Environment control method and system for preparing aluminum nitride vacuum detector

Abstract

The invention discloses an environment control method and system for aluminum nitride vacuum detector preparation, and relates to the technical field of vacuum detector preparation, and the method comprises the steps: obtaining the characteristics of an aluminum nitride vacuum detector preparation environment, constructing a monitoring system, carrying out the real-time monitoring and filtering of key parameters, and storing the data to a database. And predicting key parameters based on the database, calculating the deviation between a predicted value and an actual value, inputting the deviation into the optimization control model to obtain a strategy, and carrying out adaptive control on the preparation environment according to the strategy. The technical problems that in the prior art, an aluminum nitride vacuum detector preparation environment is not accurate in monitoring, lacks effective prediction and control and is difficult to adapt to complex environment changes are solved, accurate monitoring, effective prediction and self-adaptive control over the aluminum nitride vacuum detector preparation environment are achieved, and the production efficiency of the aluminum nitride vacuum detector is improved. The technical effects of improving the preparation quality and stability of the detector and improving the efficiency and controllability of the preparation process are achieved.

Description

Technical Field

[0001] The present application relates to the technical field of vacuum detector preparation, and in particular to an environmental control method and system for preparing aluminum nitride vacuum detectors. Background Art

[0002] In the preparation scenario of aluminum nitride vacuum detectors, the problem of precise control of the preparation environment is particularly prominent, and the demand for advanced environmental control methods is also more prominent. Achieving efficient environmental control and better meeting the needs of high-quality preparation of detectors have become a crucial link in the preparation of aluminum nitride vacuum detectors. Traditional environmental control for the preparation of aluminum nitride vacuum detectors is often relatively simple and localized, lacking a comprehensive and accurate environmental parameter monitoring system, insufficient overall control of the preparation environment, lacking in-depth analysis and utilization of historical data on key environmental parameters, and difficult to accurately predict environmental change trends. There are unreasonable situations in the allocation of environmental control resources, which makes it impossible to adapt to the complex and changeable preparation environment. The formulation of environmental control plans is relatively fixed and cannot cope well with the ever-changing actual conditions during the preparation process.

[0003] At present, relevant technologies have the following technical problems: inaccurate environmental monitoring in the preparation of aluminum nitride vacuum detectors, lack of effective prediction and control, and difficulty in adapting to complex environmental changes. Summary of the invention

[0004] This application provides an environmental control method and system for the preparation of aluminum nitride vacuum detectors, and uses the environmental characteristics of aluminum nitride vacuum detectors to construct a monitoring system, monitor and filter key parameters in real time, and store data in a database. Based on the database, key parameters are predicted, the deviation between the predicted value and the actual value is calculated, and the deviation is input into the optimization control model to obtain a strategy. The preparation environment is adaptively controlled according to the strategy. By realizing accurate monitoring, effective prediction and adaptive control of the preparation environment of aluminum nitride vacuum detectors, the technical effect of improving the preparation quality and stability of the detector and improving the efficiency and controllability of the preparation process is achieved.

[0005] The present application provides an environmental control method for preparing an aluminum nitride vacuum detector, comprising:

[0006] Acquire the characteristics of the aluminum nitride vacuum detector preparation environment, and construct an environmental parameter monitoring system based on the characteristics of the aluminum nitride vacuum detector preparation environment; use the environmental parameter monitoring system to monitor and filter the key parameters of the detector preparation environment in real time, obtain historical data and real-time data of the key parameters of the detector preparation environment, and store them in a preparation environment database; predict the key parameters of the detector preparation environment based on the preparation environment database to obtain predicted values ​​of the key parameters of the detector preparation environment within a variable time window; calculate the numerical deviation between the predicted values ​​of the key parameters of the detector preparation environment and the actual values ​​of the key parameters of the detector preparation environment; input the numerical deviation into a preset optimization control model for analysis to obtain an optimization control strategy for the detector preparation environment, the optimization control strategy including optimization control parameters of the key parameters of the detector preparation environment; and adaptively control the aluminum nitride vacuum detector preparation environment according to the optimization control strategy.

[0007] The present application also provides an environmental control system for preparing an aluminum nitride vacuum detector, comprising:

[0008] An environmental parameter monitoring system construction module, the environmental parameter monitoring system construction module is used to obtain the characteristics of the aluminum nitride vacuum detector preparation environment, and to construct an environmental parameter monitoring system based on the characteristics of the aluminum nitride vacuum detector preparation environment; a preparation environment database storage module, the preparation environment database storage module is used to perform real-time monitoring and data filtering on the key parameters of the detector preparation environment through the environmental parameter monitoring system, obtain the historical data and real-time data of the key parameters of the detector preparation environment, and store them in the preparation environment database; a key parameter prediction module, the key parameter prediction module is used to predict the key parameters of the detector preparation environment based on the preparation environment database, and obtain variable time The predicted values ​​of the key parameters of the detector preparation environment within the window; a numerical deviation calculation module, the numerical deviation calculation module is used to calculate the numerical deviation between the predicted values ​​of the key parameters of the detector preparation environment and the actual values ​​of the key parameters of the detector preparation environment; an optimization control strategy acquisition module, the optimization control strategy acquisition module is used to input the numerical deviation into a preset optimization control model for analysis, and obtain the optimization control strategy of the detector preparation environment, the optimization control strategy includes the optimization control parameters of the key parameters of the detector preparation environment; an adaptive control module, the adaptive control module is used to adaptively control the aluminum nitride vacuum detector preparation environment according to the optimization control strategy.

[0009] The environmental control method and system for the preparation of aluminum nitride vacuum detectors proposed in this application first obtains the environmental characteristics of aluminum nitride vacuum detector preparation to build a monitoring system, monitors and filters key parameters in real time, and stores data in a database. Based on the database, key parameters are predicted, the deviation between the predicted value and the actual value is calculated, and the deviation is input into the optimization control model to obtain a strategy. The preparation environment is adaptively controlled according to the strategy. By realizing accurate monitoring, effective prediction and adaptive control of the preparation environment of aluminum nitride vacuum detectors, the technical effect of improving the preparation quality and stability of the detector and improving the efficiency and controllability of the preparation process is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solution of the embodiment of the present invention, the accompanying drawings of the embodiment of the present invention will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0011] Figure 1 A schematic flow chart of an environmental control method for preparing an aluminum nitride vacuum detector provided in an embodiment of the present application;

[0012] Figure 2 A schematic diagram of the structure of an environmental control system for preparing an aluminum nitride vacuum detector provided in an embodiment of the present application.

