Beam uniformity adjusting method and device, storage medium and electronic equipment

Abstract

The invention provides a beam uniformity adjustment method and device, a storage medium and electronic equipment. The method comprises the following steps: acquiring historical adjustment data; determining a current training state vector from the historical adjustment data, calling an application adjustment model, and determining a device adjustment vector corresponding to the current training state vector; determining an adjustment model loss value of the application adjustment model based on the device adjustment vector; model parameters in the application adjustment model are updated according to the direction of reducing the loss value of the adjustment model; and the current training state vector determined from the historical adjustment data is iteratively executed until the target model training condition is met, so that model training of the application adjustment model through the historical adjustment data is completed, and the application adjustment model support is used for guiding beam uniformity adjustment of the target ion implanter. According to the embodiment of the invention, the accuracy of beam uniformity adjustment can be improved, so that the beam uniformity adjustment efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of semiconductor technology, and in particular to a beam uniformity adjustment method, device, storage medium and electronic equipment. Background Art

[0002] At present, with the development of the information age, the semiconductor industry has been widely used in various fields. With the continuous improvement of the performance requirements of semiconductor chips, the manufacturing process of semiconductor integrated circuits has developed rapidly, and higher requirements have been placed on the performance of related manufacturing equipment. Among them, the ion implanter, as the core equipment in the manufacturing process of semiconductor devices, can dope the surface area of ​​the semiconductor with ions, change the carrier concentration and conductivity type of the semiconductor, and its beam uniformity has a significant impact on the consistency of device performance in the wafer; therefore, the control of beam uniformity becomes particularly important during the ion implantation process, and beam uniformity has also become an important indicator for judging the performance of ion implanters. The relevant technology obtains the device adjustment vector by solving a linear equation group to adjust the ion implanter, thereby achieving the adjustment of beam uniformity; however, the influence of the device adjustment amount on the beam uniformity is nonlinear, which makes it difficult to obtain an accurate adjustment amount by solving a linear equation group, and thus makes the adjustment efficiency low. Based on this, how to improve the accuracy of beam uniformity adjustment to improve the adjustment efficiency currently has no good solution. Summary of the invention

[0003] In view of this, an embodiment of the present invention provides a beam uniformity adjustment method, device, storage medium and electronic device to solve the problem of low efficiency of beam uniformity adjustment caused by related technologies; that is, an embodiment of the present invention can perform model training on an application adjustment model through historical adjustment data, thereby obtaining a pre-trained application adjustment model, which can effectively improve the model performance of the application adjustment model, and then the application adjustment model with better model performance can be used to adjust the beam uniformity of a target ion implanter, which can effectively improve the accuracy of beam uniformity adjustment, so as to effectively improve the adjustment efficiency of beam uniformity.

[0004] According to one aspect of an embodiment of the present invention, a beam uniformity adjustment method is provided, the method comprising: Acquire historical adjustment data, wherein the historical adjustment data includes a plurality of historical state vectors of a target ion implanter, wherein one historical state vector includes a beam state vector and a device state vector of the target ion implanter at a historical acquisition moment; Determine a current training state vector from the historical adjustment data, and call an application adjustment model to determine a device adjustment vector corresponding to the current training state vector; Based on the device adjustment vector, determining an adjustment model loss value of the application adjustment model; and updating a model parameter in the application adjustment model in a direction of reducing the adjustment model loss value; The process of determining the current training state vector from the historical adjustment data is iteratively executed until the target model training condition is met, so as to complete the model training of the application adjustment model through the historical adjustment data, wherein the application adjustment model supports guiding the beam uniformity adjustment of the target ion implanter.

[0005] According to another aspect of an embodiment of the present invention, a beam uniformity adjustment device is provided, the device comprising: An acquisition unit, used for acquiring historical adjustment data, wherein the historical adjustment data includes a plurality of historical state vectors of a target ion implanter, and one historical state vector includes a beam state vector and a device state vector of the target ion implanter at a historical acquisition moment; A processing unit, configured to determine a current training state vector from the historical adjustment data, and call an application adjustment model to determine a device adjustment vector corresponding to the current training state vector; The processing unit is further configured to determine a regulation model loss value of the application regulation model based on the device regulation vector; and update a model parameter in the application regulation model in a direction of reducing the regulation model loss value; The processing unit is also used to iteratively execute the determination of the current training state vector from the historical adjustment data until the target model training conditions are met, so as to complete the model training of the application adjustment model through the historical adjustment data, and the application adjustment model supports guiding the beam uniformity adjustment of the target ion implanter.

[0006] According to another aspect of an embodiment of the present invention, an electronic device is provided, comprising a processor and a memory storing a program, wherein the program comprises instructions, which, when executed by the processor, enable the processor to perform the above-mentioned method.

[0007] According to another aspect of an embodiment of the present invention, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable a computer to execute the above-mentioned method.

[0008] The embodiment of the present invention can obtain historical adjustment data, and the historical adjustment data includes multiple historical state vectors of the target ion implanter, and one historical state vector includes the beam state vector and the device state vector of the target ion implanter at a historical acquisition moment; based on this, the current training state vector can be determined from the historical adjustment data, and the application adjustment model can be called to determine the device adjustment vector corresponding to the current training state vector. Then, correspondingly, the adjustment model loss value of the application adjustment model can be determined based on the device adjustment vector; and the model parameters in the application adjustment model are updated in the direction of reducing the adjustment model loss value. Further, the above determination of the current training state vector from the historical adjustment data can be iteratively executed until the target model training conditions are met, so as to complete the model training of the application adjustment model through the historical adjustment data, and the application adjustment model support is used to guide the beam uniformity adjustment of the target ion implanter. It can be seen that the embodiment of the present invention can train the application adjustment model through historical adjustment data to obtain a pre-trained application adjustment model, which can effectively improve the model performance of the application adjustment model, and then the application adjustment model with better model performance can be used to adjust the beam uniformity of the target ion implanter, which can effectively improve the accuracy of the beam uniformity adjustment, that is, the current device adjustment vector with higher accuracy can be obtained through the application adjustment model to effectively improve the adjustment efficiency of the beam uniformity. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Further details, features and advantages of the invention are disclosed in the following description of exemplary embodiments in conjunction with the accompanying drawings, in which: Figure 1 A schematic flow chart of a beam uniformity adjustment method according to an exemplary embodiment of the present invention is shown; Figure 2 A schematic diagram showing the effect of a change in magnetic pole position on a beam according to an exemplary embodiment of the present invention is shown; Figure 3 A schematic flow chart of another beam uniformity adjustment method according to an exemplary embodiment of the present invention is shown; Figure 4 A schematic flow chart of another beam uniformity adjustment method according to an exemplary embodiment of the present invention is shown; Figure 5 A schematic diagram of a parallel lens according to an exemplary embodiment of the present invention is shown; Figure 6 A schematic block diagram of a beam uniformity adjustment device according to an exemplary embodiment of the present invention is shown; Figure 7 A block diagram of an exemplary electronic device that can be used to implement an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0010] Embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as being limited to the embodiments described herein, which are instead provided for a more thorough and complete understanding of the present invention. It should be understood that the drawings and embodiments of the present invention are only for exemplary purposes and are not intended to limit the scope of protection of the present invention.

[0011] It should be understood that the various steps described in the method embodiments of the present invention may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.

[0012] The term "including" and its variations used in this document are open inclusions, that is, "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one other embodiment"; the term "some embodiments" means "at least some embodiments". Relevant definitions of other terms will be given in the following description. It should be noted that the concepts of "first", "second", etc. mentioned in the present invention are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0013] It should be noted that the modifications of "one" and "plurality" mentioned in the present invention are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0014] The names of the messages or information exchanged between multiple devices in the embodiments of the present invention are only used for illustrative purposes, and are not used to limit the scope of these messages or information.

[0015] It should be noted that the execution subject of the beam uniformity adjustment method provided in the embodiment of the present invention may be one or more electronic devices, which is not limited by the present invention; wherein, the electronic device may be a terminal (i.e., a client) or a server, then when the execution subject includes multiple electronic devices, and the multiple electronic devices include at least one terminal and at least one server, the beam uniformity adjustment method provided in the embodiment of the present invention may be jointly executed by the terminal and the server. Accordingly, the terminal mentioned here may include but is not limited to: smart phones, tablet computers, laptop computers, desktop computers, smart watches, smart voice interaction devices, smart home appliances, vehicle terminals, aircraft, etc. The server mentioned here may be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud services, cloud databases, cloud computing (cloud computing), cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN (Content Delivery Network), and basic cloud computing services such as big data and artificial intelligence platforms, etc.

