Method and system for pressure and temperature coupling monitoring in radioactivity column separation

Abstract

The application discloses a pressure and temperature coupling monitoring method and system in radioactive column separation, and belongs to the field of microfluidic monitoring, and the method comprises the following steps: injecting a pressure oscillation with a predetermined frequency into a microcolumn; detecting deformation signals generated at the inlet and outlet of the microcolumn due to internal pressure changes; extracting an inlet response amplitude, an outlet response amplitude and a phase difference between the inlet response and the outlet response corresponding to the pressure oscillation frequency from the deformation signals; calculating a propagation time of the pressure oscillation from the inlet to the outlet based on the phase difference; and determining the propagation time and inherent structural parameters of the microcolumn. The pressure oscillation generating device can be designed by using a non-invasive sensing method and is adapted to the microscale, thereby avoiding the problems of blockage and flow field interference of the microcolumn flow channel caused by the invasive arrangement of traditional macroscopic sensors, and effectively eliminating the measurement distortion caused by the temperature and pressure coupling at the microscale through the calculation of parameters such as the phase difference and the equivalent sound speed, so that the accuracy of the internal pressure monitoring of the microcolumn is improved.

Description

Technical Field

[0001] This invention relates to the field of microfluidic monitoring, and more specifically, to a method and system for pressure and temperature coupling monitoring in radioactive column separation. Background Technology

[0002] In the separation and purification of radiolabeled biological samples, microfluidic radioparticle column technology has advantages such as low reagent consumption, high separation efficiency, and easy integration. Microfluidic radioparticle columns are essentially microfluidic-scale chromatographic columns. This technology relies on microchannels with a size of less than 500 micrometers and a packing medium with a pore size of less than 100 micrometers to achieve efficient capture and elution of target radioactive components in trace amounts of blood and tissue samples. Due to the extremely small size of the channels and pores, the internal fluid flow and heat transfer behavior differ from that of the macroscopic system. Strong local effects are generated inside the packing pores, forming a temperature difference of 2 to 3°C and micro-pressure fluctuations at the level of ±0.5 kPa. These micro-region temperature and pressure gradients directly affect the fluid viscosity, diffusion coefficient, and phase equilibrium state, and also affect the adsorption and desorption kinetics of radioactive targets, ultimately adversely affecting the separation and recovery rate.

[0003] Currently, most monitoring technologies in this field are derived from macroscopic chromatography systems, relying on discrete physical sensors such as thermocouples and piezoresistive probes for point measurements. These sensors are combined with temperature-pressure coupling algorithms based on the assumption of a macroscopic uniform field to complete signal correction and compensation. However, at the mesoscopic scale of microfluidic columns, this technological paradigm has certain limitations: On the one hand, the sensor probe size is usually comparable to or even larger than the flow channel or pore scale. Invasive placement methods can severely interfere with the flow field distribution, block microchannels, and alter the local heat and mass transfer state, leading to severe distortion of measurement data. On the other hand, macroscopic temperature-pressure coupling algorithms do not consider the strong correlation between temperature and pressure fields at the microscale due to the modulation of microstructures. This can amplify the inherent thermal drift errors in the micropressure signal, causing the corrected data to deviate from the physical reality and fail to accurately reflect the actual process state within the pores.

[0004] The problem with current technology is that it directly applies the point measurement independent monitoring logic applicable to macroscopic continuous media to the microfluidic physical field of mesoscopic discrete multiphase and strongly coupled temperature and pressure. This monitoring method is fundamentally mismatched with the physical nature of the monitored object. The macroscopic sensing and calculation methods originally introduced to pursue the accuracy of micro-area monitoring have instead destroyed the intrinsic state and correlation characteristics of the micro-area physical field, ultimately leading to the failure of the monitoring function. Summary of the Invention

[0005] To address the problems existing in the prior art, the present invention aims to provide a method and system for monitoring pressure and temperature coupling in radioactive column separation. This method and system can achieve pressure oscillation generation using a non-invasive sensing design and a microscale-adapted device, avoiding the blockage and flow field interference problems caused by the invasive deployment of traditional macroscopic sensors in the micro-column flow channels. At the same time, by calculating parameters such as phase difference and equivalent sound velocity, the method effectively eliminates the measurement distortion caused by temperature and pressure coupling at the microscale, thereby improving the accuracy of pressure monitoring inside the micro-column.

[0006] To solve the above problems, the present invention adopts the following technical solution:

[0007] The first aspect involves a pressure-temperature coupled monitoring method for radioactive column separation, including:

[0008] Step 1: Inject pressure oscillations at a predetermined frequency into the microcolumn; detect the deformation signals generated at the inlet and outlet of the microcolumn due to internal pressure changes; extract the inlet response amplitude, outlet response amplitude, and phase difference between the inlet and outlet responses corresponding to the pressure oscillation frequency from the deformation signals.

[0009] Step 2: Calculate the propagation time of the pressure oscillation from the inlet to the outlet based on the phase difference; determine the equivalent sound velocity of the fluid inside the micro-column based on the propagation time and the inherent structural parameters of the micro-column.

[0010] Step 3: Obtain the process pressure measurement value of the micro-column; combine the equivalent sound velocity and process pressure measurement value to construct a mapping model with the equivalent sound velocity and process pressure measurement value as input and the internal pressure of the micro-column as output.

[0011] Step 4: Input the real-time equivalent sound velocity and process pressure measurement values ​​into the mapping model to obtain the pressure estimate inside the micropillar;

[0012] Step 5: Compare the pressure estimate with the target pressure value to obtain the pressure deviation; adjust the frequency of pressure oscillation based on the relationship between the characteristics of the pressure deviation and the oscillation response characteristics of the inlet and outlet response amplitudes.

[0013] Step 6: At multiple frequencies, obtain the energy attenuation ratio of the pressure oscillation from the inlet to the outlet to form an attenuation spectrum; compare the attenuation spectrum with the pre-stored reference attenuation spectrum, and evaluate the state of the microcolumn based on the comparison results.

[0014] Furthermore, the inlet response amplitude, outlet response amplitude, and phase difference between the inlet and outlet responses corresponding to the pressure oscillation frequency are extracted, including:

[0015] Step 11: Inject a pressure wave with a continuously varying frequency into the microcolumn, simultaneously detect the deformation signals at the inlet and outlet, analyze the frequency response spectrum based on the deformation signal at the outlet, and select a characteristic resonance frequency from the frequency response spectrum.

[0016] Step 12: Inject phase-locked pressure oscillation into the micropillar at the characteristic resonant frequency. After reaching a steady state, detect the first deformation signal at the inlet and the second deformation signal at the outlet respectively. Subtract the first deformation signal from the second deformation signal to obtain the differential signal.

[0017] Step 13: Perform zero-crossing detection on the differential signal to determine its real-time frequency, and perform analog multiplication operations on the differential signal with the reference sine wave and reference cosine wave of the same frequency, and then perform low-pass filtering. Calculate the amplitude and phase of the differential signal based on the filtering results to obtain the input response amplitude, the output response amplitude, and the phase difference between the two.

[0018] Further, determining the equivalent sound velocity of the fluid within the microcolumn includes:

[0019] Step 21: Switch the fluid inside the micro-column to a reference fluid with a known sound velocity-temperature relationship and maintain it at a known temperature. Obtain the phase difference between the inlet and outlet based on the reference fluid. Calculate the propagation time based on the phase difference and oscillation frequency. Combine the known sound velocity of the reference fluid at a known temperature to calculate the equivalent sound propagation path length of the micro-column in reverse.

[0020] Step 22: Switch the fluid inside the micro-column to the working fluid, obtain the real-time phase difference between the inlet and outlet based on the working fluid, calculate the real-time propagation time based on the phase difference, and determine the equivalent sound velocity of the working fluid using the equivalent sound propagation path length and the real-time propagation time.

[0021] Furthermore, the process pressure measurements of the microcolumn are obtained, including:

[0022] Step 31: Control the pressure of the micro-column at multiple known reference pressure points, and under constant temperature conditions, obtain the equivalent sound velocity of the working fluid at each reference pressure point. Based on the correspondence between the equivalent sound velocity and the reference pressure point, generate a sound velocity-pressure correlation calibration curve.

