Semiconductor processing apparatus failure diagnosis method, electronic device, and storage medium
By collecting current operating data of semiconductor processing equipment and utilizing pre-trained fault prediction and diagnosis models, accurate prediction of fault module type, time, and cause is achieved, solving the problem of low equipment maintenance efficiency in existing technologies and reducing production losses.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-23
- Publication Date
- 2026-06-23
Smart Images
Figure CN122264145A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of semiconductor processing and manufacturing technology, and in particular to a method for diagnosing faults in semiconductor processing equipment, an electronic device, and a readable storage medium. Background Technology
[0002] The operational status of semiconductor processing equipment has a significant impact on the production capacity of semiconductor companies. Furthermore, semiconductor processing equipment is extremely expensive, and equipment failures can result in substantial financial losses. Improving equipment maintenance efficiency and reducing downtime are crucial for increasing production capacity and controlling costs. Therefore, providing convenient, practical, and reliable fault diagnosis for semiconductor processing equipment has become an urgent need for semiconductor companies. However, the increasingly complex structure of semiconductor processing equipment also presents significant challenges to fault diagnosis.
[0003] It should be noted that the information disclosed in the background section of this invention is intended only to enhance the understanding of the general background of this invention, and should not be construed as an admission or in any way implying that the information constitutes prior art known to those skilled in the art. Summary of the Invention
[0004] The purpose of this invention is to provide a method for diagnosing faults in semiconductor processing equipment, an electronic device, and a readable storage medium that can predict in advance the type of faulty module, the time of failure, and the cause of failure in semiconductor processing equipment, so that equipment maintenance personnel can plan countermeasures in advance, thereby reducing losses in the semiconductor production process.
[0005] To achieve the above objectives, the present invention provides a method for fault diagnosis of semiconductor processing equipment, comprising:
[0006] Collect current operating data of the semiconductor processing equipment to be diagnosed;
[0007] The current operating data of the semiconductor processing equipment to be diagnosed is input into the pre-trained fault prediction model to obtain the fault module type prediction result and fault time prediction result of the semiconductor processing equipment to be diagnosed.
[0008] Based on the predicted fault module type of the semiconductor processing equipment to be diagnosed, the current operating data is diagnosed using a pre-created fault diagnosis model for the corresponding module to obtain the fault cause diagnosis result of the semiconductor processing equipment to be diagnosed.
[0009] Optionally, the current operating data includes current process control data, current process control restriction violation data, and current process control alarm data.
[0010] Optionally, the fault prediction model is trained through the following process:
[0011] Historical data of each sample semiconductor processing equipment of the same model as the semiconductor processing equipment to be diagnosed is obtained within a preset time period. The historical data includes the equipment operation data and equipment status data of the sample semiconductor processing equipment at different sampling times.
[0012] Based on the historical data and multiple preset sampling time intervals, multiple sample sets are constructed. Each sample set includes multiple samples. Each sample includes sample device operation data and corresponding tags. The tags include fault module type tags and fault time tags.
[0013] The pre-built neural network model is trained using the multiple sample sets to obtain multiple trained neural network models.
[0014] The neural network model with the highest accuracy among the multiple trained neural network models is selected as the fault prediction model.
[0015] Optionally, the step of constructing multiple sample sets based on the historical data and multiple preset sampling time intervals includes:
[0016] For each of the sample semiconductor processing devices, based on the device status data of the sample semiconductor processing device at different sampling times, the corresponding device fault information is obtained, and the device fault information includes the fault module type and the fault occurrence time;
[0017] For each semiconductor processing equipment, based on the equipment fault information corresponding to the semiconductor processing equipment, the equipment operation data of the semiconductor processing equipment before the fault occurred is selected from the historical data corresponding to the semiconductor processing equipment according to different preset sampling time intervals as sample equipment operation data;
[0018] For each sample of equipment operation data corresponding to each semiconductor processing equipment, based on the equipment fault information corresponding to the semiconductor processing equipment and the sampling time of the sample equipment operation data, the fault module type label and fault time label corresponding to the sample operation data are determined.
[0019] For each preset sampling time interval, the sample set for that preset sampling time interval is obtained based on the operating data of all sample devices under that preset sampling time interval and their respective corresponding fault module type labels and fault time labels.
[0020] Optionally, training the pre-built neural network model using the multiple sample sets includes:
[0021] The pre-built neural network model is trained using multiple sample sets through multi-fold cross-validation.
