A photoresist filling state monitoring system and method

Abstract

The present application relates to the technical field of industrial process state monitoring, in particular to a photoresist filling state monitoring system and method; the system comprises: a micro-pressure pulse control unit acquires a reflected pressure change signal, establishes a corresponding relationship between liquid column length and pressure, and outputs a pressure fracture judgment result; an acoustic perception discrimination unit acquires a liquid column fracture sound pressure signal, constructs a non-contact liquid column end state recognition mechanism, and outputs an acoustic fracture judgment result; a multi-channel intelligent decision unit uses a dual-state suspicious self-proving mechanism to judge the authenticity of the pressure fracture judgment result and combines the acoustic fracture judgment result to output a final liquid column fracture judgment signal; and a filling process data recording unit records end point judgment related parameters, fracture judgment results and end point control feedback information in each filling process. The present application combines micro-pressure pulse application and reflected pressure signal analysis, and introduces acoustic determination to accurately monitor the end liquid column end state of photoresist filling.

Description

Technical Field

[0001] This invention relates to the field of industrial process status monitoring technology, specifically to a photoresist filling status monitoring system and method. Background Technology

[0002] Photoresist is a key photosensitive material widely used in precision processes such as semiconductor manufacturing and micro-nano fabrication. Its filling process has a direct impact on the quality of the finished product and the stability of production. Especially in the final stage of photoresist transportation from the storage container to the dispensing container through pipelines, the liquid column state changes are sensitive and the influencing factors are complex. The accuracy of the endpoint determination not only affects the consistency of the single filling volume, but also directly affects the uniformity and stability of subsequent coating processes. Therefore, how to accurately identify the liquid column breakage state at the end of the filling stage is an important technical link in the quality control of photoresist filling.

[0003] Near the end of the photoresist filling process, two issues affecting accurate filling remain: First, the photoresist liquid itself has high viscosity and low flow rate. When using pressure signal feedback analysis, the pressure signal feedback is easily interfered with by non-fracture factors such as bubble retention, end-of-pipe liquid backflow, and micro-vibration of the inner wall of the pipe. These factors may form a fluctuation pattern similar to actual fracture in the pressure reflection signal, leading to misjudgment. At the same time, the pressure signal of actual fracture may not be reflected in time due to transmission delay, resulting in missed judgments where the liquid column has broken but is still judged as not broken. Second, in the endpoint determination, the response characteristics of pressure signal and acoustic signal to fracture differ. When the two produce conflicting judgment results at the same time, if it cannot be immediately determined which result is more reliable, an incorrect endpoint signal may be directly output, causing problems such as overfilling, underfilling, or droplet residue, which adversely affects the quality of photoresist and production stability. Summary of the Invention

[0004] The purpose of this invention is to provide a photoresist filling status monitoring system and method to solve the two problems affecting the accurate filling of photoresist mentioned in the background art.

[0005] To achieve the above objectives, the present invention aims to provide a photoresist filling status monitoring system, comprising:

[0006] The micro-pressure pulse control unit is used to apply micro-pressure pulses to the filling pipeline, while collecting reflected pressure change signals, establishing the correspondence between liquid column length and pressure, and issuing a fracture trigger command when the liquid column length is close to the critical length threshold range, and outputting the pressure fracture judgment result of the liquid column fracture state.

[0007] The acoustic sensing and discrimination unit is used to collect the acoustic pressure signal of liquid column fracture, and calculate the changes in acoustic spectrum energy, amplitude fluctuation rate and duration to construct a non-contact liquid column endpoint state recognition mechanism, and output the acoustic fracture judgment result of liquid column fracture state.

[0008] The multi-channel intelligent decision-making unit uses a dual-state skeptical self-verification mechanism to dynamically judge the authenticity of the pressure fracture judgment result and combines it with the acoustic fracture judgment result to output the final liquid column fracture judgment signal.

[0009] The filling process data recording unit is used to record the endpoint determination parameters, breakage judgment results and endpoint control feedback information for each filling process.

[0010] Preferably, the micro-pressure pulse is a periodic disturbance signal with an amplitude lower than the static pressure of the liquid column and a pulse width of less than 1 second, used to achieve disturbance of the liquid column state; the micro-pressure pulse is applied to the filling pipeline according to the micro-pressure pulse frequency setting rules and the pressure application cycle adjustment strategy.

[0011] Preferably, the reflected pressure change signal includes an initial echo response value, a multi-order attenuated echo amplitude sequence, and echo delay change data, used to determine whether the liquid column is close to breaking.

[0012] In the micro-pressure pulse control unit, the correspondence between the liquid column length and the pressure is established. The specific steps are as follows:

[0013] The reflected pressure change signal is segmented according to the pulse number, and each segment of the reflected pressure change signal is mapped to an interval of the liquid column length range to form a pressure length pre-calibration dataset. The least squares method is used to establish a multivariate nonlinear mapping function between the liquid column length and the feature vector of the reflected pressure response signal, which is the correspondence between the liquid column length and the pressure.