[0023] Step 32: Input the equivalent sound velocity of the working fluid obtained in real time into the sound velocity-pressure correlation calibration curve to obtain a preliminary pressure estimate, and obtain the real-time temperature observation value. Use the real-time temperature observation value and the pre-stored sound velocity-temperature coefficient of the working fluid to perform temperature compensation calculation on the preliminary pressure estimate value to obtain the process pressure measurement value.

[0024] Furthermore, the construction of the mapping model includes:

[0025] Step 33: Under multiple preset working conditions, measure the local pressure at multiple different axial positions inside the micro-pillar, and simultaneously acquire the equivalent sound velocity and process pressure measurement values ​​at each working condition. Use the equivalent sound velocity and process pressure measurement values ​​at each working condition as coordinates, and the corresponding multiple local pressures as pressure field vectors to form a set of probe points.

[0026] Step 34: Triangulate all points in the probe point set on a two-dimensional plane with equivalent sound velocity and process pressure measurement values ​​as axes to form a triangular mesh. Determine the target triangle to which the point belongs in the triangular mesh based on the real-time input equivalent sound velocity and process pressure measurement values. Calculate the centroid coordinates of the real-time input point relative to the three vertices of the target triangle. Based on the centroid coordinates, perform a weighted superposition of the pressure field vectors stored at the three vertices to obtain the predicted value of the internal pressure of the micropillar.

[0027] Furthermore, the real-time obtained equivalent sound velocity and process pressure measurements are input into the mapping model to obtain the pressure estimate inside the micropillar, including:

[0028] Step 41: Analyze the geometric position of the real-time input equivalent sound velocity and process pressure measurement values ​​in the historical calibration data space composed of probe point sets. Based on its distance from the calibration point and its relative positional relationship with the convex hull formed by the calibration point, select the corresponding solution strategy.

[0029] Step 42: Based on the selected solution strategy, call the corresponding solver to process the real-time input equivalent sound velocity and process pressure measurement values ​​to obtain the pressure estimate inside the micro-column.

[0030] Furthermore, the pressure estimate is compared with the target pressure value to obtain the pressure deviation, including:

[0031] Step 51: Based on the continuously acquired pressure estimate and target pressure value, generate a time series of pressure deviation, and extract morphological feature values ​​including the rising slope, falling slope, duration of steady-state offset plateau, and number of zero crossings from the time series.

[0032] Step 52: Perform a linkage analysis between the morphological characteristic value and the oscillation response attenuation mode characteristics from the inlet response amplitude and the outlet response amplitude, and separate the pressure deviation into deviation components corresponding to different physical sources according to the pre-stored judgment logic.

[0033] Furthermore, based on the relationship between the characteristics of the pressure deviation and the oscillation response characteristics of the inlet and outlet response amplitudes, the frequency of the pressure oscillation is adjusted, including:

[0034] Step 53: Based on the identification of the dominant component of the deviation and the fluctuation characteristics of the outlet response amplitude under the current pressure oscillation frequency, evaluate the suppression efficacy of the current frequency on the dominant component.

[0035] Step 54: Based on the evaluation results of the suppression effectiveness, start and execute a directional frequency scan starting from the current frequency. During the scan, calculate the instantaneous suppression effectiveness at each trial frequency in real time, and lock the frequency that makes the instantaneous suppression effectiveness reach the optimal condition as the new pressure oscillation frequency.

[0036] Furthermore, the state of the microcolumns is evaluated based on the comparison results, including:

[0037] Step 61: Analyze the morphological characteristics of the pre-stored reference attenuation spectrum to identify a set of characteristic frequency bands associated with the filler state;

[0038] Step 62: For the attenuation spectrum acquired in real time, calculate the offset of its energy relative to the reference value based on the characteristic frequency band locked in step 61, and obtain the filler state change index by combining the offsets of each frequency band.

[0039] Step 63: Based on the time series of the packing state change index, analyze its change rate and acceleration to identify the dynamic mode of packing degradation, and extrapolate to predict the remaining healthy life.

[0040] Step 64: Combining the packing deterioration dynamic mode with process pressure measurements and flow rate settings, assess the risk level of the microcolumn based on the preset decision matrix and generate maintenance recommendations.

[0041] Secondly, the present invention also provides a pressure and temperature coupling monitoring system suitable for the above-mentioned radioactive column separation, comprising:

[0042] The signal extraction module is used to inject pressure oscillations with a predetermined frequency into the microcolumn; detect the deformation signals generated at the inlet and outlet of the microcolumn due to internal pressure changes; and extract the inlet response amplitude, outlet response amplitude, and phase difference between the inlet and outlet responses corresponding to the pressure oscillation frequency from the deformation signals.

[0043] The sound velocity calculation module calculates the propagation time of pressure oscillation from inlet to outlet based on phase difference; and determines the equivalent sound velocity of the fluid inside the micro-pillar based on the propagation time and the inherent structural parameters of the micro-pillar.

[0044] The model building module obtains the measured process pressure values ​​of the micropillar; by combining the equivalent sound velocity and the measured process pressure values, a mapping model is constructed with the equivalent sound velocity and the measured process pressure values ​​as inputs and the internal pressure of the micropillar as the output.

[0045] The pressure estimation module is used to input the real-time obtained equivalent sound velocity and process pressure measurement values ​​into the mapping model to obtain the pressure estimate inside the micropillar;

[0046] The frequency adjustment module is used to compare the pressure estimate with the target pressure value to obtain the pressure deviation; and to adjust the frequency of pressure oscillation based on the relationship between the characteristics of the pressure deviation and the oscillation response characteristics of the inlet response amplitude and the outlet response amplitude.

[0047] The state assessment module is used to obtain the energy attenuation ratio of pressure oscillation from inlet to outlet at multiple frequencies to form an attenuation spectrum; the attenuation spectrum is compared with the pre-stored benchmark attenuation spectrum, and the state of the microcolumn is assessed based on the comparison results.

[0048] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0049] (1) This scheme adopts a non-invasive sensing design and a pressure oscillation generator adapted to the microscale, which avoids the blockage and flow field interference problems of the traditional macroscopic sensor invasive deployment. At the same time, by calculating parameters such as phase difference and equivalent sound velocity, the measurement distortion caused by temperature and pressure coupling at the microscale is effectively eliminated, and the accuracy of pressure monitoring inside the micro-column is improved.

[0050] (2) This scheme achieves precise location of the root cause of deviation by extracting the morphological features of the pressure deviation time series and analyzing the linkage of oscillation response characteristics. Combined with the directional frequency scanning and optimal frequency locking mechanism, it can dynamically adapt to the fluid state and process changes in the microcolumn, quickly suppress the dominant deviation component, ensure that the internal pressure of the microcolumn is stable within the target range, and improve the continuity and reliability of the separation process.

[0051] (3) Based on the comparison of multi-frequency energy decay spectrum and the analysis of characteristic frequency bands, this scheme constructs a packing state change index and a deterioration dynamic pattern recognition mechanism, which can accurately quantify the degree of packing deterioration and extrapolate the remaining healthy life, providing a reliable basis for preventive maintenance.