[0016] Based on the above description, an embodiment of the present invention proposes a beam uniformity adjustment method, which can be executed by the electronic device (terminal or server) mentioned above; or, the beam uniformity adjustment method can be executed by the terminal and the server together. For the sake of convenience, the following description will take the electronic device executing the beam uniformity adjustment method as an example; Figure 1 As shown, the beam uniformity adjustment method may include the following steps S101-S104: S101, acquiring historical adjustment data, where the historical adjustment data includes a plurality of historical state vectors of a target ion implanter, and one historical state vector includes a beam state vector and a device state vector of the target ion implanter at a historical acquisition moment.

[0017] Among them, a beam state vector can be used to indicate the distribution of current at different positions, that is, a beam state vector can include the current magnitude at each sampling position among multiple sampling positions; optionally, the embodiment of the present invention does not limit the multiple sampling positions; illustratively, the multiple sampling positions can be multiple sampling positions between -180mm (millimeter) and 180mm, and the interval between any two sampling positions can be set according to experience or according to actual needs, which is not limited by the embodiment of the present invention. Optionally, the embodiment of the present invention can measure the beam state vector of the target ion implanter at any acquisition time through any measurement equipment.

[0018] Optionally, a device state vector may include device information of at least one device in the target ion implanter (i.e., device information of each device in the at least one device); optionally, at least one device may include but is not limited to at least one of the following: at least one coil (e.g., a multi-coil array) and at least one magnetic block (e.g., a multi-pole array), etc., which is not limited in the embodiment of the present invention. Optionally, when a device is a coil, the device information of the device may be the current size of the device; when a device is a magnetic block, the device information of the device may be the distance between the device and the beam (e.g., the distance between the device and the final arrival position of the beam), that is, the device information of the device may be the position information of the device. Optionally, a state vector may be a beam state vector and a device state vector spliced ​​together, such as a historical state vector may be a spliced ​​vector between a beam state vector and a device state vector at a historical acquisition moment, which may be used to reflect the environment at the corresponding acquisition moment; based on this, the spatial dimension of the state space (i.e., the space where the state vector spliced ​​together with the beam state vector and the device state vector is distributed) may depend on the number of beam sampling points (i.e., sampling positions) and the actual number of physical devices.

[0019] Optionally, the target ion implanter may be any broadband beam ion implanter (which may be used to generate a broadband ion beam), which is not limited in the embodiments of the present invention; illustratively, the target ion implanter may be one or more ion implanters with the same structure (such as the same devices, etc.), and so on.

[0020] In the embodiment of the present invention, the historical adjustment data may be obtained in the following ways, but not limited to: The first acquisition method: the electronic device stores the historical adjustment data in its own storage space. In this case, the electronic device can directly acquire the historical adjustment data from its own storage space.

[0021] The second acquisition method: the electronic device can obtain a historical adjustment data download link. In this case, the electronic device can download the historical adjustment data through the historical adjustment data download link to obtain the historical adjustment data, and so on.

[0022] S102, determining a current training state vector from historical adjustment data, and calling an application adjustment model to determine a device adjustment vector corresponding to the current training state vector.

[0023] Optionally, the electronic device may determine a historical state vector from the historical adjustment data in sequence, and use the currently determined historical state vector as the current training state vector to achieve determination of the current training state vector from the historical adjustment data; or, it may randomly determine at least one historical state vector from the historical adjustment data, and use the currently randomly determined at least one historical state vector as the current training state vector, and so on; the embodiments of the present invention are not limited to this.

[0024] In an embodiment of the present invention, the electronic device may input the current training state vector into an application adjustment model (i.e., the current application adjustment model), thereby outputting a device adjustment vector corresponding to the current training state vector, so as to call the application adjustment model and determine the device adjustment vector corresponding to the current training state vector.

[0025] Optionally, an adjustment model can be a neural network model, that is, the application adjustment model can be a neural network model; optionally, the model structure of the application adjustment model can be set according to experience or according to actual needs, and the embodiment of the present invention does not limit this. It should be understood that before the application adjustment model is trained, that is, before the application adjustment model is used for the first time, the electronic device can randomly initialize the application adjustment model, that is, randomly initialize the model parameters in the application adjustment model.

[0026] Optionally, a device adjustment vector may include adjustment information of each device in at least one device in the target ion implanter; optionally, an adjustment information may be an adjustment action information (which can be used to indicate the size of the device information adjusted for the corresponding device (i.e., the adjustment amount), such as moving 2 mm toward the beam (such as in the direction of beam travel or the final position of the beam), or a current reduction number, etc.), or an adjustment result information (which can be used to indicate the device information after adjustment of the corresponding device, such as adjusting to a position 3 mm away from the beam or adjusting to a certain current size, etc.), which is not limited to the embodiments of the present invention.

[0027] Among them, the space where the device adjustment vector is located can be a device adjustment space, that is, the space where the adjustment information of the device is located, which can ensure that each device is located in its actual adjustable range after adjustment, and its dimension is consistent with the actual number of devices. Optionally, the adjustment information range of any device indicated by the device adjustment space can be set according to demand; or, when the adjustment information is an adjustment amount, it can also be determined according to the current device state vector (such as the device state vector in the current training state vector), so that after any device is adjusted according to any value within the adjustment information range corresponding to any device, the device information of any device is located within the device information range corresponding to any device, and so on; the embodiment of the present invention is not limited to this. Optionally, the device information range corresponding to any device can be set according to experience or actual needs, and the embodiment of the present invention is not limited to this.

[0028] S103, determining an adjustment model loss value of the application adjustment model based on the device adjustment vector; and updating model parameters in the application adjustment model in a direction of reducing the adjustment model loss value.

[0029] In the embodiment of the present invention, the model parameters in the application adjustment model can be continuously updated, thereby continuously improving the model performance of the application adjustment model, which can effectively improve the accuracy of the device adjustment vector output by the application adjustment model.

[0030] S104, iteratively executing to determine the current training state vector from the historical adjustment data until the target model training conditions are met, so as to complete the model training of the application adjustment model through the historical adjustment data, and the application adjustment model support is used to guide the beam uniformity adjustment of the target ion implanter.

[0031] Optionally, the target model training conditions may include the model training conditions in the first model training stage and / or the model training conditions in the second model training stage; then when the target model training conditions include the model training conditions in the first model training stage and the model training conditions in the second model training stage, the electronic device may perform model training in the first model training stage through historical adjustment data, and perform model training in the second model training stage through historical adjustment data to complete model training, that is, complete model training of the application adjustment model through historical adjustment data. Optionally, the target model training conditions may be set according to experience or according to actual needs, and the embodiment of the present invention does not limit this; that is, the model training conditions in the first model training stage and the model training conditions in the second model training stage may be set according to experience or according to actual needs, and the embodiment of the present invention does not limit this. Among them, a model training condition can be used to indicate the end condition of model training, that is, a model training condition can also be called a model training end condition or a model training convergence condition.

[0032] Optionally, a model training condition may be that the number of iterations reaches a corresponding preset maximum number of iterations (e.g., the model training condition in the first model training stage may be that the number of iterations reaches a first preset number of iterations threshold, and the model training condition in the second model training stage may be that the number of iterations reaches a second preset number of iterations threshold); or, a model training condition may be that the historical adjustment data is completely traversed (e.g., the model training condition in the first model training stage and the model training condition in the second model training stage may be that the historical adjustment data is completely traversed, respectively), etc.; the embodiment of the present invention does not limit this. Optionally, the first preset number of iterations threshold and the second preset number of iterations threshold may be set according to experience or according to actual needs, and the embodiment of the present invention does not limit this; optionally, the first preset number of iterations threshold and the second preset number of iterations threshold may be the same or different, and the embodiment of the present invention does not limit this; optionally, when performing model training in the second model training stage, the number of iterations may start again from 1, that is, after completing the model training in the first model training stage, the electronic device may reset the number of iterations to 1.

[0033] Based on this, the electronic device can perform model training in the first model training stage through historical adjustment data, so that after the model training conditions in the first model training stage are met (such as the number of iterations reaches a first preset iteration number threshold), it can start model training in the second model training stage through historical adjustment data, so that when the model training conditions in the second model training stage are met, the target model training conditions can be met, that is, the model training process performed through historical adjustment data can meet the target model training conditions, which can also be called the model training process under the historical adjustment data meets the target model training conditions.

[0034] Optionally, the following updated beam state vector may be determined according to the updated beam state vector determination method in the first model training stage or the updated beam state vector determination method in the second model training stage, that is, in the first model training stage, the electronic device may determine the updated beam state vector according to the updated beam state vector determination method in the first model training stage, and in the second model training stage, the electronic device may determine the updated beam state vector according to the updated beam state vector determination method in the second model training stage. The accuracy of the updated beam state vector determination method in the second model training stage may be higher than the accuracy of the updated beam state vector determination method in the first model training stage, and the first model training stage precedes the second model training stage; optionally, the efficiency (i.e., convenience) of the updated beam state vector determination method in the first model training stage may be higher than the efficiency of the updated beam state vector determination method in the second model training stage.