[0014] Among them, the multivariable nonlinear mapping function is used to estimate the liquid column length in real time based on the reflected pressure change signal.

[0015] Preferably, when the liquid column length is close to the critical length threshold range, the micro-pressure pulse control unit compares the liquid column length estimated in real time by the multivariate nonlinear mapping function with the critical length threshold range, and issues a fracture trigger command when one of the fracture triggering conditions is met.

[0016] The fracture triggering conditions are as follows: the liquid column length first enters the critical length threshold range and there is no increasing trend within two consecutive pulse cycles; the liquid column length fluctuates within the critical length threshold range and the initial echo amplitude of the reflected signal shows a continuous decrease exceeding a preset ratio.

[0017] The fracture trigger command is used to initiate the fracture state identification process and output the pressure fracture judgment result of the liquid column fracture state. The pressure fracture judgment result includes the liquid column fracture state flag bit and the pulse number of the fracture occurrence time.

[0018] Preferably, the acoustic sensing and discrimination unit segments the acquired liquid column fracture sound pressure signal and calculates the acoustic spectrum energy change, amplitude fluctuation rate, and duration index of the liquid column fracture sound pressure signal respectively. The specific calculation method is as follows:

[0019] Time-frequency analysis was performed on the liquid column fracture acoustic pressure signal within the target frequency band to extract the trend of the total amplitude variation in the frequency domain. The difference between the amplitude and the static background sound spectrum was calculated to obtain the change in spectral energy. The envelope curve of the liquid column fracture acoustic pressure signal was extracted, and the number of amplitude changes per unit time was counted to obtain the amplitude fluctuation rate. Based on the start and end time intervals of the continuous high amplitude segments in the liquid column fracture acoustic pressure signal, the duration range of the liquid column fracture event was determined to obtain the duration index.

[0020] Preferably, the non-contact liquid column endpoint state recognition mechanism is constructed based on the integration of acoustic spectrum energy change, amplitude fluctuation rate and duration index, and is used to identify whether the liquid column has reached the fracture state, and to determine whether there is any backflow residue and unbroken drag wire phenomenon.

[0021] The specific construction method of the non-contact liquid column endpoint state recognition mechanism is as follows:

[0022] The acoustic spectrum energy change, amplitude fluctuation rate, and duration of the acoustic pressure signal from the fracture of the photoresist filling liquid column were calculated multiple times, and the acoustic spectrum energy change, amplitude fluctuation rate, and duration of each calculation were constructed as three-dimensional features. The fracture state of the photoresist filling liquid column was labeled for each time. A supervised multi-classification discrimination method was used to train the three-dimensional features and construct a state recognition classification model. During the real-time filling process, the three-dimensional features of the real-time liquid column fracture acoustic pressure signal were input into the state recognition classification model, and the real-time photoresist filling liquid column fracture state was output to obtain the acoustic fracture judgment result of the liquid column fracture state.

[0023] Preferably, in the multi-channel intelligent decision-making unit, the dual-state skepticism self-verification mechanism is used to assess the credibility of the pressure fracture judgment result and actively identify potential misjudgment patterns. When the pressure fracture judgment result is identified as unreliable, the acoustic fracture judgment result is combined to assist in the judgment and verification of the liquid column fracture state.

[0024] The dual-state tolerable self-verification mechanism includes a misjudgment identification model and an acoustic compensation module.

[0025] Preferably, the misjudgment identification model is constructed based on a multi-dimensional feature matching strategy, which is used to construct a tolerance analysis vector based on the pressure fracture judgment result and the filling history data, and output a tolerance score representing the credibility of the pressure fracture judgment result;

[0026] Among them, the tolerance analysis vector is a combination of multi-dimensional abnormal features extracted based on the pressure fracture judgment result, which serves as the input of the misjudgment identification model; the combination of multi-dimensional abnormal features includes pressure fluctuation amplitude, disturbance peak response time, waveform oscillation symmetry, echo attenuation gradient and disturbance response delay;

[0027] Among them, the tolerance score is the nonlinear matching error between the tolerance analysis vector and the characteristics of historical normal fracture samples, which is used to determine whether the pressure fracture judgment result is a misjudgment.

[0028] Preferably, the acoustic compensation module is used to analyze whether the doubt score exceeds a set threshold. When the doubt score exceeds the set threshold, the acoustic fracture judgment result is used to replace the pressure fracture judgment result, and the final liquid column fracture judgment signal is output.

[0029] On the other hand, the present invention provides a method for monitoring the filling status of photoresist, used in the photoresist filling status monitoring system described above, comprising the following steps:

[0030] S10.1 Apply a micro-pressure pulse to the filling pipeline, simultaneously collect the reflected pressure change signal, establish the correspondence between the liquid column length and pressure, and issue a fracture trigger command when the liquid column length is close to the critical length threshold range, and output the pressure fracture judgment result of the liquid column fracture state.