[0022] Optionally, the fault diagnosis model is a Bayesian network model.
[0023] Optionally, the Bayesian network model includes, from bottom to top, a component layer, a module layer, and a fault cause layer. The component layer includes multiple component nodes, the module layer includes one module node, and the fault cause layer includes multiple fault cause nodes.
[0024] Optionally, the semiconductor processing equipment fault diagnosis method provided by the present invention further includes:
[0025] The component nodes of the component layer and / or the fault nodes of the fault cause layer are updated periodically.
[0026] To achieve the above objectives, the present invention also provides an electronic device, including a processor and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, it implements the semiconductor processing equipment fault diagnosis method described above.
[0027] To achieve the above objectives, the present invention also provides a readable storage medium storing a computer program, which, when executed by a processor, implements the semiconductor processing equipment fault diagnosis method described above.
[0028] Compared with the prior art, the semiconductor processing equipment fault diagnosis method, electronic device, and readable storage medium provided by the present invention have the following beneficial effects:
[0029] The semiconductor processing equipment fault diagnosis method provided by this invention first collects the current operating data of the semiconductor processing equipment to be diagnosed; then, it inputs the current operating data into a pre-trained fault prediction model to obtain the fault module type prediction result and fault time prediction result of the semiconductor processing equipment to be diagnosed; finally, based on the fault module type prediction result of the semiconductor processing equipment to be diagnosed, a pre-created fault diagnosis model for the corresponding module is used to diagnose the current operating data to obtain the fault cause diagnosis result of the semiconductor processing equipment to be diagnosed. Therefore, the fault module type, fault time, and fault cause of the semiconductor processing equipment can be predicted in advance, allowing equipment maintenance personnel to plan countermeasures in advance, thereby reducing losses in the semiconductor production process. Furthermore, the semiconductor processing equipment fault diagnosis method provided by this invention can also effectively detect the health status of semiconductor processing equipment and its components.
[0030] Since the electronic device and readable storage medium provided by this invention belong to the same inventive concept as the semiconductor processing equipment fault diagnosis method provided by this invention, the electronic device and readable storage medium provided by this invention have at least all the beneficial effects of the semiconductor processing equipment fault diagnosis method provided by this invention. For details, please refer to the relevant description above. Therefore, the beneficial effects of the electronic device and readable storage medium provided by this invention will not be elaborated here. Attached Figure Description
[0031] Figure 1 A flowchart of a semiconductor processing equipment fault diagnosis method provided in one embodiment of the present invention;
[0032] Figure 2 This is a block diagram of an electronic device provided according to an embodiment of the present invention. Detailed Implementation
[0033] The following detailed description, in conjunction with the accompanying drawings and specific embodiments, further illustrates the semiconductor processing equipment fault diagnosis method, electronic device, and readable storage medium proposed in this invention. The advantages and features of this invention will become clearer from the following description. It should be noted that the drawings are in a very simplified form and use non-precise proportions, used only to facilitate and clarify the explanation of the invention's purpose. Please refer to the drawings for a clearer understanding of the invention's objectives, features, and advantages. It should be understood that the structures, proportions, sizes, etc., depicted in the accompanying drawings are only for illustrative purposes and to enable those skilled in the art to understand and read the invention, and are not intended to limit the implementation conditions of the invention. Any modifications to the structure, changes in proportions, or adjustments to the size, provided they produce the same or similar effects and achieve the same objectives as this invention, should still fall within the scope of the technical content disclosed in this invention.
[0034] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element. The singular forms “a,” “an,” and “the” include plural objects. The term “or” is generally used to mean “and / or,” the term “several” is generally used to mean “at least one,” and the term “at least two” is generally used to mean “two or more.” Furthermore, the terms “first,” “second,” and “third” are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated.
[0035] Furthermore, in the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in a suitable manner in any one or more embodiments or examples. Furthermore, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.
[0036] The core idea of this invention is to provide a fault diagnosis method for semiconductor processing equipment, an electronic device, and a readable storage medium, which can predict in advance the fault module type, fault time, and fault cause of semiconductor processing equipment, so that equipment maintenance personnel can plan countermeasures in advance, thereby reducing losses in the semiconductor production process.