[0013] Explanation of the accompanying drawings: environmental parameter monitoring system construction module 10, preparation environment database storage module 20, key parameter prediction module 30, numerical deviation calculation module 40, optimization control strategy acquisition module 50, adaptive control module 60. DETAILED DESCRIPTION

[0014] 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.

[0015] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0016] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations 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 clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0017] The present application embodiment provides an environmental control method for preparing an aluminum nitride vacuum detector, such as Figure 1 As shown, the method includes:

[0018] Step S100, obtain the characteristics of the aluminum nitride vacuum detector preparation environment, and build an environmental parameter monitoring system based on the characteristics of the aluminum nitride vacuum detector preparation environment. Specifically, in the process of building the environmental parameter monitoring system for the preparation of aluminum nitride vacuum detectors, first obtain the preparation environment characteristics, including attribute data characteristics, event data characteristics, and service data characteristics. For attribute data, clarify the environmental parameters and equipment attribute types, understand their range of changes and the impact on the performance of the detector; for event data, cover monitoring, abnormal and control events, master their occurrence frequency, impact degree and early warning signals, etc.; for service data, evaluate the methods, goals and effects of data collection, analysis and control services, select appropriate sensors and data acquisition equipment according to the characteristics, establish reliable data transmission channels and storage solutions, develop monitoring system software to realize real-time monitoring, data analysis and control functions, and finally optimize and improve through testing to improve the stability and reliability of the system and provide strong guarantees for detector preparation.

[0019] Step S200, the environmental parameter monitoring system performs real-time monitoring and data filtering on the key parameters of the detector preparation environment, obtains the historical data and real-time data of the key parameters of the detector preparation environment, and stores them in the preparation environment database. Specifically, the environmental parameter monitoring system performs real-time monitoring on the key parameters of the detector preparation environment, uses the box plot method to filter the data, groups the key parameter data in chronological order, calculates the lower, middle, and upper quartiles and the interquartile range to determine the range of abnormal values, removes the abnormal value data points, and after filtering, stores the historical data and real-time data of the key parameters in the preparation environment database. The database classifies and stores attribute data (including environmental parameter attribute data and equipment attribute data), event data (monitoring event data, abnormal event data, and control event data), and service data (data collection services, data analysis services, and control services) for subsequent targeted analysis and processing, providing a basis for problem troubleshooting and optimization control, and improving the level of preparation environment management.

[0020] In a possible implementation, the environmental parameter monitoring system is used to perform real-time monitoring and data filtering on the key parameters of the detector preparation environment, obtain historical data and real-time data of the key parameters of the detector preparation environment, and store them in the preparation environment database. Step S200 further includes step S210, grouping the key parameters of the detector preparation environment according to time sequence to obtain multiple groups of key parameter grouping data. Specifically, during the detector preparation process, the environmental parameter monitoring system continuously collects key parameter data. Since the data is continuously generated over time, in order to better analyze the distribution and change patterns of the data, the key parameters are grouped according to time sequence, and the data is divided into multiple time period groups according to certain time intervals, such as every hour, every day, etc., to facilitate the subsequent separate analysis of the data in each time period, and also to better observe the changes in the data on different time scales.

[0021] Step S220, for the multiple groups of key parameter grouped data, calculate the lower quartile, median, and upper quartile of each group of the key parameter grouped data. Specifically, for each group of key parameter grouped data, it is necessary to calculate its lower quartile, median and upper quartile. The statistics can reflect the distribution of the data. The lower quartile represents the middle value of the lower part of the data. When calculating the lower quartile, first sort the group of data from small to large, and then determine the value at one-quarter of the data. If the number of data is an odd number, the lower quartile is the value of the (n+1) / 4th data after sorting. If the number of data is an even number, the lower quartile is the average of the n / 4th data and the (n / 4+1)th data after sorting. The median is the middle value of the data. After sorting the data from small to large, if the number of data is an odd number, the median is the middle data value after sorting. If the number of data is an even number, the median is the average of the two middle data after sorting. The upper quartile represents the middle value of the higher part of the data. The calculation method is similar to that of the lower quartile. Determine the value at three-quarters of the data.

[0022] Step S230, obtaining the interquartile range of the key parameter grouping data, the interquartile range being the upper quartile minus the lower quartile. Specifically, after calculating the upper quartile and the lower quartile of each group of key parameter grouping data, the interquartile range is obtained by subtracting the two, and the interquartile range reflects the dispersion of the data. A smaller interquartile range indicates that the data is relatively concentrated, while a larger interquartile range indicates that the distribution of the data is relatively dispersed.

[0023] Step S240, determine the outlier range, determine whether the data point is within the outlier range, if so, mark the data point as an outlier data point, and remove the outlier data point from the key parameter grouping data. Specifically, determine the outlier range according to the interquartile range, and set the outlier range to be less than the lower quartile minus 1.5 times the interquartile range, or greater than the upper quartile plus 1.5 times the interquartile range. For each data point, determine whether it is within this outlier range. If the value of the data point is less than the lower quartile minus 1.5 times the interquartile range, or greater than the upper quartile plus 1.5 times the interquartile range, the data point is marked as an outlier data point. Once the outlier data point is determined, it is removed from the key parameter grouping data to remove abnormal data that may be caused by sensor failure, sudden interference, etc., to improve the quality and reliability of the data, so that more accurate analysis and processing can be performed later.

[0024] In a possible implementation, the key parameters of the detector preparation environment are monitored and filtered in real time by the environmental parameter monitoring system, and the historical data and real-time data of the key parameters of the detector preparation environment are obtained and stored in the preparation environment database. Step S200 further includes step S250, which obtains the attribute data of the key parameters of the detector preparation environment, classifies the attribute data according to the attribute type, obtains the environmental parameter attribute data and the equipment attribute data, and stores the environmental parameter attribute data and the equipment attribute data separately. Specifically, the attribute data of the key parameters of the detector preparation environment are obtained, and the key parameters in the detector preparation environment are clarified, including environmental related parameters such as ambient temperature, humidity, vacuum degree, and equipment attribute parameters such as equipment model and location. By setting corresponding sensors and monitoring equipment in the preparation environment, these attribute data are collected in real time. For example, temperature and humidity sensors can be installed in the preparation workshop to obtain ambient temperature and humidity data, and the model and location information of each detector preparation equipment are recorded at the same time. The attribute data are classified according to the attribute type, and the acquired attribute data are divided into two categories according to their properties: environmental parameter attribute data and equipment attribute data. The environmental parameter attribute data mainly includes The data includes parameters that reflect environmental conditions, such as temperature, humidity, and vacuum degree; the equipment attribute data covers equipment model, location, and other equipment-related information. For example, the data obtained through the temperature and humidity sensor is classified as environmental parameter attribute data, while the manufacturer, model, and other information of the equipment are classified as equipment attribute data. The environmental parameter attribute data and the equipment attribute data are stored separately. In order to facilitate management and query, the environmental parameter attribute data and the equipment attribute data are stored in different database tables or files. For example, create a table called "environmental parameter table" to store temperature, humidity, vacuum degree, and other data, and then create a "device information table" to store equipment model, location, and other information. At the same time, for fields that are frequently used for query and screening, such as equipment model, temperature range, etc., indexes are established to improve query efficiency.

