Energy consumption optimization method and system for semiconductor polishing equipment
By generating energy consumption factors based on wafer size information and polishing parameter timing information in semiconductor polishing equipment, adjusting polishing parameters, and building a double-constraint function, the problem of high energy consumption in traditional equipment is solved, and precise optimization of energy consumption and environmentally friendly are achieved. production process.
Patent Information
- Application Number
- CN202411302964.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-19
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2044-09-19
AI Technical Summary
Traditional semiconductor polishing equipment has problems of high energy consumption and low efficiency in energy consumption control, resulting in increased production costs and increased environmental burden.
By receiving polishing preset parameters based on wafer size information, traversing the timing information of polishing liquid flow, polishing disk speed, polishing pressure and polishing temperature, energy consumption factors are generated, and these parameters are adjusted according to the energy consumption constraints, and a double-constraint function is constructed. Optimize polishing parameters to achieve refined control of energy consumption.
It has achieved effective optimization of energy consumption of semiconductor polishing equipment, reduced energy consumption, improved production efficiency, and reduced impact on the environment.
Smart Images

Figure CN118832471B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of energy engineering technology, specifically to the field of energy consumption optimization technology, and in particular to an energy consumption optimization method and system for semiconductor polishing equipment. Background Art
[0002] With the continuous advancement of semiconductor manufacturing technology, semiconductor polishing equipment plays a vital role in the integrated circuit manufacturing process. However, with the continuous reduction in the size of semiconductor chips and the increase in integration, the performance requirements for semiconductor polishing equipment are becoming increasingly stringent, especially in terms of energy consumption optimization. Traditional energy consumption control methods for semiconductor polishing equipment often have problems such as high energy consumption and low efficiency, which not only increases production costs, but also creates unnecessary burdens on the environment. Summary of the invention
[0003] The present application provides an energy consumption optimization method and system for semiconductor polishing equipment, aiming to solve the technical problem of the lack of refined control of energy consumption in traditional semiconductor polishing equipment.
[0004] In view of the above problems, the present application provides an energy consumption optimization method and system for semiconductor polishing equipment.
[0005] The first aspect disclosed in the present application provides an energy consumption optimization method for semiconductor polishing equipment, the method comprising: according to wafer size information, interacting with semiconductor polishing equipment, receiving polishing preset parameters, wherein the polishing preset parameters include alkaline component concentration information, silica concentration information, silica particle size information, polishing liquid flow timing information, polishing disk speed timing information, polishing pressure timing information and polishing temperature timing information; traversing the polishing liquid flow timing information, the polishing disk speed timing information, the polishing pressure timing information and the polishing temperature timing information to perform energy consumption evaluation, and generate a flow regulation energy consumption factor, a speed regulation energy consumption factor, a pressure regulation energy consumption factor and a temperature regulation energy consumption factor; when the flow regulation energy consumption factor, the speed regulation energy consumption factor, the pressure regulation energy consumption factor and the temperature regulation energy consumption factor are not satisfied According to the energy consumption constraint conditions, the alkaline component concentration information, the silicon dioxide concentration information, the silicon dioxide particle size information, the polishing liquid flow timing information, the polishing disk speed timing information, the polishing pressure timing information and the polishing temperature timing information are assigned according to the wafer size information and the semiconductor polishing equipment to generate an initial optimized population; a first constraint function is constructed according to the flow adjustment energy consumption factor, the speed adjustment energy consumption factor, the pressure adjustment energy consumption factor and the temperature adjustment energy consumption factor; a second constraint function is constructed according to the flatness expected factor, the surface defect expected factor and the surface damage expected factor; according to the first constraint function and the second constraint function, the initial optimized population is double-constrained optimized to generate optimized polishing parameters, and the polishing preset parameters are updated by the optimized polishing parameters.
[0006] Another aspect disclosed in the present application provides an energy consumption optimization system for semiconductor polishing equipment, the system comprising: a parameter receiving module, the parameter receiving module being used to interact with the semiconductor polishing equipment according to wafer size information, and receive polishing preset parameters, wherein the polishing preset parameters include alkaline component concentration information, silica concentration information, silica particle size information, polishing liquid flow timing information, polishing disk speed timing information, polishing pressure timing information and polishing temperature timing information; an energy consumption evaluation module, the energy consumption evaluation module being used to traverse the polishing liquid flow timing information, the polishing disk speed timing information, the polishing pressure timing information and the polishing temperature timing information to perform energy consumption evaluation and generate a flow regulation energy consumption factor, a speed regulation energy consumption factor, a pressure regulation energy consumption factor and a temperature regulation energy consumption factor; an information assignment module, the information assignment module being used to evaluate the energy consumption when the flow regulation energy consumption factor, the speed regulation energy consumption factor, the pressure regulation energy consumption factor and the temperature regulation energy consumption factor do not meet the energy consumption requirements Constraint conditions, according to the wafer size information and the semiconductor polishing equipment, assign values to the alkaline component concentration information, the silicon dioxide concentration information, the silicon dioxide particle size information, the polishing liquid flow timing information, the polishing disk speed timing information, the polishing pressure timing information and the polishing temperature timing information to generate an initial optimization population; a first function construction module, the first function construction module is used to construct a first constraint function according to the flow regulation energy consumption factor, the speed regulation energy consumption factor, the pressure regulation energy consumption factor and the temperature regulation energy consumption factor; a second function construction module, the second function construction module is used to construct a second constraint function according to the flatness expected factor, the surface defect expected factor and the surface damage expected factor; a parameter updating module, the parameter updating module is used to perform double-constraint optimization on the initial optimization population according to the first constraint function and the second constraint function, generate optimized polishing parameters, and update the polishing preset parameters through the optimized polishing parameters.
[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:
[0008] The above-mentioned method for optimizing energy consumption of semiconductor polishing equipment interacts with the polishing equipment according to the size information of the wafer, and receives preset polishing parameters, which include the component concentration and particle size of the polishing liquid, and the timing information of the flow rate, rotation speed, pressure and temperature during the polishing process. Subsequently, a detailed energy consumption evaluation is performed on these timing information to generate energy consumption factors related to the flow rate, rotation speed, pressure and temperature adjustment. If these energy consumption factors do not meet the set energy consumption constraints, the above polishing parameters will be readjusted according to the wafer size and equipment characteristics to form an initial set of optimized parameters. Afterwards, the first constraint function is constructed using these energy consumption factors, and the second constraint function is constructed based on the expectations of the flatness, surface defects and damage of the wafer after polishing. These two constraint functions will help find a parameter combination that can meet both the energy consumption requirements and the polishing quality goals. Then, the initial optimization parameter set is optimized under dual constraints using these two constraint functions to obtain the final optimized polishing parameters. These parameters will be used to update the original preset polishing parameters, thereby improving the energy efficiency of the polishing process and reducing energy consumption. This method achieves effective optimization of the energy consumption of semiconductor polishing equipment by real-time monitoring and adjusting key parameters in the polishing process, combined with the optimization of dual constraints of energy consumption and quality.
[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS
[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0011] Figure 1 A schematic flow chart of an energy consumption optimization method for semiconductor polishing equipment in one embodiment;
[0012] Figure 2 A diagram of an energy consumption optimization system for a semiconductor polishing device in one embodiment.