[0052] (4) This solution is fully compatible with the microseparation scenario of microfluidic radioactive columns. Both the equipment and the algorithm are designed for microscale temperature and pressure coupling characteristics. It does not need to rely on macroscopic chromatographic monitoring logic, which can reduce reagent consumption and sample loss. Furthermore, through fully automated data acquisition, calculation and control, it can reduce the cost of manual intervention and improve the efficiency and standardization of radiolabeled biological sample separation and purification. Attached Figure Description

[0053] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0054] Figure 1This is a flowchart of the pressure and temperature coupling monitoring method in radioactive column separation according to the present invention;

[0055] Figure 2 This is a flowchart illustrating the relationships between various modules in the pressure and temperature coupling monitoring system for radioactive column separation in this invention. Detailed Implementation

[0056] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0057] Example 1

[0058] Please see Figure 1 A pressure-temperature coupling monitoring method for radioactive column separation includes the following steps: Step 1, injecting pressure oscillations at a predetermined frequency into the microcolumn; detecting the deformation signals at the inlet and outlet of the microcolumn caused by internal pressure changes; extracting the inlet response amplitude, outlet response amplitude, and phase difference between the inlet and outlet responses corresponding to the pressure oscillation frequency from the deformation signals. The specific operations are as follows:

[0059] For ease of reading, the microfluidic radioactive column will be referred to as a microcolumn from now on. During the radioactive column separation process, a pressure oscillation at a predetermined frequency must first be injected into the microcolumn. This predetermined frequency needs to be pre-set based on the inherent structural parameters of the microcolumn and the characteristics of the internal working fluid to ensure that the pressure oscillation can propagate stably within the microcolumn without disrupting the normal operation of the separation process. After the pressure oscillation is injected, the internal pressure of the microcolumn undergoes periodic changes. These pressure changes cause corresponding deformations in the walls or associated sensing structures at the inlet and outlet ends of the microcolumn. A suitable deformation detection device synchronously captures the deformation signals at the inlet and outlet. These deformation signals indirectly reflect the pressure change characteristics within the microcolumn. Since the detected deformation signals may contain irrelevant components such as environmental interference and equipment noise, the captured deformation signals need to be specifically processed. Feature parameters that strictly correspond to the injected pressure oscillation frequency are screened and extracted from the signals. These parameters include the inlet response amplitude, the outlet response amplitude, and the phase difference between the inlet and outlet responses.

[0060] The extraction of the inlet response amplitude, outlet response amplitude, and phase difference between the inlet and outlet responses corresponding to the pressure oscillation frequency also includes the following steps:

[0061] Step 11: Inject a pressure wave with a continuously varying frequency into the micropillar, simultaneously detect the deformation signals at the inlet and outlet, analyze the frequency response spectrum based on the deformation signal at the outlet, and select a characteristic resonant frequency from the frequency response spectrum. The specific operation is as follows:

[0062] To accurately extract the response parameters corresponding to the pressure oscillation frequency, the characteristic resonant frequency must first be selected. A continuously varying pressure wave is injected into the microcolumn. The frequency range of this pressure wave must cover the frequency bands where the microcolumn and its internal fluid may resonate, ensuring the capture of key frequency response characteristics. Simultaneously, deformation signals at the inlet and outlet of the microcolumn are collected using a deformation detection device. Frequency response analysis is primarily based on the deformation signal at the outlet. A frequency response spectrum is plotted using signal processing techniques. This spectrum clearly shows the variation of the amplitude and phase of the outlet deformation signal with the pressure wave frequency. Combining the spectral characteristics, frequencies with significant response characteristics are selected as the characteristic resonant frequency. This frequency must exhibit high sensitivity and stability of the outlet deformation signal at that frequency, and effectively distinguish signal differences caused by pressure changes within the microcolumn.

[0063] The deformation detection device consists of a miniature sensing element and a signal conditioning module. The miniature sensing element is a probe, the size of which must be less than 100 micrometers to match the pore size of the micropillars, avoiding invasive placement that could interfere with the flow field in the microchannel. Commonly used sensing elements fall into two categories: one is a micro-strain gauge, which uses the piezoresistive effect of semiconductor materials to convert minute deformations of the micropillar wall into changes in resistance; the other is a capacitive micro-pressure sensor, which uses changes in the capacitance of the microstructure to respond to pressure-induced deformation. This type of sensor does not require direct contact with the micropillar wall, enabling non-contact detection. The signal conditioning module is responsible for amplifying, filtering, and performing analog-to-digital conversion on the weak electrical signal output by the sensing element, ensuring that the signal-to-noise ratio meets the requirements of subsequent analysis. When pressure oscillations are injected into the micropillar, the periodic changes in pressure cause minute elasticity on the walls of the micropillar inlet or outlet. The deformation acts on the surface of the sensitive element of the deformation detection device. Taking a micro-strain gauge as an example, the deformation of the micro-pillar wall will cause the resistance value of the strain gauge to change. The change in resistance is converted into a voltage signal through a Wheatstone bridge circuit. Taking a capacitive micro-pressure sensor as an example, the deformation of the micro-pillar wall will change the distance between the sensor plates, thereby causing a change in capacitance. This change can be converted into a voltage signal through a resonant circuit or a bridge circuit. The signal conditioning module amplifies the converted voltage signal and filters out environmental noise and electromagnetic interference, and finally outputs a stable deformation signal for subsequent operations such as response amplitude and phase difference extraction.

[0064] Step 12: Inject phase-locked pressure oscillations into the micropillar at the characteristic resonant frequency. After reaching a steady state, detect the first deformation signal at the inlet and the second deformation signal at the outlet. Subtract the first deformation signal from the second deformation signal to obtain the differential signal. The specific operation is as follows:

[0065] Using the characteristic resonant frequency selected in step 11 as the injection frequency, phase-locked pressure oscillations are injected into the microcolumn. The phase-locking operation is achieved through a dedicated signal generation and control device to ensure that the phase of the injected pressure oscillation remains stable, avoiding deviations in subsequent signal analysis due to phase drift. Phase-locked pressure oscillations are continuously injected until the pressure field, fluid flow state, and deformation detection signal inside the microcolumn all reach a steady state. The criterion for judging the steady state is that the amplitude and phase of the deformation signals at the inlet and outlet do not fluctuate significantly within a stable range over multiple consecutive cycles. After reaching a steady state, the first deformation signal at the inlet and the second deformation signal at the outlet of the microcolumn are collected by the deformation detection device to ensure that the two collected signals are strictly synchronized in the time dimension. The collected first deformation signal and the second deformation signal are subtracted to obtain a differential signal. This operation can effectively cancel out irrelevant components such as environmental interference and equipment common-mode noise, highlighting the signal differences caused by the internal characteristics of the microcolumn during the propagation of pressure oscillations, and improving the accuracy of subsequent signal processing.

[0066] Step 13: Perform zero-crossing detection on the differential signal to determine its real-time frequency. Then, perform analog multiplication operations on the differential signal with reference sine and reference cosine waves of the same frequency, followed by low-pass filtering. Calculate the amplitude and phase of the differential signal based on the filtering results to obtain the input response amplitude, output response amplitude, and the phase difference between them. The specific operations are as follows:

[0067] Zero-crossing detection is performed on the differential signal obtained in step 12. By identifying the time node when the signal crosses the zero potential point, the time interval between two adjacent zero-crossing points is calculated to determine the real-time frequency of the differential signal. This real-time frequency is ensured to be consistent with the injected pressure oscillation frequency to verify the validity of the signal. Subsequently, the differential signal is subjected to simulated multiplication with reference sine and reference cosine waves of the same frequency. This operation essentially achieves synchronous demodulation of the differential signal and the reference signal, converting the amplitude and phase information in the differential signal into low-frequency signal components. The signal after multiplication is then subjected to low-pass filtering. The cutoff frequency of the low-pass filter needs to be set according to the reference signal frequency to ensure effective filtering of high-frequency harmonic components and residual noise generated by the multiplication operation, while retaining the low-frequency signal that reflects the characteristics of the differential signal. The amplitude and phase of the differential signal are calculated based on the result of the low-pass filtered signal. The amplitude of the inlet response and the amplitude of the outlet response are determined by the peak or effective value of the filtered signal, and the phase difference between them is obtained by comparing the phase relationship between the inlet and outlet response signals. The amplitude calculation can be achieved by the following formula:

[0068] ;

[0069] This formula is derived based on the principle of synchronous detection. It uses the amplitudes of the filtered signal obtained by multiplying the differential signal with reference sine and cosine waves as orthogonal components, and synthesizes them using the Pythagorean theorem to obtain the total response amplitude. In the formula, A represents the response amplitude, indicating the inlet or outlet. This represents the amplitude of the filtered signal after multiplying the differential signal with a reference sine wave. The amplitude and phase difference represent the signal amplitude after the differential signal is multiplied and filtered by the reference cosine wave. Then it is calculated using the following formula:

[0070] ;

[0071] This formula is derived based on the phase relationship of orthogonal signals. The phase of the response signal is obtained by using the arctangent function of the ratio of the amplitudes of two orthogonal filtered signals, and then the phase difference between the inlet response and the outlet response is determined.

