Satellite disc rotating speed detection method and system of MOCVD equipment, computer equipment and storage medium

Abstract

The invention provides a satellite disc rotating speed detection method and system of MOCVD equipment, computer equipment and a storage medium, and relates to the technical field of semiconductor detection. Weak rotating speed characteristic signals are accurately extracted from noise through time domain and frequency domain fusion and a state estimation algorithm; the detection capability and the peak position stability of the rotating speed characteristic under the strong noise background are obviously improved; by taking a fusion result as an initial value and introducing state estimation of system dynamics, millisecond-level real-time updating and high-precision rotating speed analysis can be realized; the preset system dynamics model can dynamically fuse the current observation value and the historical state, the random error caused by airflow disturbance, mechanical vibration and electromagnetic interference is effectively inhibited, the output result is smooth and stable, and the reliability is greatly enhanced.

Description

Technical Field

[0001] This application relates to the field of semiconductor testing technology, and in particular to a method, system, computer equipment, and storage medium for detecting the rotational speed of a satellite disk in an MOCVD device. Background Technology

[0002] Metal-organic chemical vapor deposition (MOCVD) is a key process for preparing compound semiconductor epitaxial materials such as GaN, GaAs, and InP, and is widely used in the manufacture of optoelectronic and microelectronic devices such as light-emitting diodes (LEDs), lasers (LDs), and high electron mobility transistors (HEMTs). As the semiconductor industry strives for higher performance and lower costs, increasingly stringent requirements are being placed on the uniformity, consistency, and defect control of MOCVD epitaxial growth.

[0003] To address the challenge of achieving uniform temperature epitaxial growth across multiple wafers, planetary (or planetary gear) reaction chamber structures have become the mainstream technology for high-end MOCVD equipment. In this structure, a planetary disk drives multiple satellite disks in a revolution, while simultaneously, through a precise mechanical transmission system such as gears or friction drives, these satellite disks rotate around their own axes. The core design purpose of this combined revolution and rotation motion mode is to eliminate the inherent asymmetry in gas flow, temperature, and precursor concentration distribution within the reaction chamber through a dynamic averaging effect, thereby achieving ultra-uniform epitaxial growth with atomically flat intra-wafer uniformity and inter-wafer uniformity on a single substrate and between multiple substrates.

[0004] The rotational speed of the satellite disk is one of the key process parameters affecting the uniformity, composition, and defect density of the epitaxial layer. Currently, the satellite disk rotational speed is usually determined based on the design of the mechanical transmission system, assuming that it maintains a fixed transmission ratio with the planetary disk rotational speed. However, in actual high-temperature and high-pressure reaction environments, due to thermal expansion, mechanical wear, and particulate congestion, the actual rotational speed of the satellite disk may deviate from the theoretical value, and this deviation is difficult to detect.

[0005] Therefore, it is necessary to provide a new method, system, computer equipment, and storage medium for detecting the satellite disk rotation speed of MOCVD equipment to solve the above-mentioned problems existing in the prior art. Summary of the Invention

[0006] The purpose of this application is to provide a method, system, computer equipment, and storage medium for detecting the rotational speed of the satellite disk in an MOCVD equipment, so as to overcome the defects of traditional technology, which usually determines the rotational speed of the satellite disk based on the design of the mechanical transmission system and assumes that it maintains a fixed transmission ratio with the planetary disk rotational speed. In the reaction environment of high temperature and high pressure, the possible rotational speed deviation may affect the process production.

[0007] Firstly, this application proposes a method for detecting the rotational speed of the satellite disk in an MOCVD device, comprising: Acquire the spectral reflectance signal of the wafer surface located within the planetary reaction chamber; Time-domain analysis is performed on the spectral reflection signal to obtain the periodic pulse signal corresponding to the satellite disk in the spectral reflection signal, thereby obtaining time-domain rotational speed information; The spectral reflection signal is analyzed in the frequency domain to extract the spectral reflection signal within a preset time period, and then subjected to Fourier transform to obtain the frequency domain signal. The peak information of the corresponding satellite disk in the frequency domain signal is identified to obtain the frequency domain rotation speed information. By fusing and analyzing the time-domain and frequency-domain rotational speed information, the initial rotational speed information is obtained; Based on a preset system dynamics model, the initial rotational speed information is dynamically fused and the state estimate is recursively updated to obtain the real-time optimal rotational speed information.

[0008] In one embodiment, the step of performing time-domain analysis on the spectral reflectance signal to obtain the periodic pulse signal corresponding to the satellite disk in the spectral reflectance signal, and obtaining time-domain rotational speed information, includes: Based on the reflectivity difference between the wafer and the satellite disk substrate, a periodic pulse sequence is identified from the spectral reflectance signal; The time interval between consecutive pulses is calculated as the pulse period, and the time-domain rotational speed information is calculated based on the mapping relationship between the pulse period and the rotational speed.

[0009] In one embodiment, identifying the peak information of the corresponding satellite disk in the frequency domain signal to obtain the frequency domain rotation speed information includes: Multiple spectral peaks are extracted within a range defined by a priori frequency band to form a candidate set. The basic spectral peak of the corresponding satellite disk is determined from the candidate set based on the amplitude and harmonic consistency of the spectral peak. Interpolation is performed based on the amplitude relationship between the basic spectral peak and its adjacent frequency points to obtain the rotation frequency. Frequency domain rotation speed information is calculated based on the mapping relationship between the rotation frequency and the rotation speed.

[0010] In one embodiment, the fusion and parsing of time-domain rotational speed information and frequency-domain rotational speed information to obtain initial rotational speed information includes: Calculate the relative deviation between the time-domain rotational speed information and the frequency-domain rotational speed information; When the relative deviation is within a preset threshold range, the time-domain rotational speed information and the frequency-domain rotational speed information are weighted and summed based on preset time-domain factors and frequency-domain factors to obtain the initial rotational speed information; If the relative deviation exceeds a preset threshold range, the time-domain rotational speed information is corrected based on the frequency-domain rotational speed information; If the correction is successful, the corrected time-domain speed information and the frequency-domain speed information are weighted and summed based on preset time-domain factors and frequency-domain factors to obtain the initial speed information; wherein, a successful correction means that the relative deviation between the corrected time-domain speed information and the frequency-domain speed information is within a preset threshold range.