[0037] It should be noted that the semiconductor processing equipment fault diagnosis method provided by this invention can be applied to the electronic equipment provided by this invention. This electronic equipment can be a personal computer, a mobile terminal, etc., and the mobile terminal can be a mobile phone, tablet computer, or other hardware device with various operating systems. It should also be noted that, as those skilled in the art will understand, the semiconductor processing equipment fault diagnosis method provided by this invention can not only diagnose faults in etching equipment but also in other semiconductor processing equipment; this invention does not limit its application to these areas.
[0038] To achieve the above-mentioned goals, this invention provides a method for fault diagnosis of semiconductor processing equipment. Please refer to [the relevant documentation]. Figure 1 This is a flowchart of a semiconductor processing equipment fault diagnosis method provided in one embodiment of the present invention. Figure 1 As shown, the semiconductor processing equipment fault diagnosis method provided by the present invention includes the following steps:
[0039] Step S100: Collect the current operating data of the semiconductor processing equipment to be diagnosed.
[0040] Step S200: Input the current operating data of the semiconductor processing equipment to be diagnosed into the pre-trained fault prediction model to obtain the fault module type prediction result and fault time prediction result of the semiconductor processing equipment to be diagnosed.
[0041] Step S200: Based on the prediction results of the fault module type of the semiconductor processing equipment to be diagnosed, the current operating data is diagnosed using a pre-created fault diagnosis model of the corresponding module to obtain the fault cause diagnosis results of the semiconductor processing equipment to be diagnosed.
[0042] Therefore, the semiconductor processing equipment fault diagnosis method provided by this invention can predict the type, time, and cause of faulty modules in semiconductor processing equipment in advance, enabling equipment maintenance personnel to plan countermeasures ahead of time and thus reduce losses in the semiconductor production process. Furthermore, the semiconductor processing equipment fault diagnosis method provided by this invention can also effectively detect the health status of semiconductor processing equipment and its components.
[0043] Specifically, by inputting the current operating data of the semiconductor processing equipment to be diagnosed into the fault prediction model, the fault module type prediction result (i.e., which module of the semiconductor processing equipment to be diagnosed will fail) and the fault time prediction result (i.e., how long after which the failure will occur) of the semiconductor processing equipment to be diagnosed can be obtained. Based on the fault module type prediction result of the semiconductor processing equipment to be diagnosed, the fault diagnosis model of the corresponding module is used to diagnose the current operating data, thereby obtaining the fault cause diagnosis result of the semiconductor processing equipment to be diagnosed (because of what reason of which component in the module caused the failure).
[0044] It should be noted that the types of fault modules that the fault prediction model can predict, as well as the number and types of pre-created fault diagnosis models, are related to the type of semiconductor processing equipment to be diagnosed. Taking the semiconductor processing equipment to be diagnosed as an etching equipment as an example, since the etching equipment includes a vacuum module, a gas transmission module, an RF power supply module, an electrostatic chuck module, a wafer transfer module, and a process module, the types of fault modules that the fault prediction model can predict include vacuum module faults, gas transmission module faults, RF power supply module faults, electrostatic chuck module faults, wafer transfer module faults, and process module faults. The pre-created fault diagnosis models include vacuum module fault diagnosis models, gas transmission module fault diagnosis models, RF power supply module fault diagnosis models, electrostatic chuck module fault diagnosis models, wafer transfer module fault diagnosis models, and process module fault diagnosis models.
[0045] In some exemplary embodiments, the current operating data includes current process control data, current process control limit violation data, and current process control alarm data. Therefore, by simultaneously inputting the current process control data, current process control limit violation data, and current process control alarm data of the semiconductor processing equipment to be diagnosed into the fault prediction model, the accuracy of the fault prediction model's prediction of the fault module type and fault time of the semiconductor processing equipment to be diagnosed can be effectively guaranteed.
[0046] Specifically, process control data can be important process parameter data provided by the Automated Process Control (APC) system for the semiconductor processing equipment under diagnosis during wafer processing. Taking an etching machine as an example, the process control data includes gas flow rate, voltage, and other process parameter data. Process control limit violation data can be data provided by the APC system regarding whether various process parameters of the semiconductor processing equipment under diagnosis exceed the corresponding upper and lower thresholds during wafer processing, along with the corresponding violation level. It should be noted that the APC system classifies the violation level of each process parameter based on the severity of the violation of the corresponding upper and lower thresholds. Limit violations can trigger the process chamber (process cavity) of the semiconductor processing equipment (e.g., etching machine) to stop working and send alert emails, among other actions. Process control alarm data can be alarm classification data provided by the APC system for the semiconductor processing equipment under diagnosis during wafer processing. For example, the alarm types provided by the APC system can include warnings, information, critical, errors, and others. Different alarm types will trigger different actions (including equipment shutdown).