[0035] Optionally, in the first model training stage, an updated beam state vector may be determined based on the uniformity simulation system, that is, the determination method of the updated beam state vector in the first model training stage may refer to the method of determining through the uniformity simulation system, and the device adjustment vector may be input into the simulation system to output the corresponding updated beam state vector. Optionally, the electronic device may also obtain beam uniformity change information corresponding to the device adjustment vector (also referred to as the benefit information or beam uniformity change amount corresponding to the updated state vector corresponding to the device adjustment vector) through the simulation system, and the embodiment of the present invention is not limited to this; Optionally, the beam uniformity change information corresponding to the device adjustment vector may be used to indicate the change of beam uniformity after adjustment according to the device adjustment vector. Optionally, a beam uniformity change information may be any value; Optionally, a beam uniformity may be described by a beam uniformity indication value, and then a beam uniformity change information may be used to indicate the change amount of the beam uniformity indication value after adjustment according to the corresponding device adjustment vector, that is, it may be used to indicate the change of beam uniformity after taking the action indicated by the device adjustment vector in the current state. Based on this, the simulation system may include but is not limited to at least one of the following: a relationship model between a device adjustment vector and a beam state vector (which can be used to determine an updated beam state vector corresponding to any device adjustment vector) and a relationship model between a device adjustment vector and beam uniformity change information (which can be used to determine beam uniformity change information corresponding to any device adjustment vector), etc., and the embodiment of the present invention does not limit this. Optionally, the relationship model between the device adjustment vector and the beam state vector may be a linear equation group constructed by using multiple device adjustment vectors and beam state vectors corresponding to each of the multiple device adjustment vectors (a beam state vector corresponding to a device adjustment vector may refer to a beam state vector adjusted according to the corresponding device adjustment vector, i.e., an updated beam state vector corresponding to a device adjustment vector), and the relationship model between the device adjustment vector and beam uniformity change information may be a linear equation group constructed by using multiple device adjustment vectors and beam uniformity change information corresponding to each of the device adjustment vectors (a beam uniformity change information corresponding to a device adjustment vector may refer to beam uniformity change information between the beam uniformity of the beam state vector adjusted according to the corresponding device adjustment vector and the beam uniformity before adjustment), etc.; the embodiment of the present invention does not limit this. Based on this, the embodiment of the present invention can quickly determine the updated beam state vector corresponding to any device adjustment vector through a simulation system, thereby quickly realizing the preliminary training of model parameters to effectively improve the model training efficiency.

[0036] Optionally, beam uniformity change information (i.e., benefit information) may be a difference between a beam uniformity indication value corresponding to an updated beam state vector and a beam uniformity indication value corresponding to a corresponding beam state vector before adjustment (such as a current state vector or a current training state vector, etc.). Optionally, the beam uniformity indicator value may be negatively correlated with the beam uniformity. In this case, a larger beam uniformity indicator value indicates a lower beam uniformity (i.e., worse, more uneven), and a smaller corresponding beam uniformity indicator value indicates a higher beam uniformity (i.e., better, and the smaller the beam uniformity indicator value, the more uniform the beam). In this case, when a beam uniformity change value is a positive value, it may indicate that the beam uniformity indicator value corresponding to the updated beam state vector is greater than the beam uniformity indicator value before adjustment (such as the beam uniformity indicator value corresponding to the current training state vector), which means that the beam uniformity corresponding to the updated beam state vector is lower. Correspondingly, when a beam uniformity change value is a negative value, it may indicate that the beam uniformity indicator value corresponding to the updated beam state vector is less than the beam uniformity indicator value before adjustment, which means that the beam uniformity corresponding to the updated beam state vector is higher. Optionally, the beam uniformity indicator value may also be positively correlated with the beam uniformity. In this case, a larger beam uniformity indicator value indicates a higher beam uniformity, and a smaller corresponding beam uniformity indicator value indicates a lower beam uniformity. In this case, when a beam uniformity change value is a positive value, it may indicate that the beam uniformity indicator value corresponding to the updated beam state vector is greater than the beam uniformity indicator value before adjustment (i.e., the beam uniformity change information may be an increase in the beam uniformity indicator value). In this case, it may indicate that the beam uniformity corresponding to the updated beam state vector is higher. Correspondingly, when a beam uniformity change value is a negative value, it may indicate that the beam uniformity indicator value corresponding to the updated beam state vector is less than the beam uniformity indicator value before adjustment (i.e., the beam uniformity change information may be a decrease in the beam uniformity indicator value). In this case, it may indicate that the beam uniformity corresponding to the updated beam state vector is lower, and so on. The embodiments of the present invention are not limited to this.

[0037] Optionally, the beam uniformity indicator value corresponding to a beam state vector may be the variance (i.e., the variance between the current magnitudes in the beam state vector) or the standard deviation (i.e., the standard deviation between the current magnitudes in the beam state vector) corresponding to the corresponding beam state vector, etc. In this case, the beam uniformity indicator value is negatively correlated with the beam uniformity, that is, the larger the variance or the standard deviation, the higher the volatility corresponding to the corresponding beam state vector, and the lower the beam uniformity; or, the beam uniformity indicator value corresponding to a beam state vector may be the inverse of the variance or the standard deviation corresponding to the corresponding beam state vector, in which case the beam uniformity indicator value may be positively correlated with the beam uniformity, etc. The embodiment of the present invention does not limit the specific method for determining the beam uniformity indicator value. For ease of explanation, the following description will be based on the example that the beam uniformity indicator value is negatively correlated with the beam uniformity.

[0038] Optionally, in the second model training stage, an updated beam state vector can be determined based on the measurement of the angle Faraday cup in the target chamber of the target ion implanter, that is, the method of determining the updated beam state vector in the second model training stage can refer to the method of determining by measuring the angle Faraday cup in the target chamber of the target ion implanter, that is, in the second model training stage, the electronic device can obtain the measurement result of the beam state vector of the angle Faraday cup in the target chamber of the target ion implanter after adjusting the target ion implanter according to any device adjustment vector, so as to determine it as the updated beam state vector; in this case, the electronic device can output any device adjustment vector or send any device adjustment vector to the target ion implanter to adjust at least one device in the target ion implanter. Based on this, the measurement of the beam state vector is almost real-time, so that an accurate updated beam state vector can be obtained, which can effectively improve the convergence rate of the model training, thereby quickly obtaining an application adjustment model with higher model performance.

[0039] In summary, the embodiments of the present invention can achieve rapid adjustment of beam uniformity by applying an adjustment model, etc.; while the prior art uses a linear relationship to adjust, resulting in low adjustment efficiency, because the actual change of the beam is not linearly related to the change of device information; for example, assuming that at least one device includes at least one magnetic block, one device information may be a device position, and the beam change (which may include the percentage change of the current magnitude at any sampling position) caused by the upper magnetic pole No. 10 of the multi-pole array moving 1mm, 2mm, 3mm and 4mm in the reference position toward the direction of the beam can be sequentially as follows: Figure 2 As shown by the line from small to large fluctuations in the middle; therefore, this makes the traditional adjustment method based on linear approximation have the problem of inaccurate adjustment and long adjustment time. In the embodiment of the present invention, a neural network is used to fit the relationship between the current environment and the device adjustment vector, giving full play to the fitting ability of the neural network, realizing the precise adjustment of the device position or the device current, so that the beam uniformity can meet the actual requirements (such as the target process requirements) after a small number of device adjustments.

[0040] The embodiment of the present invention can obtain historical adjustment data, and the historical adjustment data includes multiple historical state vectors of the target ion implanter, and one historical state vector includes the beam state vector and the device state vector of the target ion implanter at a historical acquisition moment; based on this, the current training state vector can be determined from the historical adjustment data, and the application adjustment model can be called to determine the device adjustment vector corresponding to the current training state vector. Then, correspondingly, the adjustment model loss value of the application adjustment model can be determined based on the device adjustment vector; and the model parameters in the application adjustment model are updated in the direction of reducing the adjustment model loss value. Further, the above determination of the current training state vector from the historical adjustment data can be iteratively executed until the target model training conditions are met, so as to complete the model training of the application adjustment model through the historical adjustment data, and the application adjustment model support is used to guide the beam uniformity adjustment of the target ion implanter. It can be seen that the embodiment of the present invention can train the application adjustment model through historical adjustment data, so as to obtain a pre-trained application adjustment model, which can effectively improve the model performance of the application adjustment model, and then the application adjustment model with better model performance can be used to adjust the beam uniformity of the target ion implanter, which can effectively improve the accuracy of the adjustment, that is, the current device adjustment vector with higher accuracy can be obtained through the application adjustment model, so as to effectively improve the adjustment efficiency of the beam uniformity.