[0072] In a preferred embodiment of the present invention, step 2 is further included: calculating the propagation time of the pressure oscillation from the inlet to the outlet based on the phase difference; determining the equivalent sound velocity of the fluid inside the micro-pillar based on the propagation time and the inherent structural parameters of the micro-pillar, the specific operation of which is as follows:

[0073] When pressure oscillations propagate through the fluid inside a micro-column, their propagation characteristics are directly related to the velocity of sound in the fluid. The phase difference between the inlet and outlet responses is essentially the phase lag corresponding to the time difference of the pressure oscillation's propagation from the inlet to the outlet. According to the phase propagation law of simple harmonic motion, there is a fixed correlation between the phase difference, propagation time, and oscillation frequency. Based on this, the calculation logic for the propagation time can be derived: the angular frequency of the pressure oscillation is the product of 2π and the oscillation frequency, and the phase difference is equal to the product of the angular frequency and the propagation time. Based on this physical relationship, the formula for calculating the propagation time can be derived as follows:

[0074] ;

[0075] This formula is derived from the phase lag principle of harmonic signal propagation. The phase difference is essentially the amount of phase change that the oscillating signal completes during propagation, expressed through angular frequency. By correlating frequency with phase, the propagation time can be calculated in reverse; in the formula, t represents the propagation time of the pressure oscillation from the inlet to the outlet. The phase difference between the inlet and outlet responses extracted in step 1 is represented by f, which represents the predetermined frequency of the injection pressure oscillation. After calculating the propagation time, the equivalent sound propagation path length is determined by combining the inherent structural parameters of the micropillar, specifically the equivalent sound propagation path length. This parameter needs to be obtained through calibration in step 21, rather than simply the geometric length of the micropillar. Based on the basic definition of sound velocity, i.e., the sound velocity is equal to the ratio of the propagation path length to the propagation time, the equivalent sound velocity of the fluid inside the micropillar can be determined. The equivalent sound velocity here is the effective sound velocity after considering factors such as the pore structure of the micropillar filling medium and the fluid flow state, and can truly reflect the actual sound velocity. This method effectively reflects the acoustic propagation characteristics of fluids in microscale spaces. Throughout the process, the deformation detection device used is a miniaturized, non-invasive sensing device adapted to the micro-column scale. Its probe size does not exceed 100 micrometers and matches the micro-column pore size. Using micro-strain gauges or capacitive micro-pressure sensing elements, the device can convert the minute deformation of the micro-column wall caused by pressure oscillations into an acquireable electrical signal. After amplification and filtering by the signal conditioning module, a stable deformation signal is output, ensuring the accuracy of phase difference extraction and without interfering with the flow field distribution within the micro-column or changing the local heat and mass transfer state.

[0076] Determining the equivalent sound velocity of the fluid inside the micropillar also includes the following steps:

[0077] Step 21: Switch the fluid inside the micro-column to a reference fluid with a known sound velocity-temperature relationship, and maintain it at a known temperature. Obtain the phase difference between the inlet and outlet based on the reference fluid. Calculate the propagation time based on the phase difference and oscillation frequency. Combined with the known sound velocity of the reference fluid at a known temperature, calculate the equivalent sound propagation path length of the micro-column in reverse. The specific operation is as follows:

[0078] Because the micropillar contains a filling medium and has a complex flow channel structure, the actual sound propagation path of the pressure oscillation is not the geometric axial length of the micropillar and cannot be obtained through direct measurement. Therefore, calibration is required using a reference fluid with a known sound velocity-temperature relationship. First, the original fluid inside the micropillar is drained and replaced with the reference fluid. The reference fluid must meet the requirement of a clear and stable correlation between sound velocity and temperature, such as deionized water or a standard ethanol solution of a specific concentration. Those skilled in the art can select common standard reference fluids according to actual process conditions. Subsequently, a micro temperature control module maintains the micropillar and the internal reference fluid at a known and stable temperature. This temperature must be within the valid range of the reference fluid's sound velocity data to ensure that the sound velocity of the reference fluid is a clearly known value. The micro temperature control module consists of a micro-patterned temperature sensor and a micro-scale lifting mechanism. The module consists of a temperature actuator and a closed-loop temperature control circuit, and is equipped with an insulation sleeve adapted to the micro-column structure. The temperature sensor and the temperature control actuator are non-invasively positioned against the outer wall of the micro-column, without altering the internal flow channel structure or interfering with fluid flow and pressure oscillation propagation. The temperature sensor's placement corresponds to the fluid contact area within the micro-column, ensuring the measured temperature matches the actual fluid temperature. The closed-loop temperature control circuit receives the real-time temperature signal from the sensor, filters the signal to eliminate noise interference, and adjusts the operating power of the temperature control actuator using a PID algorithm to achieve closed-loop temperature feedback control. This module has a temperature control accuracy of ≤±0.1℃ and a temperature stability of ≤±0.1℃, covering the operating temperature range of the micro-column separation process. It can accurately and stably maintain the fluid within the micro-column at a known, pre-set constant temperature.

[0079] In this state, pressure oscillations of a predetermined frequency are injected into the microcolumn according to the operation method in step 1. Deformation signals at the inlet and outlet are simultaneously acquired by a deformation detection device, and the phase difference between them is extracted. Then, using the propagation time calculation formula derived in step 2, combined with the frequency of the injected pressure oscillations, the propagation time of the pressure oscillations from the inlet to the outlet in the reference fluid is calculated. According to the basic definition of sound speed, the sound speed is equal to the ratio of the propagation path length to the propagation time. Based on this relationship, the equivalent sound propagation path length of the microcolumn can be calculated in reverse. The corresponding calculation formula is:

[0080] ;

[0081] This formula is derived from the basic physical definition of the speed of sound. The speed of sound in the reference fluid is constant at a known temperature, and the propagation time has been calculated using the phase difference. The product of these two factors is the actual sound propagation path length. In the formula, L represents the equivalent sound propagation path length of the micropillar. The known speed of sound of the reference fluid at a known temperature. This represents the propagation time of pressure oscillations in the reference fluid. The deformation detection device used here is exactly the same as in step 1. Its non-invasive design avoids interference with the flow state of the reference fluid, and its high sensitivity can accurately capture minute deformation signals, ensuring the accuracy of the phase difference and propagation time calculations, thereby ensuring the reliability of the equivalent sound propagation path length calibration results.

[0082] Step 22: Switch the fluid inside the micro-column to the working fluid. Obtain the real-time phase difference between the inlet and outlet based on the working fluid. Calculate the real-time propagation time based on the phase difference. Determine the equivalent sound velocity of the working fluid using the equivalent sound propagation path length and the real-time propagation time. The specific operations are as follows:

[0083] The working fluid is the actual fluid used in the radioactive column separation process, typically a buffer solution containing radiolabeled biological samples. Its sound velocity changes with pressure, temperature, and composition within the microcolumn, requiring real-time detection and calculation to obtain the equivalent sound velocity. First, the reference fluid within the microcolumn is emptied and replaced with the working fluid used in the actual separation process, ensuring the fluid fills the internal channels and pores of the filling medium without any air bubbles. Then, following the procedure in step 1, a phase-locked pressure oscillation is injected into the microcolumn at the characteristic resonance frequency selected in step 11. Once the fluid state and deformation signal within the microcolumn reach a steady state, the first deformation at the inlet is synchronously collected using a deformation detection device. The signal and the second deformation signal at the outlet are subtracted to obtain a differential signal. Then, through signal processing operations such as zero-crossing detection and filtering, the real-time phase difference between the inlet and outlet responses is extracted. Based on this real-time phase difference and the frequency of the injected pressure oscillation, the real-time propagation time of the pressure oscillation in the working fluid from the inlet to the outlet is calculated using the propagation time calculation formula derived in step 2. Since the equivalent sound propagation path length of the microcolumn has been calibrated in step 21, and this length is an inherent structural parameter of the microcolumn that remains unchanged during the process, the basic definition of sound velocity can be used again, combined with the real-time propagation time, to calculate the equivalent sound velocity of the working fluid. The corresponding calculation formula is:

[0084] ;

[0085] This formula is directly derived from the basic physical definition of the speed of sound. The equivalent sound propagation path length is fixed, and the effective speed of sound in the working fluid can be calculated by using the real-time propagation time. In the formula… The equivalent sound velocity of the working fluid is represented by L, which represents the equivalent sound propagation path length of the micropillar calibrated in step 21. The deformation detection device used here represents the real-time propagation time of pressure oscillations in the working fluid. It follows the selection criteria of the previous steps, with a high response speed, that is, a response time of less than 1 millisecond, which can ensure the timeliness of real-time phase difference extraction. Its radiation resistance and microscale adaptability can ensure stable operation in radioactive separation scenarios and microchannel environments, providing a reliable signal input for the real-time accurate calculation of equivalent sound velocity. Those skilled in the art can select a suitable micro non-invasive deformation sensing device to complete the signal acquisition operation based on the above description.