[0011] In one embodiment, correcting the time-domain rotational speed information based on the frequency-domain rotational speed information includes: The theoretical pulse period is calculated based on the frequency domain rotational speed information. A time-domain search window with a preset range is determined centered on the theoretical pulse period; Within the time-domain search window, pulse identification and period estimation are performed again, and candidate pulses falling outside the time-domain search window are removed to obtain the corrected time-domain rotational speed information.

[0012] In one embodiment, based on a preset system dynamics model, the initial rotational speed information is dynamically fused and the state estimate is recursively updated to obtain optimal real-time rotational speed information, including: Using a pre-defined system dynamics model and taking the initial rotational speed information as an observation, the rotational state of the satellite disk is predicted, forming a priori state and its uncertainty. The Kalman filter algorithm is used to update the initial speed information, dynamically weighting and fusing the prior state and the observed values ​​according to their respective uncertainties, suppressing process noise and measurement noise, and outputting smoothed real-time optimal speed information.

[0013] In one embodiment, the method further includes: Acquire historical operating data; wherein, the historical operating data includes estimated satellite disk rotation speed, planetary tray rotation speed, and process parameters; At preset time intervals, machine learning algorithms are used to learn the mapping relationship between each parameter in the historical running data; The system dynamics model and the Kalman filter algorithm are updated based on the mapping relationship; The updated system dynamics model and the Kalman filter algorithm are then applied to the next rotational speed state estimation. In one embodiment, the method further includes: The acquired spectral reflectance signal is preprocessed; wherein the preprocessing steps include filtering and smoothing the spectral reflectance signal, and the filtering includes filtering out high-frequency noise and / or filtering out extremely low-frequency drift.

[0014] In one embodiment, the method further includes: The system displays the real-time optimal rotational speed information and spectrum of the satellite disk, and generates an alarm signal when the real-time optimal rotational speed information exceeds the process setting range.

[0015] Secondly, this application proposes a satellite disk rotation speed detection system for an MOCVD equipment, the system comprising: The acquisition module is used to acquire the spectral reflectance signal of the wafer surface set in the planetary reaction chamber; The analysis module is used to perform time-domain analysis on the spectral reflection signal to obtain the periodic pulse signal of the corresponding satellite disk in the spectral reflection signal, thereby obtaining time-domain rotational speed information; to perform frequency-domain analysis on the spectral reflection signal to extract the spectral reflection signal within a preset time period, and to perform Fourier transform on it to obtain a frequency-domain signal, identify the peak information of the corresponding satellite disk in the frequency-domain signal, thereby obtaining frequency-domain rotational speed information; and to fuse and analyze the time-domain rotational speed information and the frequency-domain rotational speed information to obtain initial rotational speed information. The calculation module is used to dynamically fuse the initial rotational speed information and recursively update the state estimate based on a preset system dynamics model to obtain real-time optimal rotational speed information.

[0016] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method steps in the first aspect.

[0017] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the method steps of the first aspect.

[0018] The aforementioned method, system, computer equipment, and storage medium for detecting the satellite disk rotation speed of MOCVD equipment have at least the following advantages: This application acquires the spectral reflection signal of the wafer surface located within a planetary reaction chamber, performs time-domain and frequency-domain analysis on the signal, and obtains time-domain and frequency-domain rotational speed information, respectively. The time-domain and frequency-domain rotational speed information are then fused to obtain initial rotational speed information. Based on a pre-defined system dynamics model, state estimation is performed on the initial rotational speed information to obtain real-time optimal rotational speed information. Using the above scheme, this application, through time-domain and frequency-domain fusion and state estimation algorithms, can accurately extract weak rotational speed feature signals from noise, significantly improving the detection capability and peak stability of rotational speed features in strong noise backgrounds. Simultaneously, by using the fused result as the initial value and introducing system dynamics state estimation, millisecond-level real-time updates and high-precision rotational speed analysis can be achieved. Furthermore, the pre-defined system dynamics model can dynamically fuse current observations with historical states, effectively suppressing random errors caused by airflow disturbances, mechanical vibrations, and electromagnetic interference, resulting in smooth and stable output results and greatly enhancing reliability. Attached Figure Description

[0019] Figure 1 This is a diagram illustrating the application environment of a satellite disk rotation speed detection method for an MOCVD device in one embodiment. Figure 2 This is a flowchart illustrating a method for detecting the rotational speed of a satellite disk in an MOCVD device, as described in one embodiment. Figure 3 This is a flowchart illustrating the time-domain analysis steps in one embodiment; Figure 4 This is a flowchart illustrating the steps for obtaining initial rotational speed information through fusion analysis in one embodiment. Figure 5 This is a structural block diagram of the satellite disk rotation speed detection system of an MOCVD device in one embodiment; Figure 6 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0020] The following specific examples illustrate the implementation of this application. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. This application can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this application. It should be noted that, unless otherwise specified, the following embodiments and features in the embodiments can be combined with each other.

[0021] Some exemplary embodiments of this application have been described for illustrative purposes. It should be understood that this application may be implemented in other ways not specifically shown in the accompanying drawings.

[0022] The satellite disk rotation speed detection method for MOCVD equipment provided in this application embodiment can be applied to, for example... Figure 1 The application environment shown is as follows. The planetary reaction chamber 1 contains a rotatable planetary disk, on which multiple rotatable satellite disks are mounted. Each satellite disk has multiple wafer carrier disks for placing wafers. The planetary disk rotates under motor control, and each satellite disk rotates due to airflow.

[0023] Above the planetary reaction chamber 1 is a signal source and acquisition module 2, and a satellite disk rotation speed detection system 3 for the MOCVD equipment.

[0024] The signal source and acquisition module 2 includes an in-situ monitoring optical path unit, a photodetector, and a multi-channel synchronous data acquisition card.

[0025] The in-situ monitoring optical path unit is a standard configuration for MOCVD equipment. It typically includes an emitting end that emits a beam of monitoring light and a detection end that receives light reflected from the wafer and satellite disk surfaces. It is used to monitor process parameters such as film thickness, temperature, or reflectivity in real time.