[0041] Based on the above description, the embodiment of the present invention further proposes a more specific beam uniformity adjustment method. Accordingly, the beam uniformity adjustment method can be executed by the electronic device (terminal or server) mentioned above; or, the beam uniformity adjustment method can be executed by the terminal and the server together. For the sake of convenience, the following description will be given by taking the electronic device executing the beam uniformity adjustment method as an example; please refer to Figure 3 The beam uniformity adjustment method may include the following steps S301-S307: S301, acquiring historical adjustment data, where the historical adjustment data includes a plurality of historical state vectors of a target ion implanter, and one historical state vector includes a beam state vector and a device state vector of the target ion implanter at a historical acquisition moment.

[0042] S302, determining a current training state vector from historical adjustment data, and calling an application adjustment model to determine a device adjustment vector corresponding to the current training state vector.

[0043] S303, determining an updated beam state vector based on the device adjustment vector, and determining a current updated state vector corresponding to the current training state vector based on the updated beam state vector.

[0044] Among them, the updated beam state vector can be an updated beam state vector corresponding to the device adjustment vector, that is, it can be the beam state vector adjusted according to the device adjustment vector; exemplarily, in the first model training stage, the updated beam state vector can be determined based on the device adjustment vector through a simulation system; in the second model training stage, the device adjustment vector can be output (such as displaying the device adjustment vector or sending the device adjustment vector to a target ion implanter, etc.) so that the target ion implanter is adjusted according to the device adjustment vector. Then, after the target ion implanter is adjusted according to the device adjustment vector, the electronic device can obtain the current beam state vector measured by the angle Faraday cup (that is, the beam state vector measured after adjustment according to the device adjustment vector) to use the currently obtained beam state vector as the updated beam state vector to determine the updated beam state vector, and so on.

[0045] Optionally, the current updated state vector corresponding to the current training state vector may include an updated beam state vector and an updated device state vector. Based on this, when determining the current updated state vector corresponding to the current training state vector based on the updated beam state vector, the electronic device may determine the updated device state vector corresponding to the device adjustment vector. For example, when an adjustment information is an adjustment result information (i.e., the adjusted device information), the device adjustment vector may be used as the corresponding updated device state vector. When an adjustment information is an adjustment action information, the device state vector in the current training state vector may be adjusted according to the device adjustment vector (such as moving the device information of any device in the beam direction (such as the beam forward direction) by 2 mm, etc.), to obtain an updated device state vector, etc.; accordingly, the updated beam state vector and the updated device state vector may be spliced ​​to obtain the current updated state vector corresponding to the current training state vector, so as to determine the current updated state vector corresponding to the current training state vector based on the updated beam state vector.

[0046] S304, determining a current training tuple based on the current training state vector and the current update state vector; and determining at least one training sample based on the current training tuple.

[0047] Optionally, when determining the current training tuple based on the current training state vector and the current update state vector, the electronic device may also determine the benefit information (i.e., the above-mentioned beam uniformity change information) and process status information corresponding to the current update state vector, and use the current training state vector, the current update state vector, the device adjustment vector, and the benefit information and process status information corresponding to the current update state vector to determine the current training tuple. In other words, the benefit information and process status information corresponding to the current training state vector, the current update state vector, the device adjustment vector, and the current update state vector may be added to the current training tuple to achieve determination of the current training tuple.

[0048] Optionally, when determining the benefit information corresponding to the current updated state vector, the difference between the beam uniformity indication value corresponding to the beam state vector in the current updated state vector (i.e., the above-mentioned updated beam state vector) and the beam uniformity indication value corresponding to the beam state vector in the current training state vector can be used as the benefit information corresponding to the current updated state vector.

[0049] Optionally, when determining the process state information corresponding to the current updated state vector, the electronic device may determine whether the beam uniformity indication value corresponding to the updated beam state vector is less than or equal to a preset beam uniformity threshold; when the beam uniformity indication value corresponding to the updated beam state vector is less than or equal to the preset beam uniformity threshold, the process state information corresponding to the current updated state vector may be determined to be indication information that meets the target process requirement; when the beam uniformity indication value corresponding to the updated beam state vector is greater than the preset beam uniformity threshold, the process state information corresponding to the current updated state vector may be determined to be indication information that does not meet the target process requirement. Among them, the indication information that meets the target process requirement may be used to indicate that the beam uniformity meets the target process requirement, and the indication information that does not meet the target process requirement may be used to indicate that the beam uniformity does not meet the target process requirement, and the target process requirement may be a process requirement that the beam uniformity indication value meets (i.e., is less than or equal to) the preset beam uniformity threshold, and the beam uniformity indication value may be used to indicate beam uniformity, i.e., the target process requirement may refer to a process requirement that the beam uniformity meets the beam uniformity indicated by the preset beam uniformity threshold. Optionally, the preset beam uniformity threshold, the target process requirement satisfaction indication information, and the target process requirement failure indication information may be set based on experience or actual needs, which is not limited in the embodiment of the present invention.

[0050] Based on this, when the process state information corresponding to the currently updated state vector is indication information that meets the target process requirements, the process state information corresponding to the currently updated state vector can be used to indicate that the beam uniformity of the beam state vector in the currently updated state vector meets the target process requirements; when the process state information corresponding to the currently updated state vector is indication information that does not meet the target process requirements, the process state information corresponding to the currently updated state vector can be used to indicate that the beam uniformity of the beam state vector in the currently updated state vector does not meet the target process requirements.

[0051] Further, the electronic device may add the current training tuple to the experience replay set; optionally, before the first model training phase or the second model training phase begins, the electronic device may initially set the experience replay set to be empty, that is, clear the experience replay set. Based on this, the electronic device may determine at least one training sample from the experience replay set to determine at least one training sample based on the current training tuple; optionally, the number of training samples in at least one training sample may be m (m is a positive integer), in which case the electronic device may randomly select m training tuples from the experience replay set to use each selected training tuple as a training sample to determine at least one training sample. Based on this, a training sample can be a training tuple, that is, it can be a training tuple in an experience replay set; wherein, the experience replay set can be used to store the training tuples obtained in each training process (such as the current training tuple mentioned above), and a training tuple may include a training state vector, a device adjustment vector corresponding to the corresponding training state vector, an update state vector corresponding to the corresponding training state vector (that is, an update state vector corresponding to the device adjustment vector corresponding to the corresponding training state vector), profit information corresponding to the corresponding update state vector (that is, an update state vector corresponding to the corresponding training state vector), and process state information corresponding to the corresponding update state vector. In other words, a training sample may include a training state vector, a device adjustment vector, an update state vector, profit information, and process state information.

[0052] S305 , respectively calling the application scoring model to determine the application scoring value corresponding to each training sample in the at least one training sample; and calculating the adjustment model loss value of the application adjustment model based on the application scoring value corresponding to each training sample.

[0053] Optionally, a scoring model can be a neural network model, that is, the application scoring model can be a neural network model; it should be noted that the embodiment of the present invention does not limit the specific model structure of the scoring model, that is, the model structure of the application scoring model can be set according to experience or according to actual needs, and the embodiment of the present invention does not limit this. Optionally, before using the application scoring model for the first time, the electronic device can randomly initialize the application scoring model, that is, it can randomly initialize the model parameters in the application scoring model; or, the model parameters in the application scoring model can be set according to experience or actual needs, and so on.

[0054] In an embodiment of the present invention, when calling the application scoring model separately to determine the application scoring value corresponding to each training sample in at least one training sample, for any training sample in the at least one training sample, the electronic device may splice the training state vector and the device adjustment vector in any training sample to obtain a scoring estimation vector corresponding to any training sample, and may call the application scoring model (i.e., the current application scoring model) to determine the application scoring value corresponding to any training sample based on the scoring estimation vector corresponding to any training sample. That is to say, the scoring estimation vector corresponding to any training sample may be input into the application scoring model to output the application scoring value corresponding to any training sample through the application scoring model, thereby realizing the calling of the application scoring model to determine the application scoring value corresponding to any training sample.

[0055] Optionally, the electronic device may also determine the target scoring value corresponding to each training sample respectively, and determine the scoring model loss value of the application scoring model based on the target scoring value and application scoring value corresponding to each training sample; based on this, the model parameters in the application scoring model may be updated in the direction of reducing the scoring model loss value, that is, the electronic device may also update the model parameters in the application scoring model through at least one training sample. In an embodiment of the present invention, the electronic device may first update the model parameters in the application scoring model through at least one training sample, and then update the model parameters in the application adjustment model through at least one training sample (that is, at this time, before step S305, the above-mentioned determination of the target scoring value corresponding to each training sample may be triggered, and the scoring model loss value of the application scoring model may be determined based on the target scoring value and the application scoring value corresponding to each training sample, thereby updating the model parameters in the application scoring model). In this case, the application scoring value corresponding to each training sample used to calculate the adjustment model loss value may be determined by the application scoring model updated by at least one training sample, that is, the current application scoring model when determining the application scoring value corresponding to each training sample used to calculate the adjustment model loss value may be the application scoring model updated by at least one training sample; based on this, when determining the scoring model loss value of the application scoring model based on the target scoring value and the application scoring value corresponding to each training sample, the electronic device may also call the current application scoring model (here, referring to the application scoring model that has not been updated by the current at least one training sample) to determine the application scoring value corresponding to each training sample, respectively, so that the scoring model loss value of the application scoring model may be determined based on the target scoring value and the application scoring value corresponding to each training sample. It should be understood that when an electronic device calls any model, it means calling any current model, and the model parameters in any current model are the latest values. For example, after the model parameters in any model are updated, the model parameters in any current model can be the updated model parameters.