[0086] In a preferred embodiment of the present invention, step 3 is further included: obtaining the process pressure measurement value of the micropillar; combining the equivalent sound velocity and the process pressure measurement value, constructing a mapping model with the equivalent sound velocity and process pressure measurement value as input and the internal pressure of the micropillar as output, the specific operation of which is as follows:

[0087] The process pressure measurement value of the micro-column is obtained. The process pressure measurement value is a parameter reflecting the overall pressure state of the micro-column. Its acquisition needs to be achieved through calibration and temperature compensation to eliminate temperature interference and ensure measurement accuracy. The equivalent sound velocity reflects the real-time physical characteristics of the fluid inside the micro-column. Together with the process pressure measurement value, it constitutes the input parameters characterizing the pressure field state inside the micro-column. Since the pressure field inside the micro-column exhibits spatial distribution characteristics due to the influence of factors such as the filling medium and fluid flow, it is impossible to directly obtain the internal global pressure through single-point measurement. Therefore, it is necessary to construct a mapping model to establish the correlation between the input parameters and the internal pressure. This model takes the equivalent sound velocity and process pressure measurement value as input and the internal pressure of the micro-column, which can be the global pressure or the pressure at key locations, as output. Its construction process requires collecting sample data through working condition calibration, and then realizing parameter correlation through mesh generation and interpolation algorithms. Finally, a stable model that can predict the internal pressure of the micro-column in real time is formed. Throughout the process, the acquisition of all parameters relies on the non-invasive deformation detection device adapted to the microscale and the supporting temperature and pressure control module in the previous steps to ensure the accuracy and reliability of data acquisition.

[0088] The process pressure measurement value of the microcolumn also includes the following steps:

[0089] Step 31: Control the pressure of the micro-column at multiple known reference pressure points, and under constant temperature conditions, obtain the equivalent sound velocity of the working fluid at each reference pressure point. Based on the correspondence between the equivalent sound velocity and the reference pressure points, generate a sound velocity-pressure correlation calibration curve. The specific operation is as follows:

[0090] Since the equivalent velocity of sound of the working fluid changes with pressure, and this relationship is influenced by fluid properties in a fixed manner, the correspondence between the two can be established through calibration experiments at multiple reference pressures. First, the microcolumn is filled with the working fluid, and the microcolumn pressure is precisely controlled to multiple known reference pressure points using a micro-pressure control module. These reference pressure points must cover the normal operating pressure range of the microcolumn separation process, and the intervals between adjacent pressure points must be reasonably set to ensure the continuity and accuracy of the correlation curve. Simultaneously, the microcolumn and its internal working fluid are maintained at a constant temperature using a temperature control module to eliminate the interference of temperature changes on the equivalent velocity of sound, ensuring that each set of calibration data is consistent. The influence of a single pressure variable is addressed. After each known reference pressure point stabilizes, the equivalent sound velocity of the working fluid under that pressure state is obtained according to the method in step 22. This involves injecting pressure oscillations, acquiring deformation signals, extracting phase differences, and calculating propagation time to finally obtain the equivalent sound velocity value corresponding to the reference pressure. Data fitting is performed on the pressure values ​​and equivalent sound velocity values ​​corresponding to all reference pressure points to generate a sound velocity-pressure correlation calibration curve. This curve must accurately characterize the quantitative relationship between the equivalent sound velocity and pressure under conditions without temperature interference. Those skilled in the art can use common data fitting methods, such as the least squares method, to generate the curve, ensuring that the error of the fitting result is within the allowable range of the process.

[0091] Step 32: Input the equivalent sound velocity of the working fluid obtained in real time into the sound velocity-pressure correlation calibration curve to obtain a preliminary pressure estimate. Also, obtain the real-time temperature observation value. Using the real-time temperature observation value and the pre-stored sound velocity-temperature coefficient of the working fluid, perform temperature compensation calculation on the preliminary pressure estimate to obtain the process pressure measurement value. The specific operation is as follows:

[0092] The sound velocity-pressure correlation calibration curve generated in step 31 is established at a constant temperature. However, in the actual separation process, the temperature inside the microcolumn will fluctuate locally, approximately 2 to 3°C. This temperature fluctuation will cause changes in the equivalent sound velocity of the working fluid. Directly substituting this into the calibration curve to calculate the pressure will result in deviations. Therefore, the equivalent sound velocity of the working fluid obtained in real time in step 22 needs to be input into the sound velocity-pressure correlation calibration curve generated in step 31. A preliminary pressure estimate is obtained through curve interpolation. This estimate does not consider the influence of real-time temperature fluctuations and needs further correction. Subsequently, the real-time temperature observation value of the working fluid inside the microcolumn is obtained through a micro temperature sensor. This temperature sensor needs to be adapted to the microscale scene, its placement should avoid interfering with the flow field, and it should be able to accurately capture the fluid temperature. At the same time, the sound velocity temperature coefficient of the working fluid is pre-stored. This coefficient is an inherent physical parameter of the working fluid, characterizing the amount of sound velocity change caused by a unit temperature change. It can be obtained through existing literature or previous experimental measurements. Based on the real-time temperature observation value and the sound velocity temperature coefficient, the influence of temperature change on the equivalent sound velocity is calculated, and then temperature compensation is applied to the preliminary pressure estimate. The correction formula is as follows:

[0093] ;

[0094] This formula is derived from the definition of the sound velocity temperature coefficient. Temperature change is converted into pressure correction through the sound velocity temperature coefficient. The correction is then added to the preliminary pressure estimate to obtain the actual process pressure. In the formula, P represents the measured process pressure value. The value represents the preliminary pressure estimate; k represents the sound velocity temperature coefficient of the working fluid, in kPa / ℃; T represents the real-time temperature observation value, in ℃. The constant temperature during calibration in step 31 is expressed in °C. This temperature compensation calculation effectively eliminates the interference of temperature fluctuations on pressure measurement, ensuring the accuracy of process pressure measurements.

[0095] The construction of the mapping model also includes the following steps:

[0096] Step 33: Under multiple preset operating conditions, measure the local pressure at multiple different axial positions inside the micropillar, and simultaneously acquire the equivalent sound velocity and process pressure measurements at each operating condition. Use the equivalent sound velocity and process pressure measurements at each operating condition as coordinates, and the corresponding multiple local pressures as pressure field vectors to form a set of probe points. The specific operation is as follows:

[0097] The mapping model needs to accurately reflect the correlation between equivalent sound velocity, measured process pressure, and local pressure inside the microcolumn. This correlation should be consistent under different process conditions. Therefore, comprehensive sample data needs to be collected through calibration experiments at multiple preset operating points. The preset operating points need to cover all typical operating states of the microcolumn separation process, including different flow rate setpoints, target pressure values, and sample loads, ensuring that the probe point set can cover the parameter combinations that may occur in the actual process. At each preset operating point, a micro pressure sensing array, which can consist of multiple non-invasive micro pressure sensors, can be used to measure the local pressure at multiple different axial positions inside the microcolumn. These axial positions need to be evenly distributed, covering the inlet, middle, and outlet sections of the microcolumn, to fully characterize the spatial distribution characteristics of the internal pressure field. Simultaneously, synchronously... The equivalent sound velocity of the working fluid at this operating point can be calculated in real time using the method in step 22, and the process pressure measurement value can be obtained in real time using the method in step 32. This ensures that the acquisition of the three sets of data—equivalent sound velocity, process pressure measurement value, and local pressure at multiple locations—is strictly synchronized in time. For the same operating condition, the equivalent sound velocity and process pressure measurement value of each operating point are taken as a point on a two-dimensional coordinate plane. The local pressures at multiple different axial positions measured at this operating point are combined into a pressure field vector. The coordinates corresponding to each operating point and the pressure field vector together constitute a probe data point. The probe data points corresponding to all preset operating points are summarized to form a complete probe point set. This point set is the sample data for constructing the mapping model, and its completeness and accuracy directly determine the prediction accuracy of the model.