[0025] Step S260, obtain the event data of the key parameters of the detector preparation environment, the event data includes monitoring event data, abnormal event data and control event data, and store the monitoring event data, abnormal event data and control event data separately. Specifically, the event data of the key parameters of the detector preparation environment is obtained, and the event data mainly includes monitoring event data, abnormal event data and control event data. The monitoring event data is the routine monitoring result of the detector preparation environment; the abnormal event data is the data generated when the environment is abnormal; the control event data is the data generated when the environment is adjusted and controlled. For example, by continuously monitoring the ambient temperature, each temperature reading can be used as a monitoring event data. If the temperature exceeds the set range, abnormal event data will be generated. When taking measures to adjust the temperature, such as turning on the air conditioner or adjusting the ventilation system, control event data will be generated. The monitoring event data, abnormal event data and control event data are stored separately, and the three types of event data are stored in different database tables. For the monitoring event data, the timestamp can be used as the primary key to store the key parameter values ​​of each monitoring. For example, a "monitoring event table" is created, which contains a timestamp field and key parameter fields such as temperature, humidity, and vacuum. Each monitored data is stored in the table in chronological order. For abnormal event data, in addition to the timestamp and abnormality type, the degree of abnormality and related key parameter values ​​need to be recorded in detail. For example, when the temperature is too high and causes an abnormality, the timestamp, the abnormality type of "temperature is too high", the degree of abnormality (such as the specific value beyond the normal temperature range) and the temperature value at that time are recorded in the "Abnormal Event Table". For control event data, the time, method and adjusted key parameter values ​​of the control operation are stored. For example, when the temperature is lowered by turning on the air conditioner, the time and method of the control operation are recorded in the "Control Event Table" as "turn on cooling" and the adjusted temperature value.

[0026] Step S270, obtain the service data of the key parameters of the detector preparation environment, the service data includes data collection service, data analysis service and control service, and store the data collection service, data analysis service and control service separately. Specifically, obtain the service data of the key parameters of the detector preparation environment, the service data includes data collection service, data analysis service and control service, the data collection service is responsible for collecting the key parameters of the environment; the data analysis service analyzes and processes the collected data; the control service adjusts and controls the environment according to the analysis results. For example, the data collection service collects environmental temperature, humidity and other data through sensors at regular intervals; the data analysis service uses a specific algorithm to analyze the collected data to determine whether the environment is normal; the control service decides whether to turn on cooling, adjust humidity, etc. according to the analysis results. The data collection service, data analysis service and control service are stored separately, and for the data collection service, its collection frequency, data source and other configurations are stored Information, for example, creating a data collection service table to record the collection frequency (such as once every 5 minutes), data source (such as the specific sensor number or name) and other information; recording the algorithm used by the data analysis service, the time range and results of the analysis, for example, recording the name of the algorithm used (such as the mean algorithm, trend analysis algorithm, etc.), the time range of the analysis (such as the past 24 hours) and the analysis results (such as the judgment of whether the environment is normal or abnormal) in the data analysis service table; saving the control policy generation rules and execution history records of the control service, for example, recording the control policy generation rules (such as turning on the air conditioner when the temperature exceeds a certain value) and the execution history records of each control operation in the control service table, including the execution time, control method and other information.

[0027] Step S300, based on the preparation environment database, predict the key parameters of the detector preparation environment to obtain the predicted values ​​of the key parameters of the detector preparation environment within the variable time window. Specifically, based on the preparation environment database, the historical operation data and real-time operation data of the detector preparation environment are sorted out, and predictions are made using deep learning algorithms such as recurrent neural networks (RNN) and long short-term memory networks (LSTM). The historical operation data is input into the model for training to learn the law of key parameter changes, and then the real-time operation data is input into the trained model to obtain the predicted values ​​of the environmental parameters in the future. The length of the time window affects the prediction accuracy. Generally, the shorter the time window, the higher the prediction accuracy, because the changes in environmental parameters in the short term are relatively stable; the longer the time window, the greater the impact of uncertainty factors, and the lower the prediction accuracy. In practical applications, it is necessary to determine the appropriate time window according to specific needs to improve the preparation quality and efficiency of the detector.

[0028] In a possible implementation, based on the preparation environment database, the key parameters of the detector preparation environment are predicted to obtain the predicted values ​​of the key parameters of the detector preparation environment within a variable time window. Step S300 further includes step S310, constructing a key parameter prediction model, and the key parameter prediction model is constructed based on the historical data of the key parameters of the aluminum nitride vacuum detector preparation environment. Specifically, the range of the key parameters of the aluminum nitride vacuum detector preparation environment is clarified, and the key parameters include temperature, humidity, vacuum degree, gas composition, etc. A large amount of historical data is collected. The historical data is recorded by the environmental parameter monitoring system during the past detector preparation process. For the construction of the model, based on machine learning algorithms, especially time series analysis algorithms, considering the time series characteristics of the data, an algorithm suitable for processing time series data is selected, such as a long short-term memory network. The long short-term memory network model is constructed. First, the collected historical data is cleaned and preprocessed to remove outliers and noise. Statistical methods, such as the box plot method, are used to determine the range of outliers and eliminate outlier data points. The data is normalized and the data values ​​are mapped to a specific interval so that the model can better learn the data. Features, design the structure of the long short-term memory network, including determining the number of nodes in the input layer, hidden layer, and output layer. The number of nodes in the input layer usually corresponds to the number of key parameters. The hidden layer can be set to multiple levels according to the complexity of the data and the prediction requirements. Each level contains a certain number of long short-term memory network units. The number of nodes in the output layer is usually the number of key parameters that need to be predicted. The preprocessed historical data is input into the designed long short-term memory network for training. During the training process, the network weights and biases are continuously adjusted to minimize the error between the predicted value and the actual value. The back propagation algorithm and optimization algorithm, such as the stochastic gradient descent method, are used to update the network parameters. The training process usually requires multiple iterations until the model converges or reaches the preset number of training times.

