressure data set of the reaction chamber within a certain time length is obtained, and the pressure data set is subjected to filtering processing, so that abnormal data points with a large fluctuation range in the pressure data set can be filtered, the phenomenon that the leak rate fluctuates greatly due to the abnormal data points is avoided, the false alarm probability is reduced, the production efficiency of equipment is guaranteed, and the personnel maintenance cost is reduced; the fitting function is obtained by fitting the pressure filtering data, and the leak rate is determined based on the pressure increase value in unit time length in the fitting function, so that the leak rate detection of the same container in each process is relatively stable, and the leak rate detection accuracy is improved.
In an embodiment, in order to avoid that continuous oscillation of pressure values in a pressure data set affects the accuracy of leak rate calculation, the embodiment provides an implementation manner of performing filtering processing on the pressure data set to obtain pressure filtering data: and sequentially determining the optimal pressure estimation value corresponding to each pressure value in the pressure data set based on a Kalman filtering algorithm, and taking the optimal pressure estimation value corresponding to each pressure value in the pressure data set as pressure filtering data so as to optimize and remove pressure fluctuation noise data.
Based on a Kalman filtering algorithm, a linear system state equation is utilized, an optimal estimated value of a state variable of a dynamic system at the current moment is obtained through a state estimated value at the previous moment and an observed value at the current moment, pressure value data acquired during leakage rate detection is set to be X (k), k represents a kth pressure value acquisition point in a pressure data set and can also be called as k moment, referring to a Kalman filtering algorithm flow chart shown in FIG. 3, and a filtering process mainly comprises calculation of two main parameters of state estimation and covariance estimation:
in one embodiment, each pressure value in the pressure data set may be sequentially obtained based on the order of pressure value collection time, a pressure predicted value (pressure predicted value at time k) corresponding to a pressure value collected at a later time may be predicted based on an optimal pressure estimated value (optimal pressure estimated value at time k-1) corresponding to a pressure value collected at a previous time, and an error covariance predicted value (covariance predicted value at time k) corresponding to a pressure value collected at a later time may be predicted based on a covariance (covariance at time k-1) corresponding to a pressure value collected at a previous time.
Specifically, as shown in fig. 3, a predicted value X (k | k-1) of the pressure at the k time is predicted according to an optimal pressure estimation value at the k-1 time (an optimal pressure estimation value corresponding to the pressure value acquired at the k-1 time, that is, a state estimation value X (k-1) at the k-1 time); predicting a k moment error covariance predicted value P (k | k-1) according to the k-1 moment covariance P (k-1) which is the covariance corresponding to the pressure value collected at the k-1 moment:
x(k|k-1)＝Ax(k-1|k-1)+Bu(k) (1)
P(k|k-1)＝AP(k-1|k-1)A'+S (2)
wherein, X (k | k-1) is a pressure predicted value at the moment k, P (k | k-1) is an error covariance predicted value at the moment k, X (k-1) is an optimal pressure estimated value at the moment k-1, P (k-1) is a covariance at the moment k-1, A is a state matrix, A' is a transpose of the state matrix, B is a control matrix, and u (k) is a control variable of the current state; since no external control variable such as a specific gas flow input regulation pressure value is present during the leak rate detection process, the system parameter B of the control matrix u (k) is 0.
The error covariance P (k | k-1) at the current k time is predicted from the covariance P (k-1) at the k-1 time in the above formula (2), the initial value P (0 | 0) is 1, S is the pressure fluctuation noise at the time of system measurement, and the pressure fluctuation noise S in the present embodiment may be set to 0.5.
Determining a Kalman gain corresponding to the pressure value acquired at the later moment based on the error covariance predicted value corresponding to the pressure value acquired at the later moment; specifically, as shown in fig. 3, the kalman gain Kg (k) at time k is calculated from the error covariance predicted value P (k | k-1) at time k.
The predicted value of the covariance of the k-time error is input to the following equation (3), and the kalman gain at the k-time is calculated.