[0031] S10.2 Acquire the acoustic pressure signal of liquid column fracture, and calculate the acoustic spectrum energy change, amplitude fluctuation rate and duration index to construct a non-contact liquid column endpoint state recognition mechanism, and output the acoustic fracture judgment result of liquid column fracture state;

[0032] S10.3. Use a dual-state skeptical self-verification mechanism to dynamically determine the authenticity of the pressure fracture judgment result and combine it with the acoustic fracture judgment result to output the final liquid column fracture judgment signal.

[0033] S10.4 Record the relevant parameters for endpoint determination, breakage judgment results, and endpoint control feedback information for each filling process.

[0034] Compared with the prior art, the above-mentioned technical solution of the present invention has the following beneficial technical effects:

[0035] 1. In this invention, based on the establishment of the correspondence between the micro-pressure pulse application and the reflected pressure signal, and combined with the reliability analysis of the pressure fracture judgment result, it is possible to identify false fluctuations caused by air bubble retention, liquid backflow at the end of the photoresist filling process, and micro-vibration of the inner wall of the pipe, as well as missed judgments caused by the delay in the transmission of the photoresist liquid column fracture signal.

[0036] 2. In this invention, when the pressure fracture judgment result is abnormal, the acoustic fracture judgment result is introduced for cross-validation. By combining the complementarity of the two types of signals in fracture response characteristics, the endpoint judgment result is dynamically corrected, reducing the probability of overfilling, underfilling or droplet residue, and improving the photoresist filling quality and production consistency. Attached Figure Description

[0037] Figure 1 This is a schematic diagram of one embodiment of the present invention;

[0038] Reference numerals in the attached figures: 1. Micro-pressure pulse control unit; 2. Acoustic sensing and discrimination unit; 3. Multi-channel intelligent decision-making unit; 4. Filling process data recording unit. Detailed Implementation

[0039] Example 1, as Figure 1 As shown, a photoresist filling status monitoring system is provided, including: a micro-pressure pulse control unit 1, an acoustic sensing and discrimination unit 2, a multi-channel intelligent decision-making unit 3, and a filling process data recording unit 4.

[0040] In this embodiment, the micro-pressure pulse control unit 1 is used to apply micro-pressure pulses to the filling pipeline, collect reflected pressure change signals, establish the correspondence between liquid column length and pressure, issue a fracture trigger command when the liquid column length is close to the critical length threshold range, and output the pressure fracture judgment result of the liquid column fracture state.

[0041] In this embodiment, the micro-pressure pulse is a periodic disturbance signal with an amplitude lower than the static pressure of the liquid column and a pulse width of less than 1 second, used to achieve disturbance of the liquid column state; the micro-pressure pulse is applied to the filling pipeline according to the micro-pressure pulse frequency setting rules and the pressure application cycle adjustment strategy.

[0042] In this embodiment, the micro-pressure pulse is selected as a periodic disturbance signal with an amplitude lower than the hydrostatic pressure of the liquid column and a pulse width of less than 1 second. The specific reasons are as follows:

[0043] If the applied micro-pressure pulse is greater than the static pressure of the liquid column, it will cause the liquid column to move forward, retract, or bubble, interfering with the actual filling process. The liquid column is already unstable at the end of the filling process, and an excessively large pulse can easily cause premature breakage or allow disturbed bubbles to mix into the photoresist, affecting product quality. Disturbance signals below the static pressure are considered "metastable disturbances," which can be absorbed by the liquid column system without causing physical changes, while generating perceptible micro-disturbances in the reflected signal. The photoresist filling path is short, and the liquid column response time is very fast; a micro-pressure pulse with a pulse width of 10ms to 200ms can reflect the interference response. If the pulse width is too long, the signal reflection will superimpose the responses of multiple state segments, causing the echo interference to become blurred, making it difficult to distinguish whether the pressure reversal is caused by changes in the liquid column or by long-pressure interference.

[0044] In this embodiment, the micro-pressure pulse frequency setting rule is as follows: the optimal disturbance pressure frequency range is calculated and dynamically adjusted based on the average flow velocity of the liquid column and the pipeline length parameters in the historical filling data; the pressure cycle adjustment strategy is as follows: whether the pressure reflection signal has a sudden edge caused by the interruption of the liquid column continuity within the pressure cycle, and the subsequent pressure interval is adjusted according to the time position of the sudden point within the cycle.

[0045] In this embodiment, the reflected pressure change signal includes the initial echo response value, the multi-order attenuated echo amplitude sequence, and the echo time delay change data, which are used to determine whether the liquid column is close to breaking.

[0046] In the micro-pressure pulse control unit 1, the correspondence between the liquid column length and the pressure is established. The specific steps are as follows:

[0047] The reflected pressure change signal is segmented according to the pulse number, and each segment of the reflected pressure change signal is mapped to an interval of the liquid column length range to form a pressure length pre-calibration dataset. The least squares method is used to establish a multivariate nonlinear mapping function between the liquid column length and the feature vector of the reflected pressure response signal, which is the correspondence between the liquid column length and the pressure.