[0029] Step S320, using the key parameter prediction model to predict the key parameters of the detector preparation environment according to the real-time data, and obtaining the predicted values ​​of the key parameters of the detector preparation environment within the variable time window. Specifically, after the key parameter prediction model is constructed, the model is used in combination with real-time data to predict the key parameters of the detector preparation environment, and real-time data is obtained. The values ​​of the key parameters of the detector preparation environment collected in real time by the environmental parameter monitoring system are input into the trained key parameter prediction model. The model predicts the key parameters of the detector preparation environment within the variable time window according to the real-time data and the learned historical data pattern. The length of the variable time window can be adjusted according to actual needs. A shorter time window may provide a more accurate short-term prediction, while a longer time window can provide a rough estimate of future trends. The model outputs the predicted values ​​of the key parameters of the detector preparation environment within the variable time window. The predicted values ​​can provide an important reference basis for environmental control during the detector preparation process, and help operators adjust environmental parameters in a timely manner to ensure the preparation quality and stability of the detector.

[0030] In a possible implementation, based on the preparation environment database, the key parameters of the detector preparation environment are predicted to obtain the predicted values ​​of the key parameters of the detector preparation environment within a variable time window, and step S300 further includes step S330, obtaining the attribute data, event data, and service data from the preparation environment database. Specifically, the various required data are extracted from the already constructed preparation environment database, the attribute data includes environmental parameter attribute data (such as temperature, humidity, vacuum degree, etc.) and equipment attribute data (such as equipment model, location, etc.), the event data covers monitoring event data (recording key parameter values ​​for routine monitoring), abnormal event data (recording abnormal conditions and related parameter values) and control event data (recording key parameter values ​​for control operations and adjustments), and the service data includes data acquisition service data (such as configuration information such as acquisition frequency and data source), data analysis service data (recording the algorithm used, the time range and results of the analysis) and control service data (saving control strategy generation rules and execution history records).

[0031] Step S340, according to the data characteristics of the attribute data, determine the reference range of the variable time window. Specifically, analyze the data characteristics of the attribute data, which include trend, importance, sensitivity, change speed and amplitude, etc. For example, if the change trend of a key parameter is relatively stable, the change speed is slow, and the amplitude is small, then a relatively long time window reference range can be preliminarily determined. If the parameter changes frequently, the amplitude is large, and the sensitivity is high, then a shorter time window reference range may be required. By analyzing the characteristics of different attribute data, a rough reference range of a variable time window is comprehensively determined.

[0032] Step S350, obtaining sub-event data in the event data, associating data features of the sub-event data with actual event features, and obtaining event constraint conditions, wherein the sub-event data includes monitoring event data, abnormal event data, and control event data. Specifically, for monitoring event data, analyze the parameter changes and detection frequency of its data characteristics. If the monitoring frequency is high and the parameter changes are slow, a longer time range can be considered when determining the time window, because frequent monitoring can capture parameter changes in a timely manner. On the contrary, if the monitoring frequency is low, the parameter changes may change greatly between two monitorings. At this time, a shorter time window is required to ensure timely response; for abnormal event data, pay attention to the frequency of abnormal occurrence. If the frequency of abnormal events is high, it means that the environment is unstable, and a shorter time window may be required to promptly discover and handle abnormal situations. If the frequency of abnormal occurrence is low, the time window can be appropriately relaxed; for control event data, consider the time delay between control instructions and parameter changes. If the parameters change quickly after the control instructions are issued and the time delay is short, a shorter time window can be selected to achieve more timely control. If the time delay is long, a longer time window is required to ensure the effectiveness of the control. The event constraints are obtained by analyzing the correlation between these sub-event data features and the actual event features.

[0033] Step S360, obtaining sub-service data in the service data, associating the data features of the sub-service data with the actual service features, and obtaining service constraints. The sub-service data includes data collection service data, data analysis service data, and control service data. Specifically, for data collection service data, the collection frequency is considered. If the collection frequency is high, more real-time data can be provided, and a shorter time window can be supported for more accurate analysis and prediction. If the collection frequency is low, the data update is slow, and a longer time window is required; for data analysis service data, the adaptability of the key parameter prediction model to the variable time window is analyzed. If the model shows good prediction accuracy and stability in a shorter time window, a shorter time window can be given priority. If the model can better capture trend changes under a longer time window, a longer time window can be considered; for control service data, its response capability under different variable time windows is analyzed. If the control service can quickly respond and adjust environmental parameters in a shorter time window, a shorter time window can be selected. If the response speed is slow, a longer time window may be required to prepare for control in advance; service constraints are obtained by analyzing the association between the sub-service data features and the actual service features.

[0034] Step S370, constrain the reference range of the variable time window according to the event constraint and the service constraint, and obtain multiple time window lengths. Specifically, the event constraint and the service constraint are comprehensively considered, and the previously determined variable time window reference range is further constrained. According to different combinations of constraint conditions, multiple possible time window lengths can be obtained. For example, a shorter time window length can be obtained by combining a higher frequency of abnormal occurrence in the event constraint and a faster data collection frequency and control response capability in the service constraint. If the frequency of abnormal occurrence in the event constraint is low and the model in the service constraint performs better under a longer time window, a longer time window length may be obtained.

[0035] Step S380, judging the actual effects of the multiple time window lengths, and determining the optimal time window length. Specifically, the actual effects of the multiple time window lengths obtained are judged, and the prediction accuracy, control effect, etc. under different time window lengths are evaluated through simulation experiments, historical data analysis, etc. For example, in a simulated detector preparation environment, different time window lengths are used for prediction and control, and the changes in parameters and the preparation quality of the detector are observed. According to the evaluation results, the time window length that can achieve the best balance in terms of prediction accuracy, control effect, response speed, etc. is selected as the optimal time window length.

[0036] In a possible implementation, the actual effects of the multiple time window lengths are judged, and the optimal time window length is determined. Step S380 further includes step S381, which periodically evaluates the optimal time window length and calculates a prediction error index. Specifically, after determining the optimal time window length, it is necessary to periodically evaluate it and select appropriate prediction error indicators, such as root mean square error, mean absolute error, etc. The indicators can measure the accuracy of the prediction, and the prediction error index is calculated by comparing the difference between the predicted value and the actual value. The historical data and real-time data in the prepared environment database are used to compare the actual environmental parameter value with the predicted value under the current optimal time window length, and the calculation is performed according to the calculation formula of the error index.

[0037] Step S382, compare the prediction error index with the preset prediction accuracy threshold, if the prediction error index is less than the preset prediction accuracy threshold, formulate the optimal solution of the window time length adjustment strategy. Specifically, set a preset prediction accuracy threshold, which represents an acceptable prediction error range, and compare the calculated prediction error index with the preset threshold. If the prediction error index is less than the preset prediction accuracy threshold, it means that the current optimal time window length is within the acceptable range and the prediction accuracy is high. If the prediction error index is greater than the preset threshold, it means that the prediction accuracy has decreased and the time window length needs to be adjusted.