[0098] Step 34: Triangulate all points in the probe point set on a two-dimensional plane with equivalent sound velocity and process pressure measurement values ​​as axes to form a triangular mesh. Determine the target triangle to which each point belongs in the triangular mesh based on the real-time input equivalent sound velocity and process pressure measurement values. Calculate the centroid coordinates of the real-time input point relative to the three vertices of the target triangle. Weighted summation of the pressure field vectors stored at the three vertices based on the centroid coordinates yields the predicted value of the internal pressure of the micropillar. The specific operations are as follows:

[0099] First, all points in the probe point set formed in step 33 are projected onto a two-dimensional plane with the equivalent sound velocity as the vertical axis and the process pressure measurement value as the horizontal axis. Triangulation is then performed on all probe points on this plane. The triangulation must follow the optimal triangulation principle, ensuring that the interior angles of each triangle are moderate and avoiding extreme obtuse or acute angles. Those skilled in the art can use common triangulation algorithms, such as the Delaunay triangulation algorithm, to complete this operation. After triangulation, a triangular mesh covering the entire probe point set distribution range is formed. The three vertices of each triangle are valid data points in the probe point set. During real-time model operation, the real-time acquired equivalent sound velocity and process pressure measurement value are compared... Given an input point on a two-dimensional plane, a spatial position determination algorithm identifies the target triangle to which the input point belongs, i.e., which triangle the input point is located inside or on. Then, the centroid coordinates of the input point relative to the three vertices of the target triangle are calculated. The centroid coordinates characterize the relative position of the input point within the triangle; the sum of the three coordinate values ​​is 1, and each coordinate value is non-negative. Based on the centroid coordinates, the pressure field vectors stored at the three vertices of the target triangle are weighted and superimposed, with the weights being the corresponding centroid coordinate values. The resulting pressure field vector is the predicted pressure value at each axial position inside the micro-pillar. This predicted value accurately reflects the internal pressure distribution state corresponding to the input parameters. The formula for the weighted superposition is:

[0100] ;

[0101] This formula is derived from the principle of triangle interpolation, using the centroid coordinates. , , This ensures that the predicted value at the input point is within a reasonable range of the pressure field vectors at the three vertices; in the formula, This represents the predicted pressure field vector. , , This represents the centroid coordinates of the input point relative to the three vertices A, B, and C of the target triangle. , , These represent the pressure field vectors corresponding to the three vertices. Through this process, the mapping model is constructed and real-time pressure prediction is completed, ensuring that the model output can accurately reflect the actual pressure state inside the micropillar.

[0102] In a preferred embodiment of the present invention, step 4 is further included, in which the real-time obtained equivalent sound velocity and process pressure measurement values ​​are input into the mapping model to obtain the pressure estimate inside the micropillar. The specific operation is as follows:

[0103] During the real-time monitoring of the radioactive column separation process, the equivalent sound velocity of the working fluid is continuously acquired using the method in step 22, while the process pressure measurement value of the microcolumn is simultaneously acquired using the method in step 32. This ensures that the acquisition times of the two sets of parameters are strictly synchronized, corresponding to the same working state of the microcolumn. These two sets of real-time parameters are used as input variables and substituted into the mapping model constructed in steps 33 and 34. This model has established the correlation between the equivalent sound velocity, the process pressure measurement value, and the local pressure inside the microcolumn through multi-condition calibration, enabling accurate mapping from input parameters to internal pressure. To ensure the reliability and adaptability of the pressure estimate, an appropriate solution strategy needs to be selected through geometric position analysis in step 41, and then the corresponding solver is called in step 42 to complete the data processing. Finally, the pressure estimate value that can truly reflect the pressure state at each axial position inside the microcolumn is output. The entire process relies on a stable signal acquisition module and data processing unit to ensure the real-time performance of parameter transmission and solution, meeting the dynamic response requirements of process monitoring.

[0104] Step 4 also includes the following sub-steps:

[0105] Step 41: Analyze the geometric position of the real-time input equivalent sound velocity and process pressure measurement values ​​in the historical calibration data space composed of probe point sets. Based on their distance from the calibration points and their relative positional relationship with the convex hull formed by the calibration points, select the corresponding solution strategy. The specific operation is as follows:

[0106] By analyzing the geometric position of real-time input parameters in the historical calibration data space, suitable solution strategies are selected to ensure the accuracy and rationality of pressure estimation. The historical calibration data space consists of the probe point set formed in step 33. This space has the equivalent sound velocity as the vertical axis and the process pressure measurement value as the horizontal axis. All probe points are distributed in this two-dimensional plane. First, using commonly used convex hull construction algorithms in this field, such as the Graham scan method or the Andrew algorithm, all calibration points in the probe point set are connected to form a minimum convex polygon, i.e., the calibration point convex hull. This convex hull can define the effective coverage of the historical calibration data. Subsequently, the two-dimensional coordinate points corresponding to the real-time input equivalent sound velocity and the process pressure measurement value are projected into this historical calibration data space, and their geometry is analyzed. Location characteristics are considered. On one hand, the straight-line distances from the input point to all calibration points are calculated, typically using Euclidean distance for quantization. These distance values ​​are used to determine the proximity of the input point to historical calibration data. On the other hand, the relative position of the input point to the convex hull is determined by the positional relationship between the point and the convex hull edges, clarifying whether the input point is inside, on the boundary, or outside the convex hull. Combining distance characteristics and relative positional relationships, a corresponding solution strategy is selected: if the input point is inside the convex hull and close to neighboring calibration points, an interpolation-based solution strategy is chosen; if the input point is on the boundary of the convex hull, a boundary-fitting solution strategy is chosen; if the input point is outside the convex hull, an extrapolation-based solution strategy is chosen. This ensures that the solution strategy matches the location characteristics of the input point, avoiding distortion of the pressure estimate due to inappropriate strategies.

[0107] Step 42: Based on the selected solution strategy, the corresponding solver is invoked to process the real-time input equivalent sound velocity and process pressure measurement values ​​to obtain the pressure estimate inside the micro-pillar. The specific operation is as follows:

[0108] Based on the solution strategy selected in step 41, the corresponding solver is called to process the real-time input parameters, ultimately obtaining the pressure estimate inside the micropillar. The solver needs to be pre-integrated into the data processing unit. Different solution strategies correspond to different solver modules. Each module is developed based on the triangular mesh and probe point set data constructed in step 34. If an interpolation-type solution strategy is selected in step 41, it means that the input point is within the effective coverage range of historical calibration data. In this case, the triangulation interpolation solver is called. This solver first determines that the target triangle to which the input point belongs matches the triangular mesh in step 34, and then calculates the centroid coordinates of the input point relative to the three vertices of the target triangle. The pressure field vectors stored at the three vertices are weighted and superimposed using the centroid coordinates to output the pressure estimate. This process is consistent with the pressure prediction logic in step 34, ensuring the continuity and accuracy of the interpolation results. If a boundary adaptation-type solution strategy is selected in step 41, it means that the input point is located on the convex hull boundary. In this case, the edge adaptation solution strategy is called. The boundary correction interpolation solver prioritizes the two endpoints of the convex hull where the input point is located and the nearby calibration points as references. It performs boundary constraint correction on the pressure field vector before completing the interpolation calculation to avoid deviations in the boundary point estimates. If the extrapolation-type solution strategy is selected in step 41, it means that the input point exceeds the coverage of historical calibration data. In this case, the nearest neighbor extrapolation solver is called. This solver selects multiple calibration points closest to the input point, such as usually 3 to 5, analyzes the pressure field vector change trend of these calibration points, and extrapolates the pressure field vector corresponding to the input point, i.e., the pressure estimate, based on the trend. Reasonable constraints are set during the extrapolation process to avoid the extrapolation result exceeding the reasonable pressure range of the process. All solver processes are completed automatically without manual intervention. After the solution is completed, the pressure estimate values ​​of each axial position inside the micro-pillar are directly output. Those skilled in the art can realize the automatic calling of the solver and data processing through preset program instructions.