[0013] Explanation of the accompanying drawings: parameter receiving module 1, energy consumption evaluation module 2, information assignment module 3, first function construction module 4, second function construction module 5, parameter updating module 6. DETAILED DESCRIPTION
[0014] The embodiments of the present application provide a method and system for optimizing energy consumption of semiconductor polishing equipment to solve the technical problem of the lack of refined control over energy consumption in traditional semiconductor polishing equipment.
[0015] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.
[0016] It should be noted that the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or inherent to these processes, methods, products or devices.
[0017] Embodiment 1
[0018] like Figure 1 As shown, the present application provides a method for optimizing energy consumption of semiconductor polishing equipment, the method comprising:
[0019] According to the wafer size information, the semiconductor polishing device interacts to receive polishing preset parameters, wherein the polishing preset parameters include alkaline component concentration information, silicon dioxide concentration information, silicon dioxide particle size information, polishing liquid flow timing information, polishing disk speed timing information, polishing pressure timing information and polishing temperature timing information;
[0020] As chip integration continues to increase and size gradually decreases, the performance requirements for semiconductor polishing equipment are becoming higher and higher. Polishing is an important part of the semiconductor manufacturing process, and its quality and efficiency directly affect the performance and market competitiveness of the final product. However, traditional semiconductor polishing equipment often has shortcomings in energy consumption control. High energy consumption not only increases production costs, but also creates unnecessary pressure on the environment.
[0021] In an embodiment of the present application, when optimizing the energy consumption of a semiconductor polishing device, the system terminal first interacts with the semiconductor polishing device based on the size information of the wafer and receives its polishing preset parameters. These parameters include the concentration of alkaline components in the polishing liquid, the concentration of silicon dioxide, the particle size of silicon dioxide, and key timing information during the polishing process, such as the flow rate of the polishing liquid, the rotation speed of the polishing disk, the polishing pressure, and the polishing temperature. After receiving these polishing preset parameters, the system terminal will evaluate and optimize their energy consumption to ensure that the energy consumption of the polishing equipment is reduced as much as possible while meeting the polishing quality requirements. Such an optimization process will help the semiconductor manufacturing industry improve production efficiency and reduce the impact on the environment.
[0022] Traversing the polishing liquid flow timing information, the polishing disk speed timing information, the polishing pressure timing information and the polishing temperature timing information to perform energy consumption evaluation, and generate a flow regulation energy consumption factor, a speed regulation energy consumption factor, a pressure regulation energy consumption factor and a temperature regulation energy consumption factor;
[0023] In one embodiment, in the process of optimizing the energy consumption of semiconductor polishing equipment, an important step is to traverse and evaluate the key timing information in the polishing process. These key timing information include the flow rate of the polishing liquid, the rotation speed of the polishing disk, the polishing pressure and the polishing temperature, etc., which have a significant impact on the polishing effect and energy consumption over time. The greater the flow rate of the polishing liquid, the rotation speed of the polishing disk, the polishing pressure and the polishing temperature, the more energy is consumed. Specifically, the system terminal analyzes these timing information one by one, and counts the proportion of moments when their distance from the corresponding median feature is greater than or equal to the corresponding adjustment step threshold. Through this process, a series of energy consumption factors can be generated, including flow regulation energy consumption factors, rotation speed regulation energy consumption factors, pressure regulation energy consumption factors and temperature regulation energy consumption factors. These energy consumption factors reflect the degree of influence on the overall energy consumption when adjusting the corresponding parameters. In summary, the purpose of this step is to quantify the influence of different parameters on energy consumption by traversing and evaluating the key timing information in the polishing process, so as to provide guidance for subsequent energy consumption optimization. These generated energy consumption factors will help the system terminal to adjust the polishing parameters more accurately to achieve a low-energy and high-efficiency polishing process.
[0024] Further, the present application provides a method for traversing the polishing liquid flow timing information, the polishing disk speed timing information, the polishing pressure timing information and the polishing temperature timing information to perform energy consumption evaluation, and generate a flow regulation energy consumption factor, a speed regulation energy consumption factor, a pressure regulation energy consumption factor and a temperature regulation energy consumption factor, including:
[0025] Performing median analysis on the polishing liquid flow timing information, the polishing disk speed timing information, the polishing pressure timing information and the polishing temperature timing information to obtain flow median characteristics, speed median characteristics, pressure median characteristics and temperature median characteristics;
[0026] Configure flow adjustment step threshold, speed adjustment step threshold, pressure adjustment step threshold and temperature adjustment step threshold;
[0027] Preferably, in the process of optimizing the energy consumption of the semiconductor polishing equipment, the system terminal performs median analysis on the time series information such as the polishing liquid flow rate, the polishing disk speed, the polishing pressure and the polishing temperature. Median analysis is a statistical method used to find the central trend of a data set. Specifically, the system terminal arranges the time series information such as the polishing liquid flow rate, the polishing disk speed, the polishing pressure and the polishing temperature in ascending order, and calculates the median of the time series data respectively. The median is the value in the middle position after sorting a set of data from small to large. If the time series data is an odd number, the median is the middle data; if the time series data is an even number, the median is the average of the two middle data. Subsequently, the system terminal records the median of each calculated time series information to generate a flow median feature, a speed median feature, a pressure median feature and a temperature median feature. These median features represent the average level of these parameters during the polishing process, and provide the system terminal with important information about the stability and consistency of the polishing process. Afterwards, the system terminal understands the performance parameters and operating characteristics of the semiconductor polishing equipment. This includes understanding the adjustable range, adjustment accuracy and stability of the equipment in terms of flow, speed, pressure and temperature. Then, according to the process requirements of semiconductor polishing, the influence of parameters such as flow rate, speed, pressure and temperature on polishing quality is evaluated. The acceptable adjustment range of these parameters while maintaining the polishing quality is determined. Then, the parameter adjustment data in the historical polishing process is analyzed. This helps to understand the common adjustment amplitude and frequency of these parameters in actual operation, as well as the impact of these adjustments on energy consumption and polishing quality. Finally, based on equipment performance, process requirements and historical data analysis, a suitable adjustment step threshold is determined for each parameter, that is, the flow adjustment step threshold, speed adjustment step threshold, pressure adjustment step threshold and temperature adjustment step threshold are determined. These thresholds can not only ensure the stability of polishing quality, but also allow necessary parameter fine-tuning during energy consumption optimization. In summary, the typical levels of these parameters are obtained by performing median analysis on the key timing information in the polishing process. Then, the step threshold of parameter adjustment is configured to guide the adjustment range of these parameters during the optimization process. This process provides an important data basis and parameter setting basis for subsequent energy consumption optimization.
[0028] Counting the proportion of moments when the distance between the polishing liquid flow time series information and the flow median feature is greater than or equal to the flow adjustment step threshold, and setting it as the flow adjustment energy consumption factor;
[0029] Preferably, the system terminal calculates the difference between the flow value at that moment and the previously calculated flow median characteristic for each time point in the polishing liquid flow timing information. This difference represents the degree of deviation between the flow value at that moment and the flow median characteristic. Subsequently, this difference is compared with the flow adjustment step threshold. If the difference is greater than or equal to the flow adjustment step threshold, it is considered that the flow value at that moment deviates greatly from the median characteristic and needs to be adjusted. After that, for all the time points in the flow timing information, the above steps are repeated, and the number of time points that need to be adjusted is recorded. Then, the number of time points that need to be adjusted is divided by the total number of time points to calculate the proportion of the time that needs to be adjusted. Finally, this proportion is set as the flow regulation energy consumption factor. This factor reflects the proportion of the time when the flow value needs to be adjusted during the polishing process, thereby indirectly reflecting the degree of influence of flow regulation on energy consumption. Through this process, the deviation between the polishing liquid flow timing information and the flow median characteristic, as well as the influence of this deviation on energy consumption, can be quantified. The flow regulation energy consumption factor provides the system terminal with an indicator for evaluating the energy consumption of flow regulation during polishing, which helps to better optimize polishing parameters and reduce energy consumption.