[0047] In some exemplary embodiments, the current operating data also includes power supply voltage drop data. Since a sudden drop in power supply voltage can also cause semiconductor processing equipment to malfunction, by simultaneously inputting the current power supply voltage drop data along with the current process control data, current process control limit violation data, and current process control alarm data into the fault prediction model, the accuracy of the obtained fault module type prediction results and fault time prediction results can be further guaranteed.
[0048] In some exemplary embodiments, the fault prediction model is trained through the following process:
[0049] Historical data of each sample semiconductor processing equipment of the same model as the semiconductor processing equipment to be diagnosed is obtained within a preset time period. The historical data includes the equipment operation data and equipment status data of the sample semiconductor processing equipment at different sampling times.
[0050] Based on the historical data and multiple preset sampling time intervals, multiple sample sets are constructed. Each sample set includes multiple samples, and each sample includes sample device operation data and corresponding tags. The tags include fault module type tags and fault time tags.
[0051] The pre-built neural network model is trained using the multiple sample sets to obtain multiple trained neural network models.
[0052] The neural network model with the highest accuracy among the multiple trained neural network models is selected as the fault prediction model.
[0053] Since wafer fabrication processes (e.g., etching) are dynamic, historical data from various sample semiconductor processing equipment of the same model as the semiconductor processing equipment to be diagnosed, collected within a preset time period, are used to construct sample sets distributed according to different preset sampling time intervals. These samples are then used to train a pre-built neural network model, resulting in multiple trained neural network models. By selecting the trained neural network model with the highest accuracy as the fault prediction model, the prediction accuracy of the final fault prediction model can be effectively guaranteed. Experimental results show that the fault prediction model provided by this invention can provide acceptable and effective prediction results for fault module type and fault time at least 24 hours before the fault occurs, enabling equipment maintenance personnel to react in advance based on the prediction results.
[0054] It should be noted that the present invention does not limit the specific value of the preset time period; for example, the preset time period can be within 6 months. It should also be noted that the samples in the sample sets under different preset sampling time intervals are different from each other.
[0055] Furthermore, the equipment operation data may include process control data, process control limit violation data, process control alarm data, power supply voltage drop data, etc., wherein the process control data, process control limit violation data, and process control alarm data can be provided by an Automated Process Control (APC) system. Equipment status data can be provided by a real-time clock system, which records changes in the status (e.g., standby, production, fault) of semiconductor processing equipment (e.g., etching equipment) and its corresponding components. It should also be noted that, as those skilled in the art will understand, this invention does not limit the network structure of the pre-built neural network model; the pre-built neural network model can be, but is not limited to, a densely connected convolutional neural network (DenseNet) model, a long short-term memory network (LSTM) model, etc.
[0056] In some exemplary embodiments, the construction of multiple sample sets based on the historical data and multiple preset sampling time intervals includes:
[0057] For each of the sample semiconductor processing devices, based on the device status data of the sample semiconductor processing device at different sampling times, the corresponding device fault information is obtained, and the device fault information includes the fault module type and the fault occurrence time;
[0058] For each semiconductor processing equipment, based on the equipment fault information corresponding to the semiconductor processing equipment, the equipment operation data of the semiconductor processing equipment before the fault occurred is selected from the historical data corresponding to the semiconductor processing equipment according to different preset sampling time intervals as sample equipment operation data;
[0059] For each sample of equipment operation data corresponding to each semiconductor processing equipment, based on the equipment fault information corresponding to the semiconductor processing equipment and the sampling time of the sample equipment operation data, the fault module type label and fault time label corresponding to the sample operation data are determined.
[0060] For each preset sampling time interval, the sample set for that preset sampling time interval is obtained based on the operating data of all sample devices under that preset sampling time interval and their respective corresponding fault module type labels and fault time labels.