[0048] Among them, the multivariable nonlinear mapping function is used to estimate the liquid column length in real time based on the reflected pressure change signal.

[0049] In this embodiment, the initial echo response value refers to the maximum reflected pressure amplitude collected within the first time window after the micro-pressure pulse is applied, which can reflect the overall continuity of the liquid column; the multi-order attenuated echo amplitude sequence refers to the amplitude changes of the second-order, third-order, and subsequent reflected waves collected in chronological order after the initial echo, which is used to characterize the absorption and reflection characteristics of the pressure wave at the end of the liquid column; the echo delay change data refers to the time interval from the application of the micro-pressure pulse to the occurrence of the peak value of each order reflected wave, which is used to determine the change in the physical length of the liquid column.

[0050] The pulse number is a sequence number automatically generated by the system in the order of pressure application during the micro-pressure pulse control process. Each applied pulse corresponds to a unique number, which is used to associate the reflected pressure signal with the specific pressure application event during the data processing stage. The pressure length pre-calibration dataset is a sample set formed by combining the reflected pressure change signal under known liquid column length conditions with its corresponding length information, which serves as the training data source for the fitting model.

[0051] When establishing a multivariate nonlinear mapping function between the liquid column length and the feature vector of the reflected pressure response signal using the least squares method, the reflected signal features are first standardized, and then the regression coefficients of each feature and the liquid column length are calculated to minimize the squared error between the predicted value and the measured length value.

[0052] In this embodiment, when the liquid column length is close to the critical length threshold range, the micro-pressure pulse control unit 1 compares the liquid column length estimated in real time by the multivariate nonlinear mapping function with the critical length threshold range, and issues a fracture trigger command when one of the fracture triggering conditions is met.

[0053] The fracture triggering conditions are as follows: the liquid column length first enters the critical length threshold range and there is no increasing trend within two consecutive pulse cycles; the liquid column length fluctuates within the critical length threshold range and the initial echo amplitude of the reflected signal shows a continuous decrease exceeding a preset ratio.

[0054] The fracture trigger command is used to initiate the fracture state identification process and output the pressure fracture judgment result of the liquid column fracture state. The pressure fracture judgment result includes the liquid column fracture state flag bit and the pulse number of the fracture occurrence time.

[0055] In this embodiment, the critical length threshold range is the range of liquid column length obtained by statistical analysis of multiple batches of photoresist filling experimental data. This range covers the boundary area between the stable length and the unstable length before the liquid column breaks. The upper limit of the critical length threshold range corresponds to the critical state when the liquid column length is about to break, and the lower limit corresponds to the minimum length value that can be stably determined to break in most tests.

[0056] The multivariate nonlinear mapping function takes the latest acquired reflection pressure change signal as input during real-time operation, outputs an estimated value of the current length of the liquid column, and compares it with the critical length threshold range; when determining whether the liquid column length is within the range, the system simultaneously records the comparison result and the corresponding pulse number.

[0057] When determining whether the liquid column length first enters the critical length threshold range, the system determines the "first entry" state by detecting whether the length value of the previous pulse cycle is outside the range and the current length value is inside the range. The absence of an increasing trend within two consecutive pulse cycles means that the length estimate in the current pulse cycle is not greater than that in the previous cycle, thus confirming that the liquid column is in a stable or contracting phase. For cases where the liquid column length fluctuates within the critical length threshold range, the system continuously monitors the changes in the length estimate over multiple pulse cycles. When the change amplitude does not exceed the range boundary, it is considered a fluctuation within the range. Simultaneously, the system monitors the initial echo amplitude of the reflected signal, subtracts the current amplitude from the amplitude of the previous cycle, and divides by the amplitude of the previous cycle to obtain a decrease ratio. If this ratio continuously exceeds a preset ratio threshold, it is considered that the liquid column state is continuously decaying, consistent with a breakage trend.

[0058] After the fracture trigger command is issued, the system will immediately call the fracture state identification process, start the endpoint state discrimination calculation, and output the pressure fracture judgment result. In the pressure fracture judgment result, the liquid column fracture state flag is a binary flag, which is used to distinguish between fracture and non-fracture states. The pulse number at the time of fracture occurrence corresponds to the unique number of the micro-pressure pulse that triggered the fracture judgment in the pressure sequence, which is convenient for traceability and data archiving.

[0059] In the acoustic sensing and discrimination unit 2 of this embodiment, the acoustic sensing and discrimination unit 2 is used to collect the sound pressure signal of liquid column fracture, and calculate the changes in sound spectrum energy, amplitude fluctuation rate and duration index to construct a non-contact liquid column endpoint state recognition mechanism, and output the acoustic fracture judgment result of liquid column fracture state.