[0030] Counting the proportion of moments when the distance between the polishing disk speed time series information and the speed median feature is greater than or equal to the speed adjustment step threshold, and setting it as the speed adjustment energy consumption factor;
[0031] Counting the proportion of moments when the distance between the polishing pressure timing information and the pressure median feature is greater than or equal to the pressure adjustment step threshold, and setting it as the pressure adjustment energy consumption factor;
[0032] The proportion of moments when the distance between the polishing temperature timing information and the temperature median feature is greater than or equal to the temperature adjustment step threshold is counted and set as the temperature adjustment energy consumption factor.
[0033] Preferably, for the polishing disk speed timing information, polishing pressure timing information and polishing temperature timing information and their respective adjustment energy consumption factors, the system terminal calculates the distances of the polishing disk speed timing information, polishing pressure timing information and polishing temperature timing information respectively based on the same method as the above-mentioned setting of the flow adjustment energy consumption factor, and compares these distances with the corresponding adjustment step thresholds to set the speed adjustment energy consumption factor, pressure adjustment energy consumption factor and temperature adjustment energy consumption factor. Through the calculation of these energy consumption factors, the energy consumption of each parameter adjustment in the polishing process can be more comprehensively understood, providing an important basis for subsequent energy consumption optimization.
[0034] When the flow rate regulating energy consumption factor, the speed regulating energy consumption factor, the pressure regulating energy consumption factor and the temperature regulating energy consumption factor do not meet the energy consumption constraint condition, the alkaline component concentration information, the silicon dioxide concentration information, the silicon dioxide particle size information, the polishing liquid flow timing information, the polishing disk speed timing information, the polishing pressure timing information and the polishing temperature timing information are assigned values according to the wafer size information and the semiconductor polishing equipment to generate an initial optimized population;
[0035] In one embodiment, when the flow regulation energy consumption factor, the speed regulation energy consumption factor, the pressure regulation energy consumption factor and the temperature regulation energy consumption factor do not meet the predetermined energy consumption constraint condition, it indicates that the current polishing liquid formula and process parameter combination are not optimal and need to be optimized and adjusted to reduce energy consumption. In order to perform such optimization, the system terminal re-assigns a series of key factors affecting polishing performance based on wafer size information and specific characteristics of semiconductor polishing equipment, including alkaline component concentration, silicon dioxide concentration, silicon dioxide particle size, polishing liquid flow timing, polishing disk speed timing, polishing pressure timing and polishing temperature timing. This process involves the generation of an initial optimization population. The initial optimization population is a set of candidate solutions containing different parameter combinations, which represent possible optimization directions. In this step, the system terminal interacts with multiple production nodes based on known wafer size and polishing equipment characteristics to collect a polishing log set. Subsequently, discretization deletion is performed based on this polishing log set to generate an initial, diversified optimization population. This initial optimized population will serve as the starting point for subsequent dual-constraint optimization. Through iterative search, the energy consumption performance of each parameter combination will be continuously evaluated, and finally a set of parameter combinations that can meet the energy consumption constraints and have excellent performance will be found. This process can help the system terminal find a more efficient polishing liquid formula and process parameter combination, thereby reducing energy consumption in the semiconductor polishing process and improving production efficiency.
[0036] Furthermore, the present application provides that when the flow regulation energy consumption factor, the speed regulation energy consumption factor, the pressure regulation energy consumption factor and the temperature regulation energy consumption factor do not meet the energy consumption constraint condition, the method further includes:
[0037] The energy consumption constraint condition includes adjusting the energy consumption factor threshold;
[0038] When one or more of the flow regulation energy consumption factor, the speed regulation energy consumption factor, the pressure regulation energy consumption factor and the temperature regulation energy consumption factor is greater than or equal to the regulation energy consumption factor threshold, it is deemed that the flow regulation energy consumption factor, the speed regulation energy consumption factor, the pressure regulation energy consumption factor and the temperature regulation energy consumption factor do not meet the energy consumption constraint condition;
[0039] When all of the flow regulation energy consumption factor, the speed regulation energy consumption factor, the pressure regulation energy consumption factor and the temperature regulation energy consumption factor are less than the regulation energy consumption factor threshold, it is deemed that the flow regulation energy consumption factor, the speed regulation energy consumption factor, the pressure regulation energy consumption factor and the temperature regulation energy consumption factor meet the energy consumption constraint condition.
[0040] Preferably, the energy consumption constraint condition includes a plurality of specific indicators, which are the threshold values of each adjustment energy consumption factor. These threshold values are used to measure whether the adjustment frequency of parameters such as flow, rotation speed, pressure and temperature during the polishing process is too high, and then evaluate whether the energy consumption is effectively controlled. When one or more of the flow adjustment energy consumption factor, the rotation speed adjustment energy consumption factor, the pressure adjustment energy consumption factor and the temperature adjustment energy consumption factor is greater than or equal to the adjustment energy consumption factor threshold value, the system terminal determines that these adjustment energy consumption factors do not meet the energy consumption constraint condition. This means that during the polishing process, the adjustment frequency of certain parameters is too high, which may lead to unnecessary energy consumption increase. On the contrary, if all the values of the flow adjustment energy consumption factor, the rotation speed adjustment energy consumption factor, the pressure adjustment energy consumption factor and the temperature adjustment energy consumption factor are less than the adjustment energy consumption factor threshold value, then it is judged that these adjustment energy consumption factors meet the energy consumption constraint condition. This means that the adjustment frequency of each parameter during the polishing process is within an acceptable range and the energy consumption is effectively controlled. In this case, if the adjustment energy consumption factor does not meet the energy consumption constraint condition, it is necessary to optimize and adjust the parameters during the polishing process. This includes reassigning the polishing solution formulation, such as alkaline component concentration, silica concentration and particle size, and polishing process parameters, such as polishing solution flow rate, polishing disk speed, polishing pressure and temperature, based on wafer size information and the characteristics of the semiconductor polishing equipment to generate an initial optimized population.
[0041] Further, the present application provides that, according to the wafer size information and the semiconductor polishing equipment, the alkaline component concentration information, the silicon dioxide concentration information, the silicon dioxide particle size information, the polishing liquid flow timing information, the polishing disk speed timing information, the polishing pressure timing information and the polishing temperature timing information are assigned values to generate an initial optimized population, and the method also includes:
[0042] The wafer size information, the semiconductor polishing equipment model, the alkaline component concentration information, the silicon dioxide concentration information, and the silicon dioxide particle size information are used as retrieval assignment conditions, and the polishing liquid flow timing information, the polishing disk speed timing information, the polishing pressure timing information, and the polishing temperature timing information are used as retrieval assignment targets, multiple production nodes are interacted, and a polishing log set is collected, wherein any polishing log includes the polishing liquid flow record timing information, the polishing disk speed record timing information, the polishing pressure record timing information, and the polishing temperature record timing information;
[0043] Preferably, in order to optimize polishing parameters and reduce energy consumption, the system terminal uses wafer size information, semiconductor polishing equipment model, alkaline component concentration, silica concentration and silica particle size as retrieval and assignment conditions. These conditions provide the system terminal with basic requirements and equipment characteristics of the polishing process, and are an important basis for determining polishing parameters. At the same time, the polishing liquid flow timing information, polishing disk speed timing information, polishing pressure timing information and polishing temperature timing information are used as retrieval and assignment targets. These timing information represent the dynamic changes of various parameters in the polishing process, which is crucial for controlling polishing quality and energy consumption. In order to achieve the optimal assignment of these parameters, the system terminal interacts with multiple production nodes to collect polishing log sets. These production nodes include different polishing equipment, production lines or process stages. Each polishing log contains the timing information of various parameters recorded in the actual polishing process, such as polishing liquid flow records, polishing disk speed records, polishing pressure records and polishing temperature records.