[0026] A photodetector is used to receive reflected light signals and radiation signals from the in-situ monitoring optical path unit and convert them into analog electrical signals proportional to the light intensity. Specifically, in this embodiment, the analog electrical signal is the spectral reflection signal of the wafer surface disposed within the planetary reaction cavity.

[0027] A multi-channel synchronous data acquisition card is used to synchronously acquire and digitize two signals at high frequency: one is an analog electrical signal from the aforementioned photodetector, and the other is a reference trigger pulse signal provided by the MOCVD equipment, characterizing each rotation of the planetary tray.

[0028] The satellite disk rotation speed detection system 3 of the MOCVD equipment is connected to a multi-channel synchronous data acquisition card to acquire the spectral reflection signal of the wafer surface located in the planetary reaction chamber. Time-domain analysis of the spectral reflection signal is performed to obtain the periodic pulse signal corresponding to the satellite disk, thus obtaining time-domain rotation speed information. Frequency-domain analysis of the spectral reflection signal is performed to extract the spectral reflection signal within a preset time period and perform a Fourier transform to obtain the frequency-domain signal. The peak information corresponding to the satellite disk in the frequency-domain signal is identified to obtain frequency-domain rotation speed information. The time-domain and frequency-domain rotation speed information are fused and analyzed to obtain initial rotation speed information. Based on a preset system dynamics model, the initial rotation speed information is dynamically fused and the state estimate is recursively updated to obtain real-time optimal rotation speed information.

[0029] The aforementioned method for detecting the rotational speed of the satellite disk in MOCVD equipment acquires the spectral reflection signal from the wafer surface within the planetary reaction chamber, performs time-domain and frequency-domain analysis on the signal, and obtains time-domain and frequency-domain rotational speed information, respectively. The time-domain and frequency-domain rotational speed information are then fused to obtain initial rotational speed information. Based on a pre-defined system dynamics model, state estimation is performed on the initial rotational speed information to obtain real-time optimal rotational speed information. Using this approach, this application, through time-domain and frequency-domain fusion and state estimation algorithms, can accurately extract weak rotational speed characteristic signals from noise, significantly improving the detection capability and peak stability of rotational speed characteristics in strong noise backgrounds. Simultaneously, by using the fused result as the initial value and introducing system dynamics state estimation, millisecond-level real-time updates and high-precision rotational speed analysis can be achieved. Furthermore, the pre-defined system dynamics model can dynamically fuse current observations with historical states, effectively suppressing random errors caused by airflow disturbances, mechanical vibrations, and electromagnetic interference, resulting in smooth and stable output results and greatly enhancing reliability.

[0030] To further explain, the rotation speed of the satellite disk is a crucial process parameter affecting the quality of the epitaxial layer. Its stability and accuracy directly influence the growth rate, thickness uniformity, alloy composition uniformity, and defect density. For example, the rotation speed affects the thickness of the stagnant boundary layer on the substrate surface and the transport efficiency of reaction precursors, which is key to controlling the epitaxial layer growth rate and thickness uniformity. For multi-component compounds (such as InGaN and AlGaAs), the rotation speed determines the uniformity of the alloy composition radial distribution on the substrate by affecting the incorporation efficiency of temperature-sensitive precursors (such as TMI). A certain rotation speed helps to promptly remove reaction byproducts, suppress gas-phase pre-reactions, and thus reduce the density of crystal defects (such as dislocations and point defects).

[0031] However, in existing MOCVD equipment, the rotation of the satellite disk relies entirely on an open-loop mechanical transmission system. Its design is based on an idealized premise: that the rotational speed of the satellite disk and the revolution speed of the main disk maintain a constant transmission ratio. In actual industrial production, this premise is often difficult to maintain. The MOCVD reaction chamber is an extreme environment continuously subjected to high temperatures (often exceeding 1000℃), atmospheric or low pressure, and the presence of chemical corrosion byproducts. Under this environment, numerous factors cause the actual rotational speed of the satellite disk to deviate from the theoretical value, including: thermal deformation, where non-uniform thermal expansion of components within the chamber alters gear meshing clearances or friction wheel contact pressures, leading to transmission efficiency drift; mechanical wear, where long-term continuous operation causes wear on key transmission components such as gears and bearings, introducing transmission errors, and this process is slow and irreversible; and particulate contamination and jamming, where tiny particles generated during the reaction settle and intrude into the transmission pair clearances, potentially causing momentary jamming or permanent blockage, resulting in a sudden drop in satellite disk rotational speed or even complete shutdown.

[0032] A more pressing challenge is that, due to current technological limitations, existing semiconductor MOCVD equipment generally lacks effective means to monitor the real-time rotational speed of each satellite disk during the process. Engineers cannot know the true operating status of the satellite disks during operation; abnormal rotational speeds are typically only detected after the process is complete, through inspection of scrapped epitaxial wafers. This model introduces significant production risks, including but not limited to: scrapping entire batches of epitaxial wafers, resulting in expensive raw material costs (e.g., substrates, MO sources, high-purity gases) and substantial energy consumption losses; batch-to-batch product quality fluctuations and unstable yields, severely restricting the large-scale, consistent manufacturing of high-end semiconductor devices; and lagging equipment maintenance, failing to provide early warnings of potential mechanical failures, increasing the risk of unplanned downtime.

[0033] Therefore, there is an urgent need in this field for a technical solution that can monitor the rotational speed of the satellite disk in MOCVD equipment in real time and accurately, and can effectively control it based on this speed, in order to overcome the inherent defects of the existing open-loop mechanical transmission system and ensure the repeatability of the process and the yield of the product.

[0034] Please see Figure 2 In one exemplary embodiment, this application provides a method for detecting the rotational speed of a satellite disk in an MOCVD device, specifically including the following steps: Step 202: Obtain the spectral reflection signal of the wafer surface located in the planetary reaction chamber. Specifically, the spectral reflectance signal refers to the light intensity signal that varies with wavelength and / or time after the monitoring light emitted by the in-situ monitoring optical path unit illuminates the wafer surface, is reflected by the wafer surface, and is collected by the photodetector. After being normalized by the reference channel, it is expressed as a curve of reflectance as a function of wavelength and time.

[0035] Optionally, embodiments of this application further preprocess the acquired spectral reflectance signal; wherein the preprocessing steps include: filtering and smoothing the spectral reflectance signal, and the filtering includes filtering out high-frequency noise and / or filtering out extremely low-frequency drift.