[0109] In a preferred embodiment of the present invention, step 5 is further included: comparing the pressure estimate with the target pressure value to obtain the pressure deviation; adjusting the frequency of pressure oscillation based on the relationship between the characteristics of the pressure deviation and the oscillation response characteristics of the inlet response amplitude and the outlet response amplitude, as follows:

[0110] During real-time monitoring, the estimated internal pressure value of the microcolumn output in step 4 is continuously acquired, while the preset target pressure value is retrieved. This target pressure value is pre-set according to the requirements of the radioactive separation process and adapts to the pressure parameter requirements of different sample separations. The estimated pressure value and the target pressure value are compared and calculated at each time step to obtain the pressure deviation reflecting the difference between the two. Since the pressure deviation may be caused by different physical sources such as flow field interference and changes in packing state, it is not possible to accurately locate the cause and implement effective control based solely on the deviation value. Therefore, it is necessary to combine the oscillation response characteristics of the inlet response amplitude and outlet response amplitude extracted in step 1 to establish the correlation between the deviation characteristics and the oscillation characteristics. By analyzing this correlation, the effect of the current pressure oscillation frequency on the deviation suppression is determined, and then the pressure oscillation frequency is adjusted accordingly to make the internal pressure of the microcolumn quickly return to the target range, while avoiding the adverse effects of frequency adjustment on the separation process. The entire process relies on the real-time data transmission and analysis unit to achieve closed-loop control, ensuring the timeliness and accuracy of the response.

[0111] The process of comparing the estimated pressure value with the target pressure value to obtain the pressure deviation also includes the following steps:

[0112] Step 51: Based on the continuously acquired pressure estimates and target pressure values, generate a time series of pressure deviations. Extract morphological feature values ​​from the time series, including the upward slope, downward slope, duration of steady-state offset plateau, and number of zero crossings. The specific operations are as follows:

[0113] Pressure estimates and target pressure values ​​are continuously collected at fixed time intervals. These intervals must match the pressure oscillation period and the real-time requirements of process monitoring to ensure complete capture of the dynamic changes in pressure deviation. The pressure deviation value at any given moment is obtained by subtracting the target pressure value from the estimated pressure value. The pressure deviation values ​​from multiple consecutive moments are then arranged chronologically to form a time series of pressure deviations. This series visually reflects the trend of pressure deviation over time. Based on this time series, morphological feature values ​​are extracted using signal analysis and data processing techniques. These feature values ​​are used to quantify the dynamic changes in deviation. The rising slope refers to the rate of increase in pressure deviation over time. The rate of change of a segment is calculated by linearly fitting the deviation data of that segment; the rate of change of the deviation over time is the rate of change of the decreasing segment, obtained using the same linear fitting method; the duration of the steady-state offset platform refers to the duration during which the pressure deviation is stable within a certain fixed range and the fluctuation amplitude is less than a preset threshold, determined by identifying continuous periods in the time series that meet the fluctuation conditions; the number of zero crossings refers to the number of times the pressure deviation value crosses zero potential from a positive value to a negative value or from a negative value to a positive value, obtained by statistically analyzing the deviation data in the time series; the extraction of all morphological feature values ​​is automatically completed by a preset algorithm, and the extraction results are used as input parameters for deviation root cause analysis.

[0114] Step 52: Perform a linkage analysis between the morphological characteristic values ​​and the oscillation response attenuation mode characteristics from the inlet response amplitude and the outlet response amplitude. Based on the pre-stored judgment logic, separate the pressure deviation into deviation components corresponding to different physical sources. The specific operation is as follows:

[0115] First, the oscillation response decay mode features of the inlet and outlet response amplitudes extracted in step 1 are simultaneously acquired. These features refer to the decay law of the inlet and outlet response amplitudes over time at the current pressure oscillation frequency, including quantitative parameters such as decay rate, steady-state decay value, and decay fluctuation amplitude. This is obtained through trend analysis and data fitting of the response amplitude time series. Then, the pressure deviation morphology feature values ​​extracted in step 51 are linked with the aforementioned oscillation response decay mode features for analysis. This linkage analysis establishes a corresponding relationship between the two types of feature parameters; that is, different morphologies of pressure deviation correspond to specific oscillation response decay modes. (Pre-stored...) The judgment logic is based on the previous multi-condition calibration experiments. This logic includes different physical causes, such as local blockage of microchannels, changes in fluid viscosity, and saturation of packing adsorption, and the corresponding combination of deviation morphological characteristics and oscillation decay mode characteristics. Those skilled in the art can obtain sufficient sample data through the previous calibration experiments to build a complete judgment logic library. According to the pre-stored judgment logic, the current linkage analysis results are matched to separate the pressure deviation into deviation components corresponding to different physical causes. At the same time, the dominant deviation component with the greatest impact on the overall deviation is identified to ensure that subsequent frequency adjustments can be targeted at the deviation causes and improve the accuracy of control.

[0116] The process of adjusting the frequency of pressure oscillation based on the relationship between the characteristics of pressure deviation and the oscillation response characteristics of the inlet and outlet response amplitudes also includes the following steps:

[0117] Step 53: Based on the identification of the dominant deviation component and the fluctuation characteristics of the outlet response amplitude at the current pressure oscillation frequency, evaluate the suppression efficacy of the current frequency on the dominant component. The specific operation is as follows:

[0118] Based on the separation results of step 52, the dominant component of the pressure deviation is clearly identified, that is, the deviation component with the greatest impact and highest proportion on the stability of the microcolumn separation process. Simultaneously, the quantitative characteristic parameters of this dominant component, such as deviation amplitude and fluctuation frequency, are recorded. Subsequently, the fluctuation characteristics of the outlet response amplitude under the current pressure oscillation frequency are focused on. These characteristics include the fluctuation amplitude, fluctuation frequency, and synchronicity of the fluctuation with the dominant deviation component. This is obtained through real-time analysis of the outlet response amplitude time series. The smaller the fluctuation amplitude and the lower the synchronicity with the dominant deviation component, the better the suppression effect of the current frequency on the dominant deviation component. Based on the above analysis, an evaluation index system for suppression effectiveness is constructed. The evaluation indexes include the amplitude decay rate of the dominant deviation component, the stability of deviation fluctuations, and the stability of the outlet response amplitude. By quantifying and weighting each index, the suppression effectiveness value of the current frequency on the dominant deviation component is obtained. The higher the suppression effectiveness value, the more effectively the current frequency can suppress the dominant deviation component and maintain the stability of the microcolumn pressure. If the suppression effectiveness value is lower than a preset threshold, it indicates that the current frequency is not suitable and frequency adjustment is required.