[0044] According to the polishing liquid flow rate recording timing information, the polishing disk rotation speed recording timing information, the polishing pressure recording timing information and the polishing temperature recording timing information, the polishing log set is discretized and deleted to obtain the initial optimized population.
[0045] Preferably, the system terminal performs a process of discretizing and deleting the polishing log set according to the timing information of the polishing liquid flow record, the timing information of the polishing disk speed record, the timing information of the polishing pressure record, and the timing information of the polishing temperature record. Specifically, the system terminal randomly extracts two groups from the polishing log set to align the first moment, and then calculates the proportion of the moments in which the flow deviation between the two groups in the specified time zone from the first moment to the front tail moment is greater than or equal to the flow deviation threshold, and obtains the first discrete coefficient and the second discrete coefficient. Afterwards, based on these two discrete coefficients, the first discreteness is determined. Then, the above process is repeated for the timing information of the polishing disk speed record, the timing information of the polishing pressure record, and the timing information of the polishing temperature record to obtain the corresponding discreteness. Finally, based on the calculated discreteness, the system terminal performs discretization and deletion to obtain the initial optimized population. Through discretization and deletion, the system terminal can remove data that has little impact on the optimization process or is irrelevant to the problem, while retaining key information that has an important impact on the optimization results. Doing so can ensure that the distances between solutions are relatively far, that is, the solution distribution area is larger, thereby improving the globality of subsequent analysis, as well as the efficiency and accuracy of optimization, so as to better achieve the optimization of polishing parameters and the reduction of energy consumption.
[0046] Further, the present application provides a method for discretizing and deleting the polishing log set according to the polishing liquid flow rate recording timing information, the polishing disk speed recording timing information, the polishing pressure recording timing information and the polishing temperature recording timing information to obtain the initial optimized population, and the method also includes:
[0047] Obtaining first polishing liquid flow rate record timing information of the first polishing log and second polishing liquid flow rate record timing information of the second polishing log;
[0048] Aligning the first polishing liquid flow rate recording timing information with the second polishing liquid flow rate recording timing information at the first moment to obtain a tail moment deviation, which is set as a first discrete coefficient;
[0049] Optionally, the system terminal randomly extracts two different polishing logs from the polishing log set as the first polishing log and the second polishing log, and then extracts data from the two logs to obtain the first polishing liquid flow record timing information and the second polishing liquid flow record timing information. Subsequently, the system terminal uses the start time of the two polishing logs as the start time of the two polishing liquid flow record timing information. Subsequently, the start time of the two timing information is set to the same time point. This is done by intercepting the two timing information so that they start from the same time point. After the first moment is aligned, the system terminal searches for the end time of the two timing information. These end times may be different because the time lengths of different log records may be different due to various reasons in the actual polishing process, such as equipment failure, human intervention, etc. After that, the difference between the tail moments of the two timing information is calculated, that is, the tail moment of the second polishing liquid flow record timing information minus the tail moment of the first polishing liquid flow record timing information. If the calculated result is a negative value, the order of the front and back is changed and the calculation is performed again. The calculated difference is the tail moment deviation. Then, this tail moment deviation is set as the first discrete coefficient. This coefficient reflects the difference in time length between two time series information and is an important parameter that needs to be considered in subsequent analysis of the similarity of time series information or in performing time series data fusion.
[0050] Calculate the proportion of time periods in the specified time zone from the first moment to the front tail moment when the flow deviation between the first polishing liquid flow recording timing information and the second polishing liquid flow recording timing information is greater than or equal to the flow deviation threshold, and set it as a second discrete coefficient;
[0051] Optionally, the system terminal first determines a specified time zone starting from the first moment and ending at the first and last moments. This time zone should be determined by actual needs and process requirements. Subsequently, a flow deviation threshold is set based on process requirements and historical experience. This threshold will be used to determine whether the flow values of two time series information at the same moment are close enough. After that, each time point of the two time series information in the specified time zone is traversed. For each time point, the flow values of the two time series information at that time point are compared, and the flow deviation between them is calculated. Then, for each time point, check whether the flow deviation is greater than or equal to the flow deviation threshold. If it is satisfied, the system terminal increases a counter to indicate the number of moments that meet the conditions. At the same time, regardless of whether the flow deviation meets the conditions, another counter is added to record the total number of moments. Finally, the second discrete coefficient is calculated by dividing the number of moments that meet the conditions by the total number of moments. This coefficient represents the proportion of moments in the specified time zone where the flow deviation is greater than or equal to the flow deviation threshold. If the second discrete coefficient is large, it means that there is a large difference in the flow values of the two time series information in the time zone, otherwise, it is small.
[0052] When the first discrete coefficient is less than or equal to a first discrete threshold, and the second discrete coefficient is less than or equal to a second discrete threshold, the first discreteness of the first polishing liquid flow recording timing information and the second polishing liquid flow recording timing information is equal to 0;
[0053] Otherwise, the first discreteness of the first polishing liquid flow rate recording timing information and the second polishing liquid flow rate recording timing information is equal to 1;
[0054] Optionally, when comparing two polishing liquid flow record timing information, the system terminal uses the two obtained discrete coefficients to evaluate the difference between them. The two discrete coefficients are calculated based on the time length difference and the flow value difference, respectively. If the first discrete coefficient is less than or equal to the preset first discrete threshold, and the second discrete coefficient is also less than or equal to the preset second discrete threshold, then the system terminal determines that the difference between the two timing information is within an acceptable range, and therefore sets the first discreteness of the two timing information to 0, indicating that they are highly similar and have no obvious differences. On the contrary, if any discrete coefficient exceeds its corresponding threshold, the system terminal determines that there is a large difference between the two timing information, which may be due to inconsistent time lengths or excessive changes in flow values. In this case, the system terminal sets the first discreteness of the two timing information to 1, indicating that there is a discrete or inconsistent situation between them.
[0055] Traversing the polishing disk rotation speed recording timing information, the polishing pressure recording timing information and the polishing temperature recording timing information to obtain a second discreteness, a third discreteness and a fourth discreteness;
[0056] When the first discreteness, the second discreteness, the third discreteness, and the fourth discreteness are all equal to 0, randomly deleting one of the first polishing log and the second polishing log from the polishing log set;
[0057] Optionally, the system terminal traverses the timing information of the polishing disk speed record, the polishing pressure record, and the polishing temperature record, and calculates the discreteness corresponding to each timing information, that is, the second discreteness, the third discreteness, and the fourth discreteness, by the same method as the aforementioned method of obtaining the first discreteness. If these four discretenesses are all equal to 0, it means that the timing information of the two sets of polishing processes are very close in each key parameter and there is no obvious difference. Therefore, in order to reduce the redundancy of data storage and improve the representativeness of the data set, the system terminal randomly deletes one of the two polishing logs from the polishing log set. Doing so can ensure the simplicity and efficiency of the data set while retaining sufficient representativeness to support subsequent analysis and decision-making. In summary, when comparing the timing information of the two sets of polishing processes and finding that they have no significant differences in key parameters such as the polishing disk speed, polishing pressure, polishing temperature, and polishing liquid flow, delete one of the polishing logs to optimize the quality and efficiency of the data set.
[0058] The polishing log set is traversed in a loop to perform discretization deletion to obtain the initial optimized population.
[0059] Optionally, the system terminal traverses the polishing log set and evaluates their similarity by calculating the discreteness between different polishing logs through the same steps as above. If the discreteness of the two sets of polishing logs on the key parameters is 0, then you can choose to delete one of them to reduce the redundancy of the data set and maintain its representativeness. This process will continue until the polishing log set is traversed, and by deleting redundant and similar polishing logs, an optimized data set, namely the initial optimized population, is finally obtained. Each polishing log in this initial optimized population has a high uniqueness and can better represent the changes and characteristics of the entire polishing process, providing a more valuable data basis for subsequent optimization.
[0060] Constructing a first constraint function according to the flow rate regulation energy consumption factor, the speed regulation energy consumption factor, the pressure regulation energy consumption factor and the temperature regulation energy consumption factor;
[0061] In one embodiment, the process of constructing the first constraint function is essentially to integrate the flow regulation energy consumption factor, the speed regulation energy consumption factor, the pressure regulation energy consumption factor and the temperature regulation energy consumption factor into one constraint condition to ensure that when adjusting these parameters, the overall energy consumption does not exceed the preset limit. By constructing such a constraint function, the important factor of energy consumption can be taken into account during the optimization process, thereby ensuring that the obtained optimization solution not only meets the polishing performance requirements, but also has high energy efficiency. This approach helps to achieve efficient use of resources.
[0062] Furthermore, the present application provides the first constraint function, and the method further includes:
[0063] The first constraint function is:
[0064] ;
[0065] Preferably, the first constraint function is used to evaluate the energy consumption changes caused by the adjustment of various parameters during the polishing process. In this function, different parameter adjustments are assigned different weights and compared with the preset polishing parameter adjustment energy consumption factors to obtain a comprehensive energy consumption evaluation value. Characterizes the first constraint output value of the i-th individual in the initial optimization population, Characterizes the energy consumption factor of the i-th individual flow regulation, Characterize the energy consumption factor of polishing preset parameters flow regulation, Characterizes the energy consumption factor of the i-th individual speed regulation, Characterize the energy consumption factor of polishing preset parameter speed adjustment, Characterizes the energy consumption factor of the pressure regulation of the i-th individual, Characterize the polishing preset parameter pressure adjustment energy consumption factor, Characterizes the temperature regulation energy consumption factor of the i-th individual, Characterize the polishing preset parameter temperature adjustment energy consumption factor, , , and Respectively characterize the weights of the energy consumption factors for regulating flow, speed, pressure and temperature. This first constraint function compares the energy consumption factors of the four parameters of flow, speed, pressure and temperature of the individuals with the preset reference values, and performs weighted calculations according to their respective weights to obtain a comprehensive energy consumption evaluation value. In addition, when the pressure regulation energy consumption factor of the i-th individual is not less than the polishing preset parameter flow regulation energy consumption factor, or the i-th individual speed regulation energy consumption factor is not less than the polishing preset parameter speed regulation energy consumption factor, or the i-th individual pressure regulation energy consumption factor is not less than the polishing preset parameter pressure regulation energy consumption factor, or the i-th individual temperature regulation energy consumption factor is not less than the polishing preset parameter temperature regulation energy consumption factor, the system terminal directly records the first constraint output value of the i-th individual in the initial optimization population as 0. This evaluation value can be used to evaluate the energy consumption performance of each individual in the initial optimization population and provide guidance for the subsequent optimization process.
[0066] Constructing a second constraint function according to the flatness expected factor, the surface defect expected factor and the surface damage expected factor;
[0067] In one embodiment, in addition to considering energy consumption factors, the system terminal also pays attention to the quality of the polishing results, including aspects such as flatness, surface defects and surface damage. In order to incorporate these quality indicators into the optimization process, a second constraint function can be constructed. First, the flatness expectation factor is determined based on the output value of the flatness detection instrument. This factor represents the expected or target value of the surface flatness after polishing. Secondly, the surface defect expectation factor is used to characterize the maximum size of the surface defects after polishing. During the polishing process, some surface defects such as scratches, pits, etc. may be generated. The surface defect expectation factor sets a maximum allowable defect size to ensure that the quality of the polishing result meets the requirements. Finally, the surface damage expectation factor focuses on the maximum size of the surface damage that may be caused during the polishing process. Surface damage may be caused by over-polishing, using inappropriate polishing tools or improper parameter settings. The surface damage expectation factor sets an acceptable maximum damage size to prevent the polishing process from causing irreversible damage to the workpiece surface. Based on the above three expectation factors, the system terminal can construct a second constraint function. This function will consider the quality indicators of flatness, surface defects and surface damage to ensure that the quality of the polishing result meets the predetermined requirements.
[0068] Furthermore, the present application provides the second constraint function, and the method further includes:
[0069] ;
[0070] Preferably, in the second constraint function, Characterizes the second constraint output value of the i-th individual of the initial optimization population, Characterize the flatness factor of the i-th individual, Characterize the surface defect factor of the i-th individual, Characterize the surface damage factor of the i-th individual, Characterize the flatness factor weight, Characterize the weight of surface defect factors, Characterize the weight of the surface damage factor, the flatness expectation factor, the surface defect expectation factor and the surface damage expectation factor are all positive factors. When , the system terminal records the second constraint output value of the i-th individual of the initial optimization population as negative infinity. When , the system terminal calculates the relative change rates of the i-th individual flatness factor, the i-th individual surface defect factor, and the i-th individual surface damage factor, respectively, and combines them with the corresponding weights, performs weighted summation, and then divides by 3 to obtain the i-th individual second constraint output value of the initial optimization population. When there is a value greater than or equal to the corresponding preset parameter, the system terminal directly records the relative change rate of the factor as 0, and performs the same calculation process as above to obtain the corresponding i-th individual second constraint output value of the initial optimization population.
[0071] According to the first constraint function and the second constraint function, the initial optimization population is subjected to dual-constraint optimization to generate optimized polishing parameters, and the preset polishing parameters are updated by the optimized polishing parameters.
[0072] In one embodiment, in the optimization of the polishing process, the system terminal uses the first constraint function and the second constraint function for dual-constraint optimization. First, the second constraint function is used to select the population that meets the quality requirements, and then the first constraint function is used to screen the selected population again, and the population with lower weighted energy consumption is selected, and then the optimized polishing parameters are generated for updating the polishing preset parameters. In summary, the dual-constraint optimization combines the considerations of energy consumption and quality, and finds the optimal solution by adjusting the polishing parameters, thereby realizing the update and optimization of the polishing preset parameters. This helps to reduce energy consumption and costs while improving polishing efficiency and quality.
[0073] Further, the present application provides performing double-constraint optimization on the initial optimization population according to the first constraint function and the second constraint function to generate optimized polishing parameters, and the method also includes:
[0074] Configure the second constraint function output threshold;
[0075] According to the second constraint function, a first sorted population is sorted from the initial optimized population, which is less than or equal to the output threshold of the second constraint function;
[0076] The first sorting population is iterated and sorted a preset number of times according to the first constraint function, an optimal solution is output, and the optimized polishing parameters are generated.
[0077] Preferably, in the optimization of the polishing process, the system terminal uses two constructed constraint functions: a first constraint function and a second constraint function. These two functions are used to evaluate the energy consumption of the polishing process and the quality of the polishing result, respectively. First, in order to ensure the quality of the polishing result, the system terminal configures the output threshold of the second constraint function. This threshold represents the minimum requirements for the quality of the polishing result, including indicators in terms of flatness, surface defects and surface damage. By using the second constraint function, individuals that meet or exceed the quality requirements can be screened out from the initial optimization population to form a first sorting population. Subsequently, in order to reduce energy consumption while meeting the quality requirements, the system terminal uses the first constraint function to iteratively sort the first sorting population. This process gradually screens out individuals with lower energy consumption through a preset number of iterations. In each iteration, according to the evaluation result of the first constraint function, those individuals with higher energy consumption are excluded, thereby gradually approaching the optimal solution. Finally, after a preset number of iterative sorting, the optimal solution, that is, the optimized polishing parameters, is obtained. These parameters not only meet the quality requirements of the polishing results, but also reach a lower level in terms of energy consumption. By updating the polishing preset parameters to these optimized polishing parameters, a more efficient and energy-saving polishing process can be achieved. In summary, the second constraint function is used to first screen out individuals that meet the quality requirements, and then the first constraint function is used to find the optimal solution with low energy consumption through iterative sorting to generate optimized polishing parameters. This method combines the considerations of quality and energy consumption, which helps to optimize the polishing process.
[0078] In summary, the embodiments of the present application have at least the following technical effects:
[0079] The embodiment of the present application interacts with the semiconductor polishing equipment according to the wafer size information, and receives the polishing preset parameters including the concentration information of the alkaline component, the concentration information of silicon dioxide, the particle size information of silicon dioxide, etc. Then traverse the timing information of the polishing liquid flow, the polishing disk speed, the polishing pressure and the polishing temperature, perform energy consumption evaluation, and generate energy consumption factors for flow regulation, speed regulation, pressure regulation and temperature regulation. When the energy consumption factor does not meet the constraint condition, the parameter is assigned according to the equipment and wafer information to generate the initial optimized population. Subsequently, a constraint function is constructed according to the energy consumption factor and factors such as expected flatness, surface defects, and surface damage. The initial optimized population is optimized using the constraint function to generate optimized polishing parameters, thereby updating the polishing preset parameters. This method optimizes the polishing process by comprehensively considering the energy consumption and workpiece quality factors, while satisfying the energy consumption constraint. These technical effects jointly solve the technical problem that traditional semiconductor polishing equipment lacks refined control of energy consumption, and achieves the effect of accurate optimization and effective control of the energy consumption of semiconductor polishing equipment.
[0080] Embodiment 2
[0081] Based on the same inventive concept as the energy consumption optimization method of a semiconductor polishing device in the aforementioned embodiment, Figure 2 As shown, the present application provides an energy consumption optimization system for semiconductor polishing equipment, the system comprising:
[0082] Parameter receiving module 1: the parameter receiving module 1 is used to interact with the semiconductor polishing equipment according to the wafer size information and receive the polishing preset parameters, wherein the polishing preset parameters include the concentration information of the alkaline component, the concentration information of silicon dioxide, the particle size information of silicon dioxide, the flow timing information of the polishing liquid, the rotation speed timing information of the polishing disk, the polishing pressure timing information and the polishing temperature timing information;
[0083] Energy consumption evaluation module 2: the energy consumption evaluation module 2 is used to traverse the polishing liquid flow timing information, the polishing disk speed timing information, the polishing pressure timing information and the polishing temperature timing information to perform energy consumption evaluation, and generate a flow regulation energy consumption factor, a speed regulation energy consumption factor, a pressure regulation energy consumption factor and a temperature regulation energy consumption factor;
[0084] Information assignment module 3: the information assignment module 3 is used to assign the alkaline component concentration information, the silicon dioxide concentration information, the silicon dioxide particle size information, the polishing liquid flow timing information, the polishing disk speed timing information, the polishing pressure timing information and the polishing temperature timing information according to the wafer size information and the semiconductor polishing equipment when the flow regulation energy consumption factor, the speed regulation energy consumption factor, the pressure regulation energy consumption factor and the temperature regulation energy consumption factor do not meet the energy consumption constraint condition, so as to generate an initial optimized population;
[0085] First function building module 4: The first function building module 4 is used to build a first constraint function according to the flow rate regulation energy consumption factor, the speed regulation energy consumption factor, the pressure regulation energy consumption factor and the temperature regulation energy consumption factor;
[0086] Second function construction module 5: The second function construction module 5 is used to construct a second constraint function according to the flatness expected factor, the surface defect expected factor and the surface damage expected factor;
[0087] Parameter updating module 6: the parameter updating module 6 is used to perform double-constraint optimization on the initial optimization population according to the first constraint function and the second constraint function, generate optimized polishing parameters, and update the preset polishing parameters by using the optimized polishing parameters.
[0088] Furthermore, the energy consumption assessment module 2 is also used to execute the following method:
[0089] Performing median analysis on the polishing liquid flow timing information, the polishing disk speed timing information, the polishing pressure timing information and the polishing temperature timing information to obtain flow median characteristics, speed median characteristics, pressure median characteristics and temperature median characteristics;
[0090] Configure flow adjustment step threshold, speed adjustment step threshold, pressure adjustment step threshold and temperature adjustment step threshold;
[0091] Counting the proportion of moments when the distance between the polishing liquid flow time series information and the flow median feature is greater than or equal to the flow adjustment step threshold, and setting it as the flow adjustment energy consumption factor;
[0092] Counting the proportion of moments when the distance between the polishing disk speed time series information and the speed median feature is greater than or equal to the speed adjustment step threshold, and setting it as the speed adjustment energy consumption factor;
[0093] Counting the proportion of moments when the distance between the polishing pressure timing information and the pressure median feature is greater than or equal to the pressure adjustment step threshold, and setting it as the pressure adjustment energy consumption factor;
[0094] The proportion of moments when the distance between the polishing temperature timing information and the temperature median feature is greater than or equal to the temperature adjustment step threshold is counted and set as the temperature adjustment energy consumption factor.
[0095] Furthermore, the information assignment module 3 is also used to execute the following method:
[0096] The energy consumption constraint condition includes adjusting the energy consumption factor threshold;
[0097] When one or more of the flow regulation energy consumption factor, the speed regulation energy consumption factor, the pressure regulation energy consumption factor and the temperature regulation energy consumption factor is greater than or equal to the regulation energy consumption factor threshold, it is deemed that the flow regulation energy consumption factor, the speed regulation energy consumption factor, the pressure regulation energy consumption factor and the temperature regulation energy consumption factor do not meet the energy consumption constraint condition;
[0098] When all of the flow regulation energy consumption factor, the speed regulation energy consumption factor, the pressure regulation energy consumption factor and the temperature regulation energy consumption factor are less than the regulation energy consumption factor threshold, it is deemed that the flow regulation energy consumption factor, the speed regulation energy consumption factor, the pressure regulation energy consumption factor and the temperature regulation energy consumption factor meet the energy consumption constraint condition.
[0099] Furthermore, the information assignment module 3 is also used to execute the following method:
[0100] The wafer size information, the semiconductor polishing equipment model, the alkaline component concentration information, the silicon dioxide concentration information, and the silicon dioxide particle size information are used as retrieval assignment conditions, and the polishing liquid flow timing information, the polishing disk speed timing information, the polishing pressure timing information, and the polishing temperature timing information are used as retrieval assignment targets, multiple production nodes are interacted, and a polishing log set is collected, wherein any polishing log includes the polishing liquid flow record timing information, the polishing disk speed record timing information, the polishing pressure record timing information, and the polishing temperature record timing information;
[0101] According to the polishing liquid flow rate recording timing information, the polishing disk rotation speed recording timing information, the polishing pressure recording timing information and the polishing temperature recording timing information, the polishing log set is discretized and deleted to obtain the initial optimized population.
[0102] Furthermore, the information assignment module 3 is also used to execute the following method:
[0103] Obtaining first polishing liquid flow rate record timing information of the first polishing log and second polishing liquid flow rate record timing information of the second polishing log;
[0104] Aligning the first polishing liquid flow rate recording timing information with the second polishing liquid flow rate recording timing information at the first moment to obtain a tail moment deviation, which is set as a first discrete coefficient;
[0105] Calculate the proportion of time periods in the specified time zone from the first moment to the front tail moment when the flow deviation between the first polishing liquid flow recording timing information and the second polishing liquid flow recording timing information is greater than or equal to the flow deviation threshold, and set it as a second discrete coefficient;
[0106] When the first discrete coefficient is less than or equal to a first discrete threshold, and the second discrete coefficient is less than or equal to a second discrete threshold, the first discreteness of the first polishing liquid flow recording timing information and the second polishing liquid flow recording timing information is equal to 0;
[0107] Otherwise, the first discreteness of the first polishing liquid flow rate recording timing information and the second polishing liquid flow rate recording timing information is equal to 1;
[0108] Traversing the polishing disk rotation speed recording timing information, the polishing pressure recording timing information and the polishing temperature recording timing information to obtain a second discreteness, a third discreteness and a fourth discreteness;
[0109] When the first discreteness, the second discreteness, the third discreteness, and the fourth discreteness are all equal to 0, randomly deleting one of the first polishing log and the second polishing log from the polishing log set;
[0110] The polishing log set is traversed in a loop to perform discretization deletion to obtain the initial optimized population.
[0111] Furthermore, the first function building module 4 is also used to execute the following method:
[0112] The first constraint function is:
[0113] ;
[0114] in, Characterizes the first constraint output value of the i-th individual in the initial optimization population, Characterizes the energy consumption factor of the i-th individual flow regulation, Characterize the energy consumption factor of polishing preset parameters flow regulation, Characterizes the energy consumption factor of the i-th individual speed regulation, Characterize the energy consumption factor of polishing preset parameter speed adjustment, Characterizes the energy consumption factor of the pressure regulation of the i-th individual, Characterize the polishing preset parameter pressure adjustment energy consumption factor, Characterizes the temperature regulation energy consumption factor of the i-th individual, Characterize the polishing preset parameter temperature adjustment energy consumption factor, , , and Characterize the weights of energy consumption factors for flow, speed, pressure and temperature regulation respectively.
[0115] Furthermore, the second function construction module 5 is also used to execute the following method:
[0116] The second constraint function is:
[0117] ;
[0118] in, Characterizes the second constraint output value of the i-th individual of the initial optimization population, Characterize the flatness factor of the i-th individual, Characterize the surface defect factor of the i-th individual, Characterize the surface damage factor of the i-th individual, Characterize the flatness factor weight, Characterize the weight of surface defect factors, Characterizing the weight of the surface damage factor, the flatness expectation factor, the surface defect expectation factor and the surface damage expectation factor are all positive factors.
[0119] Furthermore, the parameter updating module 6 is also used to execute the following method:
[0120] Configure the second constraint function output threshold;
[0121] According to the second constraint function, a first sorted population is sorted from the initial optimized population, which is less than or equal to the output threshold of the second constraint function;
[0122] The first sorting population is iterated and sorted a preset number of times according to the first constraint function, an optimal solution is output, and the optimized polishing parameters are generated.
[0123] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order and continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0124] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.
[0125] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.
Claims
1. A method for optimizing energy consumption of semiconductor polishing equipment, characterized in that: include: According to the wafer size information, the semiconductor polishing device interacts to receive polishing preset parameters, wherein the polishing preset parameters include alkaline component concentration information, silicon dioxide concentration information, silicon dioxide particle size information, polishing liquid flow timing information, polishing disk speed timing information, polishing pressure timing information and polishing temperature timing information; Traversing the polishing liquid flow timing information, the polishing disk speed timing information, the polishing pressure timing information and the polishing temperature timing information to perform energy consumption evaluation, and generate a flow regulation energy consumption factor, a speed regulation energy consumption factor, a pressure regulation energy consumption factor and a temperature regulation energy consumption factor; When the flow rate regulating energy consumption factor, the speed regulating energy consumption factor, the pressure regulating energy consumption factor and the temperature regulating energy consumption factor do not meet the energy consumption constraint condition, the alkaline component concentration information, the silicon dioxide concentration information, the silicon dioxide particle size information, the polishing liquid flow timing information, the polishing disk speed timing information, the polishing pressure timing information and the polishing temperature timing information are assigned values according to the wafer size information and the semiconductor polishing equipment to generate an initial optimized population; Constructing a first constraint function according to the flow rate regulation energy consumption factor, the speed regulation energy consumption factor, the pressure regulation energy consumption factor and the temperature regulation energy consumption factor; Constructing a second constraint function according to the flatness expected factor, the surface defect expected factor and the surface damage expected factor; According to the first constraint function and the second constraint function, performing double-constraint optimization on the initial optimization population to generate optimized polishing parameters, and updating the preset polishing parameters by using the optimized polishing parameters; Among them, traversing the polishing liquid flow timing information, the polishing disk speed timing information, the polishing pressure timing information and the polishing temperature timing information to perform energy consumption evaluation, generating a flow regulation energy consumption factor, a speed regulation energy consumption factor, a pressure regulation energy consumption factor and a temperature regulation energy consumption factor, including: Performing median analysis on the polishing liquid flow timing information, the polishing disk speed timing information, the polishing pressure timing information and the polishing temperature timing information to obtain flow median characteristics, speed median characteristics, pressure median characteristics and temperature median characteristics; Configure flow adjustment step threshold, speed adjustment step threshold, pressure adjustment step threshold and temperature adjustment step threshold; Counting the proportion of moments when the distance between the polishing liquid flow time series information and the flow median feature is greater than or equal to the flow adjustment step threshold, and setting it as the flow adjustment energy consumption factor; Counting the proportion of moments when the distance between the polishing disk speed time series information and the speed median feature is greater than or equal to the speed adjustment step threshold, and setting it as the speed adjustment energy consumption factor; Counting the proportion of moments when the distance between the polishing pressure timing information and the pressure median feature is greater than or equal to the pressure adjustment step threshold, and setting it as the pressure adjustment energy consumption factor; The proportion of moments when the distance between the polishing temperature timing information and the temperature median feature is greater than or equal to the temperature adjustment step threshold is counted and set as the temperature adjustment energy consumption factor.
2. The method according to claim 1, characterized in that When the flow rate regulation energy consumption factor, the speed regulation energy consumption factor, the pressure regulation energy consumption factor and the temperature regulation energy consumption factor do not meet the energy consumption constraint condition, it includes: The energy consumption constraint condition includes adjusting the energy consumption factor threshold; When one or more of the flow regulation energy consumption factor, the speed regulation energy consumption factor, the pressure regulation energy consumption factor and the temperature regulation energy consumption factor is greater than or equal to the regulation energy consumption factor threshold, it is deemed that the flow regulation energy consumption factor, the speed regulation energy consumption factor, the pressure regulation energy consumption factor and the temperature regulation energy consumption factor do not meet the energy consumption constraint condition; When all of the flow regulation energy consumption factor, the speed regulation energy consumption factor, the pressure regulation energy consumption factor and the temperature regulation energy consumption factor are less than the regulation energy consumption factor threshold, it is deemed that the flow regulation energy consumption factor, the speed regulation energy consumption factor, the pressure regulation energy consumption factor and the temperature regulation energy consumption factor meet the energy consumption constraint condition.
3. The method according to claim 1, characterized in that According to the wafer size information and the semiconductor polishing equipment, the alkaline component concentration information, the silicon dioxide concentration information, the silicon dioxide particle size information, the polishing liquid flow timing information, the polishing disk speed timing information, the polishing pressure timing information and the polishing temperature timing information are assigned values to generate an initial optimized population, including: The wafer size information, the semiconductor polishing equipment model, the alkaline component concentration information, the silicon dioxide concentration information, and the silicon dioxide particle size information are used as retrieval assignment conditions, and the polishing liquid flow timing information, the polishing disk speed timing information, the polishing pressure timing information, and the polishing temperature timing information are used as retrieval assignment targets, multiple production nodes are interacted, and a polishing log set is collected, wherein any polishing log includes the polishing liquid flow record timing information, the polishing disk speed record timing information, the polishing pressure record timing information, and the polishing temperature record timing information; According to the polishing liquid flow rate recording timing information, the polishing disk rotation speed recording timing information, the polishing pressure recording timing information and the polishing temperature recording timing information, the polishing log set is discretized and deleted to obtain the initial optimized population.
4. The method according to claim 3, characterized in that According to the polishing liquid flow rate record timing information, the polishing disk speed record timing information, the polishing pressure record timing information and the polishing temperature record timing information, the polishing log set is discretized and deleted to obtain the initial optimized population, including: Obtaining first polishing liquid flow rate record timing information of the first polishing log and second polishing liquid flow rate record timing information of the second polishing log; Aligning the first polishing liquid flow rate recording timing information with the second polishing liquid flow rate recording timing information at the first moment to obtain a tail moment deviation, which is set as a first discrete coefficient; Calculate the proportion of time periods in the specified time zone from the first moment to the front tail moment when the flow deviation between the first polishing liquid flow recording timing information and the second polishing liquid flow recording timing information is greater than or equal to the flow deviation threshold, and set it as a second discrete coefficient; When the first discrete coefficient is less than or equal to a first discrete threshold, and the second discrete coefficient is less than or equal to a second discrete threshold, the first discreteness of the first polishing liquid flow recording timing information and the second polishing liquid flow recording timing information is equal to 0; Otherwise, the first discreteness of the first polishing liquid flow rate recording timing information and the second polishing liquid flow rate recording timing information is equal to 1; Traversing the polishing disk rotation speed recording timing information, the polishing pressure recording timing information and the polishing temperature recording timing information to obtain a second discreteness, a third discreteness and a fourth discreteness; When the first discreteness, the second discreteness, the third discreteness, and the fourth discreteness are all equal to 0, randomly deleting one of the first polishing log and the second polishing log from the polishing log set; The polishing log set is traversed in a loop to perform discretization deletion to obtain the initial optimized population.
5. The method according to claim 1, characterized in that The first constraint function is: ; in, Characterizes the first constraint output value of the i-th individual in the initial optimization population, Characterizes the energy consumption factor of the i-th individual flow regulation, Characterize the energy consumption factor of polishing preset parameter flow regulation, Characterizes the energy consumption factor of the i-th individual speed regulation, Characterize the energy consumption factor of polishing preset parameter speed adjustment, Characterizes the energy consumption factor of the pressure regulation of the i-th individual, Characterize the polishing preset parameter pressure adjustment energy consumption factor, Characterizes the temperature regulation energy consumption factor of the i-th individual, Characterize the polishing preset parameter temperature adjustment energy consumption factor, , , and Characterize the weights of energy consumption factors for flow, speed, pressure and temperature regulation respectively.
6. The method according to claim 1, characterized in that The second constraint function is: ; in, Characterizes the second constraint output value of the i-th individual of the initial optimization population, Characterizes the flatness factor of the i-th individual, Characterize the flatness expectation factor, Characterize the surface defect factor of the i-th individual, Characterize the expected factor of surface defects, Characterize the surface damage factor of the i-th individual, Characterize the expected factor of surface damage, Characterize the flatness factor weight, Characterize the weight of surface defect factors, Characterizing the weight of the surface damage factor, the flatness expectation factor, the surface defect expectation factor and the surface damage expectation factor are all positive factors.
7. The method according to claim 1, characterized in that According to the first constraint function and the second constraint function, the initial optimization population is subjected to double constraint optimization to generate optimized polishing parameters, including: Configure the second constraint function output threshold; According to the second constraint function, a first sorted population is sorted from the initial optimized population, which is less than or equal to the output threshold of the second constraint function; The first sorting population is iterated and sorted a preset number of times according to the first constraint function, an optimal solution is output, and the optimized polishing parameters are generated.
8. An energy consumption optimization system for semiconductor polishing equipment, characterized in that: The steps for implementing the energy consumption optimization method of a semiconductor polishing equipment as described in any one of claims 1 to 7 include: Parameter receiving module: according to the wafer size information, interacting with the semiconductor polishing equipment, receiving the polishing preset parameters, wherein the polishing preset parameters include the concentration information of the alkaline component, the concentration information of silicon dioxide, the particle size information of silicon dioxide, the flow timing information of the polishing liquid, the rotation timing information of the polishing disk, the polishing pressure timing information and the polishing temperature timing information; Energy consumption evaluation module: traverses the polishing liquid flow timing information, the polishing disk speed timing information, the polishing pressure timing information and the polishing temperature timing information to perform energy consumption evaluation, and generates a flow regulation energy consumption factor, a speed regulation energy consumption factor, a pressure regulation energy consumption factor and a temperature regulation energy consumption factor; Information assignment module: when the flow rate regulating energy consumption factor, the speed regulating energy consumption factor, the pressure regulating energy consumption factor and the temperature regulating energy consumption factor do not meet the energy consumption constraint condition, the alkaline component concentration information, the silicon dioxide concentration information, the silicon dioxide particle size information, the polishing liquid flow timing information, the polishing disk speed timing information, the polishing pressure timing information and the polishing temperature timing information are assigned according to the wafer size information and the semiconductor polishing equipment to generate an initial optimized population; A first function construction module: constructing a first constraint function according to the flow rate regulation energy consumption factor, the speed regulation energy consumption factor, the pressure regulation energy consumption factor and the temperature regulation energy consumption factor; Second function construction module: constructing the second constraint function according to the flatness expected factor, the surface defect expected factor and the surface damage expected factor; Parameter updating module: performing double-constraint optimization on the initial optimization population according to the first constraint function and the second constraint function, generating optimized polishing parameters, and updating the preset polishing parameters by using the optimized polishing parameters.