And determining an optimal pressure estimation value corresponding to the pressure value acquired at the later moment based on the pressure value acquired at the later moment, the corresponding pressure prediction value and the Kalman gain, and determining a covariance corresponding to the pressure value acquired at the later moment based on the Kalman gain and the error covariance prediction value corresponding to the pressure value acquired at the later moment until the optimal pressure estimation value corresponding to all the pressure values in the pressure data set is obtained. Specifically, the optimal pressure estimation value X (k | k) at the time k is calculated according to the pressure prediction value X (k | k-1) at the time k, the pressure value Z (k) acquired at the time k (i.e., the kth data in the pressure data set), and the kalman gain Kg (k) at the time k. The covariance P (k | k) at the time k is calculated from the Kalman gain Kg (k) at the time k and the error covariance predicted value P (k | k-1) at the time k.
In the above formula (3), kg (k) is the kalman gain at the time k, R is the noise average value (the value may be 0.5) of the pressure sensor, H is the measurement matrix, and H' is the transpose. And acquiring a pressure value Z (k) measured at the current k moment, and inputting a predicted pressure value X (k | k-1), the pressure value Z (k) and a Kalman gain Kg (k) at the k moment into the following formula (4) to calculate an optimal pressure estimation value X (k | k) at the k moment.
X(k|k)＝X(k|k-1)+Kg(k)(Z(k)-HX(k|k-1)) (4)
The kalman gain Kg (k) at the time k and the error covariance prediction value O (k | k-1) are input to the following equation (5) to calculate the covariance P (k | k) at the time k for subsequent iterative calculations.
P(k|k)＝(1-Kg(k)H)P(k|k-1) (5)
After the optimal pressure estimation value at the k moment and the covariance at the k moment are obtained through calculation, the optimal pressure estimation value at the k +1 moment and the covariance at the k +1 moment are continuously calculated according to the optimal pressure estimation value at the k moment and the covariance at the k moment, so that the algorithm can perform autoregressive operation until the optimal pressure estimation values corresponding to all pressure values in the pressure data set are obtained, and filtering of all pressure values in the pressure data set is completed.
In an embodiment, in order to improve the accuracy of leak rate calculation, the present embodiment provides a specific implementation manner of performing curve fitting on the pressure filtering data to obtain a fitting function: and establishing a regression model corresponding to the pressure filtering data based on a least square method, and solving a fitting function matched with the pressure filtering data based on the regression model. And the least square method is adopted for curve fitting, the optimal function matching of the pressure filtering data can be calculated through the square of the data minimization error, and the accuracy of leakage rate calculation is improved.
In a specific embodiment, since the pressure data continuously increases during the leak detection process of the PECVD apparatus and meets a linear condition, the fitting function may be set as a linear equation, and the horizontal and vertical coordinates of the fitting function are the data arrangement number and the pressure value corresponding to the data arrangement number, respectively.
Let the equation of the fitting function be the following equation (6):
y＝K·x+b (6)
the slope of the fitted curve is
After the slope is calculated, the value can be based on>
Determining an intercept b, wherein>
And &>
The mean of the data set x and the mean of the data set y, respectively, and n is the number of data sets (n =240 when the frequency of acquisition of pressure values is 4 Hz). The x-coordinate represents the number of acquired pressure values and the y-coordinate represents the corresponding real-time pressure value.
Determining a straight line by 'residual square sum minimum' based on the principle of a least square method, and firstly determining a regression model of pressure data, namely a functional relation between a pressure value and a sampling point:
wherein the variable x of the function Q i Denotes the ith point, y i And (3) expressing the pressure value of the ith point, wherein the pressure value can be determined by a regression model, the current main solving objects are K and b, and the following formula can be obtained by differentiating the regression model:
the optimal parameters K and b corresponding to the function Q taking the minimum value can be solved through the above formula (8) and formula (9):
according to the formula, a fitting function of the pressure value in the leakage rate detection process can be obtained, and the actual leakage rate value can be calculated through the fitting function.
The related leakage rate detection technology generally takes a pressure change value at the starting time and after 1min as a leakage rate detection value, and gives an alarm when the leakage rate detection value is larger than a certain threshold value, but the related leakage rate detection technology is easy to cause fluctuation of the leakage rate detection value due to abnormal detection value of a sensor, the leakage rate detection accuracy is low, and the problem of false alarm is easy to occur in the leakage rate detection process.
In one embodiment, in order to improve the leak rate detection accuracy and avoid the problem of false alarm, the present embodiment provides two embodiments of determining the leak rate of the reaction chamber based on the fitting function and the preset frequency, which can be specifically performed with reference to the following embodiment one or embodiment two:
the first implementation mode comprises the following steps: if the fitting function is a linear equation, acquiring the slope of the fitting function; determining the pressure value collection quantity per minute based on a preset frequency; and calculating the product of the slope of the fitting function and the pressure value collection quantity per minute to obtain the leakage rate of the reaction chamber.
