CN117828275A - Prediction method and device for plasma chamber alignment - Google Patents
Prediction method and device for plasma chamber alignment Download PDFInfo
- Publication number
- CN117828275A CN117828275A CN202410032425.4A CN202410032425A CN117828275A CN 117828275 A CN117828275 A CN 117828275A CN 202410032425 A CN202410032425 A CN 202410032425A CN 117828275 A CN117828275 A CN 117828275A
- Authority
- CN
- China
- Prior art keywords
- probability
- information
- power supply
- radio frequency
- cavity
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 35
- 238000009826 distribution Methods 0.000 claims abstract description 15
- 238000001514 detection method Methods 0.000 claims abstract description 12
- 238000007781 pre-processing Methods 0.000 claims abstract description 9
- 238000007499 fusion processing Methods 0.000 claims abstract description 7
- 238000004422 calculation algorithm Methods 0.000 claims description 11
- 238000005314 correlation function Methods 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000012545 processing Methods 0.000 claims description 4
- 230000009471 action Effects 0.000 claims description 3
- 230000004927 fusion Effects 0.000 claims description 3
- 210000002381 plasma Anatomy 0.000 abstract description 31
- 239000007789 gas Substances 0.000 description 28
- XKRFYHLGVUSROY-UHFFFAOYSA-N Argon Chemical compound [Ar] XKRFYHLGVUSROY-UHFFFAOYSA-N 0.000 description 4
- 239000004065 semiconductor Substances 0.000 description 4
- 238000004140 cleaning Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 229910052786 argon Inorganic materials 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 150000001875 compounds Chemical class 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000000151 deposition Methods 0.000 description 1
- 230000008021 deposition Effects 0.000 description 1
- 238000005530 etching Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000000623 plasma-assisted chemical vapour deposition Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
Landscapes
- Plasma Technology (AREA)
Abstract
The invention discloses a method and a device for predicting a plasma chamber in the field of plasmas. The prediction method comprises the following steps: s1, acquiring state information of a radio frequency power supply, power information of a cavity and gas information in the cavity; s2, preprocessing the state information, the power information and the gas information to obtain a parameter data set; s3, carrying out fusion processing on parameter data in the parameter data set, constructing a Bayesian network, acquiring probability distribution of each parameter, verifying probability of detection of the correlation, and acquiring a probability model; and S4, detecting various data through a probability model, and controlling the radio frequency power supply when the probability of the probability generation is greater than a set value. The prediction method improves the probability of detecting the arong, and based on the mode of the arong, the arong can be perceived in advance, so that the condition of the arong is blocked, the probability of the arong is greatly reduced, and the safety of equipment is ensured.
Description
Technical Field
The invention relates to the technical field of plasmas, in particular to a method and a device for predicting a plasma chamber.
Background
The generation and application of radio frequency plasmas is very widespread in semiconductor devices, such as PVD and PECVD for material deposition using plasmas in semiconductor fabrication, cleaning using plasmas, etching, and the like. In a chamber for plasma generation and application of the apparatus, an Arc phenomenon may occur, and Arc is usually generated and terminated in a short process of several microseconds to several tens of microseconds, which is represented by a discharge phenomenon between the plasma and the electrode, and between the plasma and the chamber sidewall. Arcing can cause instability of the plasma, damage semiconductor equipment and fittings that make the plasma, and produce unwanted compounds that contaminate the semiconductor product. Therefore, the damage caused by the arc is avoided to the greatest extent in the plasma chamber, and the arc is observed and captured.
After the arc is generated, electrons are released, which macroscopically appear as a decrease in voltage and current amplitude. The conventional method is to detect the change of the voltage and current. However, this method is prone to two problems, namely, the sampling rate is often not fast enough relative to the time of occurrence of the ranging signal, so that the ranging signal is lost, and the signal to noise ratio of the test parameter is not high enough, so that the real ranging signal is not easily identified, and the monitoring of the ranging signal fails. In addition, in the process of detecting the occurrence of arc, the chamber has normal signal transient and signal instability, which all cause false alarm of detected arc.
Disclosure of Invention
The method and the device for predicting the arc of the plasma chamber solve the problem that the arc omission occurs easily in the traditional method in the prior art, and improve the probability that the arc phenomenon is detected.