[0060] In this embodiment, the acoustic sensing and discrimination unit 2 performs segmented processing on the acquired liquid column fracture sound pressure signal, and calculates the acoustic spectrum energy change, amplitude fluctuation rate, and duration index of the liquid column fracture sound pressure signal respectively. The specific calculation method is as follows:

[0061] Time-frequency analysis was performed on the liquid column fracture acoustic pressure signal within the target frequency band to extract the trend of the total amplitude variation in the frequency domain. The difference between the amplitude and the static background sound spectrum was calculated to obtain the change in spectral energy. The envelope curve of the liquid column fracture acoustic pressure signal was extracted, and the number of amplitude changes per unit time was counted to obtain the amplitude fluctuation rate. Based on the start and end time intervals of the continuous high amplitude segments in the liquid column fracture acoustic pressure signal, the duration range of the liquid column fracture event was determined to obtain the duration index.

[0062] In this embodiment, the acquisition process of the liquid column fracture sound pressure signal is completed by a high-sensitivity acoustic sensor. The sampling frequency is selected to cover the main energy frequency band of the sound wave at the moment of liquid column fracture. The sampling signal is preprocessed by an anti-aliasing filter to remove high-frequency noise and background noise from the process environment.

[0063] In the calculation of the acoustic spectrum energy change, the target frequency band is determined based on the analysis results of a large number of filling tests on the typical fracture acoustic spectrum. The frequency band with significant differences between the fracture event signal and the background noise frequency distribution is preferentially selected. The time-frequency analysis adopts short-time Fourier transform to decompose the continuous time signal into a series of fixed time windows of spectrum. The sum of squares of the frequency domain amplitude of the target frequency band in each time window is calculated to obtain the energy value sequence. The energy change value is obtained by subtracting the average energy value of the static background acoustic spectrum from this sequence.

[0064] In the calculation of amplitude fluctuation rate, the envelope curve of the sound pressure signal is first extracted by envelope detection. The envelope curve reflects the amplitude change trend of the signal in the time domain. Then, within the specified analysis time window, the number of changes between adjacent peaks and valleys on the envelope curve is counted and divided by the length of the time window to obtain the number of amplitude changes per unit time.

[0065] In the calculation of the duration index, the amplitude threshold is set as the mean of the background noise plus a multiple of the noise standard deviation. When the envelope curve is continuously higher than the amplitude threshold, its start and end time points are recorded, and the difference between the two is the duration of a liquid column breakage event. In order to avoid misjudging multiple independent high-amplitude disturbances as a continuous event, a time interval judgment is introduced in the duration calculation process. If the interval between two high-amplitude signals exceeds the preset time threshold, they are recorded as two independent duration events.

[0066] In this embodiment, the non-contact liquid column endpoint state recognition mechanism is constructed based on the fusion of acoustic spectrum energy change, amplitude fluctuation rate and duration index, and is used to identify whether the liquid column has reached the fracture state, and to determine whether there is any backflow residue and unbroken drag wire phenomenon.

[0067] The specific construction method of the non-contact liquid column endpoint state recognition mechanism is as follows:

[0068] The acoustic spectrum energy change, amplitude fluctuation rate, and duration of the acoustic pressure signal from the fracture of the photoresist filling liquid column were calculated multiple times, and the acoustic spectrum energy change, amplitude fluctuation rate, and duration of each calculation were constructed as three-dimensional features. The fracture state of the photoresist filling liquid column was labeled for each time. A supervised multi-classification discrimination method was used to train the three-dimensional features and construct a state recognition classification model. During the real-time filling process, the three-dimensional features of the real-time liquid column fracture acoustic pressure signal were input into the state recognition classification model, and the real-time photoresist filling liquid column fracture state was output to obtain the acoustic fracture judgment result of the liquid column fracture state.

[0069] In this embodiment, the process of marking the fracture state of the photoresist filling liquid column is completed manually in combination with high frame rate video recording and signal characteristic curves. The fracture state of the photoresist filling liquid column includes three states: "fractured", "residual after retraction", and "filament not broken". The non-contact liquid column endpoint state recognition mechanism uses the liquid column fracture sound pressure signal obtained by acoustic sensing for analysis, without physical contact with the liquid column or filling pipeline, avoiding mechanical contact from interfering with the stability and cleanliness of the filling fluid.

[0070] In this embodiment, the training phase of the state recognition classification model adopts a supervised multi-class discrimination method, uses a cross-validation strategy to evaluate the generalization ability of the model, optimizes the classification boundary parameters, and introduces a feature selection algorithm to remove redundant feature components with low correlation to state discrimination, thereby reducing the impact of acoustic environmental noise on the model accuracy.

[0071] In this embodiment, the multi-channel intelligent decision-making unit 3 uses a dual-state skeptical self-verification mechanism to dynamically determine the authenticity of the pressure fracture judgment result and combines it with the acoustic fracture judgment result to output the final liquid column fracture judgment signal.

[0072] In this embodiment, the dual-state skepticism self-verification mechanism in the multi-channel intelligent decision-making unit 3 is used to assess the credibility of the pressure fracture judgment result and actively identify potential misjudgment patterns. When the pressure fracture judgment result is identified as unreliable, the acoustic fracture judgment result is combined to assist in the judgment and verification of the liquid column fracture state.