The preset frequency is the pressure value collection number per second, the pressure value collection number per minute = the preset frequency 60, when the preset frequency is 4Hz, the pressure value collection number per minute is 240, and the leakage rate of the reaction chamber = the slope K.
The second embodiment: acquiring coordinates of any two points from the fitting function, and calculating a pressure difference value and a data sequence number difference value corresponding to the coordinates of any two points; determining the acquisition time difference corresponding to the coordinates of any two points based on the data sequence number difference and the preset frequency; and determining the leakage rate of the reaction chamber based on the pressure difference value and the acquisition time difference.
Assuming that coordinates of any two points obtained are (x 1, y 1) and (x 2, y 2), a data sequence number difference = x2-x1, and a pressure difference = y2-y1, an acquisition time interval between x2 and x1 may be determined according to the data sequence number difference x2-x1 and a sampling frequency (i.e., a preset frequency) of the pressure values, such as, when the preset frequency is 4Hz, acquiring 4 pressure values per second, and if x2-x1=40, indicating that an acquisition time difference corresponding to the coordinates of the two points is 10s, a leakage rate = (y 2-y 1) × (60 s/10 s) of the reaction chamber, i.e., a leakage rate = [ (y 2-y 1)/(x 2-x 1) ] -a preset frequency = 60.
According to the leak rate detection method provided by the embodiment, the pressure increase value per minute is determined through the fitting function based on the pressure value, the leak rate calculation fluctuation caused by continuous oscillation of the collected pressure value is avoided, the leak rate obtained by calculation of the same container every time is relatively stable, the problem of false alarm caused by easy fluctuation of the calculated leak rate is solved, and the stable operation of equipment and the production yield of customers are ensured; meanwhile, the accuracy of each leakage rate detection is ensured, help is provided for prejudging and maintaining the operation condition of the field equipment, and the cost of unnecessary personnel is reduced.
For example, this embodiment provides an example of calculating the leak rate of the pressure value detected in the furnace tube of the PECVD apparatus in fig. 1 by using the leak rate detection method described above:
the pressure data set shown in fig. 1 is subjected to filtering processing based on a kalman filtering algorithm to obtain pressure filtering data, see a comparison graph before and after filtering of the pressure data set shown in fig. 4, a dotted line in fig. 4 shows a pressure value trend before filtering, a solid line in fig. 4 shows pressure filtering data after kalman filtering, and as can be seen from fig. 4, the filtered pressure data is more concentrated, and fluctuation deviation of the pressure value is repaired.
The pressure filtering data in fig. 4 is subjected to curve fitting based on a least square method to obtain a fitting function of y =0.0404 × x +0.5728, see the pressure value fitting curve diagram shown in fig. 5, the dotted line in fig. 5 shows the pressure filtering data, the solid line in fig. 5 shows the fitting function curve obtained by the pressure filtering data through the least square method, since the collection frequency of the pressure values is 4hz, 240 pressure values can be collected within 1 minute, and the leak rate =0.0404 × 240= 9.6966 pa/min can be calculated according to the slope of the fitting function.
In order to verify the stability of the leak rate detection in the above embodiment, the present embodiment provides a comparison experiment for performing multiple leak rate detections on a furnace tube of a PECVD apparatus by using a related leak rate (denoted as an original leak rate) detection method and the leak rate (denoted as a test leak rate) detection method provided in the above embodiment, see the comparison diagram of the leak rate detection experiment shown in fig. 6, where a dotted line in fig. 6 shows the results of performing 13 leak rate detections on the furnace tube by using the existing leak rate (original leak rate) detection technique, a solid line in fig. 6 shows the results of performing 13 leak rate detections on the furnace tube by using the leak rate (test leak rate) detection method provided in the above embodiment, through experimental comparison, the leakage rate value obtained by the existing leakage rate (original leakage rate) detection technology fluctuates between 5.19 and 12.37, and the leakage rate value obtained by the leakage rate (test leakage rate) detection method provided by the application can be stabilized between 10.03 and 10.92, so that the leakage rate detection method provided by the application can effectively solve the problem that the detection of the leakage rate value of the equipment is fluctuated due to abnormal detection values of the sensors, the problem that the software calculation leakage rate value is larger than the actual deviation due to abnormal data sampling in the alarm detection process is avoided, false alarms are generated by the equipment, and more accurate data support is provided for equipment stability prediction and field maintenance.