[0038] Step S383, re-determine the optimal window time length according to the window time length adjustment strategy optimization solution. Specifically, according to different trigger conditions, select the corresponding adjustment strategy. If the prediction accuracy decreases, you can consider shortening the time window to capture parameter changes more timely, because a longer time window may lead to untimely response to recent parameter changes. Shortening the time window can improve the sensitivity to rapidly changing environmental parameters. If the control effect is not good, you can try to adjust the time window to better adapt to the control needs. For example, if the response time of the control instruction under the current time window is too long or the control effect is not obvious, you can adjust the time window length appropriately to find a more suitable time range for control. For new periodic changes in environmental parameters, you can adjust the time window according to the cycle length. If new periodic fluctuations in environmental parameters are found, analyze the cycle length, and then adjust the time window length accordingly to match it with the cycle length, so as to better predict and control environmental parameters; small-scale step-by-step adjustment. In order to avoid the instability caused by large-scale adjustment of the time window, a small-scale step-by-step adjustment method is adopted. For example, a certain proportion of the time window length is adjusted each time, and a more suitable time window length is gradually explored without causing drastic fluctuations in the system; before the adjusted time window is actually applied, historical data or simulation data are used for simulation prediction and control effect verification, and a part of historical data is selected from the preparation environment database. Historical data, or use the simulation data generator to generate simulation data. In the simulation environment, use the adjusted time window for prediction and control, and observe the prediction results and control effects. If the simulation results show that the adjusted time window can improve the prediction accuracy and control effect, continue to apply it; otherwise, adjust the strategy and try again; apply the adjusted time window to the actual environmental control, and continuously monitor the effect. If the simulation verification passes, apply the adjusted time window to the actual aluminum nitride vacuum detector preparation environment control, continuously monitor the changes in environmental parameters and control effects, and collect actual operation data to timely discover problems and make adjustments. Feedback on actual effects, continuously optimize and adjust strategies, so that the system can better adapt to different environmental changes. According to the feedback on the effects in actual applications, analyze whether the adjusted time window has really improved the prediction accuracy and control effect. If the effect is good, continue to use the current adjustment strategy; if new problems arise or the effect is not good, re-analyze the reasons, adjust the strategy, and continuously optimize the performance of the system; record the process and results of each adjustment for subsequent analysis and learning, and improve the system's adaptive ability. During the entire adjustment process, record in detail the time window length, trigger conditions, adjustment strategy, simulation verification results, and actual application effects of each adjustment. These records can provide valuable experience for subsequent analysis and learning, help the system continuously improve its adaptive ability, and better cope with various environmental changes.

[0039] Step S400, calculate the numerical deviation between the predicted value of the key parameter of the detector preparation environment and the actual value of the key parameter of the detector preparation environment. Specifically, in the process of calculating the numerical deviation between the predicted value and the actual value of the key parameter of the detector preparation environment, first determine the key parameters including temperature, humidity, vacuum degree, etc., obtain the actual value through the environmental parameter monitoring system, and use the prediction model to obtain the predicted value. The deviation can be calculated by using absolute deviation, relative deviation, root mean square error, etc. For example, for the temperature parameter, compare the actual measured value with the predicted value at a certain moment, and obtain the deviation value through a specific calculation method. The size of the numerical deviation can evaluate the accuracy of the prediction model, guide the environmental control strategy, and continuously improve the performance of the environmental parameter monitoring system and the prediction model.

[0040] Step S500, input the numerical deviation into a preset optimization control model for analysis, and obtain an optimization control strategy for the detector preparation environment, wherein the optimization control strategy includes optimization control parameters for key parameters of the detector preparation environment. Specifically, in the scheme, the key parameters of the detector preparation environment are first clarified, the actual values ​​are collected through the environmental parameter monitoring system, and the predicted values ​​are obtained using the prediction model, and the numerical deviation between the two is calculated. The deviation is input into a preset optimization control model, which comprehensively considers factors such as environmental conditions, equipment operation conditions, and preparation process requirements, analyzes the deviation, determines the optimization control parameters including specific values, control methods, and timing, etc., forms an optimization control strategy and outputs it, provides specific guidance for high-quality preparation of detectors, and improves the performance and quality of detectors by adjusting environmental parameters to make them close to the ideal state.

[0041] Step S600, according to the optimization control strategy, the preparation environment of the aluminum nitride vacuum detector is adaptively controlled. Specifically, in the scheme for adaptively controlling the preparation environment of the aluminum nitride vacuum detector, the optimization control strategy obtained by the preset optimization control model includes the optimization control parameters of key parameters and the timing of the control method, etc., identifies different stages in the preparation process, such as material growth, device processing and packaging, etc., and each stage has different process requirements. The system makes judgments by monitoring specific parameter combinations, change trends, equipment operating status and other factors, and adaptively controls the requirements of specific stages according to the optimization control strategy, adjusts the corresponding control equipment to meet the environmental parameter requirements, and continuously monitors in real time. If parameter deviations are found, feedback is immediately adjusted. As the preparation process proceeds, the system continuously accumulates data experience, optimizes its own control strategy, and can also interact with the outside to learn new methods and improve adaptive capabilities, thereby improving the quality and efficiency of detector preparation.

[0042] In a possible implementation, according to the optimization control strategy, the preparation environment of the aluminum nitride vacuum detector is adaptively controlled, and step S600 further includes step S610, establishing a preparation stage feature library, and the preparation stage feature library includes environmental parameter characteristics of different preparation stages. Specifically, the preparation process of the aluminum nitride vacuum detector is deeply analyzed and divided into different preparation stages, mainly including material growth, device processing and packaging stages. For each stage, the relevant environmental parameter characteristics are collected and sorted. For the material growth stage, the environmental parameter characteristics of this stage are determined by analyzing the previous preparation data and communicating with experts. The typical range of environmental parameters may include a specific temperature range, a certain vacuum range, and a specific gas component ratio. For example, the temperature may need to be maintained in a relatively stable high temperature range to promote the growth of the material, and the vacuum degree also needs to reach a certain value to ensure the purity of the growth environment. The change trend of the key parameters may be that as the growth time goes by, the temperature gradually increases and then stabilizes at a specific value, and the vacuum degree needs to remain relatively stable throughout the growth process. The operating status of a specific device may include the power output of the growth device and the flow stability of the gas supply device etc. For the device processing stage, the environmental parameter characteristics will be different. The typical range may be more precise temperature control and lower humidity levels. The changing trend of key parameters may be that during the processing, the temperature needs to be fine-tuned according to different processing steps, and the humidity should always be kept at a low level to prevent the device from being corroded. The operating status of specific equipment may include the precision adjustment of processing equipment, the operating efficiency of the cooling system, etc. For the packaging stage, the environmental parameter characteristics change again. The typical range may include a strictly controlled humidity range and specific cleanliness requirements. The changing trend of key parameters is that during the packaging process, the humidity must be quickly reduced to an extremely low level, and the cleanliness should always be maintained at a high level. The operating status of specific equipment includes the sealing of the packaging equipment, the operating status of the air purification equipment, etc. The environmental parameter characteristics of different preparation stages are sorted out and summarized, and a preparation stage feature library is established to provide a basis for subsequent optimization target determination and control strategy formulation.