The embodiment of the application provides a prediction method of plasma chamber alignment, which comprises the following steps:
s1, acquiring state information of a radio frequency power supply, power information of a cavity and gas information in the cavity;
s2, preprocessing the state information, the power information and the gas information to obtain a parameter data set;
s3, carrying out fusion processing on parameter data in the parameter data set, constructing a Bayesian network, acquiring probability distribution of each parameter, verifying probability of detection of the correlation, and acquiring a probability model;
and S4, detecting various data through a probability model, and controlling the radio frequency power supply when the probability of the probability generation is greater than a set value.
The beneficial effects of the above embodiment are that: the prediction method improves the probability of detecting the arong, and based on the mode of the arong, the arong can be perceived in advance, so that the condition of the arong is blocked, the probability of the arong is greatly reduced, and the safety of equipment is ensured.
Based on the above embodiments, the present application may be further improved, specifically as follows:
in one embodiment of the present application, the status information includes a start-up S of the radio frequency power supply, an amplitude a of a signal, a type Ty, a duration Tc, and a current power supply temperature Te information. The status information is read directly from the radio frequency power supply.
In one embodiment of the present application, the power information is amplitude and phase information of the voltage and current or forward power and reverse power information, the power information is obtained by a sensor disposed between the matching network and the cavity, the sensor is a VI sensor or a directional coupler sensor, the VI sensor is used to obtain amplitude and phase information of the voltage and current of the cavity, and the directional coupler sensor is used to obtain forward power Pf and reflected power Pr information. The radio frequency power supply acquires information such as whether power is mismatched or not through a sensor connected between the matcher and the cavity, wherein the sensor is of two types, namely a VI sensor, acquires the state of the cavity by detecting current and voltage information, monitors the generation of the ranging, and a directional coupler sensor, and acquires information by detecting forward power Pf and reflected power Pr.
In one embodiment of the present application, the gas information includes gas temperature Ta and flow Fv information, and the gas information is acquired by a gas detection sensor disposed at a gas outlet of the cavity. The plasma chamber is provided with a pump for injecting a plasma-generating or cleaning gas, such as CF4, argon, etc., into the chamber, and a gas detection sensor, which may be a mass spectrometer, or a temperature, flow rate detector, etc., is connected in series to the gas outlet, the mass spectrometer being adapted to detect changes in the composition of the gas, and the temperature sensor and flow rate sensor being adapted to detect changes in the gas environment in the plasma chamber.
In one embodiment of the present application, the step S3 specifically includes:
s3.1, obtaining a correlation function among parameters, and establishing a connection edge among parameters with calculation results of the correlation function larger than a set value to obtain a directed acyclic graph of the Bayesian network so as to complete a learning structure;
and S3.2, performing parameter learning through a Bayesian estimation algorithm according to the parameter data set and the learning structure to obtain a probability distribution table among all nodes in the Bayesian network, and verifying the probability of detecting the probability according to the probability distribution table to obtain a probability model.
In one embodiment of the present application, the controlling the radio frequency power supply includes: and turning off the radio frequency power supply or setting the radio frequency power supply as a pulse signal with larger duty ratio. Both of these can reduce the magnitude of the rf power output, preventing or interrupting the plasma generation.
The embodiment of the application also provides a prediction device for plasma chamber alignment, which comprises:
the data sensing module is used for acquiring state information of the radio frequency power supply, power information of the cavity and gas information in the cavity;
the data processing module is used for preprocessing the state information, the power information and the gas information to obtain a parameter data set;
the data fusion module is used for carrying out fusion processing on parameter data in the parameter data set, constructing a Bayesian network, acquiring probability distribution of each parameter, verifying the probability of detecting the probability, and acquiring a probability model;
and the action execution module is used for detecting various data through a probability model, and controlling the radio frequency power supply when the probability of the generation of the ranging is greater than a set value.
According to the prediction method and the prediction device for the arc of the plasma chamber, the probability of arc detection can be improved, the generation of arc can be perceived in advance based on the mode of arc generation, so that the generation condition is blocked, the probability of arc generation is greatly reduced, and the safety of equipment is guaranteed.
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. Like elements or portions are generally identified by like reference numerals throughout the several figures. In the drawings, elements or portions thereof are not necessarily drawn to scale.
FIG. 1 is a flow chart showing the steps of a method for predicting plasma chamber alignment in an embodiment;
FIG. 2 is a schematic diagram of a connection structure between a plasma chamber and a power supply in an embodiment;
FIG. 3 is a network diagram of a Bayesian network in an embodiment;
fig. 4 is a block diagram of a prediction apparatus for plasma chamber alignment in an embodiment.