On the basis of the foregoing embodiments, the present embodiment provides an example of performing leak rate calculation on a furnace tube of a PECVD apparatus by applying the leak rate detection method, which may be specifically performed by referring to the following steps:
The furnace tube pressure control of the PECVD equipment mainly comprises two parts: the pressure sensor transmits the real-time pressure value in the furnace into the butterfly valve controller through 0-10A current. Referring to the pressure value acquisition flow chart shown in fig. 7, control software of the lower computer sends a control command to the butterfly valve controller by using a sampling frequency of a serial port communication pressure value, the butterfly valve controller samples a pressure value from real-time pressure values sent by the pressure sensor based on the sampling frequency, and sends the pressure value obtained based on the sampling frequency to the control software, so that the control software obtains a pressure data set under the sampling frequency, and the control software can be installed in electronic equipment such as a computer.
Setting the sampling frequency of the pressure value in the control software, optimizing the serial port command response time, modifying the serial port attribute of the software pressure value reading thread into the blocking property, and quickly sending a second pressure reading command after one command is sent and received. The control software can also schedule serial port resources during leak rate detection, in a normal production process, a pressure thread needs to read commands such as butterfly valve angles, pressure, actual set values and the like, the serial port resources are only used for reading real-time pressure values when the leak rate is detected through resource scheduling, and other commands do not respond, so that the serial port response speed is increased.
The response time of serial ports sending commands can be optimized through the sampling frequency of the pressure values, the serial ports are scheduled to use resources, the serial ports are only used for continuously and rapidly reading the pressure values in the leak rate detection process, 4 pressure values can be collected every second when the leak detection process steps are carried out, and the data collection speed is greatly improved.
And 2, filtering the pressure data set based on a Kalman filtering algorithm to remove noise in the data and restore the real data.
And 3, performing curve fitting on the filtered pressure data set based on a least square method to obtain a linear equation, determining the pressure value collection number of 1 minute based on the sampling frequency of the pressure values, and calculating the product of the slope of the linear equation and the pressure value collection number of 1 minute to obtain the current leakage rate of the furnace tube.
Corresponding to the method provided by the foregoing embodiment, an embodiment of the present invention further provides a semiconductor processing apparatus, including a reaction chamber and a control unit, where the control unit is configured to execute the leak rate detection method provided by the foregoing embodiment to detect a leak rate of the reaction chamber in the semiconductor processing apparatus.
The reaction chamber can be a furnace tube of PECVD equipment, and the leak rate of the furnace tube can be detected based on the control unit before plasma deposition so as to detect whether the leak rate of the furnace tube meets the requirement before the process.
In one embodiment, the control unit is in serial port communication with the lower computer, so that the lower computer collects the pressure value of the reaction chamber at a preset frequency; the serial port attribute can be set to be blocking.
The control unit can be an upper computer provided with control software, the lower computer can comprise a butterfly valve controller and a pressure sensor, the butterfly valve controller is respectively in communication connection with the pressure sensor and a butterfly valve, the pressure sensor is used for detecting the pressure in the furnace, and the real-time pressure value in the furnace is transmitted into the butterfly valve controller in a current mode. In the normal production process, the control unit executes commands such as reading a butterfly valve angle, a pressure value and an actual set value (such as a pressure set value or a butterfly valve angle set value) from the butterfly valve controller through the serial port, when the leakage rate detection method is executed, the serial port is only used for reading a real-time pressure value, other commands do not respond, concretely, the control unit sends a pressure reading command to the butterfly valve controller through the serial port and receives the pressure value, and sends a second pressure reading command after one command is sent and received, so that a large amount of pressure value data can be collected quickly, and reliable data support is provided for leakage rate detection.
Embodiments of the present invention provide a computer-readable medium, wherein the computer-readable medium stores computer-executable instructions, which, when invoked and executed by a processor, cause the processor to implement the method of the above-described embodiments.
It can be clearly understood by those skilled in the art that, for convenience and simplicity of description, the specific working process of the system described above may refer to the corresponding process in the foregoing embodiment, and details are not described herein again.