[0043] Step S620, determining the optimization targets of different preparation stages in the preparation stage feature library, and formulating environmental parameter control indicators for the different preparation stages. Specifically, based on actual needs and past experience, the optimization goals of different preparation stages are determined. For the material growth stage, the optimization goal is to obtain high-quality material crystal structure and improve the purity and performance of the material. It is required that the environmental parameters at this stage can promote material growth to the greatest extent and remain stable. For example, it is necessary to ensure that parameters such as temperature, vacuum degree and gas composition are always in the most favorable state for material growth, and avoid parameter fluctuations from having adverse effects on the growth process. For the device processing stage, the optimization goal may be to achieve high-precision device processing and improve the performance and reliability of the device. It is necessary to strictly control the environmental parameters at this stage to ensure processing accuracy and stability. For example, the temperature is controlled within a very small fluctuation range to prevent the influence of thermal expansion and contraction on processing accuracy; maintain a low humidity level to prevent the device from being corroded by moisture. For the packaging stage, the optimization goal may be to ensure the sealing and reliability of the packaging and improve the service life and stability of the detector. It is required to strictly control the environmental parameters at this stage to meet the special requirements of the packaging. For example, the humidity is controlled at an extremely low level to prevent moisture from entering the interior of the package; maintain high cleanliness to prevent dust and impurities from affecting the packaging quality.

[0044] Step S630, formulate an adaptive control strategy according to the environmental parameter characteristics and environmental parameter control indicators of different preparation stages. Specifically, for different preparation stages, according to their optimization goals and environmental parameter characteristics, formulate specific environmental parameter control indicators. For the material growth stage, the formulated environmental parameter control indicators include the allowable temperature fluctuation range, such as within plus or minus a few degrees Celsius; the target value of vacuum degree, such as reaching a specific Pascal value; the ratio requirement of gas components, such as the content of a certain gas within a certain percentage range, etc. For the device processing stage, the environmental parameter control indicators include a stricter temperature control range, such as within plus or minus a few tenths of a degree; lower humidity requirements, such as relative humidity below a certain percentage; specific air cleanliness indicators, etc. For the packaging stage, the environmental parameter control indicators may include an extremely strict humidity control range, such as relative humidity close to zero; high cleanliness requirements, such as the air The number of particles is within a specific range, etc. According to the environmental parameter characteristics and environmental parameter control indicators of different preparation stages, an adaptive control strategy is formulated to analyze the key environmental parameters and their changing trends in each preparation stage. For the material growth stage, temperature and vacuum may be key parameters, and corresponding control strategies need to be formulated according to their changing trends. For example, if the temperature has a tendency to rise, the cooling system should be adjusted in advance to prevent the temperature from being too high. For the device processing stage, the accuracy requirements may be high, and the control strategy needs to be adjusted in time according to the operating status of the equipment and the slight changes in the environmental parameters. For example, if the accuracy of the processing equipment deviates, the environmental parameters should be adjusted in time to ensure the processing accuracy. For the packaging stage, humidity and cleanliness are key parameters, and the control strategy needs to be adjusted in time according to the environmental changes during the packaging process. For example, if the humidity suddenly rises, the dehumidification equipment is started immediately, and the specific control method and timing are determined in combination with the environmental parameter control indicators. For example, when the temperature exceeds the allowable fluctuation range, what control measures should be taken; when the vacuum degree does not reach the target value, how to adjust the vacuum pump and other equipment.

[0045] Step S640, according to the adaptive control strategy, the preparation environment of the aluminum nitride vacuum detector is adaptively controlled. Specifically, in the actual preparation process, according to the formulated adaptive control strategy, the preparation environment of the aluminum nitride vacuum detector is controlled in real time, and the environmental parameter monitoring system is used to continuously monitor the changes in the environmental parameters. When it is monitored that the environmental parameters deviate from the control index, the relevant equipment is automatically adjusted according to the adaptive control strategy to ensure that the environmental parameters are always within the target range. For example, if the temperature rise is monitored to exceed the allowable fluctuation range of the material growth stage, the cooling equipment is automatically started to cool down; if the humidity is found to rise during the packaging stage, the dehumidification equipment is immediately started to reduce the humidity. The adaptive control strategy is continuously optimized and adjusted according to the actual control effect. If it is found that some control measures are not effective, the reasons are analyzed in time and the strategy is adjusted to improve the accuracy and efficiency of the control. Through continuous adaptive control, it is ensured that the aluminum nitride vacuum detector can be in the best environmental conditions at different preparation stages, and the preparation quality and performance of the detector are improved.

[0046] The embodiment of the present application adopts the method of obtaining the characteristics of the aluminum nitride vacuum detector preparation environment to construct a monitoring system, monitor and filter the key parameters in real time, and store the data in a database. The key parameters are predicted based on the database, the deviation between the predicted value and the actual value is calculated, the deviation is input into the optimization control model to obtain the strategy, and the preparation environment is adaptively controlled according to the strategy. By realizing accurate monitoring, effective prediction and adaptive control of the aluminum nitride vacuum detector preparation environment, the technical effect of improving the preparation quality and stability of the detector and improving the efficiency and controllability of the preparation process is achieved.

[0047] In the above, refer to Figure 1 The environmental control method for preparing an aluminum nitride vacuum detector according to an embodiment of the present invention is described in detail. Figure 2 An environmental control system for preparing an aluminum nitride vacuum detector according to an embodiment of the present invention is described.

[0048] The environmental control system for the preparation of aluminum nitride vacuum detectors according to the embodiment of the present invention is used to solve the technical problems of inaccurate environmental monitoring, lack of effective prediction and control, and difficulty in adapting to complex environmental changes in the prior art of aluminum nitride vacuum detector preparation. By realizing accurate monitoring, effective prediction and adaptive control of the aluminum nitride vacuum detector preparation environment, the technical effect of improving the preparation quality and stability of the detector and improving the efficiency and controllability of the preparation process is achieved. The environmental control system for the preparation of aluminum nitride vacuum detectors includes: an environmental parameter monitoring system construction module 10, a preparation environment database storage module 20, a key parameter prediction module 30, a numerical deviation calculation module 40, an optimization control strategy acquisition module 50, and an adaptive control module 60.