Detailed Description
The present invention is further illustrated below in conjunction with the specific embodiments, it being understood that these embodiments are meant to be illustrative of the invention only and not limiting the scope of the invention, and that modifications of the invention, which are equivalent to those skilled in the art to which the invention pertains, will fall within the scope of the invention as defined in the claims appended hereto.
According to the method and the device for predicting the arc of the plasma chamber, the problem that the arc is easy to miss in the traditional method in the prior art is solved, and the probability that the arc phenomenon is detected is improved.
The technical scheme in the embodiment of the application aims to solve the problems, and the overall thought is as follows:
example 1:
as shown in fig. 1, a method for predicting a plasma chamber, comprising the steps of:
s1, acquiring state information of a radio frequency power supply, power information of a cavity and gas information in the cavity. The method comprises the following steps:
the basic structure of the plasma chamber and the power supply connected together is schematically shown in fig. 2. The state information comprises the information of the starting S of the radio frequency power supply, the amplitude A of the signal, the type Ty, the duration Tc and the current power supply temperature Te. The status information is read directly from the radio frequency power supply. The power information is amplitude and phase information of the voltage and current or forward power and reverse power information, the power information is obtained through a sensor arranged between the matching network and the cavity, the sensor is a VI sensor or a directional coupler sensor, the VI sensor is used for obtaining the amplitude and phase information of the voltage and current of the cavity, and the directional coupler sensor is used for obtaining forward power Pf and reflected power Pr information. The radio frequency power supply acquires information such as whether power is mismatched or not through a sensor connected between the matcher and the cavity, wherein the sensor is of two types, namely a VI sensor, acquires the state of the cavity by detecting current and voltage information, monitors the generation of the ranging, and a directional coupler sensor, and acquires information by detecting forward power Pf and reflected power Pr. The gas information comprises gas temperature Ta and flow Fv information, and the gas information is acquired through a gas detection sensor arranged at a gas outlet of the cavity. The plasma chamber is provided with a pump for injecting a plasma-generating or cleaning gas, such as CF4, argon, etc., into the chamber, and a gas detection sensor, which may be a mass spectrometer, or a temperature, flow rate detector, etc., is connected in series to the gas outlet, the mass spectrometer being adapted to detect changes in the composition of the gas, and the temperature sensor and flow rate sensor being adapted to detect changes in the gas environment in the plasma chamber.
S2, preprocessing state information, power information and gas information to obtain a parameter data set. The method comprises the following steps:
the preprocessing mainly comprises the steps of removing abnormal values and blank values from data acquired from three places, filling missing values, and performing parameterization on the acquired data, so that modeling processing in the later step is facilitated, and event parameters are increased. Assuming that the occurrence of the ranging is defined as event Ac, a parametric dataset about ranging is obtained after preprocessing, and the parametric dataset has { S, a, ty, tc, te, pf, pr, ta, fv, ac }, where the forward power Pf and the reverse power Pr information are used for the subsequent power information.
And S3, carrying out fusion processing on parameter data in the parameter data set, constructing a Bayesian network, acquiring probability distribution of each parameter, verifying the probability of detecting the correlation, and acquiring a probability model. The method comprises the following steps:
s3.1, obtaining a correlation function among parameters, and establishing a connection edge among parameters with calculation results of the correlation function larger than a set value to obtain a directed acyclic graph of the Bayesian network so as to complete a learning structure;
the data are fused, i.e. a relation between parameters is established. The Bayesian network is utilized to mainly complete two parts, the first step is to complete structure learning, namely, a directed acyclic graph of the Bayesian network is obtained, each node in the graph is an event that each parameter in a data set reaches a certain threshold value, and an algorithm used is a method relying on statistical analysis. The specific process is to count all events, locate 1 when the event occurs more than a threshold value and set 0 when the event does not occur more than the threshold value, so as to obtain a sequence of events, and then calculate a correlation function between any two parameters X1 and X2:
;
wherein the method comprises the steps ofRepresenting covariance +_>And->Representing the variance;
and analyzing the dependence among the parameters according to the calculation result of the correlation function, and establishing a connecting edge for the parameters with large dependence, so that an undirected graph is gradually obtained, and the structure learning is completed.
And S3.2, performing parameter learning through a Bayesian estimation algorithm according to the parameter data set and the learning structure to obtain a probability distribution table among all nodes in the Bayesian network, and verifying the probability of detecting the probability according to the probability distribution table to obtain a probability model.