The leak rate detection method and the computer program product of the semiconductor process equipment provided by the embodiment of the invention include a computer readable storage medium storing program codes, instructions included in the program codes can be used for executing the method described in the foregoing method embodiment, and specific implementation can refer to the method embodiment, and details are not described herein.
In addition, in the description of the embodiments of the present invention, unless otherwise explicitly specified or limited, the terms "mounted," "connected," and "connected" are to be construed broadly, e.g., as meaning either a fixed connection, a removable connection, or an integral connection; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood in specific cases to those skilled in the art.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, and are only for convenience of description and simplicity of description, but do not indicate or imply that the device or element being referred to must have a particular orientation, be constructed and operated in a particular orientation, and thus, should not be construed as limiting the present invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
Finally, it should be noted that: the above-mentioned embodiments are only specific embodiments of the present invention, which are used for illustrating the technical solutions of the present invention and not for limiting the same, and the protection scope of the present invention is not limited thereto, although the present invention is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: any person skilled in the art can modify or easily conceive the technical solutions described in the foregoing embodiments or equivalent substitutes for some technical features within the technical scope of the present disclosure; such modifications, changes or substitutions do not depart from the spirit and scope of the embodiments of the present invention, and they should be construed as being included therein. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. A leak rate detection method for detecting the leak rate of a reaction chamber in semiconductor processing equipment is characterized by comprising the following steps:
acquiring a pressure data set of the reaction chamber in a closed state within a preset time; the pressure data set is a pressure value set acquired under a preset frequency;
filtering the pressure data set to obtain pressure filtering data;
performing curve fitting on the pressure filtering data to obtain a fitting function;
determining a pressure increase per unit time based on the fit function, and determining a leak rate of the reaction chamber based on the pressure increase per unit time.
2. The method of claim 1, wherein the step of filtering the pressure data set to obtain pressure filtered data comprises:
and sequentially determining the optimal pressure estimation value corresponding to each pressure value in the pressure data set based on a Kalman filtering algorithm, and taking the optimal pressure estimation value corresponding to each pressure value in the pressure data set as the pressure filtering data so as to optimize and remove pressure fluctuation noise data.
3. The method of claim 2, wherein the step of sequentially determining an optimal pressure estimation value corresponding to each pressure value in the pressure data set based on a kalman filter algorithm comprises:
sequentially acquiring each pressure value in the pressure data set based on the sequence of the pressure value acquisition time, predicting a pressure predicted value corresponding to a pressure value acquired at the later time based on an optimal pressure estimated value corresponding to a pressure value acquired at the previous time, and predicting an error covariance predicted value corresponding to a pressure value acquired at the later time based on a covariance corresponding to a pressure value acquired at the previous time;
determining a Kalman gain corresponding to the pressure value acquired at the later moment based on the error covariance predicted value corresponding to the pressure value acquired at the later moment;
and determining an optimal pressure estimation value corresponding to the pressure value acquired at the later moment based on the pressure value acquired at the later moment, the corresponding pressure prediction value and the Kalman gain, and determining a covariance corresponding to the pressure value acquired at the later moment based on the Kalman gain corresponding to the pressure value acquired at the later moment and the error covariance prediction value until the optimal pressure estimation value corresponding to all the pressure values in the pressure data set is obtained.
4. The method of claim 1, wherein said step of curve fitting said pressure filtered data to obtain a fitting function comprises:
and establishing a regression model corresponding to the pressure filtering data based on a least square method, and solving a fitting function matched with the pressure filtering data based on the regression model.
5. The method according to any one of claims 1 to 4, wherein the fitting function is a linear equation, and the abscissa and ordinate of the fitting function are the data rank number and the pressure value corresponding to the data rank number, respectively.
6. The method of claim 5, wherein the step of determining the leak rate of the reaction chamber based on the fitting function and the predetermined frequency comprises:
obtaining the slope of the fitting function;
determining the pressure value collection quantity per minute based on the preset frequency;
and calculating the product of the slope of the fitting function and the pressure value collection quantity per minute to obtain the leakage rate of the reaction chamber.
7. The method of claim 1, wherein the step of determining the leak rate of the reaction chamber based on the fitted function and the predetermined frequency comprises:
acquiring coordinates of any two points from the fitting function, and calculating a pressure difference value and a data sorting number difference value corresponding to the coordinates of any two points;
determining the acquisition time difference corresponding to the coordinates of any two points based on the data sequence number difference and the preset frequency;
and determining the leakage rate of the reaction chamber based on the pressure difference value and the acquisition time difference.