[0061] Therefore, after determining the equipment fault information of each sample semiconductor processing equipment within a preset time period (e.g., six months), based on the equipment fault information, the equipment operation data of the sample semiconductor processing equipment before the fault occurred can be selected from the historical data corresponding to the sample semiconductor processing equipment at different preset sampling time intervals. Specifically, for each preset sampling time interval, according to the fault occurrence time of the sample semiconductor processing equipment, equipment operation data at a sampling time before the fault occurred can be selected from the historical data corresponding to the sample semiconductor processing equipment at each preset sampling time interval as sample equipment operation data. Thus, multiple sample equipment operation data of the sample semiconductor processing equipment under the preset sampling time interval can be obtained. For each sample equipment operation data corresponding to each sample semiconductor processing equipment, the fault time tag corresponding to the sample equipment operation data can be determined according to the time interval between the fault occurrence time of the sample semiconductor processing equipment and the sampling time of the sample equipment operation data; the fault module type tag corresponding to the sample equipment operation data can be determined according to the fault module type of the sample semiconductor processing equipment. For each preset sampling time interval, the sample set for that preset sampling time interval can be obtained by using the operating data of all sample devices under that preset sampling time interval and their respective corresponding fault module type labels and fault time labels.
[0062] It should be noted that the present invention does not limit the specific value of the preset sampling time interval, and the specific value of the preset sampling time interval can be set according to the actual situation.
[0063] In some exemplary embodiments, training the pre-built neural network model using the multiple sample sets respectively includes:
[0064] The pre-built neural network model is trained using multiple sample sets through multi-fold cross-validation.
[0065] Multi-fold cross-validation divides the dataset into k (e.g., 4) mutually exclusive subsets of equal size. Each subset serves as a validation dataset to validate the model, while the other (k-1) subsets serve as training datasets to train the model. Thus, for each sample set, the pre-built neural network model is trained using multi-fold cross-validation (k-fold cross-validation) on that sample set. This avoids dependence on a single experiment, allowing for better evaluation and comparison of the performance of different models on a limited dataset, effectively improving the model's generalization ability and predictive performance.
[0066] In some exemplary implementations, the fault diagnosis model is a Bayesian network model.
[0067] Because Bayesian networks have the characteristics of multi-source information representation and fusion, they are suitable for knowledge representation and reasoning under uncertain information conditions. Therefore, this invention uses a Bayesian network model as a fault diagnosis model to diagnose fault causes, which can effectively ensure the accuracy of fault cause reasoning results.
[0068] Specifically, after obtaining the fault module type prediction result of the semiconductor processing equipment to be diagnosed, the current operating data of the semiconductor processing equipment to be diagnosed is input into the fault diagnosis model (Bayesian network model) of the module corresponding to the fault module type prediction result, so as to obtain the occurrence probability (posterior probability) of each fault cause. The fault cause with the highest occurrence probability (posterior probability) is the fault cause diagnosis result of the semiconductor processing equipment to be diagnosed.
[0069] In some exemplary implementations, the Bayesian network model comprises, from bottom to top, a component layer, a module layer, and a fault cause layer. The component layer includes multiple component nodes, the module layer includes one module node, and the fault cause layer includes multiple fault cause nodes.
[0070] Specifically, for each module in the semiconductor processing equipment to be diagnosed, the component nodes and fault cause nodes in the Bayesian network model (fault diagnosis model) corresponding to the module can be defined according to the component composition of the module and its fault causes. Taking the vacuum module in the etching equipment as an example, the component nodes in the component layer include vacuum valves, vacuum sensors, throttle valves, pressure relief valves, etc., and the fault cause nodes in the fault cause layer include pressure relief valve leakage, poor vacuum conduction, abnormal pressure relief valve position, abnormal throttle valve position, abnormal vacuum valve position, etc.
[0071] It should be noted that, for each module in the semiconductor processing equipment to be diagnosed, the content on how to determine the prior probability and conditional probability of each node in the Bayesian network model (fault diagnosis model) corresponding to the module can refer to relevant content in related fields known to those skilled in the art, and will not be elaborated here.
[0072] In some exemplary embodiments, the semiconductor processing equipment fault diagnosis method provided by the present invention further includes:
[0073] The component nodes of the component layer and / or the fault nodes of the fault cause layer are updated periodically.
[0074] Since the semiconductor manufacturing process is dynamic, the accuracy of the fault cause inference results of the fault diagnosis model (Bayesian network model) can be effectively guaranteed by periodically updating the component nodes of the component layer and / or the fault nodes of the fault cause layer.