[0036] Step 204: Perform time-domain analysis on the spectral reflection signal to obtain the periodic pulse signal of the corresponding satellite disk in the spectral reflection signal, and obtain the time-domain rotational speed information.

[0037] Specifically, the spectral reflection signal usually includes periodic high-frequency pulses caused by the rotation of each satellite disk and periodic low-frequency pulse signals caused by the revolution of the planetary disk. The two have different frequencies and shapes.

[0038] Time-domain analysis refers to the time-domain processing of spectral reflectance signals that vary over time. It involves identifying a periodic high-frequency pulse sequence formed by the alternating passage of a high-reflectance wafer and a low-reflectance satellite disk through a monitoring point under the influence of the satellite disk's rotation. The pulse interval is then calculated to obtain the time-domain rotational speed information of the satellite disk. The periodic low-frequency pulse signal caused by the planetary disk's revolution is used for gating and verification and is not included in the pulse interval calculation.

[0039] Step 206: Perform frequency domain analysis on the spectral reflection signal, extract the spectral reflection signal within a preset time period, perform Fourier transform on it to obtain the frequency domain signal, identify the peak information of the corresponding satellite disk in the frequency domain signal, and obtain the frequency domain rotation speed information.

[0040] Specifically, frequency domain analysis refers to processing the spectral reflectance signal collected within a preset time period using window functions and performing Fast Fourier Transform (FFT) to convert it from the time domain to the frequency domain. By identifying the characteristic peaks of the corresponding satellite disk in the spectrum, frequency information related to the rotational speed is obtained, and frequency domain rotational speed information is calculated accordingly. Step 208: Fuse and analyze the time-domain rotational speed information and the frequency-domain rotational speed information to obtain the initial rotational speed information.

[0041] Specifically, the purpose of fusion is to obtain more accurate, continuous, and stable rotational speed observations under noisy and occluded conditions, providing highly reliable input for subsequent state estimation and significantly improving the real-time performance and reliability of detection.

[0042] Step 210: Based on the preset system dynamics model, the initial speed information is dynamically fused and the state estimate is recursively updated to obtain the real-time optimal speed information.

[0043] Specifically, the system dynamics model is a mathematical model used to describe the evolution of the actual rotational speed of the satellite disk over time. Generally speaking, the rotational speed of the satellite disk includes angular velocity and rotational speed. In this embodiment, the initial rotational speed information is used as the observation of the system dynamics model, and the rotational speed state of the satellite disk is used as the state variable of the system dynamics model. The system dynamics model is then used to describe the correspondence between the observation and the state variable.

[0044] Furthermore, in this embodiment, state prediction is performed based on a preset system dynamics model, and a filter is used to observe, update, and dynamically fuse the initial rotational speed information to obtain the real-time optimal rotational speed estimate; wherein, the filter includes Kalman filtering and / or extended Kalman filtering.

[0045] The above-mentioned method for detecting the rotational speed of the satellite disk in MOCVD equipment acquires the spectral reflection signal of the wafer surface set in the planetary reaction chamber, performs time-domain analysis and frequency-domain analysis on it respectively to obtain time-domain rotational speed information and frequency-domain rotational speed information; fuses the time-domain rotational speed information and frequency-domain rotational speed information to obtain initial rotational speed information; and then, based on a preset system dynamics model, performs state estimation on the initial rotational speed information to obtain real-time optimal rotational speed information. By employing the above-mentioned scheme, this application, through time-domain and frequency-domain fusion and state estimation algorithms, can accurately extract weak rotational speed characteristic signals from noise, significantly improving the detection capability and peak stability of rotational speed characteristics under strong noise backgrounds. Simultaneously, by using the fusion result as an initial value and introducing system dynamics state estimation, millisecond-level real-time updates and high-precision rotational speed analysis can be achieved. Furthermore, the pre-set system dynamics model can dynamically fuse current observations with historical states, effectively suppressing random errors caused by airflow disturbances, mechanical vibrations, and electromagnetic interference, resulting in smooth and stable outputs that greatly enhance reliability. Furthermore, the frequency domain analysis is insensitive to signal amplitude fluctuations, giving the system greater tolerance to common engineering problems such as optical path window contamination and light source aging, ensuring the stability of long-term measurements.

[0046] Please see Figure 3 Optionally, time-domain analysis is performed on the spectral reflectance signal to obtain the periodic pulse signal corresponding to the satellite disk in the spectral reflectance signal, thereby obtaining time-domain rotational speed information, including: Step 302: Based on the reflectivity difference between the wafer and the satellite disk substrate, a periodic pulse sequence is identified from the spectral reflectance signal.

[0047] Step 304: Calculate the time interval between consecutive pulses as the pulse period, and calculate the time-domain rotational speed information based on the mapping relationship between the pulse period and the rotational speed.

[0048] Specifically, the wafer region has relatively high reflectivity, while the satellite disk region has relatively low reflectivity. The two regions alternate at the monitoring point as the satellite disk rotates, causing the waveform of the spectral reflection signal to exhibit periodic high and low changes. Based on the above high and low changes and equal intervals, the pulse sequence generated by the transition between the wafer and satellite disk edges can be identified by the peak detection algorithm.

[0049] Furthermore, the time interval between consecutive pulses is calculated, and its average value is taken as the pulse period. Then, the time-domain rotational speed information is calculated based on the mapping relationship. The mapping relationship is expressed as: RPM = 60 / T, where T is the pulse period and RPM is the rotational speed of the satellite disk. If the time-domain rotational speed information also includes angular velocity, then the mapping relationship is expressed as: angular velocity = 2π / T.

[0050] Using the above scheme, since the pulse signal is generated by the physical contrast between the high-reflection area and the low-reflection area, it has a naturally high signal-to-noise ratio and strong discriminative power, which can effectively suppress slow-varying interference such as light source drift and window contamination; by performing robust statistics on the pulse interval, the output time-domain rotation speed information has small jitter and high accuracy.