And the second step is to perform parameter learning after the result of the data set and the structure is obtained, namely obtaining a probability distribution table among all nodes in the network. The algorithm mainly used here is a bayesian estimation algorithm. Assuming that one of the parameters is θ and the sample set is D, the posterior probability of that parameter:
;
according to the algorithm, a probability distribution table of each parameter event occurrence in parameter sets { S, A, ty, tc, te, pf, pr, ta, fv and Ac } can be obtained. A network schematic can be obtained using a bayesian network as shown in fig. 3. The established Bayesian network can be used for solving the probability of parameters in various states, namely the joint probability of parameters in a data set, wherein the value of the joint probability is the product of the conditional probabilities of the parameters:
;
the probability of occurrence of the bearing in various states can be analyzed, and an analyzed model is obtained.
And S4, detecting various data through a probability model, and controlling the radio frequency power supply when the probability of probability generation is greater than a set value.
Controlling the radio frequency power supply, comprising: the radio frequency power supply is turned off or set to a pulse signal with a larger duty cycle. Both of these can reduce the magnitude of the rf power output, preventing or interrupting the plasma generation.
The method can continuously count the probability of detecting the rating in the application period of the method, and judge whether to correct the model according to the expected requirement. Thus, the steps are also included:
and S5, counting the probability of detection of the ranging in the period, if the probability meets the requirement, continuously applying the probability model, otherwise, returning to the step S3, modifying model parameters, and optimizing structure learning and parameter learning algorithms.
Modifying model parameters refers to modifying probability of the probability table according to experimental results, and optimizing algorithm refers to adopting different structure learning and parameter learning algorithms.
The beneficial effects of the above embodiment are that: the prediction method improves the probability of detecting the arong, and based on the mode of the arong, the arong can be perceived in advance, so that the condition of the arong is blocked, the probability of the arong is greatly reduced, and the safety of equipment is ensured.
Example 2:
as shown in fig. 4, a prediction apparatus for plasma chamber ranging includes:
the data sensing module is used for acquiring state information of the radio frequency power supply, power information of the cavity and gas information in the cavity;
the data processing module is used for preprocessing the state information, the power information and the gas information to obtain a parameter data set;
the data fusion module is used for carrying out fusion processing on parameter data in the parameter data set, constructing a Bayesian network, acquiring probability distribution of each parameter, verifying the probability of detecting the probability, and acquiring a probability model;
and the action execution module is used for detecting various data through a probability model, and controlling the radio frequency power supply when the probability of the generation of the ranging is greater than a set value.
And the monitoring module is used for counting the probability of detection of the ranging in the period, continuously applying the probability model if the probability meets the requirement, otherwise returning to the step S3 and modifying model parameters, and optimizing structure learning and parameter learning algorithms.
The functional roles of the above modules are consistent with those in the corresponding method embodiments, and are not described herein.
While embodiments of the present invention have been shown and described above, it will be understood that the above embodiments are illustrative and not to be construed as limiting the invention, and that variations, modifications, alternatives and variations may be made to the above embodiments by one of ordinary skill in the art within the scope of the invention.
Claims (8)
1. A method for predicting plasma chamber alignment, comprising the steps of:
s1, acquiring state information of a radio frequency power supply, power information of a cavity and gas information in the cavity;
s2, preprocessing the state information, the power information and the gas information to obtain a parameter data set;
s3, carrying out fusion processing on parameter data in the parameter data set, constructing a Bayesian network, acquiring probability distribution of each parameter, verifying probability of detection of the correlation, and acquiring a probability model;
and S4, detecting various data through a probability model, and controlling the radio frequency power supply when the probability of the probability generation is greater than a set value.
2. The prediction method according to claim 1, characterized in that: the state information comprises the information of the starting S of the radio frequency power supply, the amplitude A of a signal, the type Ty, the duration Tc and the current power supply temperature Te.
3. The prediction method according to claim 1, characterized in that: the power information is amplitude and phase information of voltage and current or forward power and reverse power information, the power information is obtained through a sensor arranged between a matching network and the cavity, the sensor is a VI sensor or a directional coupler sensor, the VI sensor is used for obtaining the amplitude and phase information of the voltage and current of the cavity, and the directional coupler sensor is used for obtaining forward power Pf and reflected power Pr information.
4. The prediction method according to claim 1, characterized in that: the gas information comprises gas temperature Ta and flow Fv information, and is obtained through a gas detection sensor arranged at a gas outlet of the cavity.