[0075] Based on the same inventive concept, the present invention also provides an electronic device, please refer to [reference needed]. Figure 2 This is a block diagram of an electronic device provided in one embodiment of the present invention. Figure 2 As shown, the electronic device includes a processor 101 and a memory 103. The memory 103 stores a computer program. When the computer program is executed by the processor 101, it implements the semiconductor processing equipment fault diagnosis method described above. Since the electronic device provided by this invention and the semiconductor processing equipment fault diagnosis method provided by this invention belong to the same inventive concept, the electronic device provided by this invention possesses at least all the beneficial effects of the semiconductor processing equipment fault diagnosis method provided by this invention. Therefore, the beneficial effects of the electronic device provided by this invention can be referred to the relevant descriptions of the beneficial effects of the semiconductor processing equipment fault diagnosis method provided by this invention above, and will not be repeated here.
[0076] Please continue to refer to this. Figure 2 ,like Figure 2As shown, the electronic device also includes a communication interface 102 and a communication bus 104, wherein the processor 101, the communication interface 102, and the memory 103 communicate with each other through the communication bus 104. The communication bus 104 can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus 104 can be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is used to represent it in the figure, but this does not indicate that there is only one bus or one type of bus. The communication interface 102 is used for communication between the aforementioned electronic device and other devices.
[0077] It should be noted that the processor 101 referred to in this invention can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor. The processor 101 is the control center of the electronic device, connecting various parts of the electronic device through various interfaces and lines.
[0078] It should also be noted that the memory 103 can be used to store the computer program, and the processor 101 implements various functions of the electronic device by running or executing the computer program stored in the memory 103 and calling the data stored in the memory 103. The memory 103 may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable memory (PROM), electrically programmable memory (EPROM), electrically erasable programmable memory (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, random access memory is available in a variety of forms, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous random access memory (SDRAM), dual data rate synchronous random access memory (DDRSDRAM), enhanced synchronous random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), memory bus direct random access memory (RDRAM), direct memory bus dynamic random access memory (DRDRAM), and memory bus dynamic random access memory (RDRAM), etc.
[0079] This invention also provides a readable storage medium storing a computer program. When executed by a processor, the computer program can implement the semiconductor processing equipment fault diagnosis method described above. Since the readable storage medium provided by this invention and the semiconductor processing equipment fault diagnosis method provided by this invention belong to the same inventive concept, the readable storage medium provided by this invention possesses at least all the beneficial effects of the semiconductor processing equipment fault diagnosis method provided by this invention. Therefore, the beneficial effects of the readable storage medium provided by this invention can be referred to the relevant descriptions of the beneficial effects of the semiconductor processing equipment fault diagnosis method provided by this invention above, and will not be repeated here.
[0080] Specifically, the readable storage medium provided by this invention can be any combination of one or more computer-readable media. The readable medium can be a computer-readable signal medium or a computer-readable storage medium. Computer-readable storage media can be, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof. More specific examples (not exhaustive) of computer-readable storage media include: electrical connections having one or more wires, portable computer hard drives, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, apparatus, or device.
[0081] Furthermore, the computer-readable signal medium may include data signals propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, etc., or any suitable combination thereof.
[0082] In summary, compared with the prior art, the semiconductor processing equipment fault diagnosis method, electronic device, and readable storage medium provided by the present invention have the following beneficial effects:
[0083] This invention first collects the current operating data of the semiconductor processing equipment to be diagnosed; then, it inputs this data into a pre-trained fault prediction model to obtain predictions of the fault module type and fault time; finally, based on the fault module type predictions, a pre-created fault diagnosis model for the corresponding module is used to diagnose the current operating data to obtain a fault cause diagnosis. This allows for the prediction of the fault module type, fault time, and fault cause of the semiconductor processing equipment in advance, enabling maintenance personnel to plan countermeasures ahead and reduce losses during semiconductor production. Furthermore, this invention can effectively detect the health status of semiconductor processing equipment and its components.
[0084] It should be noted that computer program code for performing the operations of this invention can be written in one or more programming languages or a combination thereof. These programming languages include object-oriented programming languages such as Java, Smalltalk, and C++, as well as conventional procedural programming languages such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).
[0085] It should be noted that the apparatus and methods disclosed in the embodiments herein can also be implemented in other ways. The apparatus embodiments described above are merely illustrative; for example, the flowcharts and block diagrams in the accompanying drawings show the architecture, functionality, and operation of possible implementations of apparatus, methods, and computer program products according to various embodiments herein. In this regard, each block in a flowchart or block diagram may represent a module, program, or part of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system to perform the specified function or action, or can be implemented using a combination of dedicated hardware and computer instructions. In addition, the functional modules in the various embodiments of this article can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.