[0073] The dual-state tolerable self-verification mechanism includes a misjudgment identification model and an acoustic compensation module.

[0074] In this embodiment, the dual-state doubt-tolerant self-verification mechanism operates in the liquid column endpoint determination stage. Its purpose is to verify the credibility of the pressure fracture judgment result within the framework of the pressure judgment channel and the acoustic judgment channel, so as to avoid misjudgment caused by the characteristics of the pressure signal from directly entering the final output.

[0075] In this embodiment, the misjudgment identification model is constructed based on a multi-dimensional feature matching strategy. It is used to construct a tolerance analysis vector based on the pressure fracture judgment result and the filling history data, and output a tolerance score representing the credibility of the pressure fracture judgment result.

[0076] Among them, the tolerance analysis vector is a combination of multi-dimensional abnormal features extracted based on the pressure fracture judgment result, which serves as the input of the misjudgment identification model; the combination of multi-dimensional abnormal features includes pressure fluctuation amplitude, disturbance peak response time, waveform oscillation symmetry, echo attenuation gradient and disturbance response delay;

[0077] Among them, the tolerance score is the nonlinear matching error between the tolerance analysis vector and the characteristics of historical normal fracture samples, which is used to determine whether the pressure fracture judgment result is a misjudgment.

[0078] In this embodiment, the historical filling data includes the original waveforms of pressure reflection signals collected during previous batches of photoresist filling, key feature values ​​extracted after preprocessing, the actual observed state at the filling endpoint, and the corresponding filling environment parameters.

[0079] In this embodiment, the method for constructing the tolerance analysis vector based on the pressure fracture judgment result and the filling history data is as follows: five dimensions of abnormal features are extracted from the pressure reflection signal corresponding to the current pressure fracture judgment result, namely, pressure fluctuation amplitude, disturbance peak response time, waveform oscillation symmetry, echo attenuation gradient and disturbance response delay; these five features are normalized with similar features in the filling history data sample; these five features are combined in a fixed order to form a set of multi-dimensional abnormal feature combinations, which are used as the input of the misjudgment identification model.

[0080] In this embodiment, the specific calculation method of the tolerance score is as follows: a nonlinear matching metric based on kernel function is used to map the current tolerance analysis vector to the same feature space as the historical normal fracture sample, and the weighted Euclidean distance from it to the feature center of the normal fracture sample is calculated as the residual value; the residual value is mapped to a tolerance score between 0 and 1 by the sigmoid function. The closer the value is to 1, the greater the difference from the normal fracture pattern, and the higher the possibility of misjudgment.

[0081] In this embodiment, the specific method for determining whether the pressure fracture judgment result is a misjudgment is as follows: when the doubt score is greater than the doubt threshold, the system determines that the current pressure fracture judgment result is a high-risk misjudgment and triggers the acoustic compensation module to perform cross-channel verification; when the doubt score is lower than the threshold, the current result is considered credible and the pressure fracture judgment result is directly output as the final endpoint judgment signal; in order to avoid false triggering caused by instantaneous anomalies, the system will perform two consecutive score confirmations when the doubt score exceeds the threshold, and the acoustic compensation process will only be executed when both consecutive scores exceed the threshold.

[0082] In this embodiment, the acoustic compensation module is used to analyze whether the doubt score exceeds a set threshold. When the doubt score exceeds the set threshold, the acoustic fracture judgment result is used to replace the pressure fracture judgment result, and the final liquid column fracture judgment signal is output.

[0083] In this embodiment, after receiving the doubt score output by the misjudgment identification model, the acoustic compensation module first compares it with a preset doubt threshold. This threshold is obtained from statistical analysis of multiple batches of experiments to ensure that normal fluctuations and abnormal fluctuations can be distinguished under actual filling conditions. If the doubt score does not exceed the threshold, the system directly outputs the pressure fracture judgment result as the final judgment signal. If the doubt score exceeds the threshold, the acoustic compensation process is initiated. In the acoustic compensation process, the system calls the real-time fracture judgment result of the acoustic sensing path and matches the acoustic signal judgment result with the corresponding pressure signal event in the time series to ensure that the two correspond to the same liquid column end event. To avoid misjudgment of the acoustic path under short-term strong noise interference, the acoustic compensation module also performs a consistency verification of the acoustic fracture judgment result, including a rapid comparison of the similarity between the peak position and duration of the acoustic spectrum energy and the typical fracture template. Only when the acoustic result passes the consistency verification will it replace the original pressure fracture judgment result and be output as the final liquid column fracture judgment signal.

[0084] In this embodiment, the output final liquid column breakage judgment signal contains two types of information: one is the breakage status flag, which indicates whether the liquid column has broken; the other is the event timestamp and pulse number, which record the time when the breakage judgment is triggered, so that the filling control system can execute the stop filling action or enter the end-of-line procedure.