[0051] Optionally, peak information of the corresponding satellite disk in the frequency domain signal is identified to obtain frequency domain rotational speed information, including: Multiple spectral peaks are extracted within a range defined by a priori frequency band to form a candidate set. The basic spectral peak of the corresponding satellite disk is determined from the candidate set based on the consistency of the amplitude and harmonics of the spectral peak. Interpolation is performed based on the amplitude relationship between the basic spectral peak and its adjacent frequency points to obtain the rotation frequency. Frequency domain rotation speed information is calculated based on the mapping relationship between the rotation frequency and the rotation speed.

[0052] Specifically, in this embodiment, a reasonable search interval is set based on process settings or an estimate from the previous moment. Multiple spectral peaks with amplitudes greater than a preset threshold are extracted within this search interval to form a candidate set. From the candidate set, a spectral peak that simultaneously satisfies both amplitude and harmonic consistency conditions is selected as the base spectral peak, which is the most significant peak. Simultaneously satisfying both conditions means that the spectral peak has the highest signal-to-noise ratio, and that a corresponding filter exists near an integer multiple of this spectral peak. Since frequency domain signals are discrete frequency points, the true peak usually falls between two points. Using the amplitude asymmetry of the base spectral peak and its adjacent frequency points for interpolation can significantly improve frequency resolution and obtain a more accurate rotation frequency.

[0053] Furthermore, frequency domain rotational speed information is calculated based on the mapping relationship between rotational frequency and rotational speed; where the expression for the mapping relationship is: RPM_freq = 60 × f_freq, where RPM_freq is the frequency domain rotational speed and f_freq is the rotational frequency.

[0054] Using the above scheme, the prior frequency band constraint limits the search to a reasonable range. Combined with the consistency of amplitude and harmonics to screen the basic spectral peak, it can effectively distinguish the rotational basic peak from harmonics, sidebands and environmental pseudo-peaks, reducing the false detection rate. At the same time, interpolation is performed on the basic spectral peak and its adjacent frequency points, which can improve the accuracy of frequency resolution and significantly reduce the deviation of rotational speed calculation.

[0055] Please see Figure 4 Optionally, the initial rotational speed information is obtained by fusing and analyzing the time-domain rotational speed information and the frequency-domain rotational speed information, including: Step 402: Calculate the relative deviation between the time-domain rotational speed information and the frequency-domain rotational speed information.

[0056] Step 404: If the relative deviation is within a preset threshold range, the time-domain rotational speed information and the frequency-domain rotational speed information are weighted and summed based on preset time-domain factors and frequency-domain factors to obtain the initial rotational speed information.

[0057] Step 406: If the relative deviation exceeds the preset threshold range, correct the time-domain rotational speed information based on the frequency-domain rotational speed information.

[0058] Step 408: If the correction is successful, the corrected time-domain speed information and frequency-domain speed information are weighted and summed based on the preset time-domain factor and frequency-domain factor to obtain the initial speed information; wherein, the correction is successful when the relative deviation between the corrected time-domain speed information and frequency-domain speed information is within the preset threshold range.

[0059] Specifically, the initial speed information is the optimal joint analytical speed value output by performing consistency verification and adaptive weighted fusion of time-domain speed information (RPM_time) and frequency-domain speed information (RPM_freq).

[0060] Furthermore, the expression for the relative deviation is: δ = |RPM_time - RPM_freq| / RPM_freq.

[0061] If the relative deviation is within a preset threshold range, the two results are considered to corroborate each other and pass the verification. The process then proceeds to the adaptive weighted fusion process, which involves performing a confidence-based weighted fusion algorithm on both results to output the optimal joint analytical value and obtain the initial rotational speed information.

[0062] For example, the calculation formula for weighted fusion is expressed as: RPM_joint = α × RPM_time + β × RPM_freq, where α and β are the time-domain factor and the frequency-domain factor, respectively, and satisfy α + β = 1.

[0063] Furthermore, the determination of the aforementioned time-domain and frequency-domain factors is jointly determined by one or more of the following confidence indices used in real-time evaluation. For the time-domain confidence indices, a higher signal-to-noise ratio (SNR) of the preprocessed reflectivity signal and better waveform regularity of the identified pulse sequence result in a correspondingly larger α value. For the frequency-domain confidence indices, a higher peak-to-noise ratio, higher peak sharpness, and better harmonic consistency with the orbital frequency corresponding to the satellite disk's rotation frequency result in a correspondingly larger β value. For system state indices, at extremely high or low rotational speeds, or when the system's motion stability is poor (i.e., when the variance of the rotational speed estimate is large), α is automatically reduced and β is increased, relying more heavily on noise-resistant frequency-domain analysis.

[0064] Alternatively, in another embodiment, since frequency domain analysis has a stronger ability to resist random noise, the frequency domain rotational speed information obtained by frequency domain analysis can be preferentially used as the initial rotational speed information, and the time domain rotational speed information can be used as a reliability reference.

[0065] Furthermore, if the relative deviation exceeds a preset threshold, error correction mode is activated. Since frequency domain analysis is more resistant to random noise, time domain analysis is corrected based on frequency domain rotational speed information. After successful correction, the aforementioned adaptive weighted fusion process is executed again based on the frequency domain rotational speed information and the corrected time domain rotational speed information. If correction still fails, the current time domain rotational speed information is directly output, triggering a warning that indicates a potential systemic fault in the time domain channel.

[0066] Optionally, the above-mentioned correction of the time-domain rotational speed information based on the frequency-domain rotational speed information includes: The theoretical pulse period is calculated based on the frequency domain rotational speed information; a time domain search window with a preset range is determined with the theoretical pulse period as the center; pulse identification and period estimation are performed again within the time domain search window, and candidate pulses falling outside the time domain search window are removed to obtain the corrected time domain rotational speed information.

[0067] Specifically, based on the frequency domain rotational speed information (RPM_freq) obtained from frequency domain analysis, a more accurate theoretical value of the pulse period can be calculated, expressed as: T_theoretical = 60 / RPM_freq. Using this theoretical pulse period value as the center, a narrower, dynamic time search window is defined, for example, with a window length of T_theoretical ± 10%. Within this narrow window constrained by the frequency domain results, pulses are re-searched and identified. If a stable pulse sequence is found within this narrow window, the corrected time-domain rotational speed information is calculated using this new pulse period. This narrow window constrained by the frequency domain results effectively avoids false detections and missed detections of spurious pulses caused by noise.