[0049] The environmental parameter monitoring system construction module 10 is used to obtain the characteristics of the aluminum nitride vacuum detector preparation environment and to construct an environmental parameter monitoring system based on the characteristics of the aluminum nitride vacuum detector preparation environment;

[0050] The preparation environment database storage module 20 is used to perform real-time monitoring and data filtering of the key parameters of the detector preparation environment through the environmental parameter monitoring system, obtain the historical data and real-time data of the key parameters of the detector preparation environment, and store them in the preparation environment database;

[0051] The key parameter prediction module 30 is used to predict the key parameters of the detector preparation environment based on the preparation environment database to obtain the predicted values ​​of the key parameters of the detector preparation environment within a variable time window;

[0052] The numerical deviation calculation module 40 is used to calculate the numerical deviation between the predicted value of the key parameter of the detector preparation environment and the actual value of the key parameter of the detector preparation environment;

[0053] The optimization control strategy acquisition module 50 is used to input the numerical deviation into a preset optimization control model for analysis to obtain an optimization control strategy for the detector preparation environment, wherein the optimization control strategy includes optimization control parameters for key parameters of the detector preparation environment;

[0054] The adaptive control module 60 is used to adaptively control the aluminum nitride vacuum detector preparation environment according to the optimization control strategy.

[0055] The specific configuration of the preparation environment database storage module 20 will be described in detail below. As described above, the key parameters of the detector preparation environment are monitored in real time and data filtered by the environmental parameter monitoring system, and the historical data and real-time data of the key parameters of the detector preparation environment are obtained and stored in the preparation environment database. The preparation environment database storage module 20 further includes: a parameter grouping unit, the parameter grouping unit is used to group the key parameters of the detector preparation environment according to the time sequence, and obtain multiple groups of key parameter grouping data; a grouping data calculation unit, the grouping data calculation unit is used to calculate the lower quartile, median, and upper quartile of each group of the key parameter grouping data for the multiple groups of key parameter grouping data; an interquartile range acquisition unit, the interquartile range acquisition unit is used to obtain the interquartile range of the key parameter grouping data, and the interquartile range is the upper quartile minus the lower quartile; an abnormal value range determination unit, the abnormal value range determination unit is used to determine the abnormal value range, judge whether the data point is in the abnormal value range, if it is, mark the data point as an abnormal value data point, and remove the abnormal value data point in the key parameter grouping data.

[0056] Among them, the preparation environment database storage module 20 further includes: an attribute data storage unit, the attribute data storage unit is used to obtain the attribute data of the key parameters of the detector preparation environment, classify the attribute data according to the attribute type, obtain the environment parameter attribute data and the equipment attribute data, and store the environment parameter attribute data and the equipment attribute data separately; an event data storage unit, the event data storage unit is used to obtain the event data of the key parameters of the detector preparation environment, the event data includes monitoring event data, abnormal event data and control event data, and the monitoring event data, abnormal event data and the control event data are stored separately; a service data storage unit, the service data storage unit is used to obtain the service data of the key parameters of the detector preparation environment, the service data includes data acquisition service, data analysis service and control service, and the data acquisition service, data analysis service and control service are stored separately.

[0057] The specific configuration of the key parameter prediction module 30 will be described in detail below. As described above, based on the preparation environment database, the key parameters of the detector preparation environment are predicted to obtain the predicted values ​​of the key parameters of the detector preparation environment within the variable time window. The key parameter prediction module 30 further includes: a prediction model construction unit, the prediction model construction unit is used to construct a key parameter prediction model, the key parameter prediction model is constructed based on the historical data of the key parameters of the aluminum nitride vacuum detector preparation environment; a prediction value acquisition unit, the prediction value acquisition unit is used to use the key parameter prediction model to predict the key parameters of the detector preparation environment according to the real-time data, and obtain the predicted values ​​of the key parameters of the detector preparation environment within the variable time window.

[0058] Among them, the key parameter prediction module 30 further includes: a data acquisition unit, the data acquisition unit is used to acquire the attribute data, event data, and service data from the preparation environment database; a reference range determination unit, the reference range determination unit is used to determine the reference range of the variable time window according to the data characteristics of the attribute data; an event constraint condition acquisition unit, the event constraint condition acquisition unit is used to acquire sub-event data in the event data, associate the data characteristics of the sub-event data with the actual event characteristics, and acquire event constraints, the sub-event data includes monitoring event data, abnormal event data, and control event data; a service constraint condition acquisition unit The service constraint condition acquisition unit is used to obtain sub-service data in the service data, associate the data characteristics of the sub-service data with the actual service characteristics, and obtain service constraints, wherein the sub-service data includes data collection service data, data analysis service data, and control service data; a time window length acquisition unit, wherein the time window length acquisition unit is used to constrain the reference range of the variable time window according to the event constraints and the service constraints, and obtain multiple time window lengths; an optimal time window length determination unit, wherein the optimal time window length determination unit is used to judge the actual effects of the multiple time window lengths and determine the optimal time window length.

[0059] Among them, the actual effects of the multiple time window lengths are judged to determine the optimal time window length, and the optimal time window length determination unit further includes: a prediction error index calculation subunit, the prediction error index calculation subunit is used to regularly evaluate the optimal time window length and calculate the prediction error index; an adjustment strategy optimal solution formulation subunit, the adjustment strategy optimal solution formulation subunit is used to compare the prediction error index with a preset prediction accuracy threshold, if the prediction error index is less than the preset prediction accuracy threshold, formulate a window time length adjustment strategy optimal solution; an optimal window time length re-determination subunit, the optimal window time length re-determination subunit is used to redetermine the optimal window time length according to the window time length adjustment strategy optimal solution.

[0060] The specific configuration of the adaptive control module 60 will be described in detail below. As described above, according to the optimization control strategy, the preparation environment of the aluminum nitride vacuum detector is adaptively controlled, and the adaptive control module 60 further includes: a preparation stage feature library construction unit, the preparation stage feature library construction unit is used to establish a preparation stage feature library, and the preparation stage feature library includes environmental parameter characteristics of different preparation stages; an optimization target determination unit, the optimization target determination unit is used to determine the optimization targets of different preparation stages in the preparation stage feature library, and formulate environmental parameter control indicators for the different preparation stages; an adaptive control strategy formulation unit, the adaptive control strategy formulation unit is used to formulate an adaptive control strategy according to the environmental parameter characteristics and environmental parameter control indicators of different preparation stages; an adaptive control unit, the adaptive control unit is used to adaptively control the preparation environment of the aluminum nitride vacuum detector according to the adaptive control strategy.