[0085] In this embodiment, the filling process data recording unit 4 is used to record the endpoint determination parameters, breakage judgment results and endpoint control feedback information for each filling process.

[0086] In this embodiment, the endpoint determination parameters include micro-pressure pulse application parameters, reflected pressure signal characteristic values, liquid column length estimation values, critical length threshold range, acoustic signal characteristic values, and tolerance analysis vectors; the fracture determination results include pressure fracture determination results, acoustic fracture determination results, and final endpoint determination signals; the final endpoint determination signals include the irrigation stop command trigger time, actual irrigation stop response time, abnormal event records, and data archiving information.

[0087] Example 2: This invention proposes a method for monitoring the filling status of photoresist, used in the photoresist filling status monitoring system described in Example 1 above, comprising the following steps:

[0088] S10.1 Apply a micro-pressure pulse to the filling pipeline, simultaneously collect the reflected pressure change signal, establish the correspondence between the liquid column length and pressure, and issue a fracture trigger command when the liquid column length is close to the critical length threshold range, and output the pressure fracture judgment result of the liquid column fracture state.

[0089] S10.2 Acquire the acoustic pressure signal of liquid column fracture, and calculate the acoustic spectrum energy change, amplitude fluctuation rate and duration index to construct a non-contact liquid column endpoint state recognition mechanism, and output the acoustic fracture judgment result of liquid column fracture state;

[0090] S10.3. Use a dual-state skeptical self-verification mechanism to dynamically determine the authenticity of the pressure fracture judgment result and combine it with the acoustic fracture judgment result to output the final liquid column fracture judgment signal.

[0091] S10.4 Record the relevant parameters for endpoint determination, breakage judgment results, and endpoint control feedback information for each filling process.

[0092] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited thereto. Various changes can be made within the scope of knowledge possessed by those skilled in the art without departing from the spirit of the present invention.

Claims

1. A photoresist filling status monitoring system, characterized in that, include: The micro-pressure pulse control unit (1) is used to apply micro-pressure pulses to the filling pipeline, collect the reflected pressure change signal, establish the correspondence between the liquid column length and the pressure, and issue a fracture trigger command when the liquid column length is close to the critical length threshold range, and output the pressure fracture judgment result of the liquid column fracture state. Acoustic sensing and discrimination unit (2) is used to collect the sound pressure signal of liquid column fracture, and calculate the changes in sound spectrum energy, amplitude fluctuation rate and duration index to construct a non-contact liquid column endpoint state recognition mechanism, and output the acoustic fracture judgment result of liquid column fracture state. Multi-channel intelligent decision-making unit (3) uses a dual-state doubt-tolerant self-verification mechanism to dynamically judge the authenticity of the pressure fracture judgment result and combines it with the acoustic fracture judgment result to output the final liquid column fracture judgment signal. In the multi-channel intelligent decision-making unit (3), the dual-state skepticism self-verification mechanism is used to assess the credibility of the pressure fracture judgment result and actively identify potential misjudgment patterns. When the pressure fracture judgment result is identified as unreliable, the liquid column fracture state is verified by combining the acoustic fracture judgment result. The dual-state skepticism self-verification mechanism includes a misjudgment identification model and an acoustic compensation module. The filling process data recording unit (4) is used to record the endpoint determination parameters, breakage judgment results and endpoint control feedback information for each filling process.

2. The photoresist filling status monitoring system according to claim 1, characterized in that, The micro-pressure pulse is a periodic disturbance signal with an amplitude lower than the static pressure of the liquid column and a pulse width of less than 1 second, used to achieve disturbance of the liquid column state; the micro-pressure pulse is applied to the filling pipeline according to the micro-pressure pulse frequency setting rules and the pressure application cycle adjustment strategy.

3. The photoresist filling status monitoring system according to claim 2, characterized in that, The reflected pressure change signal includes the initial echo response value, the multi-order attenuated echo amplitude sequence, and the echo time delay change data, which are used to determine whether the liquid column is close to breaking. In the micro-pressure pulse control unit (1), the correspondence between the liquid column length and the pressure is established. The specific steps are as follows: The reflected pressure change signal is segmented according to the pulse number, and each segment of the reflected pressure change signal is mapped to an interval of the liquid column length range to form a pressure length pre-calibration dataset. The least squares method is used to establish a multivariate nonlinear mapping function between the liquid column length and the feature vector of the reflected pressure response signal, which is the correspondence between the liquid column length and the pressure. Among them, the multivariable nonlinear mapping function is used to estimate the liquid column length in real time based on the reflected pressure change signal.