[0068] By employing the above scheme, when the relative deviation exceeds a preset threshold, the theoretical period is calculated back from the frequency domain results, and the time domain pulse is re-examined within a narrow time window. This effectively eliminates spurious pulses and reduces deviations and jumps caused by window-crossing counting. Simultaneously, by weighting the output joint value based on confidence level, the low latency advantage of the time domain and the high resolution advantage of the frequency domain are combined, significantly reducing the variance and jitter of the rotational speed estimation. Furthermore, since frequency domain analysis is insensitive to signal amplitude fluctuations, correcting the time domain analysis based on frequency domain analysis allows the system to have greater tolerance to common engineering problems such as optical path window contamination and light source aging, ensuring the stability of long-term measurements.

[0069] Optionally, based on a preset system dynamics model, the initial rotational speed information is dynamically fused and the state estimate is recursively updated to obtain the optimal real-time rotational speed information, including: Using a pre-defined system dynamics model and taking initial rotational speed information as the observation, the rotational state of the satellite disk is predicted to form a priori state and its uncertainty. The Kalman filter algorithm is then used to update the initial rotational speed information. The priori state and the observation are dynamically weighted and fused according to their respective uncertainties to suppress process noise and measurement noise, and output smoothed real-time optimal rotational speed information.

[0070] Specifically, a system dynamics model is used to advance the satellite disk's rotation state from the previous moment to the next moment, obtaining the current prior state and its covariance. This covariance characterizes the uncertainty of the prior prediction; a larger value indicates a lower expected accuracy of the prior prediction. Furthermore, this application employs a Kalman filter algorithm to construct an innovation based on the observations and their variance, then calculates the Kalman gain. Based on this Kalman gain, the observations are optimally weighted according to uncertainty and injected into the prior state to obtain the posterior state and posterior covariance, thereby obtaining the real-time optimal estimate in the sense of minimum variance.

[0071] By employing the above scheme, which integrates time and frequency domains and incorporates a state estimation algorithm, weak rotational speed characteristic signals are accurately extracted from noise, significantly improving the detection capability and peak stability of rotational speed characteristics in strong noise environments. Using the fusion result as an initial value and introducing state estimation based on system dynamics, millisecond-level real-time updates and high-precision rotational speed analysis can be achieved. The dynamic fusion mechanism, employing a system dynamics model and Kalman recursive estimation, can output smooth real-time rotational speed in the sense of minimum variance, significantly reducing fluctuations caused by random noise and intermittent false detections, and effectively suppressing random errors caused by airflow disturbances, mechanical vibrations, and electromagnetic interference. When observation quality deteriorates or is lost for a short period, continuous and uninterrupted output is maintained based on model prediction, demonstrating strong robustness. Recursive updates are performed according to the sampling rhythm, eliminating the need for long window accumulation and meeting the real-time requirements of online control.

[0072] Optionally, the method for detecting the satellite disk rotation speed of the above-mentioned MOCVD equipment further includes: Historical operating data is acquired, including estimated satellite disk rotation speed, planetary tray rotation speed, and process parameters. At preset intervals, machine learning algorithms are used to learn the mapping relationship between the parameters in the historical operating data. The model parameters of the system dynamics model and Kalman filter algorithm are updated based on the mapping relationship. The updated system dynamics model and Kalman filter algorithm are then applied to the next rotation speed state estimation.

[0073] Specifically, the embodiments of this application also continuously collect historical operating data, periodically or triggeredly start the model training process, and use machine learning algorithms such as linear regression or time series analysis to learn the intrinsic mapping relationship between the above parameters, and the regression model of satellite disk rotation speed and airflow of process parameters.

[0074] Once the trained new model is validated, the system dynamics model and Kalman filter algorithm parameters are published and updated in real time. The updated model is then used for the next state estimation, thereby achieving online automatic compensation and calibration for slowly varying system errors such as mechanical wear and thermal expansion, forming a continuously self-optimizing closed-loop system.

[0075] By adopting the above scheme, the system dynamics model and Kalman filter parameters are updated periodically based on historical data, which can ensure that the speed estimation maintains high accuracy and robustness under long-term operation and multiple working conditions.

[0076] Optionally, the method for detecting the satellite disk rotation speed of the above-mentioned MOCVD equipment further includes: Displays the real-time optimal rotational speed information and spectrum of the satellite disk, and generates an alarm signal when the real-time optimal rotational speed information exceeds the process setting range.

[0077] Specifically, this application also provides users with real-time optimal speed information, spectrum diagrams and other analysis results through a human-machine interface, which facilitates users to quickly identify anomalies, trigger correction or interlocking in advance, and shorten processing delay.

[0078] The above-mentioned method for detecting the rotational speed of the satellite disk in MOCVD equipment acquires the spectral reflection signal of the wafer surface set in the planetary reaction chamber, performs time-domain analysis and frequency-domain analysis on it respectively to obtain time-domain rotational speed information and frequency-domain rotational speed information; fuses the time-domain rotational speed information and frequency-domain rotational speed information to obtain initial rotational speed information; and then, based on a preset system dynamics model, performs state estimation on the initial rotational speed information to obtain real-time optimal rotational speed information. By employing the above-mentioned scheme, this application, through time-domain and frequency-domain fusion and state estimation algorithms, can accurately extract weak rotational speed characteristic signals from noise, significantly improving the detection capability and peak stability of rotational speed characteristics under strong noise backgrounds. Simultaneously, by using the fusion result as an initial value and introducing system dynamics state estimation, millisecond-level real-time updates and high-precision rotational speed analysis can be achieved. Furthermore, the pre-set system dynamics model can dynamically fuse current observations with historical states, effectively suppressing random errors caused by airflow disturbances, mechanical vibrations, and electromagnetic interference, resulting in smooth and stable outputs that greatly enhance reliability. Furthermore, the frequency domain analysis is insensitive to signal amplitude fluctuations, giving the system greater tolerance to common engineering problems such as optical path window contamination and light source aging, ensuring the stability of long-term measurements. It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0079] Based on the same inventive concept, this application also provides a satellite disk rotation speed detection system for MOCVD equipment. This system is applicable to the above-mentioned method for detecting the satellite disk rotation speed of MOCVD equipment. The solution provided by this system is similar to the solution described in the above-mentioned method. Therefore, the specific limitations of one or more device embodiments provided below can be found in the limitations of the method above, and will not be repeated here.