8. The method according to any one of claims 1 to 4, wherein the predetermined frequency is 4Hz.
9. A semiconductor processing apparatus comprising a reaction chamber and a control unit for performing the method of any one of claims 1 to 8 to detect a leak rate of the reaction chamber.
10. The semiconductor processing equipment according to claim 9, wherein the control unit communicates with a lower computer through a serial port to enable the lower computer to acquire the pressure value of the reaction chamber at a preset frequency; and the serial port attribute is blocking.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211438988.0A CN115979533A (en) | 2022-11-17 | 2022-11-17 | Leak rate detection method and semiconductor process equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211438988.0A CN115979533A (en) | 2022-11-17 | 2022-11-17 | Leak rate detection method and semiconductor process equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115979533A true CN115979533A (en) | 2023-04-18 |
Family
ID=85963641
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211438988.0A Pending CN115979533A (en) | 2022-11-17 | 2022-11-17 | Leak rate detection method and semiconductor process equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115979533A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116335925A (en) * | 2023-05-19 | 2023-06-27 | 山东海纳智能装备科技股份有限公司 | Data enhancement-based intelligent regulation and control system for underground coal mine emulsification pump station |
-
2022
- 2022-11-17 CN CN202211438988.0A patent/CN115979533A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116335925A (en) * | 2023-05-19 | 2023-06-27 | 山东海纳智能装备科技股份有限公司 | Data enhancement-based intelligent regulation and control system for underground coal mine emulsification pump station |
| CN116335925B (en) * | 2023-05-19 | 2023-08-04 | 山东海纳智能装备科技股份有限公司 | Data enhancement-based intelligent regulation and control system for underground coal mine emulsification pump station |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7058470B2 (en) | Method of monitoring and/or controlling a semiconductor manufacturing apparatus and a system therefor | |
| Jiang et al. | Weighted kernel principal component analysis based on probability density estimation and moving window and its application in nonlinear chemical process monitoring | |
| Zhang et al. | Fault detection strategy based on weighted distance of $ k $ nearest neighbors for semiconductor manufacturing processes | |
| CN115979533A (en) | Leak rate detection method and semiconductor process equipment | |
| KR102033438B1 (en) | Plasma processing apparatus and plasma processing apparatus state prediction method | |
| CN110928262B (en) | Quality-related fault online monitoring method for efficiently updating model in time-varying system | |
| CN110112442B (en) | Fuel cell system control method and device | |
| US20210397169A1 (en) | Information processing apparatus and monitoring method | |
| CN115931055A (en) | Rural water supply operation diagnosis method and system based on big data analysis | |
| EP2347311A1 (en) | Robust filtering and prediction using switching models for machine condition monitoring | |
| CN105675320B (en) | A kind of mechanical system running status method for real-time monitoring based on acoustic signal analysis | |
| CN116086537A (en) | Equipment state monitoring method, device, equipment and storage medium | |
| CN113589085B (en) | Detection and mitigation method for false data injection attack of generator frequency modulation system | |
| EP3349084B1 (en) | Process monitoring device, method and program | |
| CN105718733B (en) | Fault prediction method based on fuzzy nearness and particle filter | |
| CN116878590B (en) | Intelligent service collaborative operation management system | |
| CN116757553B (en) | Agricultural organic waste resource utilization digital information management system | |
| CN116448263B (en) | Method for detecting running state of boehmite production equipment | |
| CN111471984A (en) | Control method and control system for film coating rate and storage medium | |
| Thieullen et al. | Application of Principal Components Analysis to improve fault detection and diagnosis on semiconductor manufacturing equipment | |
| US9666417B2 (en) | Plasma processing apparatus and method for monitoring plasma processing apparatus | |
| CN113445992B (en) | Method and device for processing movement displacement of oil pumping unit | |
| CN112234601A (en) | Method and system for identifying low-frequency oscillation characteristic parameters of power system on line | |
| Geng et al. | Fault detection of large-scale process control system with higher-order statistical and interpretative structural model | |
| JP2002099328A (en) | Monitoring device, processing equipment, monitoring method, data collecting method and semiconductor device production method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |