Novel polypeptide synthesizer

Abstract

The invention discloses a novel polypeptide synthesizer, which relates to the technical field of polypeptide synthesis, and comprises a ternary cooperative signal modeling module, a bubble time sequence analysis module, a photoacoustic coupling compensation module, a bubble discharge and circulation control module, a dynamic stability adjustment module and a self-correction optical path shaping module, establishing a ternary collaborative baseline of an ultraviolet detection signal, a pressure detection signal and a flow detection signal, reconstructing a flow cell optical path distribution model, extracting optical path refraction fingerprints, and forming a gas-liquid interface coupled dynamic monitoring reference system. The method is oriented to the field of biological medicine manufacturing, bubble recognition, flow path purification and optical path self-correction are achieved through the photoacoustic coupling detection and phase conjugate optical path shaping technology, a closed-loop control system with memory and feedback functions is constructed, and the accuracy, stability and automation level of the polypeptide synthesis process are remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of polypeptide synthesis, and particularly relates to a novel polypeptide synthesizer. BACKGROUND

[0002] The novel polypeptide synthesizer is a solid-phase polypeptide automatic synthesis device integrating gas path driving, liquid path metering, automatic valve control, online monitoring and intelligent temperature control for the field of biopharmaceutical manufacturing. The device uses inert gas constant pressure as a power source, accurately delivers amino acids and solvents to a synthesis column through a quantitative metering unit and a mass flow controller, and realizes reagent selection, forward and reverse purging and waste liquid diversion and other operations by using a multi-position multi-way valve matrix. The system is configured with a filter and a check unit at the gas source, liquid path and exhaust end to prevent blockage caused by impurities and backflow. An ultraviolet detection module is arranged in the post-column branch to monitor the deprotection reaction process in real time, and a multi-point pressure and flow sensor is used to build a closed-loop diagnostic system. The temperature control part uses a single-mode ring microwave resonant cavity, cooperates with optical fiber temperature sensing and reflection monitoring to realize uniform heating. The whole machine is cooperatively operated by an upper computer control strategy and a lower computer execution unit, and is provided with leakage, overpressure, overtemperature and waste liquid level interlocking protection, so that the automatic, high-precision and repeatable operation of multi-step solid-phase polypeptide synthesis in biopharmaceutical manufacturing can be completed under safe conditions.

[0003] The prior art has the following disadvantages: In the polypeptide automatic synthesis device of the prior art, the deprotection stage usually monitors the characteristic absorption peak of the Fmoc deprotection product in the reaction solution through a post-column ultraviolet detection module to determine whether the reaction is completed. However, during the dynamic liquid path switching and solvent bubbling process, micro-bubbles are easily generated and retained in the flow cell, especially under the conditions of temperature fluctuation or liquid viscosity change, the bubbles will form a stable retention layer above the flow cell, change the optical path length and local refractive index distribution, and cause instantaneous distortion of the signal received by the ultraviolet detector. The distortion often shows a false absorption peak value, causing the system to misjudge that the deprotection reaction has been completed, and the next coupling operation is performed in advance. Since part of the amino acid sites are still in the protection state at this time, the subsequent coupling cannot be performed, forming reaction residues and causing the peptide chain to break, resulting in failure of the entire batch of polypeptide synthesis products.

[0004] The above information disclosed in the background section is only used to enhance the understanding of the background of the present disclosure, and therefore it can include information that does not constitute prior art known to those of ordinary skill in the art. SUMMARY

[0005] The purpose of the present application is to provide a novel polypeptide synthesizer to solve the problems in the background.

[0006] In order to achieve the above object, the present application provides the following technical scheme: A novel polypeptide synthesizer is provided with a temperature control module for building a temperature field balanced environment between a synthesis column and a post-column detection area, adopts a multi-zone independent heating structure, each zone is composed of a heating element, a temperature sensor and a control driving unit; each heating element is coupled with the corresponding fluid pipeline, the outer wall of the synthesis column and the detection cell shell through a heat conducting medium or a heating jacket to form a continuous heat coupling channel; the temperature sensor is arranged at the inlet of the synthesis column, the middle section of the column body, the outlet and the periphery of the detection cell respectively for collecting multi-point temperature signals; the control driving unit adjusts the heating power of each zone in real time according to the collected temperature data through a PID control or a fuzzy control algorithm to realize independent constant temperature control of the multi-zone temperature and overall temperature field coordination, characterized in that it comprises a ternary collaborative signal modeling module, a bubble time sequence analysis module, a photoacoustic coupling compensation module, a bubble discharge and flow control module, a dynamic stability adjustment module and a self-correcting optical path shaping module: The ternary collaborative signal modeling module establishes a ternary collaborative baseline of ultraviolet detection signals, pressure detection signals and flow detection signals, reconstructs a flow cell optical path distribution model, extracts an optical path refraction fingerprint and forms a dynamic monitoring reference system of gas-liquid interface coupling; The bubble time sequence analysis module performs time reversal and differential reenactment on the detection signals according to the optical path refraction fingerprint, captures bubble nucleation and migration tracks, generates a bubble retention risk distribution map and establishes a spatio-temporal evolution relationship of bubble formation; The photoacoustic coupling compensation module constructs a double optical path differential detection unit according to the bubble retention risk distribution map, realizes synchronous comparison of ultraviolet signals and acoustic signals by photoacoustic coupling detection, generates an optical path compensation curve and establishes a refraction disturbance quantization model; The bubble discharge and flow control module links a multi-position multi-way valve and a mass flow controller according to the optical path compensation curve to control the exhaust bypass to form a bubble discharge window and realize directional bubble discharge and flow channel purification; The dynamic stability adjustment module adjusts the purge rhythm and microwave power duty cycle adaptively according to the detection residual density after the bubble discharge window is formed, calculates the refractive index fluctuation gradient in real time and generates an optical path stability criterion, and sends an optical path balance trigger signal when the refractive index fluctuation gradient is lower than a threshold value; The self-correcting optical path shaping module performs phase conjugate optical path shaping under the action of the optical path balance trigger signal, combines the time sequence of the reversible time grating migration multi-position multi-way valve and the purge frequency to form an optical path self-correcting closed loop and realize false peak elimination of the ultraviolet detection signal and optical path adaptive regulation.

[0007] Preferably, the optical path refraction fingerprint extraction step is as follows: Collect time sequence data of ultraviolet detection signals, gas pressure detection signals and liquid flow detection signals, the sampling frequencies of the three groups of signals are consistent and have time stamp marks; The time domain registration and low-pass filtering processing are performed on three groups of signals, and the steady-state coordination segments of the three groups of signals are extracted as the coordination baseline by aligning the pressure and the change time of the flow with the ultraviolet absorption sudden rising point. The dynamic deviation comparison is performed on the ultraviolet detection signal and the coordination baseline, the abnormal segments with the absorption deviation exceeding the threshold are identified, and the optical path disturbance trajectory is constructed combined with the synchronous change trend of the pressure and the flow; The waveform feature index is extracted and coded and classified according to the time period, and the refraction fingerprint with unique characteristics is formed, which is used for subsequent identification and comparison of abnormal detection behavior.

[0008] Preferably, the bubble retention risk distribution map generation step is as follows: The absorption abnormal section is selected from the ultraviolet detection signal matched with the optical path refraction fingerprint, and the gas pressure detection signal and the liquid flow detection signal in the corresponding time period are extracted to form three groups of aligned time sequence data sets; The time reversal processing is performed on the three groups of aligned time sequence data sets, the potential inducement before the interference event occurs is reversed, and a structured time model is constructed; The difference replay operation is performed on the signal set after time reversal, the rate characteristics of absorption value mutation, pressure sudden drop and flow rebound are extracted, and a three-dimensional feature vector of each interference event is formed; Based on the liquid flow rate and the pipeline geometric parameters, the three-dimensional feature vector is mapped to the physical space position, and the interference intensity, duration and type are combined to generate a bubble retention risk distribution map.

[0009] Preferably, the optical path compensation curve generation process is as follows: Two independent light paths with the same structure are arranged in parallel at the high-interference position marked in the bubble retention risk distribution map, which are respectively used for working liquid detection and stable reference liquid detection, and a piezoelectric ceramic piece is fixed on the outer wall of the light path to collect sound wave signals; The ultraviolet detection signals of the working light path and the reference light path and the sound wave signals output by the piezoelectric ceramic piece are synchronously collected, and the three groups of signals are time-aligned and compared to identify the signal offset and sound wave response event caused by optical path disturbance; The optical offset, sound wave amplitude, duration and delay are recorded to construct an optical path compensation curve describing the relationship between acoustic disturbance and optical offset, and a refraction disturbance quantization model is generated accordingly; In the detection process, the refraction disturbance quantization model is called in real time, the current sound wave amplitude is substituted into the compensation curve to calculate the standard optical offset, which is compared with the actual ultraviolet offset to identify the signal offset caused by bubbles and perform comparison and discrimination operation.

[0010] Preferably, the bubble discharge window formation process is as follows: According to the disturbance amplitude in the optical path compensation curve, it is judged whether the set threshold is exceeded. If it is exceeded, the control logic of the multi-position multi-way valve state combination is triggered, and two channels connecting the outlet of the synthesis column and the exhaust branch are opened in turn, the detection channel is kept in the minimum open state, and the waste liquid collection channel is closed; The linkage mass flow controller gradually increases the gas flow rate in a slow ramp mode, controls the bubble discharge speed through phased amplitude control, and prevents secondary disturbance from forming; The bubble migration path is limited in the vertical upward channel through structural arrangement, an exhaust outlet with a hydrophobic membrane is arranged at the upper end, and a closed collection bottle with back pressure control is connected to guide the bubble discharge; During the bubble discharge window duration, the ultraviolet detection signal and the acoustic disturbance signal changes are compared in real time to determine whether the bubble is discharged and the optical path is restored to stability; After the bubble discharge event is completed, the exhaust channel is automatically closed and all valves are restored to the original state, the slope limitation of the mass flow controller is removed, and all parameters of this event are written into the reaction control process.

[0011] Preferably, the mass flow controller controls the slope during the gas flow rate increase in the exhaust start-up stage through a stepped segmented speed increase mode. Before each stage is increased, the real-time flow rate and the target value of the previous stage are compared. If the deviation exceeds the preset range, the next stage is delayed.

[0012] Preferably, after the bubble discharge window is formed, the purge rhythm and the microwave power duty cycle are adjusted according to the detected residual density, the refractive index fluctuation gradient is calculated, the optical path stability criterion is generated, and the optical path balance trigger signal is sent when the fluctuation gradient is lower than the threshold. The steps are as follows: According to the optical path compensation curve after the bubble discharge is completed, the offset residual of the ultraviolet absorption value and the acoustic signal is extracted, and a two-dimensional residual density map is constructed to identify the disturbance area; According to the residual density map, the inert gas purge rhythm is set, and the purge frequency and duration of different positions are controlled through the multi-way valve to adjust the output flow rate of the mass flow controller to control the purge intensity; According to the residual density, the duty cycle of the microwave heating device is adjusted, and the heating period is controlled to suppress thermal disturbance and maintain stable reaction temperature; After the purge rhythm and the microwave power are stable, the refractive index fluctuation gradient is calculated based on the real-time absorption signal to determine whether the optical path stability condition is met; After continuously meeting the optical path stability condition, the optical path balance trigger signal is sent, the liquid path is controlled to enter the stable state, and the current control parameters are recorded for subsequent learning and optimization.

[0013] Preferably, under the action of the optical path balance trigger signal, phase conjugate optical path shaping is performed. Combined with the timing and purge frequency control of the multi-position multi-way valve by reversible time grid migration, the optical path self-calibration closed loop is constructed as follows: After receiving the optical path balance trigger signal, an optical reference path with reverse phase characteristics is constructed. Through beam splitting, delay, refractive crystal adjustment and phase difference control, phase conjugate interference operation is performed to cancel the ultraviolet signal shift caused by liquid disturbance. While performing optical shaping operations, a reversible time grid strategy based on the recorded valve position switching sequence is invoked to implement reverse dwelling timing control on multi-position multi-way valves, generating a reverse pulse flow of liquid to drive the interference area away from the detection position. After the optical phase compensation and liquid circuit time control are synchronized, the ultraviolet signal is periodically scanned and analyzed to determine whether the changes in absorption intensity and signal curve meet the spurious peak extinguishing criterion. The shaping parameters and detection results are recorded as feedback templates for subsequent use.

[0014] The technical effects and advantages provided by the present invention in the above technical solution are as follows: This invention achieves highly sensitive identification and response to bubble boundaries through photoacoustic coupling detection, establishing a complete quantitative model of optical path disturbance. Subsequently, a controllable exhaust window is constructed by linking a multi-position multi-way valve and a mass flow controller, enabling directional discharge of trapped bubbles and dynamic purification of the flow channel, effectively avoiding optical path distortion caused by microbubbles. Simultaneously, the system dynamically optimizes the purging process through joint adjustment of residual density and microwave power, improving the overall stability of the optical path. Finally, a closed-loop control structure with memory function and self-correction capability is constructed through phase conjugate optical path shaping and reversible time grid control strategies, achieving online identification and suppression of spurious absorption peaks. This significantly improves the accuracy of deprotection criteria and the controllability of the reaction process in the automated peptide synthesis process of biopharmaceutical manufacturing, avoiding coupling failure and product failure caused by signal misinterpretation, and comprehensively enhancing the automation level, production consistency, and industrial reliability of synthesis instruments in the field of biopharmaceutical manufacturing. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0016] Figure 1 This is a schematic diagram of a module of a novel polypeptide synthesizer according to the present invention.

[0017] Figure 2This is a schematic diagram of the gas path control principle of a novel polypeptide synthesizer according to the present invention. Detailed Implementation

[0018] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0019] This invention provides, for example Figure 1 The novel peptide synthesizer shown includes a temperature control module for creating a temperature field equilibrium environment between the synthesis column and the post-column detection area. It employs a multi-zone independent heating structure, with each zone consisting of a heating element, a temperature sensor, and a control drive unit. Each heating element is coupled to a corresponding fluid pipeline, the outer wall of the synthesis column, and the outer shell of the detection cell via a heat-conducting medium or heating jacket, forming a continuous thermal coupling channel. Temperature sensors are respectively arranged at the synthesis column inlet, the middle section of the column, the outlet, and the periphery of the detection cell to collect multi-point temperature signals. Based on the collected temperature data, the control drive unit adjusts the heating power of each zone in real time using PID control or fuzzy control algorithms, achieving independent constant temperature control of multiple zones and overall temperature field coordination. The system is characterized by including a ternary collaborative signal modeling module, a bubble timing analysis module, a photoacoustic coupling compensation module, a bubble discharge and flow control module, a dynamic stabilization adjustment module, and a self-correcting optical path shaping module. The temperature control module is used to create a stable temperature field between the synthesis column and the post-column detection area. This module employs a multi-zone independent heating structure, with an inlet heating zone, a middle heating zone, an outlet heating zone, and a detection pool heating zone along the fluid path. Each zone includes heating elements, temperature sensors, and a control drive unit. The heating elements consist of resistance heating elements, flexible heating films, or metal heating tubes, which are in close contact with the fluid pipeline, the outer wall of the synthesis column, and the outer shell of the detection pool through a heat-conducting oil circulation layer or a high-thermal-conductivity silicone heating jacket, forming a continuous heat conduction channel. The temperature sensors, using thermocouples or platinum resistance elements, are installed at the synthesis column inlet, the middle of the column, the column outlet, and the outer wall of the detection pool, respectively, to collect real-time temperature data at each location. The control drive unit consists of a microcontroller, a power drive module, and a data acquisition circuit. After receiving the temperature signal, it executes PID control or fuzzy logic algorithms to independently adjust the power output of each heating element, and changes the power supply cycle through pulse width modulation to achieve closed-loop stable temperature control. The system continuously compares the deviation between the target temperature and the measured temperature during the sampling period, automatically adjusts the heating power distribution of each zone, maintains consistent temperature across multiple zones, and ensures that the temperature gradient along the fluid direction is within a preset range, thereby achieving dynamic coordination and long-term stability of the overall temperature field.

[0020] The ternary synergistic signal modeling module establishes a ternary synergistic baseline for ultraviolet detection signals, pressure detection signals, and flow detection signals. Based on the ternary synergistic baseline, it reconstructs the optical path distribution model of the flow cell, analyzes the refraction trajectory characteristics of the light path, extracts the optical path refraction fingerprint, and forms a dynamic monitoring reference system for gas-liquid interface coupling to provide a benchmark framework for anomaly identification. To address the misjudgment problem caused by UV absorption signal distortion due to bubble retention during solid-phase peptide synthesis, a method for optical path distribution modeling based on the collaborative construction of three sets of signals—UV, pressure, and flow rate—is proposed. This method is used to identify and analyze refraction behavior and extract characteristic fingerprints for interference identification. The specific implementation steps are as follows: Time-series data of ultraviolet (UV) detection signals, gas pressure signals, and liquid flow signals were collected and established. The UV detection signal was obtained from a UV absorbance spectrophotometer installed in the waste liquid discharge channel after the synthesis column. This spectrophotometer uses a corrosion-resistant flow cell with a quartz window and is equipped with a high-intensity deuterium lamp as the light source. The detector's receiving wavelength is set between 290 and 310 nm, specifically for identifying diphenylmethylene absorption peaks generated during the Fmoc group deprotection reaction. The gas pressure signal was acquired from three piezoelectric pressure sensors located at different positions: the inert gas stabilization section, the metering unit, and the inlet pipeline before the synthesis column. These sensors were used to monitor gas source stability, liquid delivery pressure changes, and column resistance. The liquid flow signal was acquired by an electromagnetic flowmeter placed in the middle of the liquid path between the metering unit and the synthesis column. This flowmeter outputs instantaneous volumetric flow rate and cumulative flow rate in real time. All signals were connected to the upper-level control unit via a data acquisition unit. The sampling frequency was uniformly set to 5 Hz, recording one set of three-dimensional data points per second, and high-precision timestamps were used to ensure time-series consistency. Data recording begins five seconds before the start of a deprotection reaction and ends ten seconds after the post-column UV signal plateaus, ensuring that the data sequence completely covers the entire reaction process.

[0021] A ternary cooperative baseline for subsequent modeling is constructed based on three sets of signals. First, the three sets of signals undergo time-domain registration. By identifying peaks, inflection points, or plateau start points, and using a sudden increase in ultraviolet absorption as the primary control indicator, the timing of pressure rises and flow rate mutations is aligned forward to form a cooperative reference starting point. Then, each set of signals is low-pass filtered, and a third-order moving average window is used to remove high-frequency interference. The registered signal segments are smoothed using a sliding window method, with each window width set to 10 seconds and a step size of 1 second. The local mean and standard deviation of each signal are extracted to form signal stability criteria. In the ultraviolet signals, time periods with light absorption rate changes below 0.01 AU / s, pressure fluctuations below 0.3 kPa, and flow rate changes below 5 μL / s are selected and defined as the steady-state cooperative region. The three signal curves within this time period constitute the cooperative baseline for that reaction cycle, serving as a benchmark behavior template for normal fluid transport and the absence of optical path disturbances. Unlike traditional methods that rely solely on a single ultraviolet waveform feature to determine reaction nodes, this embodiment improves the specificity of abnormal event detection through three-signal coupling.

[0022] After establishing the cooperative baseline, the optical path distribution modeling and refraction trajectory reconstruction stage begins. In this step, the dynamic deviation of all UV signal curves from the unprotected phase to the cooperative baseline is calculated, and all absorption anomaly segments with deviations exceeding 0.05 AU and durations exceeding 2 seconds are extracted as candidate distortion segments. For each candidate segment, the trends of pressure and flow rate signals within the same time period are retrieved. If a sharp drop in pressure is observed accompanied by a sudden increase in flow rate, combined with the gas-driven mechanism, it can be preliminarily determined to be a rapid bubble passage phenomenon. If both pressure and flow rate exhibit periodic oscillations, it may be due to optical path refraction disturbance caused by a bubble's short-term residence at the top of the flow cell. The starting point, maximum absorption value point, and recovery plateau point are selected from the anomaly segments, and the rise rate, maximum amplitude, and recovery time are calculated. The optical path disturbance trajectory line is then plotted on a two-dimensional coordinate system, with the vertical axis representing the absorption value and the horizontal axis representing the time node. The actual optical path change range is then calculated by combining the geometric parameters of the reaction column and the flow cell. For example, a 0.08 AU instantaneous shift within a 10 mm optical path cell indicates a local refractive index change exceeding 5%. Finally, the optical path perturbation trajectories corresponding to each ultraviolet anomalous segment are superimposed to form a distribution map of all optical path deflection events within that period. This map is used to visually display the optical anomalous behavior throughout the entire reaction process and also lays the foundation for the next step of feature extraction.

[0023] Typical refractive features are extracted from the optical path perturbation spectrum to form a refractive fingerprint for subsequent comparison. In this stage, the optical path perturbation spectrum is divided into multiple non-overlapping segments along the time axis, each segment being 5 seconds long. Feature values ​​are extracted from the perturbation waveforms within these segments, including maximum absorption deviation, extreme slope, fluctuation frequency, and average recovery time. Segments with significant characteristic indicators are labeled, such as "single-peak refractive jump," "multi-step recovery," and "periodic refractive rebound," etc. For each type of code, a set of vector data containing the aforementioned indicators is generated, and this vector serves as a unique fingerprint template representing the optical performance corresponding to a single bubble perturbation. The refractive fingerprint set is stored in a comparison library. During real-time detection in subsequent reaction cycles, if any new abnormal ultraviolet signal segment matches an existing fingerprint template, the system determines it as a perturbation of the same type and decides whether to perform bubble removal, process pause, or continue operation accordingly. By visualizing the entire process of optical path perturbation as a trajectory with specific characteristics and encoding it as an identifiable fingerprint, this step effectively enhances the ability of peptide synthesis equipment to judge the true performance of the reaction under complex fluid conditions, and improves the accuracy of the control strategy and the safety of the reaction.

[0024] The bubble timing analysis module performs time reversal playback of continuous detection signals based on optical path refraction fingerprints, performs differential replay of the temporal changes of each detection signal, captures the nucleation location and migration trajectory of bubbles, generates a bubble retention risk distribution map, and establishes the spatiotemporal evolution relationship of bubble formation. To identify the nucleation source, migration route, and retention region of bubbles in the ultraviolet detection path, after extracting the optical path refraction fingerprint, the continuous detection signals are further analyzed temporally using time inversion and differential replay to form a risk map with spatial location and temporal attributes, which is used to present the true evolution process of bubble formation. The specific steps are as follows: With the optical path refraction fingerprint successfully extracted, all absorption anomaly segments matching the refraction fingerprint were selected from the complete ultraviolet detection signal data. These anomaly segments typically exhibit signal morphologies such as abrupt changes in ultraviolet absorbance, continuous oscillations, or short-term interference, and are all highly correlated with actual bubble activity. For each identified anomaly segment, the time point at which its absorbance first deviated from the baseline was used as the starting point, and a time window was established by tracing back at least five seconds and extending forward at least ten seconds to ensure coverage of the entire process of bubble formation, development, retention, and dissipation. Within this time range, gas pressure and liquid flow rate data recorded synchronously with the ultraviolet signal were extracted, forming three sets of time-series signals including absorption, pressure, and flow rate. The three sets of signals were aligned at the sampling points based on timestamp information to ensure consistency in subsequent operations.

[0025] Time inversion processing is performed on three sets of signals within a selected time window, which involves reversing the original time series along the time axis. The purpose of inversion processing is to trace the final manifestation of an event back to its potential causes, facilitating the identification of precursor signals that may have existed before the interference occurred. For example, if a sudden drop in the pressure signal or a momentary rebound in the flow signal is observed three seconds before the sudden rise of the ultraviolet absorption peak, it can usually be inferred that the bubble has nucleated at that time point and gradually floated to the detection optical path in the subsequent time. In the inversion sequence, the fluctuation start point, the point of maximum amplitude, and the recovery point of each set of signals are recorded to construct a structured time model of the event's start, development, and recovery processes. All inversion points are marked with their original time axis positions for subsequent remapping and comparison.

[0026] After the inversion data processing is completed, differential replay is performed on each set of signals to reveal the microscopic processes of signal changes and the intensity of interference. In differential processing, the signal change amplitude within each second time slice is calculated as the difference between two consecutive sampling points, forming a continuous rate curve. Taking the ultraviolet absorption signal as an example, if the differential curve shows an abnormal rise at a certain time point, and the pressure curves before and after it show a synchronous decline, while the flow curve shows a rapid rebound, then this time point can be identified as a typical event node where a bubble enters the detection optical path. Furthermore, if the oscillation duration of the ultraviolet differential curve exceeds three seconds, and the fluctuation amplitude is less than a set threshold, while the pressure and flow changes slowly but exhibit periodic fluctuations, it indicates that the bubble briefly lingers in the detection optical path and has not completely detached. By analyzing these differential characteristics, the nature, duration, and interference level of the event are determined. A three-dimensional feature vector is generated for each interference event, containing six indicators: peak ultraviolet change rate, pressure abrupt change, flow rebound value, interference duration, pre-peak delay, and post-peak recovery time, used for subsequent spatial mapping and risk level labeling.

[0027] Based on the time calibration and differential feature analysis results, spatial mapping and risk maps were generated for each bubble event. The spatial location was determined based on the liquid transport rate and flow structure geometry parameters in the pipeline. Using the pipe length from the synthesis column outlet to the UV detection location as a basis, combined with the current liquid flow rate (e.g., 200 μL / s) and the duration of the flow change before the event (e.g., 2 seconds), the initial bubble formation location could be estimated to be approximately 400 μL of liquid volume upstream of the detection window. Considering a pipe diameter of 2 mm, this volume corresponds to a physical distance of approximately 8 cm, thus locating the bubble event at a specific position from the synthesis column outlet to the middle of the detection section. The physical locations and corresponding time points of all events were synchronously recorded in a two-dimensional matrix. The horizontal axis represents the physical length of the flow path, and the vertical axis represents the time progress. Each event was visualized using an icon with color, size, and shape coding. Color represents the interference intensity, size represents the duration, and shape reflects the interference type (e.g., penetrating, stagnant, rebound). All events formed a bubble stagnation risk map on the graph, combining distribution density and intensity levels.

[0028] The photoacoustic coupling compensation module constructs a dual-optical-path differential reference detection unit based on the bubble retention risk distribution map. It uses photoacoustic coupling detection to achieve synchronous comparison between ultraviolet response signals and acoustic disturbance signals, calibrates the bubble boundary position, generates optical path compensation curves, and establishes a quantitative model of refractive disturbance. After constructing the bubble retention risk distribution map, to further accurately identify the initial boundary and intensity of optical path interference and achieve real-time correction of ultraviolet detection errors, a method based on two independent optical path structures combined with acoustic signal detection is proposed. This method compares the optical path difference caused by bubble interference. By simultaneously analyzing ultraviolet and acoustic signals, an optical path compensation curve for accurate correction is generated, and a quantitative model of refraction disturbance is established. The specific steps are as follows: Based on the locations of high-incidence optical interference marked on the retention risk distribution map, a section of the liquid path with a structure that allows for parallel optical paths was selected, and two independent but structurally equivalent detection paths were set up. The first detection path is used to transfer the reaction liquid. This path flows through the solution after the deprotection reaction, carrying potential bubbles or regions of unstable liquid refractive index, forming the actual path of the detection beam. The second detection path does not contact the reaction liquid but is filled with a calibration liquid with viscosity, density, and refractive index similar to the reaction liquid, forming a stable reference environment. Both paths are equipped with light sources, incident lenses, flow cells, and receiving detectors, and the distances, angles, and materials between all optical components are kept consistent. The flow cell is made of quartz glass with a surface coating to reduce reflection loss. A deuterium lamp is used as the light source with a fixed wavelength of 300 nanometers. The detector is a high-sensitivity photodiode array, and the sampling frequency is set to 10 times per second to ensure the ability to distinguish rapid ultraviolet changes.

[0029] Piezoelectric ceramic sheets, each approximately 1 square centimeter in area and 1.2 millimeters thick, were placed on the outer walls of the liquid contact areas of the two optical path structures. These sheets were fixed to the outer surface of the quartz using adhesive, and the contact points were polished to a mirror finish to improve acoustic response sensitivity. The piezoelectric sheets were connected to a charge amplifier, and the output analog signal was then fed into an analog-to-digital converter. The signal sampling rate was set to 1000 times per second, significantly higher than the optical signal sampling frequency, to capture the high-speed propagation of bubble disturbance sound waves. In the experiment, it was found that when bubbles nucleate, move, or burst in the liquid, a certain degree of local stress release occurs. This stress is transmitted through the liquid medium to the flow cell wall, forming a weak acoustic signal that can be sensed by the piezoelectric sheets. To avoid low-frequency mechanical resonance interference, a high-pass filter was added at the signal acquisition front end to filter out signals below 20 Hz. Furthermore, before each experiment, sound velocity calibration was performed by injecting a standard bubble of known size to determine the mapping relationship between acoustic signal propagation delay and physical location.

[0030] A synchronous comparison of the dual-optical-path signals and acoustic signals is performed to identify the boundary location and optical path disturbance intensity of bubble events. Specifically, the UV detection curve of each working optical path segment is compared with the reference optical path detection curve at each time point, and the difference between the two is calculated. If the working optical path signal shows a significant shift while the reference optical path remains stable within a certain time period, it is preliminarily determined that the shift is caused by optical path disturbance. Within the same time period, the acoustic signal is checked for abnormal peaks. If a sudden rise in the acoustic signal coincides with the optical shift time point, a real bubble crossing or retention event can be confirmed. The time point of this event, the optical shift amplitude, the acoustic response intensity, and the duration of the entire event are recorded. Multiple event datasets are compiled into a five-dimensional data table, recording: UV absorption difference, acoustic voltage peak value, acoustic waveform width, optical signal recovery time, and the time delay between the two. After grouping and analyzing all event data according to bubble interference type, a curve describing the mapping relationship between acoustic disturbance intensity and optical shift, i.e., the optical path compensation curve, is generated using a weighted average method. This curve is used to quantitatively express the expected ultraviolet signal offset value under a specific acoustic disturbance amplitude, and is input into the control process as the basis for real-time signal correction.

[0031] After the optical path compensation curve is established, a refractive perturbation quantification model is constructed based on this curve, and this model is used to distinguish between the false signal caused by bubbles and the actual reaction signal. During implementation, when the ultraviolet signal shift occurs again in the detection process, the response intensity of the current acoustic signal is simultaneously read and input into the established compensation curve to calculate the corresponding standard optical shift value. This value is compared with the actual ultraviolet detection signal shift value. If the two are similar, it can be determined that the ultraviolet shift is mainly caused by bubbles and should not be treated as a change in reaction state; if the two deviate significantly, it indicates that it is a manifestation of enhanced absorption during the deprotection reaction in the actual reaction process, and the subsequent reaction stage should proceed. To further improve reliability, the model also introduces the event duration as an auxiliary criterion. If the ultraviolet signal recovers rapidly within a short time while the acoustic signal dissipates slowly, it tends to be judged as a bubble slip event; if the ultraviolet signal continues to rise and there is no significant interference from the acoustic wave, it is judged as an deepening of the reaction process. In practical applications, the refractive perturbation quantization model serves as a comparison basis, supporting not only real-time denoising and correction of optical signals, but also assisting in making process control decisions, such as delayed coupling, adding purging, and pausing microwave heating, thereby improving the process control accuracy in peptide synthesis and preventing chain breakage or reaction interruption caused by bubble misjudgment.

[0032] The bubble discharge and flow control module, based on the optical path compensation curve, links the positional distribution of the multi-position multi-way valve to control the on / off state of the exhaust bypass and limit the rise slope of the mass flow controller, forming a bubble discharge window to achieve directional discharge of bubbles and dynamic purification of the flow channel. After synchronously comparing the ultraviolet absorption signal and the acoustic disturbance signal and generating the optical path compensation curve, this compensation curve needs to be used as the basis for bubble interference identification and response control. This involves actively linking the positional combinations of multiple channel valves and synchronously regulating the rate change of gas drive, thereby forming a controllable bubble removal window. This ensures that bubbles can be guided and directionally discharged within a specific time period, thus guaranteeing the optical path stability and measurement accuracy of the optical monitoring section in the liquid path. The specific steps are as follows: The disturbance amplitude in the optical path compensation curve is used as a quantitative indicator, and the host computer analyzes whether the disturbance value exceeds a predefined stability criterion threshold. When the disturbance value reaches or exceeds the set threshold, the scheduling logic for valve control is triggered. This logic is based on known optical path refraction anomaly modes and calls a set of predefined solenoid valve on / off sequences. Specifically, a combination state of a set of 4-channel solenoid valves is set, where the first channel is connected to the column outlet, the second channel is connected to the flow detection section, the third channel is connected to the exhaust branch, and the fourth channel is connected to the waste liquid collection unit. After the disturbance event is confirmed, the first and third channels of the valve open synchronously within 0.2 seconds, the second channel is kept in a minimum open state for signal monitoring, and the fourth channel remains closed to prevent bubbles from accidentally entering the waste liquid area and causing misjudgment. Through the above operation, the bubbles that were originally suppressed in the flow path by the pressure gradient can naturally float up under the impact of no pressure wave and be discharged into the atmospheric guide bottle through the third channel, completing the first stage of exhaust path preparation.

[0033] To prevent rapid gas injection from causing bubble breakage and secondary disturbances in the liquid path, the inert gas flow rate entering the reaction liquid path must be controlled with a gradual increase in slope when starting the degassing path. In this embodiment, the mass flow controller is set to gradually increase the airflow in steps of 10% per second for the first 5 seconds after startup, from a baseline of 0.1 liters per minute to a target flow rate of 0.5 liters per minute. During this process, the flow control device collects feedback flow rate data every second and compares it with the target value sent by the host computer. If the error exceeds ±0.02 liters per minute, the next step increase in speed is delayed until the system stabilizes. This gradual increase strategy can effectively suppress liquid column breakage or microbubble implosion caused by high instantaneous flow rates, making the entire degassing process smoother and extending the cleaning time window of the optical path, thus improving the overall signal quality.

[0034] To achieve directional migration and effective removal of bubbles along their physical path, the liquid path structure must be rationally arranged to ensure that the bubble removal channel is located upstream or above the potential bubble retention point. Specifically, a vertically rising stainless steel capillary tube, approximately 20 mm long, is installed between the synthesis column and the detection flow cell. A T-junction is connected at its highest point. The first branch of the T-junction connects to the original reaction liquid outlet, the second branch connects to the exhaust bypass path, and the third branch connects to the negative pressure outlet pipe. At the end of the exhaust path, a filter cartridge containing a 0.2-micron polytetrafluoroethylene hydrophobic membrane is connected. The filter cartridge's outer shell is made of polypropylene and is connected to a closed collection bottle equipped with a back pressure regulating valve to maintain a minimum pressure difference of no more than 10 kPa during the exhaust process. This structural configuration ensures that when the bubble removal window is activated, bubbles automatically rise to the T-junction due to density differences and are smoothly discharged, effectively preventing bubbles from repeatedly traveling back and forth in the UV detection section, thus avoiding signal overlap or false peaks.

[0035] During the entire bubble removal window, the changes in optical and acoustic disturbance signals are monitored synchronously to determine whether the bubbles have been completely guided out. If the ultraviolet absorbance value stabilizes within 0.005 absorbance units of the original baseline within 3 seconds, and the high-frequency signal peak in the acoustic detection channel disappears or its amplitude returns to the background fluctuation range, the bubble disturbance event can be considered terminated. At this time, the control logic automatically issues a recovery command, closes the exhaust channel, restores the second channel to the original reaction liquid output direction, and resets all solenoid valve channels to the initial reaction state. The mass flow controller also simultaneously releases the slope limit and resets to the standard bubbling or reaction propulsion flow rate. The entire bubble removal process does not exceed 10 seconds, and all switching steps are confirmed by a mechanical limit feedback device to prevent valve positions from deviating from preset values ​​due to controller step loss.

[0036] After the debubbling process is completed and the liquid path is confirmed to be stable, all detection parameters of this bubble event, including the trigger threshold, valve opening and closing time, mass flow controller rate curve, recovery time, photoacoustic signal comparison results, and final reset success flag, are written into the reaction control flow table as a data sequence, serving as a comparison sample for subsequent events. By analyzing the triggering patterns and intervention response efficiency after multiple bubble events, the triggering conditions for the next debubbling window can be further optimized, transforming it from static threshold-based identification to predictive action adjustment.

[0037] The dynamic stabilization adjustment module automatically adjusts the purging rhythm and microwave heating power duty cycle according to the detected residual density after the bubble discharge window is formed. It calculates the refractive index fluctuation gradient in real time and generates an optical path stability criterion. When the refractive index fluctuation gradient is lower than the stability threshold, it sends an optical path balance trigger signal and enters the optical path balance state. After bubble removal and purification of the flow path, to further stabilize the optical propagation path of the liquid, the ratio of the inert gas purging frequency to the on / off time of the microwave heating device needs to be dynamically adjusted to address residual bubbles, temperature disturbances, and minor interface fluctuations. This establishes a quantitative criterion for refractive index stability, ensuring the flow cell is in a controllable and balanced optical path state. The specific steps are as follows: Based on the optical path compensation curve formed after bubble removal, the offset residual between the ultraviolet absorption value and the acoustic reflection signal at this stage is extracted. Data is continuously collected every 0.2 seconds within a set time window. The absorbance deviating from the reference optical path and the acoustic interference signal are synchronously summarized to form a two-dimensional residual density map. This residual density map records the deviation amplitude and duration of the optical and acoustic signals in each time segment, and is a comprehensive indicator for judging whether there are unremoved microbubbles, unbalanced temperature disturbances, or residual pressure fluctuations. The higher the density, the more unstable the liquid path state. The data points in the residual map are not averaged, but are classified according to their location to identify the strength of their interference with the optical path, which facilitates differentiated control of subsequent response measures.

[0038] Based on the coordinates and intensity of the density concentration areas in the residual density map, the trigger rhythm and action mode of gas purging are determined. If the residual in the upstream region of the density map (i.e., from the synthesis column outlet to the detection section inlet) is frequent and the amplitude is greater than 0.005 absorbance units, rapid forward purging is immediately initiated, with the valve opening once every 4 seconds and each opening lasting for 0.8 seconds. If the density distribution is in the center of the detection section and the residual does not exceed 0.004 absorbance units, the purging frequency is changed to once every 6 seconds, with each purging lasting for 1.2 seconds, to avoid strong airflow disturbances that have reached near equilibrium. If the residual appears downstream of the flow cell and is accompanied by small pressure fluctuations, the next liquid injection action is gradually delayed by 2 seconds to allow for natural decay time, and back pressure-assisted venting is initiated to allow microbubbles to float to the top tee for exit. All purging is driven by a nitrogen source, and the mass flow controller is set to a flow rate not exceeding 0.6 liters per minute to prevent the liquid column from being cut off.

[0039] To minimize changes in the liquid's refractive index caused by the purging flow field disturbance, while maintaining the reaction temperature within an effective range, the power duty cycle of the microwave heating device needs to be dynamically adjusted based on the residual density. The standard duty cycle of the microwave resonant cavity is 50%. If the residual density exceeds the preset warning value of 0.007 (based on absorbance units), the duty cycle is immediately reduced to 35%, meaning it operates for only 7 seconds in every 20-second cycle, with the remaining time used for thermal equalization cooling. If the residual density is between 0.004 and 0.007, the duty cycle is maintained at 40%. When the residual density stabilizes below 0.003 for more than 3 consecutive seconds, it is restored to 50%. This adjustment is triggered by a host signal, with the slave device controlling the microwave pulse width generator. After each heating cycle, temperature sensor data and feedback absorption signals are collected to verify the temperature rise response. This approach suppresses thermal disturbances without causing insufficient reaction temperature.

[0040] With the purge rhythm and microwave power stabilized, the refractive index fluctuation gradient is calculated using real-time acquired absorption signal data. Specifically, the difference between the maximum and minimum ultraviolet signal values ​​is measured within a sliding time window (updated every second), and this value is divided by the width of the time window to obtain the gradient of optical signal fluctuation per unit time. If this gradient value is less than 0.002 absorbance units per second for three consecutive seconds, and the acoustic interference signal has no significant peak, the liquid path is considered to have entered an optical path stabilization state. Upon successful determination, the upper control unit sends an optical path balance trigger signal. This signal serves as a prerequisite for subsequent amino acid coupling or cleaning steps, ensuring that the detection conditions before the start of all reaction steps are met and preventing erroneous interpretations due to bubbles or thermal disturbances.

[0041] After the optical path balance trigger signal is issued, the entire liquid circuit control logic enters a steady state. The purging rhythm returns to the normal 8-second interval, the microwave heating power returns to the standard 50% duty cycle, and the mass flow controller returns to the reaction-set propulsion flow rate, such as 0.8 liters per minute. Simultaneously, the residual density map, purging action records, historical microwave power curves, and optical fluctuation gradient data for the current stage are retained in real time for subsequent feedback learning. By comparing the parameter change trends across multiple stages, the adaptive thresholds of various controls can be further trained. Over long-term operation, a personalized liquid circuit stabilization strategy library can be gradually established, thereby improving the repeatability and throughput consistency of different batches of synthesis.

[0042] The self-calibrating optical path shaping module performs phase conjugate optical path shaping operation under the action of the optical path balance trigger signal. Combined with the dwell time and purge frequency of the reversible time grid migration multi-position multi-way valve, it forms an optical path self-calibration closed loop, realizing the online extinguishing of false peaks in the ultraviolet detection signal and the adaptive dynamic control of the optical path. Once the optical path is confirmed to be in equilibrium and a stable signal is received, an active optical path shaping process needs to be initiated based on prior purging and compensation measures. This involves precisely controlling the optical signal path and liquid path timing structure to establish a closed-loop controllable circuit capable of extinguishing interference peaks in real time, thereby enhancing the accuracy of ultraviolet detection. The specific steps are as follows: After the optical path balance trigger signal is identified, an optical reference path with reverse phase characteristics is immediately constructed to correct the ultraviolet signal offset caused by fluctuations in the liquid refractive index in real time. The specific construction process is as follows: using the wavelength of the light source used in the main ultraviolet detection path as a reference, a set of auxiliary beam paths are arranged in parallel in the detection section after the synthesis column. This path separates the main beam into two beams through a beam splitter. One beam maintains its original path and enters the main detector, while the other beam is reverse-combined after passing through an optical delay unit and enters the detection channel. An adjustable refractive crystal element is configured on this auxiliary path, whose refractive index is finely adjusted with the electrical signal, allowing for precise delay adjustment of the optical path. By adjusting the crystal thickness and optical path length, the auxiliary beam and the main beam generate a π (180-degree) phase difference, thus constructing a complete phase conjugate interference structure. When the main beam undergoes a phase transition due to disturbance by microbubbles or microliquid films, the conjugate beam will perform real-time phase cancellation in the interference region, effectively eliminating spurious peak signals caused by the disturbance, ensuring that the final detector only receives the shaped, true, and effective signal. The entire optical path structure must ensure, in terms of mechanical structure, that the two beams have equal incident angles and consistent propagation media, so as to avoid phase difference errors introduced by external physical structures.

[0043] To further consolidate the optical stability brought about by the aforementioned interference compensation, a set of fluid path adjustment steps based on a reversible time grid control strategy needs to be executed simultaneously. Utilizing the multi-position multi-way valve structure used previously, a reversible dwell time sequence is constructed to synchronously control the subtle hydraulic changes caused by fluid flow disturbances and valve switching. Specifically, in the recorded fluid switching sequence, the time order of each valve position switch is reversed, creating a "historical playback" sequence in the fluid path during venting, feeding, and cleaning processes. For example, if the original sequence is A→B→C, then the current sequence is C→B→A. Each valve position switch has precise dwell time control; for example, C position is held for 0.5 seconds, B position for 0.7 seconds, and A position for 0.4 seconds. This sequence, combined with previous fluid inertia and pressure feedback parameters, generates a reverse pulse flow during reverse switching. This pulse drives residual small air bubbles or local refractive index abrupt changes in the pipeline to slowly move in the opposite direction of the fluid path, eventually removing them from the detection area. This reversible action must be performed in sync with optical shaping to prevent a new round of false signals from being generated due to the inconsistency between hydraulic disturbance and optical path adjustment.

[0044] After the coordinated optical phase conjugate shaping and liquid path time structure control are completed, the ultraviolet signal is periodically scanned and analyzed to confirm whether the shaping effect meets the criterion for spurious peak extinguishing. This scanning process uses the standard ultraviolet band, with the detection wavelength range set between 280 nm and 320 nm to cover the main absorption peak range of the Fmoc deprotection reaction products. The scanning period is set to once every 2 seconds, for a continuous 10 seconds. Within any period, if the change in absorption intensity of the detected signal is less than 0.002 absorbance units, and the signal curve does not show a unidirectional shift greater than 3 points, it is determined that the spurious peak has been successfully extinguished. At this time, the current beam phase structure, multi-way valve position sequence, liquid pressure distribution, and optical path delay parameters are recorded and written into the device control command set as a standard feedback curve. This feedback content will be loaded as a reference state before subsequent deprotection judgment or coupling start. If a similar disturbance occurs in a future synthesis cycle, only this configuration needs to be reverted to achieve optical path rebalancing without reshaping.

[0045] This invention achieves highly sensitive identification and response to bubble boundaries through photoacoustic coupling detection, establishing a complete quantitative model of optical path disturbance. Subsequently, a controllable exhaust window is constructed by linking a multi-position multi-way valve and a mass flow controller, enabling directional discharge of trapped bubbles and dynamic purification of the flow channel, effectively avoiding optical path distortion caused by microbubbles. Simultaneously, the system dynamically optimizes the purging process through joint adjustment of residual density and microwave power, improving the overall stability of the optical path. Finally, a closed-loop control structure with memory function and self-correction capability is constructed through phase conjugate optical path shaping and reversible time grid control strategies, achieving online identification and suppression of spurious absorption peaks. This significantly improves the accuracy of deprotection criteria and the controllability of the reaction process in the automated peptide synthesis process of biopharmaceutical manufacturing, avoiding coupling failure and product failure caused by signal misinterpretation, and comprehensively enhancing the automation level, production consistency, and industrial reliability of synthesis instruments in the field of biopharmaceutical manufacturing.

[0046] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A novel polypeptide synthesizer, provided with a temperature control module for constructing a temperature field balanced environment between the synthesis column and the post-column detection area, adopting a multi-zone independent heating structure, each zone consisting of a heating element, a temperature sensor and a control driving unit; each heating element is coupled with the corresponding fluid pipeline, the outer wall of the synthesis column and the detection cell shell through a heat conducting medium or a heating jacket, forming a continuous heat coupling channel; the temperature sensor is arranged at the inlet of the synthesis column, the middle section of the column body, the outlet and the periphery of the detection cell respectively, for collecting multi-point temperature signals; the control driving unit adjusts the heating power of each zone in real time according to the collected temperature data through PID control or fuzzy control algorithm, characterized in that, The three-element cooperative signal modeling module, the bubble timing analysis module, the photoacoustic coupling compensation module, the bubble discharge and flow control module, the dynamic stability adjustment module, and the self-correcting optical path shaping module are included. The three-element cooperative signal modeling module establishes a three-element cooperative baseline of ultraviolet detection signals, pressure detection signals, and flow detection signals, reconstructs an optical path distribution model of a flow cell, extracts an optical path refraction fingerprint, and forms a dynamic monitoring reference system of gas-liquid interface coupling. The bubble timing analysis module performs time reversal and differential reenactment on the detection signals according to the optical path refraction fingerprint, captures bubble nucleation and migration trajectories, generates a bubble retention risk distribution map, and establishes the spatiotemporal evolution relationship of bubble formation. The photoacoustic coupling compensation module constructs a double optical path differential detection unit based on the bubble retention risk distribution map, uses photoacoustic coupling detection to realize synchronous comparison of ultraviolet signals and acoustic signals, generates an optical path compensation curve, and establishes a refraction disturbance quantification model. The bubble discharge and flow control module links the multi-position multi-way valve and the mass flow controller according to the optical path compensation curve to control the exhaust bypass to form a bubble discharge window. The dynamic stability adjustment module adjusts the purge rhythm and microwave power duty cycle adaptively according to the detection residual density after the bubble discharge window is formed, calculates the refraction index fluctuation gradient in real time, generates an optical path stability criterion, and sends an optical path balance trigger signal when the refraction index fluctuation gradient is below a threshold value. The self-correcting optical path shaping module performs phase conjugate optical path shaping under the action of the optical path balance trigger signal, combines the timing of the reversible time grid migration multi-position multi-way valve and the purge frequency, and forms an optical path self-correcting closed loop.