[0080] Please see Figure 5 In one embodiment, the satellite disk rotation speed detection system of the MOCVD equipment includes: an acquisition module, an analysis module, and a calculation module.

[0081] The acquisition module is used to acquire the spectral reflection signal of the wafer surface located in the planetary reaction chamber.

[0082] The analysis module is used to perform time-domain analysis on the spectral reflection signal to obtain the periodic pulse signal of the corresponding satellite disk in the spectral reflection signal, and obtain the time-domain rotational speed information; to perform frequency-domain analysis on the spectral reflection signal, extract the spectral reflection signal within a preset time period, and perform Fourier transform on it to obtain the frequency-domain signal, identify the peak information of the corresponding satellite disk in the frequency-domain signal, and obtain the frequency-domain rotational speed information; and to fuse and analyze the time-domain rotational speed information and the frequency-domain rotational speed information to obtain the initial rotational speed information.

[0083] The calculation module is used to dynamically fuse the initial speed information based on the preset system dynamics model and recursively update the state estimate to obtain the real-time optimal speed information.

[0084] Optionally, the analysis module performs time-domain analysis on the spectral reflectance signal to obtain the periodic pulse signal of the corresponding satellite disk in the spectral reflectance signal, and obtains time-domain rotational speed information, including: identifying the periodic pulse sequence from the spectral reflectance signal based on the reflectivity difference between the wafer and the satellite disk substrate; calculating the time interval between consecutive pulses as the pulse period, and calculating the time-domain rotational speed information according to the mapping relationship between the pulse period and the rotational speed.

[0085] Optionally, the analysis module identifies the peak information of the corresponding satellite disk in the frequency domain signal to obtain frequency domain rotational speed information, including: Multiple spectral peaks are extracted within a range defined by a priori frequency band to form a candidate set. The basic spectral peak of the corresponding satellite disk is determined from the candidate set based on the consistency of the amplitude and harmonics of the spectral peak. Interpolation is performed based on the amplitude relationship between the basic spectral peak and its adjacent frequency points to obtain the rotation frequency. Frequency domain rotation speed information is calculated based on the mapping relationship between the rotation frequency and the rotation speed.

[0086] Optionally, the analysis module integrates and analyzes time-domain and frequency-domain speed information to obtain initial speed information, including: calculating the relative deviation between the time-domain and frequency-domain speed information; if the relative deviation is within a preset threshold range, performing a weighted sum of the time-domain and frequency-domain speed information based on preset time-domain and frequency-domain factors to obtain initial speed information; if the relative deviation exceeds the preset threshold range, correcting the time-domain speed information based on the frequency-domain speed information; if the correction is successful, performing a weighted sum of the corrected time-domain and frequency-domain speed information based on preset time-domain and frequency-domain factors to obtain initial speed information; wherein, successful correction means that the relative deviation between the corrected time-domain and frequency-domain speed information is within the preset threshold range.

[0087] Optionally, the analysis module corrects the time-domain rotational speed information based on the frequency-domain rotational speed information, including: calculating the theoretical pulse period based on the frequency-domain rotational speed information; determining a preset range of time-domain search window centered on the theoretical pulse period; re-identifying the pulse and estimating the period within the time-domain search window, and removing candidate pulses that fall outside the time-domain search window to obtain the corrected time-domain rotational speed information.

[0088] Optionally, the calculation module dynamically fuses the initial rotational speed information and recursively updates the state estimate based on a preset system dynamics model to obtain the optimal real-time rotational speed information, including: Using a pre-defined system dynamics model and taking initial rotational speed information as the observation, the rotational state of the satellite disk is predicted to form a priori state and its uncertainty. The Kalman filter algorithm is then used to update the initial rotational speed information. The priori state and the observation are dynamically weighted and fused according to their respective uncertainties to suppress process noise and measurement noise, and output smoothed real-time optimal rotational speed information.

[0089] Optionally, the satellite disk rotation speed detection system of the above-mentioned MOCVD equipment also includes a model update module.

[0090] The model update module is used to acquire historical operating data, which includes estimated satellite disk rotation speed, planetary tray rotation speed, and process parameters. At preset intervals, a machine learning algorithm is used to learn the mapping relationship between the parameters in the historical operating data. Based on the mapping relationship, the model parameters of the system dynamics model and the Kalman filter algorithm are updated. The updated system dynamics model and Kalman filter algorithm are then applied to the next rotation speed state estimation.

[0091] Optionally, the satellite disk rotation speed detection system of the above-mentioned MOCVD equipment also includes a display module.

[0092] The display module is used to display the real-time optimal rotational speed information and spectrum of the satellite disk, and to generate an alarm signal when the real-time optimal rotational speed information exceeds the process setting range.

[0093] The satellite disk rotation speed detection system of the aforementioned MOCVD equipment acquires the spectral reflection signal of the wafer surface set in the planetary reaction chamber, performs time-domain analysis and frequency-domain analysis on it respectively to obtain time-domain rotation speed information and frequency-domain rotation speed information; fuses the time-domain rotation speed information and frequency-domain rotation speed information to obtain initial rotation speed information; and then, based on the preset system dynamics model, performs state estimation on the initial rotation speed information to obtain real-time optimal rotation speed information. By employing the above-mentioned scheme, this application, through time-domain and frequency-domain fusion and state estimation algorithms, can accurately extract weak rotational speed characteristic signals from noise, significantly improving the detection capability and peak stability of rotational speed characteristics under strong noise backgrounds. Simultaneously, by using the fusion result as an initial value and introducing system dynamics state estimation, millisecond-level real-time updates and high-precision rotational speed analysis can be achieved. Furthermore, the pre-set system dynamics model can dynamically fuse current observations with historical states, effectively suppressing random errors caused by airflow disturbances, mechanical vibrations, and electromagnetic interference, resulting in smooth and stable outputs that greatly enhance reliability. Furthermore, the frequency domain analysis is insensitive to signal amplitude fluctuations, giving the system greater tolerance to common engineering problems such as optical path window contamination and light source aging, ensuring the stability of long-term measurements. The various modules in the satellite disk rotation speed detection system of the aforementioned MOCVD equipment can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of the computer device in hardware form or independent of it, or stored in the memory of the computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0094] In one feasible embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 6 As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements the satellite disk rotation speed detection method of the aforementioned MOCVD equipment. The display unit is used to form a visually visible image and can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0095] Those skilled in the art will understand that Figure 6 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0096] In one feasible embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method steps in the above-described method for detecting the satellite disk rotation speed of an MOCVD device.

[0097] In one feasible embodiment, a computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the method steps in the above-described method for detecting the rotational speed of the satellite disk in an MOCVD device.

[0098] In one feasible embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the method steps in the above-described method for detecting the rotational speed of the satellite disk in an MOCVD device.

[0099] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0100] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0101] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for detecting the rotational speed of a satellite disk in an MOCVD device, characterized in that, The method includes: Acquire the spectral reflectance signal of the wafer surface located within the planetary reaction chamber; Time-domain analysis is performed on the spectral reflection signal to obtain the periodic pulse signal corresponding to the satellite disk in the spectral reflection signal, thereby obtaining time-domain rotational speed information; The spectral reflection signal is analyzed in the frequency domain to extract the spectral reflection signal within a preset time period, and then subjected to Fourier transform to obtain the frequency domain signal. The peak information of the corresponding satellite disk in the frequency domain signal is identified to obtain the frequency domain rotation speed information. By fusing and analyzing the time-domain and frequency-domain rotational speed information, the initial rotational speed information is obtained; Based on a preset system dynamics model, the initial rotational speed information is used as an observation to predict the rotational state of the satellite disk, forming a priori state and its uncertainty. The Kalman filter algorithm is used to update the initial rotational speed information, and the prior state and the observed values ​​are dynamically weighted and fused according to their respective uncertainties to obtain the real-time optimal rotational speed information.

2. The method according to claim 1, characterized in that, The step of performing time-domain analysis on the spectral reflectance signal to obtain the periodic pulse signal corresponding to the satellite disk in the spectral reflectance signal, and obtaining time-domain rotational speed information, includes: Based on the reflectivity difference between the wafer and the satellite disk substrate, a periodic pulse sequence is identified from the spectral reflectance signal; The time interval between consecutive pulses is calculated as the pulse period, and the time-domain rotational speed information is calculated based on the mapping relationship between the pulse period and the rotational speed.

3. The method according to claim 1, characterized in that, The step of identifying the peak information of the corresponding satellite disk in the frequency domain signal to obtain the frequency domain rotation speed information includes: Multiple spectral peaks are extracted within a range defined by a priori frequency band to form a candidate set. The basic spectral peak of the corresponding satellite disk is determined from the candidate set based on the amplitude and harmonic consistency of the spectral peak. Interpolation is performed based on the amplitude relationship between the basic spectral peak and its adjacent frequency points to obtain the rotation frequency. Frequency domain rotation speed information is calculated based on the mapping relationship between the rotation frequency and the rotation speed.

4. The method according to claim 1, characterized in that, The fusion and analysis of time-domain and frequency-domain rotational speed information yields initial rotational speed information, including: Calculate the relative deviation between the time-domain rotational speed information and the frequency-domain rotational speed information; When the relative deviation is within a preset threshold range, the time-domain rotational speed information and the frequency-domain rotational speed information are weighted and summed based on preset time-domain factors and frequency-domain factors to obtain the initial rotational speed information; If the relative deviation exceeds a preset threshold range, the time-domain rotational speed information is corrected based on the frequency-domain rotational speed information; If the correction is successful, the corrected time-domain speed information and the frequency-domain speed information are weighted and summed based on preset time-domain factors and frequency-domain factors to obtain the initial speed information; wherein, a successful correction means that the relative deviation between the corrected time-domain speed information and the frequency-domain speed information is within a preset threshold range.

5. The method according to claim 4, characterized in that, The step of correcting the time-domain rotational speed information based on the frequency-domain rotational speed information includes: The theoretical pulse period is calculated based on the frequency domain rotational speed information. A time-domain search window with a preset range is determined centered on the theoretical pulse period; Within the time-domain search window, pulse identification and period estimation are performed again, and candidate pulses falling outside the time-domain search window are removed to obtain the corrected time-domain rotational speed information.

6. The method according to claim 1, characterized in that, The method further includes: Acquire historical operating data; wherein, the historical operating data includes estimated satellite disk rotation speed, planetary tray rotation speed, and process parameters; At preset time intervals, machine learning algorithms are used to learn the mapping relationship between each parameter in the historical running data; The system dynamics model and the Kalman filter algorithm are updated based on the mapping relationship; The updated system dynamics model and the Kalman filter algorithm are then applied to the next rotational speed state estimation.

7. The method according to claim 1, characterized in that, The method further includes: The acquired spectral reflectance signal is preprocessed; wherein the preprocessing steps include filtering and smoothing the spectral reflectance signal, and the filtering includes filtering out high-frequency noise and / or filtering out extremely low-frequency drift.

8. The method according to claim 1, characterized in that, The method further includes: The system displays the real-time optimal rotational speed information and spectrum of the satellite disk, and generates an alarm signal when the real-time optimal rotational speed information exceeds the process setting range.

9. A satellite disk rotation speed detection system for an MOCVD equipment, characterized in that, The system includes: The acquisition module is used to acquire the spectral reflectance signal of the wafer surface set in the planetary reaction chamber; The analysis module is used to perform time-domain analysis on the spectral reflection signal to obtain the periodic pulse signal of the corresponding satellite disk in the spectral reflection signal, thereby obtaining time-domain rotational speed information; to perform frequency-domain analysis on the spectral reflection signal to extract the spectral reflection signal within a preset time period, and to perform Fourier transform on it to obtain a frequency-domain signal, identify the peak information of the corresponding satellite disk in the frequency-domain signal, thereby obtaining frequency-domain rotational speed information; and to fuse and analyze the time-domain rotational speed information and the frequency-domain rotational speed information to obtain initial rotational speed information. The calculation module is used to predict the rotation state of the satellite disk based on a preset system dynamics model and the initial rotation information as an observation, forming a priori state and its uncertainty; and to update the initial rotation information by observation using a Kalman filter algorithm, dynamically weighting and fusing the prior state and the observation according to their respective uncertainties to obtain the real-time optimal rotation information.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1-8.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1-8.

Images