[0061] The environmental control system for preparing aluminum nitride vacuum detectors provided in an embodiment of the present invention can execute the environmental control method for preparing aluminum nitride vacuum detectors provided in any embodiment of the present invention, and has functional modules and beneficial effects corresponding to the execution method.

[0062] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0063] The above specific implementation manner does not constitute a limitation to the protection scope of the present application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be performed in an order different from that in the embodiment and can still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. An environmental control method for preparing an aluminum nitride vacuum detector, characterized in that: The method comprises: Acquire characteristics of the aluminum nitride vacuum detector preparation environment, and construct an environmental parameter monitoring system based on the characteristics of the aluminum nitride vacuum detector preparation environment; Real-time monitoring and data filtering of the key parameters of the detector preparation environment are performed through the environmental parameter monitoring system, and historical data and real-time data of the key parameters of the detector preparation environment are obtained and stored in the preparation environment database; Based on the preparation environment database, predicting the key parameters of the detector preparation environment, and obtaining predicted values ​​of the key parameters of the detector preparation environment within a variable time window; Calculating the numerical deviation between the predicted value of the key parameter of the detector preparation environment and the actual value of the key parameter of the detector preparation environment; Inputting the numerical deviation into a preset optimization control model for analysis to obtain an optimization control strategy for the detector preparation environment, wherein the optimization control strategy includes optimization control parameters for key parameters of the detector preparation environment; According to the optimization control strategy, the aluminum nitride vacuum detector preparation environment is adaptively controlled.

2. The environmental control method for preparing aluminum nitride vacuum detector according to claim 1, characterized in that: The data filtering includes: Grouping the key parameters of the detector preparation environment according to the time sequence to obtain multiple groups of key parameter grouping data; For the multiple groups of key parameter grouped data, calculate the lower quartile, median, and upper quartile of each group of key parameter grouped data; Obtaining the interquartile range of the key parameter grouping data, where the interquartile range is the upper quartile minus the lower quartile; Determine the outlier range, determine whether the data point is within the outlier range, and if so, mark the data point as an outlier data point, and remove the outlier data point from the key parameter grouping data.

3. The environmental control method for preparing aluminum nitride vacuum detector according to claim 1, characterized in that: The storing in the preparation environment database includes: Acquire attribute data of key parameters of the detector preparation environment, classify the attribute data according to attribute type, obtain environment parameter attribute data and device attribute data, and store the environment parameter attribute data and the device attribute data separately; Acquire event data of key parameters of the detector preparation environment, wherein the event data includes monitoring event data, abnormal event data and control event data, and store the monitoring event data, abnormal event data and control event data separately; The service data of the key parameters of the detector preparation environment are obtained, wherein the service data includes data acquisition service, data analysis service and control service, and the data acquisition service, data analysis service and control service are stored separately.

4. The environmental control method for preparing aluminum nitride vacuum detector according to claim 3, characterized in that: Based on the preparation environment database, key parameters of the detector preparation environment are predicted, including: Constructing a key parameter prediction model, wherein the key parameter prediction model is constructed based on historical data of key parameters of the aluminum nitride vacuum detector preparation environment; The key parameter prediction model is used to predict the key parameters of the detector preparation environment according to the real-time data, and the predicted values ​​of the key parameters of the detector preparation environment within the variable time window are obtained.

5. The environmental control method for preparing aluminum nitride vacuum detector according to claim 4, characterized in that: Determining the variable time window includes: Acquire the attribute data, event data, and service data from the preparation environment database; Determining a reference range of the variable time window according to data characteristics of the attribute data; Obtaining sub-event data in the event data, associating data features of the sub-event data with actual event features, and obtaining event constraint conditions, wherein the sub-event data includes monitoring event data, abnormal event data, and control event data; Acquire sub-service data in the service data, associate data features of the sub-service data with actual service features, and acquire service constraint conditions, wherein the sub-service data includes data collection service data, data analysis service data, and control service data; According to the event constraint and the service constraint, constrain the reference range of the variable time window to obtain multiple time window lengths; The actual effects of the multiple time window lengths are judged to determine the optimal time window length.

6. The environmental control method for preparing aluminum nitride vacuum detector according to claim 5, characterized in that: include: Regularly evaluating the optimal time window length and calculating a prediction error indicator; Comparing the prediction error index with a preset prediction accuracy threshold, and formulating an optimal solution for the window time length adjustment strategy if the prediction error index is less than the preset prediction accuracy threshold; The optimal window time length is re-determined according to the optimal solution of the window time length adjustment strategy.

7. The environmental control method for preparing aluminum nitride vacuum detector according to claim 1, characterized in that: Adaptively controlling the aluminum nitride vacuum detector preparation environment includes: Establishing a preparation stage feature library, wherein the preparation stage feature library includes environmental parameter features of different preparation stages; Determining the optimization targets of different preparation stages in the preparation stage feature library, and formulating environmental parameter control indicators for the different preparation stages; Formulate adaptive control strategies based on environmental parameter characteristics and environmental parameter control indicators at different preparation stages; According to the adaptive control strategy, the aluminum nitride vacuum detector preparation environment is adaptively controlled.

8. An environmental control system for the preparation of aluminum nitride vacuum detectors, characterized in that: The system is used to implement the environmental control method for preparing an aluminum nitride vacuum detector according to any one of claims 1 to 7, and the system comprises: An environmental parameter monitoring system construction module, wherein the environmental parameter monitoring system construction module is used to obtain characteristics of the aluminum nitride vacuum detector preparation environment and to construct an environmental parameter monitoring system based on the characteristics of the aluminum nitride vacuum detector preparation environment; A preparation environment database storage module, the preparation environment database storage module is used to perform real-time monitoring and data filtering of the key parameters of the detector preparation environment through the environmental parameter monitoring system, obtain historical data and real-time data of the key parameters of the detector preparation environment, and store them in the preparation environment database; A key parameter prediction module, the key parameter prediction module is used to predict the key parameters of the detector preparation environment based on the preparation environment database, and obtain the predicted values ​​of the key parameters of the detector preparation environment within a variable time window; A numerical deviation calculation module, the numerical deviation calculation module is used to calculate the numerical deviation between the predicted value of the key parameter of the detector preparation environment and the actual value of the key parameter of the detector preparation environment; An optimization control strategy acquisition module, wherein the optimization control strategy acquisition module is used to input the numerical deviation into a preset optimization control model for analysis to obtain an optimization control strategy for the detector preparation environment, wherein the optimization control strategy includes optimization control parameters for key parameters of the detector preparation environment; An adaptive control module is used to adaptively control the aluminum nitride vacuum detector preparation environment according to the optimization control strategy.

Images