2. The novel polypeptide synthesizer according to claim 1, characterized in that, The optical path refraction fingerprint extraction steps are as follows: Collect time series data of ultraviolet detection signals, gas pressure detection signals, and liquid flow detection signals, and the sampling frequencies of the three groups of signals are consistent and have time stamp markers; Perform time domain registration and low pass filtering on the three groups of signals, and align the ultraviolet absorption jump point with the pressure and flow change time, extract the steady-state cooperative segment of the three groups of signals as the cooperative baseline; Perform dynamic deviation comparison of the ultraviolet signal and the cooperative baseline, identify abnormal segments with absorption deviation exceeding the threshold, and construct the optical path disturbance trajectory combined with the synchronous change trend of pressure and flow; Extract waveform feature indexes by time period and encode classification to form a refraction fingerprint with unique characteristics for subsequent identification and comparison of abnormal detection behavior.

3. The novel polypeptide synthesizer according to claim 2, characterized in that, The bubble retention risk distribution map generation steps are as follows: Select the absorption abnormal section from the ultraviolet detection signal matched with the optical path refraction fingerprint, and extract the gas pressure detection signal and liquid flow detection signal in the corresponding time period to form a set of three aligned time series data; Perform time reversal processing on the three sets of aligned time series data to reverse the potential causes before the occurrence of interference events and construct a structured time model; Perform differential reenactment on the time-reversed signal set to extract the rate characteristics of absorption value mutation, pressure drop, and flow rebound, forming a three-dimensional feature vector for each interference event; Map the three-dimensional feature vector to the physical space location based on the liquid flow rate and pipeline geometric parameters, and generate a bubble retention risk distribution map combined with the interference intensity, duration, and type.

4. The novel polypeptide synthesizer according to claim 3, characterized in that, The optical path compensation curve generation process is as follows: The positions with high interference risk marked in the bubble residence risk distribution map are connected with two independent optical paths with the same structure, which are used for working liquid detection and stable reference liquid detection respectively, and a piezoelectric ceramic piece is fixed on the outer wall of the optical path to collect acoustic signals; The ultraviolet detection signals of the working optical path and the reference optical path and the acoustic signals output by the piezoelectric ceramic piece are synchronously collected, and the three groups of signals are time-aligned and compared to identify the signal offset and acoustic response events caused by optical path disturbance; The optical offset, acoustic amplitude, duration and delay are recorded to construct the optical path compensation curve describing the relationship between acoustic disturbance and optical offset, and a refractive disturbance quantification model is generated accordingly; In the detection process, the refractive disturbance quantification model is called in real time, the current acoustic amplitude is substituted into the compensation curve to calculate the standard optical offset, and the actual ultraviolet offset is compared to identify the signal offset caused by bubbles and perform comparison and discrimination operations.

5. The novel polypeptide synthesizer according to claim 4, characterized in that, The bubble discharge window formation process is as follows: According to the disturbance amplitude in the optical path compensation curve, it is judged whether it exceeds the set threshold, if it exceeds, the control logic of the multi-position multi-way valve state combination is triggered, and the two channels connected to the outlet of the synthesis column and the exhaust branch are opened in turn, the detection channel is kept in the minimum open state, and the waste liquid collection channel is closed; The mass flow controller gradually increases the gas flow rate in a slow ramp mode, and controls the bubble discharge speed by increasing the amplitude in stages; The bubble migration path is limited in the vertical upward channel through structural arrangement, an exhaust outlet with a hydrophobic membrane is arranged at the upper end, and a closed collection bottle with back pressure control is connected to guide the bubble discharge; During the bubble discharge window period, the changes of ultraviolet detection signals and acoustic disturbance signals are compared in real time to judge whether the bubbles are discharged and the optical path is restored to stability; After the bubble discharge event is completed, the exhaust channel is automatically closed and all valves are restored to the original state, the mass flow controller is released from the slope limitation, and all parameters of this event are written into the reaction control process.

6. The novel polypeptide synthesizer according to claim 5, wherein, During the gas flow rate increasing process of the mass flow controller in the exhaust starting stage, the slope is controlled by a stepwise segmented speed increasing mode, and before each stage is increased, the real-time flow rate and target value of the previous stage are compared, if the deviation exceeds the preset range, the next stage is delayed.

7. The novel polypeptide synthesizer according to claim 5, wherein, After the bubble discharge window is formed, the sweep rhythm and microwave power duty cycle are adjusted according to the residual density of the detection, the refractive index fluctuation gradient is calculated, the optical path stability criterion is generated, and the optical path balance trigger signal is sent when the fluctuation gradient is lower than the threshold. The steps are as follows: According to the offset residual of ultraviolet absorption value and acoustic signal extracted from the optical path compensation curve after bubble discharge, a two-dimensional residual density map is constructed to identify the disturbance area; According to the area position and intensity in the residual density map, the inert gas sweep rhythm is set, and the sweep frequency and duration of different positions are controlled by the multi-way valve to adjust the output flow rate of the mass flow controller to control the sweep intensity; The duty cycle of the microwave heating device is adjusted dynamically according to the residual density, and the heating period is controlled to suppress thermal disturbance and maintain stable reaction temperature; After the sweep rhythm and microwave power are stable, the refractive index fluctuation gradient is calculated based on the real-time absorption signal to judge whether the optical path stability condition is met; The optical path balance trigger signal is sent out after continuously meeting the optical path stability condition, the liquid path enters the stable state is controlled, and the current control parameter is recorded for subsequent learning and optimization.

8. The novel polypeptide synthesizer according to claim 7, characterized in that, Under the action of the optical path balance trigger signal, phase conjugate optical path shaping is performed, the timing of the multi-position multi-way valve and the purge frequency are controlled in combination with reversible time grid migration, and the following steps are constructed for the optical path self-correction closed loop: After receiving the optical path balance trigger signal, an optical reference path with reverse phase characteristics is constructed, phase conjugate interference operation is performed through beam splitting, delay, refractive crystal adjustment and phase difference control to offset the ultraviolet signal offset caused by liquid disturbance; At the same time of performing optical shaping operation, the reversible time grid strategy based on the recorded valve position switching sequence is called to implement reverse dwell timing control on the multi-position multi-way valve, and reverse pulse flow of liquid flow is generated to drive the disturbance area away from the detection position; After the optical phase compensation and the liquid path time regulation are completed synchronously, the ultraviolet signal is periodically scanned and analyzed, whether the absorption intensity and the signal curve change meet the false peak extinction criterion is judged, and the shaping parameters and the detection results are recorded as feedback templates for subsequent calling.

Images