[0086] It should also be noted that the above description is merely a description of preferred embodiments of the present invention and is not intended to limit the scope of the present invention in any way. Any changes or modifications made by those skilled in the art based on the above disclosure are within the protection scope of the present invention. Obviously, those skilled in the art can make various modifications and variations to the present invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the present invention and its equivalents, the present invention also intends to include these modifications and variations.
Claims
1. A method for fault diagnosis of semiconductor processing equipment, characterized in that, include: Collect current operating data of the semiconductor processing equipment to be diagnosed; The current operating data of the semiconductor processing equipment to be diagnosed is input into the pre-trained fault prediction model to obtain the fault module type prediction result and fault time prediction result of the semiconductor processing equipment to be diagnosed. Based on the predicted fault module type of the semiconductor processing equipment to be diagnosed, the current operating data is diagnosed using a pre-created fault diagnosis model for the corresponding module to obtain the fault cause diagnosis result of the semiconductor processing equipment to be diagnosed.
2. The semiconductor processing equipment fault diagnosis method according to claim 1, characterized in that, The current operating data includes current process control data, current process control restriction violation data, and current process control alarm data.
3. The semiconductor processing equipment fault diagnosis method according to claim 1, characterized in that, The fault prediction model is trained through the following process: Historical data of each sample semiconductor processing equipment of the same model as the semiconductor processing equipment to be diagnosed is obtained within a preset time period. The historical data includes the equipment operation data and equipment status data of the sample semiconductor processing equipment at different sampling times. Based on the historical data and multiple preset sampling time intervals, multiple sample sets are constructed. Each sample set includes multiple samples. Each sample includes sample device operation data and corresponding tags. The tags include fault module type tags and fault time tags. The pre-built neural network model is trained using the multiple sample sets to obtain multiple trained neural network models. The neural network model with the highest accuracy among the multiple trained neural network models is selected as the fault prediction model.
4. The semiconductor processing equipment fault diagnosis method according to claim 3, characterized in that, The step of constructing multiple sample sets based on the historical data and multiple preset sampling time intervals includes: For each of the sample semiconductor processing devices, based on the device status data of the sample semiconductor processing device at different sampling times, the corresponding device fault information is obtained, and the device fault information includes the fault module type and the fault occurrence time; For each semiconductor processing equipment, based on the equipment fault information corresponding to the semiconductor processing equipment, the equipment operation data of the semiconductor processing equipment before the fault occurred is selected from the historical data corresponding to the semiconductor processing equipment according to different preset sampling time intervals as sample equipment operation data; For each sample of equipment operation data corresponding to each semiconductor processing equipment, based on the equipment fault information corresponding to the semiconductor processing equipment and the sampling time of the sample equipment operation data, the fault module type label and fault time label corresponding to the sample operation data are determined. For each preset sampling time interval, the sample set for that preset sampling time interval is obtained based on the operating data of all sample devices under that preset sampling time interval and their respective corresponding fault module type labels and fault time labels.
5. The semiconductor processing equipment fault diagnosis method according to claim 3, characterized in that, The step of training the pre-built neural network model using the multiple sample sets includes: The pre-built neural network model is trained using multiple sample sets through multi-fold cross-validation.
6. The semiconductor processing equipment fault diagnosis method according to claim 1, characterized in that, The fault diagnosis model is a Bayesian network model.
7. The semiconductor processing equipment fault diagnosis method according to claim 6, characterized in that, The Bayesian network model comprises, from bottom to top, a component layer, a module layer, and a fault cause layer. The component layer includes multiple component nodes, the module layer includes one module node, and the fault cause layer includes multiple fault cause nodes.
8. The semiconductor processing equipment fault diagnosis method according to claim 7, characterized in that, The method further includes: The component nodes of the component layer and / or the fault nodes of the fault cause layer are updated periodically.
9. An electronic device, characterized in that, It includes a processor and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, it implements the semiconductor processing equipment fault diagnosis method according to any one of claims 1 to 8.
10. A readable storage medium, characterized in that, The readable storage medium stores a computer program, which, when executed by a processor, implements the semiconductor processing equipment fault diagnosis method according to any one of claims 1 to 8.