[0056] Optionally, a training tuple also includes revenue information and process state information corresponding to the updated state vector in the corresponding training tuple, and a training sample is a training tuple; based on this, when determining the target score value corresponding to each training sample respectively, for any training sample in at least one training sample, the electronic device can judge whether the beam uniformity corresponding to any training sample (the beam uniformity corresponding to any training sample may refer to the beam uniformity corresponding to the beam state vector included in the updated state vector in any training sample) meets the target process requirement based on the process state information in any training sample; if the beam uniformity corresponding to any training sample meets the target process requirement, the revenue information in any training sample is used as the target score value corresponding to any training sample; if the beam uniformity corresponding to any training sample does not meet the target process requirement, the reference score model can be called to determine the reference score value corresponding to any training sample based on the updated state vector in any training sample, and the target score value corresponding to any training sample can be determined based on the revenue information in any training sample and the reference score value corresponding to any training sample. Optionally, the electronic device may determine a reference scoring coefficient, and use the revenue information in any training sample, the reference scoring coefficient, and the reference scoring value corresponding to any training sample to determine the target scoring value corresponding to any training sample; optionally, the reference scoring coefficient may be multiplied by the reference scoring value corresponding to any training sample to obtain a multiplication result, and the sum of the revenue information in any training sample and the multiplication result may be used as the target scoring value corresponding to any training sample. Optionally, the reference scoring coefficient may be set according to experience or according to actual needs, and the embodiment of the present invention is not limited to this.

[0057] Optionally, when calling the reference scoring model and determining the reference scoring value corresponding to any training sample based on the updated state vector in any training sample, the electronic device may call the reference adjustment model to determine the updated device adjustment vector corresponding to the updated state vector in any training sample (that is, the updated state vector in any training sample may be input into the reference adjustment model (that is, the current reference adjustment model) to output the updated device adjustment vector corresponding to the updated state vector in any training sample through the reference adjustment model), and may call the reference scoring model to determine the reference scoring value corresponding to any training sample based on the updated device adjustment vector (that is, the updated device adjustment vector may be input into the reference scoring model (that is, the current reference scoring model) to output the reference scoring value corresponding to any training sample through the reference scoring model).

[0058] Optionally, the model structure of the reference adjustment model may be the same as the model structure of the application adjustment model, and the model parameters in the reference adjustment model may be initialized to the initialization results of the model parameters in the application adjustment model (also referred to as the initialized model parameters), that is, before model training, the initialized model parameters in the application adjustment model may be used as the initialized model parameters in the reference adjustment model. Optionally, the model structure of the reference scoring model may be the same as the model structure of the application scoring model, and the model parameters in the reference scoring model may be initialized to the initialization results of the model parameters in the application scoring model, that is, before model training, the initialized model parameters in the application scoring model may be used as the initialized model parameters in the reference scoring model.

[0059] Optionally, assuming that the number of training samples in the at least one training sample is m, when determining the scoring model loss value of the application scoring model based on the target scoring value and the application scoring value corresponding to each training sample, the electronic device may use formula 1.1 to determine the scoring model loss value of the application scoring model: Formula 1.1 Where L can represent the scoring model loss function, which is used to calculate the scoring model loss value, y j It can represent the target score value corresponding to the jth training sample in at least one training sample, Q(S j ,A j ,b) can represent the application score value corresponding to the jth training sample, S j It can represent the training state vector in the jth training sample, A j can represent the device adjustment vector in the jth training sample, and b can represent the model parameters of the application scoring model; that is, the training state vector and the device adjustment vector in the jth training sample can be spliced ​​to obtain the score estimation vector corresponding to the jth training sample, and the application scoring model can be called to determine the application scoring value corresponding to the jth training sample based on the score estimation vector corresponding to the jth training sample. Based on this, the electronic device can calculate the gradient through the back propagation algorithm to update the model parameters in the application scoring model in the direction of reducing the loss value of the scoring model.

[0060] Optionally, in other embodiments, the application score value corresponding to each training sample used to calculate the adjustment model loss value may also be determined by an application score model that has not been updated by at least one current training sample, and the present invention is not limited to this.

[0061] Optionally, when calculating the adjustment model loss value of the application adjustment model based on the application score value corresponding to each training sample, the electronic device may use formula 1.2 to calculate the adjustment model loss value of the application adjustment model: Formula 1.2 Wherein, a may represent a model parameter in the application adjustment model, and J may represent an adjustment model loss function for calculating the adjustment model loss value. Based on this, the electronic device may calculate the gradient through a back propagation algorithm to implement the following update of the model parameters in the application adjustment model in the direction of reducing the adjustment model loss value.

[0062] S306: Update the model parameters in the application adjustment model in a direction of reducing the loss value of the adjustment model.

[0063] In an embodiment of the present invention, the adjustment model loss value may be determined by applying a scoring model, the model parameters in the application scoring model support being updated, and the scoring model loss value of the application scoring model may be determined based on at least one of the application scoring model, the reference adjustment model, and the reference scoring model. Based on this, after each update of the application adjustment model and the model parameters in the application scoring model, the electronic device may also determine whether the reference update condition is met; if the reference update condition is met, the model parameters in the reference adjustment model may be updated based on the adjustment smoothing coefficient and the model parameters in the application adjustment model, and the model parameters in the reference scoring model may be updated based on the scoring smoothing coefficient and the model parameters in the application scoring model; if the reference update condition is not met, the model parameters in the reference adjustment model and the reference scoring model are not updated. Optionally, the electronic device may determine whether the reference update condition is met based on the current number of iterations; illustratively, when the remainder between the current number of iterations and the first value is equal to the second value (i.e., the current number of iterations% the first value == the second value), it may be determined that the reference update condition is met, and when the remainder between the current number of iterations and the first value is not equal to the second value, it may be determined that the reference update condition is not met, and so on; the embodiment of the present invention is not limited to this. Optionally, both the first value and the second value may be set according to experience or according to actual needs, which is not limited in the embodiment of the present invention.

[0064] Optionally, the electronic device may update the model parameters in the reference adjustment model and update the model parameters in the reference scoring model based on Formula 1.3: Formula 1.3 Wherein, b' may represent the model parameters in the reference scoring model, a' may represent the model parameters in the reference adjustment model, It can represent the score smoothing coefficient, It can represent the adjustment smoothing coefficient. Optionally, the adjustment smoothing coefficient and the scoring smoothing coefficient can be set according to experience or according to actual needs, which is not limited in the embodiment of the present invention; Optionally, the adjustment smoothing coefficient and the scoring smoothing coefficient can be the same or different, which is not limited in the embodiment of the present invention.

[0065] S307, iteratively executing to determine the current training state vector from the historical adjustment data until the target model training conditions are met, so as to complete the model training of the application adjustment model through the historical adjustment data, and the application adjustment model support is used to guide the beam uniformity adjustment of the target ion implanter.

[0066] Optionally, after adding the current training tuple to the experience playback set, the electronic device may also use the current updated state vector as the current training state vector, that is, the current training state vector may be updated to the current updated state vector. Based on this, after completing a model training process (i.e., completing a parameter update of an application adjustment model and an application scoring model, and / or satisfying a reference update condition to complete the parameter update of a reference adjustment model and a reference scoring model), the electronic device may also determine whether the current training state vector has completed the current model training process. If it is determined that the current training state vector has completed the current model training process, the above-mentioned iterative execution may be triggered to determine the current training state vector from the historical adjustment data. That is, at this time, the current training state vector may be re-determined from the historical adjustment data, such as the next training state vector in the historical adjustment data may be used as the current training state vector; if it is determined that the current training state vector has not completed the current model training process, the application adjustment model (i.e., the current application adjustment model) may be continued to be called to determine the device adjustment vector corresponding to the current training state vector to continue updating the model parameters in the application adjustment model, etc., until it is determined that the current training state vector has completed the current model training process.

[0067] Optionally, when determining whether the current training state vector completes the current model training process, the electronic device may determine whether the beam uniformity corresponding to the beam state vector in the current training state vector (which may be the above-mentioned updated beam state vector here) meets the target process requirements; if it is determined that the beam uniformity corresponding to the beam state vector in the previous training state vector meets the target process requirements, then it may be determined that the current training state vector completes the current model training process; if it is determined that the beam uniformity corresponding to the beam state vector in the previous training state vector does not meet the target process requirements, then it may be determined that the current training state vector does not complete the current model training process. Optionally, when judging whether the current training state vector completes the current model training process, the electronic device may judge whether the beam uniformity corresponding to the beam state vector in the current training state vector meets the target process requirement, and may judge whether the beam uniformity indication value corresponding to the updated beam state vector is less than or equal to a preset beam uniformity threshold; or, may judge whether the current process state information (i.e., the process state information in the current training tuple) is indication information that meets the target process requirement; or, when the indication information that meets the target process requirement is greater than 1 and the indication information that does not meet the target process requirement is equal to 0, it may judge whether the current process state information is true, so as to judge whether the beam uniformity corresponding to the beam state vector in the current training state vector meets the target process requirement, and so on; the embodiments of the present invention are not limited to this.

[0068] Further, after the target model training conditions are met, the electronic device can obtain the current beam state vector in the target ion implanter at every preset detection time to determine the current beam uniformity in the target ion implanter (i.e., calculate the current beam uniformity), that is, the beam uniformity indicator value corresponding to the current beam state vector can be calculated to indicate the current beam uniformity through the beam uniformity indicator value, so as to detect whether the current beam uniformity in the target ion implanter meets the target process requirements, that is, to detect whether the beam uniformity indicator value corresponding to the current beam state vector is less than or equal to the preset beam uniformity threshold, so as to detect whether the current beam uniformity in the target ion implanter meets the target process requirements (i.e., determine whether the current beam uniformity meets the requirements). Optionally, the preset detection time can be set according to experience or according to actual needs, and the embodiment of the present invention is not limited to this. Optionally, the current beam state vector can be measured by moving the Faraday cup (i.e., it can be collected by moving the Faraday cup) to obtain an accurate current beam state vector. Optionally, when it is detected that the current beam uniformity meets the target process requirements, the following uniformity adjustment step may not be performed, that is, the following calling application adjustment model and determining the current device adjustment vector corresponding to the current state vector may not be triggered.

[0069] Optionally, when it is detected that the current beam uniformity in the target ion implanter does not meet the target process requirements, the electronic device may also determine the current state vector, which includes the beam state vector and the device state vector of the target ion implanter at the current acquisition moment, such as the electronic device may also obtain the current device state vector to determine the current state vector based on the current beam state vector and the current device state vector. Based on this, the application adjustment model may be called to determine the current device adjustment vector corresponding to the current state vector, and the current device adjustment vector is used to guide the adjustment of at least one device in the target ion implanter; optionally, the electronic device may output the current device adjustment vector (such as displaying the current device adjustment vector) so that the target object (such as the manager of the target ion implanter, etc.) performs a device adjustment operation on the target ion implanter according to the current device adjustment vector, so as to adjust at least one device in the target ion implanter based on the current device adjustment vector, or the electronic device may send the current device adjustment vector to the target ion implanter so that the target ion implanter adjusts at least one device according to the current device adjustment vector, etc., and the embodiments of the present invention are not limited to this.

[0070] Furthermore, the electronic device may iteratively execute the above-mentioned determination of the current state vector, that is, after the target ion implanter is adjusted, the current beam state vector and the current device state vector may be reacquired to iteratively determine the current state vector, thereby obtaining the current device adjustment vector for guiding the adjustment of the target ion implanter, so as to realize continuous adjustment of the target ion implanter until the current beam uniformity in the target ion implanter meets the target process requirements; wherein, in the process of adjusting at least one device in the target ion implanter, the application of the adjustment model supports parameter updates when the model update conditions are met.

[0071] Optionally, before calling the application adjustment model and determining the device adjustment vector corresponding to the current state vector, the electronic device may determine whether the application adjustment model meets the model update condition; if it is determined that the application adjustment model meets the model update condition, the adjustment model loss value of the application adjustment model (i.e., the current adjustment model loss value) and the scoring model loss value of the application scoring model (i.e., the current scoring model loss value) may be determined based on the current state vector and the previous state vector of the current state vector (i.e., the state vector at the previous acquisition time before the current acquisition time), so as to optimize the model parameters in the application adjustment model in the direction of reducing the adjustment model loss value, and optimize the model parameters in the application scoring model in the direction of reducing the scoring model loss value, so as to implement parameter update of the application adjustment model and the application scoring model; accordingly, when it is determined that the reference update condition is met, the model parameters in the reference adjustment model may be updated based on the adjustment smoothing coefficient and the model parameters in the application adjustment model; and the model parameters in the reference scoring model may be updated based on the scoring smoothing coefficient and the model parameters in the application scoring model, and so on. Based on this, the embodiment of the present invention can implement model training in the third model training stage to further improve the model performance of the model. Optionally, if it is determined that the application adjustment model does not satisfy the model update condition, the application adjustment model may be directly triggered to be executed to determine the device adjustment vector corresponding to the current state vector.

[0072] Optionally, when determining the adjustment model loss value of the application adjustment model and the scoring model loss value of the application scoring model based on the current state vector and the previous state vector of the current state vector, the electronic device may determine the current benefit information corresponding to the current state vector (i.e., the difference between the beam uniformity indicator value corresponding to the beam state vector in the current state vector and the beam uniformity indicator value corresponding to the beam state vector in the previous state vector), and judge whether the beam uniformity corresponding to the beam state vector in the current state vector meets the target process requirement; if the beam uniformity corresponding to the beam state vector in the current state vector meets the target process requirement, the current benefit information is used as the current target scoring value; if the beam uniformity corresponding to the beam state vector in the current state vector does not meet the target process requirement, the reference adjustment model is called to determine the current device adjustment vector corresponding to the current state vector, and the reference scoring model is called to determine the current reference scoring value based on the current state vector and the current device adjustment vector corresponding to the current state vector (the same as the determination method of the reference scoring value mentioned above, which will not be repeated in the embodiment of the present invention), and then the current target scoring value may be determined based on the current benefit information and the current reference scoring value. Furthermore, the difference between the current target score value and the application score value corresponding to the previous state vector may be used to determine the score model loss value of the application score model. Accordingly, the adjustment model loss value of the application adjustment model may be determined based on the application score value corresponding to the previous state vector. In an embodiment of the present invention, the determination method of the score model loss value and the adjustment model loss value may be the same as the determination method in the first model training stage and the second model training stage, and the embodiment of the present invention will not be repeated here.

[0073] Optionally, when judging whether the application adjustment model satisfies the model update condition, it can be determined whether the previous state vector exists. If the previous state vector exists, it can be determined that the application adjustment model satisfies the model update condition. If the previous state vector does not exist, it can be determined that the application adjustment model does not satisfy the model update adjustment. Alternatively, when it is first detected that the current beam uniformity in the target ion implanter does not meet the target process requirements every preset detection time, it can be determined that the current number of iterations is 0, and after adjusting the target ion implanter once, the current number of iterations can be increased by 1 to update the current number of iterations. In this case, it can be determined whether the current number of iterations is equal to 0. If the current number of iterations is equal to 0, it can be determined that the application adjustment model does not satisfy the model update condition. If the current number of iterations is not equal to 0, it can be determined that the application adjustment model satisfies the model update condition, so as to judge whether the application adjustment model satisfies the model update condition. Figure 4 As shown, etc.; the embodiments of the present invention are not limited to this. Among them, Figure 4 The count in it indicates the current number of iterations.

[0074] In the embodiment of the present invention, since the target ion implanter can be any ion implanter, that is, the embodiment of the present invention does not limit at least one device in the target ion implanter, which can be a multi-coil array and / or a multi-pole array; based on this, the embodiment of the present invention can have good generalizability, and is applicable to the adjustment of multi-coil arrays and multi-pole arrays, and is also applicable to various modes such as symmetric adjustment and free adjustment of magnetic poles in multi-pole arrays. For example, Figure 5 As shown, the multi-pole array and the multi-coil array are respectively located at the front end and the rear end of the parallel lens beam forward direction; in practical applications, the multi-coil array controls the magnetic field in the optical path by changing the current of each coil, and the multi-pole array controls the magnetic field by changing the distance between it and the beam; although the control principles are different, the ideas of regulating the magnetic field are the same. The regulation method proposed in the present invention is applicable to both devices, and the difference lies only in the information related to the device position or current size in the device state vector and the device adjustment vector.

[0075] In summary, it can be seen that the embodiments of the present invention may involve neural network models such as application adjustment models, and the model parameters in these neural network models can be updated during the actual adjustment process. That is, when it is detected in the actual application process that the current beam uniformity in the target ion implanter does not meet the target process requirements and the model update conditions are met, the model training in the third model training stage is implemented, so that the application adjustment model can give the most accurate device adjustment vector according to the actual situation of the current beam. It should be noted that the embodiment of the present invention proposes a framework method, in which the reinforcement learning model involved can be independently selected during use, such as the reinforcement learning algorithm such as double-delay deep deterministic policy gradient can also be used for model training; illustratively, the number of application scoring models and reference scoring models can be multiple, that is, the embodiment of the present invention can perform model training through an application adjustment model, multiple application scoring models, a reference adjustment model and multiple reference scoring models, such as determining a target score value through the minimum value of the reference score values ​​under each reference scoring model in multiple reference scoring models, so as to determine the score model loss value under the corresponding application scoring model based on the target score value and the application score value under each application scoring model (such as the application score value corresponding to each training sample under any application scoring model) to update the corresponding application scoring model, and determining the adjustment model loss value of the application adjustment model through the mean value between the application score values ​​under each application scoring model (such as the mean value between the application score values ​​corresponding to any training sample under each application scoring model can be the application score value corresponding to any training sample), and so on; the embodiment of the present invention is not limited to this.

[0076] The embodiment of the present invention can obtain historical adjustment data, which includes multiple historical state vectors of the target ion implanter, and one historical state vector includes the beam state vector and the device state vector of the target ion implanter at a historical acquisition moment; so as to determine the current training state vector from the historical adjustment data, and call the application adjustment model to determine the device adjustment vector corresponding to the current training state vector. Then, the updated beam state vector can be determined based on the device adjustment vector, and the current updated state vector corresponding to the current training state vector can be determined based on the updated beam state vector. Based on this, the current training tuple can be determined based on the current training state vector and the current updated state vector, and at least one training sample can be determined based on the current training tuple; thus, the application scoring model can be called respectively to determine the application scoring value corresponding to each training sample in at least one training sample, and the adjustment model loss value of the application adjustment model can be calculated based on the application scoring value corresponding to each training sample; so as to update the model parameters in the application adjustment model in the direction of reducing the adjustment model loss value. Furthermore, it is possible to iteratively determine the current training state vector from the historical adjustment data until the target model training conditions are met, so as to complete the model training of the application adjustment model through the historical adjustment data, and the application adjustment model supports the use of the historical adjustment data to guide the beam uniformity adjustment of the target ion implanter. It can be seen that the embodiment of the present invention can effectively improve the model performance through the historical adjustment data, so that the relationship between the state vector and the device adjustment vector can be better fitted through the application adjustment model, so as to accurately obtain the device adjustment vector, so that the beam uniformity meets the target process requirements after a smaller number of device adjustments; and the embodiment of the present invention proposes a three-step parameter training method for the reinforcement learning model (such as including the above-mentioned first model training stage, the second model training stage and the third model training stage), which can make full use of the historical adjustment data of uniformity adjustment and effectively improve the training speed of the model.

[0077] Based on the description of the above-mentioned embodiments of the beam uniformity adjustment method, the embodiment of the present invention further proposes a beam uniformity adjustment device, which may be a computer program (including program code) running in an electronic device; Figure 6 As shown, the beam uniformity adjustment device may include an acquisition unit 601 and a processing unit 602. The beam uniformity adjustment device may perform Figure 1 or Figure 3 The beam uniformity adjustment method shown, that is, the beam uniformity adjustment device can operate the above-mentioned unit: An acquisition unit 601 is used to acquire historical adjustment data, wherein the historical adjustment data includes a plurality of historical state vectors of a target ion implanter, and a historical state vector includes a beam state vector and a device state vector of the target ion implanter at a historical acquisition moment; The processing unit 602 is used to determine a current training state vector from the historical adjustment data, and call the application adjustment model to determine a device adjustment vector corresponding to the current training state vector; The processing unit 602 is further configured to determine a regulation model loss value of the application regulation model based on the device regulation vector; and update the model parameters in the application regulation model in a direction of reducing the regulation model loss value; The processing unit 602 is also used to iteratively execute the determination of the current training state vector from the historical adjustment data until the target model training conditions are met, so as to complete the model training of the application adjustment model through the historical adjustment data, and the application adjustment model supports guiding the beam uniformity adjustment of the target ion implanter.

[0078] In one implementation, when the processing unit 602 determines the adjustment model loss value of the application adjustment model based on the device adjustment vector, it may be specifically configured to: Determining an updated beam state vector based on the device adjustment vector, and determining a current updated state vector corresponding to the current training state vector based on the updated beam state vector; Determine a current training tuple based on the current training state vector and the current updated state vector; and determine at least one training sample based on the current training tuple; The application scoring models are respectively called to determine the application scoring value corresponding to each training sample in the at least one training sample; and based on the application scoring value corresponding to each training sample, the adjustment model loss value of the application adjustment model is calculated.

[0079] In another implementation, the processing unit 602 may also be used to: Determine the target scoring value corresponding to each training sample respectively, and determine the scoring model loss value of the application scoring model based on the target scoring value and the application scoring value corresponding to each training sample; The model parameters in the application scoring model are updated in a direction of reducing the loss value of the scoring model.

[0080] In another implementation, a training tuple further includes revenue information and process state information corresponding to the updated state vector in the corresponding training tuple, and a training sample is a training tuple; when the processing unit 602 determines the target score value corresponding to each training sample respectively, it can be specifically used to: For any training sample among the at least one training sample, based on the process state information in the any training sample, determining whether the beam uniformity corresponding to the any training sample meets the target process requirement; If the beam uniformity corresponding to any of the training samples meets the target process requirement, the revenue information in any of the training samples is used as the target score value corresponding to any of the training samples; If the beam uniformity corresponding to any of the training samples does not meet the target process requirements, the reference scoring model is called to determine the reference scoring value corresponding to any of the training samples based on the updated state vector in any of the training samples; and the target scoring value corresponding to any of the training samples is determined based on the revenue information in any of the training samples and the reference scoring value corresponding to any of the training samples.

[0081] In another embodiment, the target model training conditions include model training conditions in a first model training stage and / or model training conditions in a second model training stage; the updated beam state vector is determined according to a method for determining an updated beam state vector in the first model training stage or a method for determining an updated beam state vector in the second model training stage, the accuracy of the method for determining the updated beam state vector in the second model training stage is higher than the accuracy of the method for determining the updated beam state vector in the first model training stage, and the first model training stage precedes the second model training stage.

[0082] In another implementation, the adjustment model loss value is determined by applying a scoring model, the model parameters in the application scoring model support being updated, and the scoring model loss value of the application scoring model is determined based on at least one of the application scoring model, the reference adjustment model, and the reference scoring model, and the processing unit 602 may also be used to: After each update of the model parameters in the application adjustment model and the application scoring model, determining whether a reference update condition is met; If the reference update condition is met, the model parameters in the reference adjustment model are updated based on the adjustment smoothing coefficient and the model parameters in the application adjustment model; and the model parameters in the reference scoring model are updated based on the scoring smoothing coefficient and the model parameters in the application scoring model; If the reference update condition is not met, the model parameters in the reference adjustment model and the reference scoring model are not updated.

[0083] In another implementation, the processing unit 602 may also be used to: When it is detected that the current beam uniformity in the target ion implanter does not meet the target process requirement, determining a current state vector, wherein the current state vector includes a beam state vector and a device state vector of the target ion implanter at the current acquisition time; Calling the application adjustment model to determine a current device adjustment vector corresponding to the current state vector, wherein the current device adjustment vector is used to guide adjustment of at least one device in the target ion implanter; The determination of the current state vector is iteratively performed until the current beam uniformity in the target ion implanter meets the target process requirement; wherein, in the process of adjusting at least one device in the target ion implanter, the application adjustment model supports parameter update when the model update condition is met.

[0084] According to one embodiment of the present invention, Figure 6 Each unit in the beam uniformity adjustment device shown can be separately or completely combined into one or several other units to form a structure, or one (or some) of the units can be further divided into multiple functionally smaller units to form a structure, which can achieve the same operation without affecting the realization of the technical effects of the embodiments of the present invention. The above-mentioned units are divided based on logical functions. In practical applications, the function of one unit can also be realized by multiple units, or the functions of multiple units can be realized by one unit. In other embodiments of the present invention, any beam uniformity adjustment device can also include other units. In practical applications, these functions can also be implemented with the assistance of other units, and can be implemented by the collaboration of multiple units.

[0085] According to another embodiment of the present invention, the program can be executed by running a program on a general electronic device such as a computer including a central processing unit (CPU), a random access memory medium (RAM), a read-only memory medium (ROM), and other processing elements and storage elements. Figure 1 or Figure 3 A computer program (including program code) for each step involved in the corresponding method shown in Figure 6 The beam uniformity adjustment device shown in the embodiment of the present invention is used to implement the beam uniformity adjustment method of the embodiment of the present invention. The computer program can be recorded on a computer storage medium, for example, and loaded into the above electronic device through the computer storage medium and run therein.

[0086] The embodiment of the present invention can obtain historical adjustment data, and the historical adjustment data includes multiple historical state vectors of the target ion implanter, and one historical state vector includes the beam state vector and the device state vector of the target ion implanter at a historical acquisition moment; based on this, the current training state vector can be determined from the historical adjustment data, and the application adjustment model can be called to determine the device adjustment vector corresponding to the current training state vector. Then, correspondingly, the adjustment model loss value of the application adjustment model can be determined based on the device adjustment vector; and the model parameters in the application adjustment model are updated in the direction of reducing the adjustment model loss value. Further, the above determination of the current training state vector from the historical adjustment data can be iteratively executed until the target model training conditions are met, so as to complete the model training of the application adjustment model through the historical adjustment data, and the application adjustment model support is used to guide the beam uniformity adjustment of the target ion implanter. It can be seen that the embodiment of the present invention can train the application adjustment model through historical adjustment data, so as to obtain a pre-trained application adjustment model, which can effectively improve the model performance of the application adjustment model, and then the application adjustment model with better model performance can be used to adjust the beam uniformity of the target ion implanter, which can effectively improve the accuracy of the adjustment, that is, the current device adjustment vector with higher accuracy can be obtained through the application adjustment model, so as to effectively improve the adjustment efficiency of the beam uniformity.

[0087] Based on the description of the above method embodiment and device embodiment, the exemplary embodiment of the present invention further provides an electronic device, including: at least one processor; and a memory connected to the at least one processor in communication. The memory stores a computer program that can be executed by the at least one processor, and the computer program is used to enable the electronic device to perform the method according to the embodiment of the present invention when executed by the at least one processor.

[0088] Exemplary embodiments of the present invention also provide a non-transitory computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor of a computer, is used to cause the computer to perform a method according to an embodiment of the present invention.

[0089] An exemplary embodiment of the present invention further provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor of a computer, the computer is used to enable the computer to perform a method according to an embodiment of the present invention.

[0090] refer to Figure 7, a block diagram of an electronic device 700 that can be used as a server or client of the present invention will now be described, which is an example of a hardware device that can be applied to various aspects of the present invention. The electronic device is intended to represent various forms of digital electronic computer devices, such as laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0091] like Figure 7 As shown, the electronic device 700 includes a computing unit 701, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 702 or a computer program loaded from a storage unit 708 into a random access memory (RAM) 703. In the RAM 703, various programs and data required for the operation of the electronic device 700 can also be stored. The computing unit 701, the ROM 702, and the RAM 703 are connected to each other via a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0092] A plurality of components in the electronic device 700 are connected to the I / O interface 705, including: an input unit 706, an output unit 707, a storage unit 708, and a communication unit 709. The input unit 706 may be any type of device capable of inputting information to the electronic device 700, and the input unit 706 may receive input digital or character information, and generate key signal inputs related to user settings and / or function control of the electronic device. The output unit 707 may be any type of device capable of presenting information, and may include but is not limited to a display, a speaker, a video / audio output terminal, a vibrator, and / or a printer. The storage unit 708 may include but is not limited to a disk, an optical disk. The communication unit 709 allows the electronic device 700 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks, and may include but is not limited to a modem, a network card, an infrared communication device, a wireless communication transceiver, and / or a chipset, such as a Bluetooth™ device, a WiFi device, a WiMax device, a cellular communication device, and / or the like.

[0093] The computing unit 701 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 701 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 701 performs the various methods and processes described above. For example, in some embodiments, the beam uniformity adjustment method may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 708. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 700 via the ROM 702 and / or the communication unit 709. In some embodiments, the computing unit 701 may be configured to perform the beam uniformity adjustment method in any other appropriate manner (e.g., by means of firmware).

[0094] Furthermore, it should be understood that what is disclosed above is only a preferred embodiment of the present invention, and certainly cannot be used to limit the scope of rights of the present invention. Therefore, equivalent changes made according to the claims of the present invention are still within the scope of the present invention.

Claims

1. A beam uniformity adjustment method, characterized in that: include: Acquire historical adjustment data, wherein the historical adjustment data includes a plurality of historical state vectors of a target ion implanter, wherein one historical state vector includes a beam state vector and a device state vector of the target ion implanter at a historical acquisition moment; Determine a current training state vector from the historical adjustment data, and call an application adjustment model to determine a device adjustment vector corresponding to the current training state vector; Determining a regulation model loss value of the application regulation model based on the device regulation vector; And updating the model parameters in the application adjustment model in a direction of reducing the loss value of the adjustment model; The process of determining the current training state vector from the historical adjustment data is iteratively executed until the target model training condition is met, so as to complete the model training of the application adjustment model through the historical adjustment data, wherein the application adjustment model supports guiding the beam uniformity adjustment of the target ion implanter.

2. The method according to claim 1, characterized in that The determining, based on the device adjustment vector, an adjustment model loss value of the application adjustment model includes: Determining an updated beam state vector based on the device adjustment vector, and determining a current updated state vector corresponding to the current training state vector based on the updated beam state vector; Determine a current training tuple based on the current training state vector and the current updated state vector; and determine at least one training sample based on the current training tuple; The application scoring models are respectively called to determine the application scoring value corresponding to each training sample in the at least one training sample; and based on the application scoring value corresponding to each training sample, the adjustment model loss value of the application adjustment model is calculated.

3. The method according to claim 2, characterized in that The method further comprises: Determine the target scoring value corresponding to each training sample respectively, and determine the scoring model loss value of the application scoring model based on the target scoring value and the application scoring value corresponding to each training sample; The model parameters in the application scoring model are updated in a direction of reducing the loss value of the scoring model.

4. The method according to claim 3, characterized in that A training tuple also includes the revenue information and process state information corresponding to the updated state vector in the corresponding training tuple. One training sample is one training tuple. The respectively determining the target score value corresponding to each training sample comprises: For any training sample among the at least one training sample, based on the process state information in the any training sample, determining whether the beam uniformity corresponding to the any training sample meets the target process requirement; If the beam uniformity corresponding to any of the training samples meets the target process requirement, the revenue information in any of the training samples is used as the target score value corresponding to any of the training samples; If the beam uniformity corresponding to any of the training samples does not meet the target process requirements, the reference scoring model is called to determine the reference scoring value corresponding to any of the training samples based on the updated state vector in any of the training samples; and the target scoring value corresponding to any of the training samples is determined based on the revenue information in any of the training samples and the reference scoring value corresponding to any of the training samples.

5. The method according to claim 2, characterized in that: The target model training conditions include model training conditions in a first model training stage and / or model training conditions in a second model training stage; the updated beam state vector is determined according to a method for determining an updated beam state vector in the first model training stage or a method for determining an updated beam state vector in the second model training stage, the accuracy of the method for determining an updated beam state vector in the second model training stage is higher than the accuracy of the method for determining an updated beam state vector in the first model training stage, and the first model training stage precedes the second model training stage.

6. The method according to any one of claims 1 to 5, characterized in that: The adjustment model loss value is determined by applying a scoring model, a model parameter in the application scoring model supports being updated, and a scoring model loss value of the application scoring model is determined based on at least one of the application scoring model, a reference adjustment model, and a reference scoring model, and the method further includes: After each update of the model parameters in the application adjustment model and the application scoring model, determining whether a reference update condition is met; If the reference update condition is met, the model parameters in the reference adjustment model are updated based on the adjustment smoothing coefficient and the model parameters in the application adjustment model; and the model parameters in the reference scoring model are updated based on the scoring smoothing coefficient and the model parameters in the application scoring model; If the reference update condition is not met, the model parameters in the reference adjustment model and the reference scoring model are not updated.

7. The method according to any one of claims 1 to 5, characterized in that: The method further comprises: When it is detected that the current beam uniformity in the target ion implanter does not meet the target process requirement, determining a current state vector, wherein the current state vector includes a beam state vector and a device state vector of the target ion implanter at the current acquisition time; Calling the application adjustment model to determine a current device adjustment vector corresponding to the current state vector, wherein the current device adjustment vector is used to guide adjustment of at least one device in the target ion implanter; The determination of the current state vector is iteratively performed until the current beam uniformity in the target ion implanter meets the target process requirement; wherein, in the process of adjusting at least one device in the target ion implanter, the application adjustment model supports parameter update when the model update condition is met.

8. A beam uniformity adjustment device, characterized in that: The device comprises: An acquisition unit, used for acquiring historical adjustment data, wherein the historical adjustment data includes a plurality of historical state vectors of a target ion implanter, and one historical state vector includes a beam state vector and a device state vector of the target ion implanter at a historical acquisition moment; A processing unit, configured to determine a current training state vector from the historical adjustment data, and call an application adjustment model to determine a device adjustment vector corresponding to the current training state vector; The processing unit is further configured to determine a regulation model loss value of the application regulation model based on the device regulation vector; and update a model parameter in the application regulation model in a direction of reducing the regulation model loss value; The processing unit is also used to iteratively execute the determination of the current training state vector from the historical adjustment data until the target model training conditions are met, so as to complete the model training of the application adjustment model through the historical adjustment data, and the application adjustment model supports guiding the beam uniformity adjustment of the target ion implanter.

9. An electronic device, characterized in that: include: processor; as well as Memory for storing programs, The program includes instructions, which, when executed by the processor, cause the processor to perform the method according to any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to make a computer execute the method according to any one of claims 1-7.

Images