4. The photoresist filling status monitoring system according to claim 3, characterized in that, When the length of the liquid column is close to the critical length threshold range, the micro-pressure pulse control unit (1) estimates the length of the liquid column in real time using a multivariate nonlinear mapping function and compares it with the critical length threshold range. When one of the fracture triggering conditions is met, a fracture triggering command is issued. The fracture triggering conditions are as follows: the liquid column length first enters the critical length threshold range and there is no increasing trend within two consecutive pulse cycles; the liquid column length fluctuates within the critical length threshold range and the initial echo amplitude of the reflected signal shows a continuous decrease exceeding a preset ratio. The fracture trigger command is used to initiate the fracture state identification process and output the pressure fracture judgment result of the liquid column fracture state. The pressure fracture judgment result includes the liquid column fracture state flag bit and the pulse number of the fracture occurrence time.

5. The photoresist filling status monitoring system according to claim 4, characterized in that, The acoustic sensing and discrimination unit (2) segments the collected liquid column fracture sound pressure signal and calculates the acoustic spectrum energy change, amplitude fluctuation rate and duration index of the liquid column fracture sound pressure signal respectively. The specific calculation method is as follows: Time-frequency analysis was performed on the acoustic pressure signal of liquid column fracture in the target frequency band to extract the trend of the total frequency domain amplitude, and the difference was calculated with the static background sound spectrum to obtain the change of sound spectrum energy. The envelope curve of the liquid column fracture acoustic pressure signal is extracted, and the number of amplitude changes per unit time is counted to obtain the amplitude fluctuation rate. Based on the start and end time intervals of the continuous high amplitude segments in the liquid column fracture acoustic pressure signal, the duration range of the liquid column fracture event is determined to obtain the duration index.

6. The photoresist filling status monitoring system according to claim 5, characterized in that, The non-contact liquid column endpoint state recognition mechanism is constructed based on the integration of acoustic spectrum energy change, amplitude fluctuation rate and duration index. It is used to identify whether the liquid column has reached the fracture state and to determine whether there is any backflow residue or unbroken drag wire. The specific construction method of the non-contact liquid column endpoint state recognition mechanism is as follows: The acoustic spectrum energy change, amplitude fluctuation rate, and duration index of the acoustic pressure signal of multiple photoresist filling liquid column fractures were calculated, and the acoustic spectrum energy change, amplitude fluctuation rate, and duration index obtained from each calculation were constructed into three-dimensional features; The fracture state of the photoresist filling liquid column is marked for each filling process; a supervised multi-classification discrimination method is used to train the three-dimensional features and construct a state recognition classification model; during the real-time filling process, the three-dimensional features of the real-time liquid column fracture acoustic pressure signal are input into the state recognition classification model, and the real-time photoresist filling liquid column fracture state is output to obtain the acoustic fracture judgment result of the liquid column fracture state.

7. The photoresist filling status monitoring system according to claim 6, characterized in that, The misjudgment identification model is constructed based on a multi-dimensional feature matching strategy. It is used to construct a tolerance analysis vector based on the pressure fracture judgment result and the filling history data, and output a tolerance score that represents the credibility of the pressure fracture judgment result. Among them, the tolerance analysis vector is a combination of multi-dimensional abnormal features extracted based on the pressure fracture judgment result, which serves as the input of the misjudgment identification model; the combination of multi-dimensional abnormal features includes pressure fluctuation amplitude, disturbance peak response time, waveform oscillation symmetry, echo attenuation gradient and disturbance response delay; Among them, the tolerance score is the nonlinear matching error between the tolerance analysis vector and the characteristics of historical normal fracture samples, which is used to determine whether the pressure fracture judgment result is a misjudgment.

8. The photoresist filling status monitoring system according to claim 7, characterized in that, The acoustic compensation module is used to analyze whether the doubt score exceeds a set threshold. When the doubt score exceeds the set threshold, the acoustic fracture judgment result is used to replace the pressure fracture judgment result, and the final liquid column fracture judgment signal is output.

9. A method for monitoring the filling status of photoresist, used in a photoresist filling status monitoring system as described in any one of claims 1-8, characterized in that: Includes the following steps: S10.1 Apply a micro-pressure pulse to the filling pipeline, simultaneously collect the reflected pressure change signal, establish the correspondence between the liquid column length and pressure, and issue a fracture trigger command when the liquid column length is close to the critical length threshold range, and output the pressure fracture judgment result of the liquid column fracture state. S10.2 Acquire the acoustic pressure signal of liquid column fracture, and calculate the acoustic spectrum energy change, amplitude fluctuation rate and duration index to construct a non-contact liquid column endpoint state recognition mechanism, and output the acoustic fracture judgment result of liquid column fracture state; S10.

3. Use a dual-state skeptical self-verification mechanism to dynamically determine the authenticity of the pressure fracture judgment result and combine it with the acoustic fracture judgment result to output the final liquid column fracture judgment signal. Among them, the dual-state skepticism self-verification mechanism is used to assess the credibility of the pressure fracture judgment result and actively identify potential misjudgment patterns. When the pressure fracture judgment result is identified as unreliable, it combines the acoustic fracture judgment result to assist in the judgment and verification of the liquid column fracture state. S10.4 Record the relevant parameters for endpoint determination, breakage judgment results, and endpoint control feedback information for each filling process.

Images