5. The prediction method according to claim 1, characterized in that: the step S3 specifically comprises the following steps:
s3.1, obtaining a correlation function among parameters, and establishing a connection edge among parameters with calculation results of the correlation function larger than a set value to obtain a directed acyclic graph of the Bayesian network so as to complete a learning structure;
and S3.2, performing parameter learning through a Bayesian estimation algorithm according to the parameter data set and the learning structure to obtain a probability distribution table among all nodes in the Bayesian network, and verifying the probability of detecting the probability according to the probability distribution table to obtain the probability model.
6. The prediction method according to claim 1, characterized in that: in the step S4, the controlling the radio frequency power supply includes: and turning off the radio frequency power supply or setting the radio frequency power supply as a pulse signal with larger duty ratio.
7. The prediction method according to claim 1, further comprising the step of:
and S5, counting the probability of detection of the ranging in the period, continuously applying the probability model if the requirement is met, and otherwise, returning to the step S3.
8. A plasma chamber coding prediction apparatus, comprising:
the data sensing module is used for acquiring state information of the radio frequency power supply, power information of the cavity and gas information in the cavity;
the data processing module is used for preprocessing the state information, the power information and the gas information to obtain a parameter data set;
the data fusion module is used for carrying out fusion processing on parameter data in the parameter data set, constructing a Bayesian network, acquiring probability distribution of each parameter, verifying the probability of detecting the probability, and acquiring a probability model;
and the action execution module is used for detecting various data through a probability model, and controlling the radio frequency power supply when the probability of the generation of the ranging is greater than a set value.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410032425.4A CN117828275B (en) | 2024-01-10 | Prediction method and device for plasma chamber arcing |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410032425.4A CN117828275B (en) | 2024-01-10 | Prediction method and device for plasma chamber arcing |
Publications (2)
Publication Number | Publication Date |
---|---|
CN117828275A true CN117828275A (en) | 2024-04-05 |
CN117828275B CN117828275B (en) | 2024-07-02 |
Family
ID=
Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100201370A1 (en) * | 2008-02-14 | 2010-08-12 | Mks Instruments, Inc. | Arc Detection |
CN101843178A (en) * | 2008-02-14 | 2010-09-22 | Mks仪器有限公司 | Application of wideband sampling for arc detection with a probabilistic model for quantitatively measuring arc events |
US20130221847A1 (en) * | 2012-02-28 | 2013-08-29 | New Power Plasma Co., Ltd. | Method and apparatus for detecting arc in plasma chamber |
US20160268108A1 (en) * | 2009-08-17 | 2016-09-15 | Verity Instruments, Inc. | Method and apparatus for the detection of arc events during the plasma processing of a wafer, surface of substrate |
CN110456234A (en) * | 2018-05-07 | 2019-11-15 | 珠海格力电器股份有限公司 | Detection method, the device and system of fault electric arc |
CN113762471A (en) * | 2021-08-24 | 2021-12-07 | 浙江中辰城市应急服务管理有限公司 | Phase space reconstruction parameter estimation method based on attention mechanism and Bayesian optimization |
CN114071849A (en) * | 2021-11-15 | 2022-02-18 | 上海无线电设备研究所 | Plasma generator for hypersonic target ablation diffusion object |
CN114446752A (en) * | 2020-11-04 | 2022-05-06 | 中微半导体设备(上海)股份有限公司 | Detection method and detection device for electric arc of plasma processing chamber |
US20220383165A1 (en) * | 2021-05-18 | 2022-12-01 | State Grid Henan Electric Power Company Electric Power Research Institute | Method for Early Warning Brandish of Transmission Wire Based on Improved Bayes-Adaboost Algorithm |
CN115906942A (en) * | 2022-11-28 | 2023-04-04 | 固德威技术股份有限公司 | Arc fault identification method, device, equipment and readable storage medium |
CN116819250A (en) * | 2023-06-27 | 2023-09-29 | 深圳供电局有限公司 | Method for detecting fault arc in photovoltaic system |
Patent Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100201370A1 (en) * | 2008-02-14 | 2010-08-12 | Mks Instruments, Inc. | Arc Detection |
CN101843178A (en) * | 2008-02-14 | 2010-09-22 | Mks仪器有限公司 | Application of wideband sampling for arc detection with a probabilistic model for quantitatively measuring arc events |
US20160268108A1 (en) * | 2009-08-17 | 2016-09-15 | Verity Instruments, Inc. | Method and apparatus for the detection of arc events during the plasma processing of a wafer, surface of substrate |
US20130221847A1 (en) * | 2012-02-28 | 2013-08-29 | New Power Plasma Co., Ltd. | Method and apparatus for detecting arc in plasma chamber |
CN110456234A (en) * | 2018-05-07 | 2019-11-15 | 珠海格力电器股份有限公司 | Detection method, the device and system of fault electric arc |
CN114446752A (en) * | 2020-11-04 | 2022-05-06 | 中微半导体设备(上海)股份有限公司 | Detection method and detection device for electric arc of plasma processing chamber |
US20220383165A1 (en) * | 2021-05-18 | 2022-12-01 | State Grid Henan Electric Power Company Electric Power Research Institute | Method for Early Warning Brandish of Transmission Wire Based on Improved Bayes-Adaboost Algorithm |
CN113762471A (en) * | 2021-08-24 | 2021-12-07 | 浙江中辰城市应急服务管理有限公司 | Phase space reconstruction parameter estimation method based on attention mechanism and Bayesian optimization |
CN114071849A (en) * | 2021-11-15 | 2022-02-18 | 上海无线电设备研究所 | Plasma generator for hypersonic target ablation diffusion object |
CN115906942A (en) * | 2022-11-28 | 2023-04-04 | 固德威技术股份有限公司 | Arc fault identification method, device, equipment and readable storage medium |
CN116819250A (en) * | 2023-06-27 | 2023-09-29 | 深圳供电局有限公司 | Method for detecting fault arc in photovoltaic system |
Non-Patent Citations (3)
Title |
---|
王俊;王涛;唐成双;辛煜;: "甚高频激发容性耦合氩等离子体的电子能量分布函数的演变", 物理学报, no. 05, 31 December 2016 (2016-12-31) * |
田守卫;孙洪福;胡海天;: "高密度等离子体钝化层中电弧放电问题的改善", 集成电路应用, no. 06, 9 June 2020 (2020-06-09) * |
郝晓亮;: "贝叶斯网络在半导体设备故障诊断中的应用", 电子工业专用设备, no. 04, 20 April 2013 (2013-04-20) * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US20200210873A1 (en) | Time-series fault detection, fault classification, and transition analysis using a k-nearest-neighbor and logistic regression approach | |
US11133204B2 (en) | Chamber matching with neural networks in semiconductor equipment tools | |
US6706543B2 (en) | Method of monitoring and/or controlling a semiconductor manufacturing apparatus and a therefor | |
US20150069912A1 (en) | RF Impedance Model Based Fault Detection | |
US20210116896A1 (en) | Real-time anomaly detection and classification during semiconductor processing | |
He et al. | Principal component based k-nearest-neighbor rule for semiconductor process fault detection | |
US6826489B2 (en) | Fault classification in a plasma process chamber | |
CN111313833B (en) | Method for detecting series arc fault on direct current side of series photovoltaic system | |
CN117828275B (en) | Prediction method and device for plasma chamber arcing | |
KR20030010468A (en) | Method and system for in-line monitoring process performance using measurable equipment signals | |
US20090206822A1 (en) | Application Of Wideband Sampling For Arc Detection With A Probabilistic Model For Quantitatively Measuring Arc Events | |
CN117828275A (en) | Prediction method and device for plasma chamber alignment | |
Yan et al. | Robust event detection for residential load disaggregation | |
CN111258863B (en) | Data anomaly detection method, device, server and computer readable storage medium | |
US10309013B2 (en) | Method and system for identifying a clean endpoint time for a chamber | |
Thieullen et al. | Application of Principal Components Analysis to improve fault detection and diagnosis on semiconductor manufacturing equipment | |
WO2010024489A1 (en) | Leakage detecting method of process chamber | |
US6855209B2 (en) | Plasma chamber cleaning | |
JP2024507688A (en) | Identifying process anomalies using measurement violation analysis | |
Ahad et al. | Data-adaptive symmetric CUSUM for sequential change detection | |
TWI677264B (en) | Rf impedance model based fault detection | |
Cherry et al. | Monitoring non-normal data with principal component analysis and adaptive density estimation | |
Mac Gearailt | Process diagnostics of industrial plasma systems | |
Katari et al. | Integration of Artificial Intelligence for Real-Time Fault Detection in Semiconductor Packaging | |
CN113567733A (en) | Voltage sag detection method based on accumulation sum algorithm and tensor theory |
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 | ||
GR01 | Patent grant |