[0119] Step 54: Based on the evaluation results of the suppression effectiveness, initiate and execute a directional frequency scan starting from the current frequency. During the scan, calculate the instantaneous suppression effectiveness at each trial frequency in real time, and lock the frequency that achieves the optimal instantaneous suppression effectiveness as the new pressure oscillation frequency. The specific operation is as follows:

[0120] Based on the suppression effectiveness evaluation results of step 53, if the current frequency suppression effectiveness does not meet the preset requirements, the directional frequency scanning process is initiated. The scan starts from the current pressure oscillation frequency and proceeds in the frequency adjustment direction according to the preset frequency step size. The frequency step size needs to be set in conjunction with the frequency response characteristics of the microcolumn to ensure that the optimal frequency point can be accurately captured during the scan, while avoiding the optimal frequency being missed due to an excessively large step size or the scan efficiency being affected by an excessively small step size. During the frequency scan, each time a test frequency is switched, the pressure oscillation is maintained at that frequency and injected stably. After the pressure state and response amplitude signal inside the microcolumn reach a steady state, the characteristic parameters of the dominant deviation component and the fluctuation characteristics of the outlet response amplitude at that test frequency are acquired in real time, according to the suppression effectiveness evaluation method of step 53. Calculate the instantaneous suppression effectiveness value corresponding to each test frequency; continuously advance the frequency scan, and synchronously record the instantaneous suppression effectiveness value at each test frequency. When the scan reaches a certain test frequency, its instantaneous suppression effectiveness value reaches the preset optimal condition, that is, the suppression effectiveness value is the maximum, and the amplitude of the dominant deviation component is lower than the process allowable threshold and the outlet response amplitude fluctuation is stable. Then, lock the test frequency as the new pressure oscillation frequency. After the frequency is locked, automatically switch the injection frequency of the pressure oscillation to the optimal frequency, and continuously monitor the stability of the suppression effectiveness to ensure that the new frequency can stably suppress the dominant deviation component in the long term and maintain the internal pressure of the microcolumn within the target range. Those skilled in the art can realize the automated operation of frequency scanning, effectiveness calculation and frequency locking through preset programs.

[0121] In a preferred embodiment of the present invention, step 6 is further included: obtaining the energy attenuation ratio of the pressure oscillation from the inlet to the outlet at multiple frequencies to form an attenuation spectrum; comparing the attenuation spectrum with a pre-stored reference attenuation spectrum, and evaluating the state of the microcolumn based on the comparison results. The specific operation is as follows:

[0122] During the operation of the microcolumn separation process, periodic or on-demand status assessments are required. The assessment cycle can be set based on the sample throughput and previous process stability data. During the assessment, pressure oscillations covering a preset frequency range are injected into the microcolumn using a pressure oscillation generator. This pressure oscillation generator consists of a pressure generating unit, a frequency control unit, and a pressure regulating module. The pressure generating unit includes a piezoelectric micropump, which is composed of a piezoelectric ceramic plate and a flexible diaphragm. The flexible diaphragm is made of polydimethylsiloxane (PDMS) material. This structure enables the precise generation of minute pressure oscillations. When the piezoelectric ceramic plate undergoes a minute deformation driven by the electrical signal output from the frequency control unit, it drives the flexible diaphragm to produce periodic expansion and contraction movements, thereby injecting frequency-controllable pressure oscillations into the microcolumn. The amplitude range of the pressure oscillations is 0.1 kPa to 1 kPa, and the frequency regulation range is 1 Hz to 100 Hz, which can meet the different frequency excitation requirements of the microcolumn separation process. The frequency control unit internally stores parameters such as the characteristic resonant frequency selected in step 11 and the predetermined pressure oscillation frequency in step 1, enabling dual functions of continuous frequency variation and fixed-point frequency locking. This unit can receive external commands to control the driving frequency of the piezoelectric micropump, with a frequency control accuracy of ±0.1Hz and a frequency stability of ±0.05Hz / 10min, ensuring that the pressure oscillation frequency is completely consistent with the characteristic resonant frequency selected in step 11 and the predetermined pressure oscillation frequency in step 1. The pressure regulation module is a miniature proportional regulating valve with an electromagnetic drive structure, enabling precise adjustment of the injection pressure amplitude. This module can adjust the pressure oscillation amplitude in real time according to the pressure deviation control requirements in step 5, with an adjustment accuracy of ±0.01kPa, ensuring that the pressure oscillation amplitude is compatible with the internal process pressure of the micro-column, avoiding damage to the flow field balance inside the micro-column due to excessive pressure amplitude, or failure to effectively excite the fluid oscillation response due to insufficient pressure amplitude.

[0123] The preset frequency range must include the characteristic resonant frequency selected in step 11 and the common response frequencies of the micropillar. The frequency coverage range is typically from 1Hz to 100Hz. For each injection frequency, the response amplitude signals at the inlet and outlet are simultaneously acquired using a deformation detection device. The energy attenuation ratio of the pressure oscillation propagating from the inlet to the outlet is calculated. The energy attenuation ratio is determined based on the square ratio of the inlet and outlet response amplitudes, since signal energy is proportional to the square of the amplitude. The energy attenuation ratios corresponding to all frequencies are arranged in chronological order to form an attenuation spectrum reflecting the energy attenuation characteristics of the micropillar. This spectrum can intuitively present the attenuation law of pressure oscillation energy within the micropillar at different frequencies, and the attenuation law is related to the micropillar's energy attenuation characteristics. The core performance parameters, such as packing condition and flow channel integrity, are directly related. Subsequently, a pre-stored reference attenuation spectrum is retrieved. This reference attenuation spectrum is obtained by oscillation injection and data acquisition within the same frequency range when the microcolumn is in a brand-new or normal working state. It has a unified calibration standard. The real-time acquired attenuation spectrum is comprehensively compared with the reference attenuation spectrum. The comparison includes the consistency of the overall shape and the deviation of the attenuation ratio at characteristic frequency points. Through comparative analysis, changes in the microcolumn condition are identified, especially potential problems such as packing deterioration and slight blockage of the flow channel. The entire process relies on a precise frequency control device and a high-sensitivity deformation detection device to ensure the accuracy and repeatability of the attenuation spectrum acquisition.

[0124] The evaluation of the microcolumn's condition based on the comparison results includes:

[0125] Step 61: Analyze the morphological characteristics of the pre-stored reference attenuation spectrum to identify a set of characteristic frequency bands associated with the packing state. The specific operation is as follows:

[0126] Retrieve the pre-stored baseline attenuation spectrum, which contains energy attenuation ratio data corresponding to multiple frequency points. A deep morphological feature analysis is required, focusing on the peak and valley positions, abrupt changes in attenuation rate, and attenuation stability within characteristic frequency ranges. These features are all related to the physical state of the packing material inside the micropillar. For example, when the packing particles are well-uniform, the attenuation rate remains stable within a specific frequency range, while the attenuation characteristics change significantly when the packing material agglomerates or deteriorates. By comparing multiple sets of baseline attenuation spectra of the micropillar in virgin, slightly deteriorated, and severely deteriorated states, the frequency range where the attenuation characteristics change most significantly with changes in packing condition is identified. This range is the characteristic frequency band associated with the packing condition. Locking a characteristic frequency band requires meeting two conditions: first, the sensitivity of the energy attenuation ratio in this frequency band to changes in the state of the packing material must be higher than that of other frequency bands, with a change range of not less than 10%; second, the attenuation characteristics of this frequency band must not be affected by fluctuations in process parameters such as fluid flow rate and temperature, or the interference can be eliminated through prior calibration. The locking process can be completed automatically through feature extraction algorithms. Those skilled in the art can use common algorithms such as spectral peak detection and attenuation rate gradient analysis, combined with prior multi-condition calibration data, to determine one or more characteristic frequency bands.

[0127] Step 62: For the attenuation spectrum acquired in real time, calculate the energy offset relative to the reference value based on the characteristic frequency bands locked in step 61, and obtain the filler state change index by combining the offsets of each frequency band. The specific operation is as follows:

[0128] For each set of characteristic frequency bands locked in step 61, extract the energy attenuation ratio data corresponding to all frequency points within that band in the real-time attenuation spectrum. Simultaneously, extract the baseline energy attenuation ratio value of the corresponding frequency points in the same characteristic frequency band from the baseline attenuation spectrum. For each characteristic frequency band, calculate the average energy attenuation ratio of all frequency points within that band, using this as the overall energy attenuation characteristic value for that band. The difference between the real-time characteristic value and the baseline characteristic value is the energy offset of that band. A positive offset indicates that the real-time attenuation ratio is greater than the baseline value, meaning the energy attenuation is more significant; a negative offset indicates that the real-time attenuation ratio is less than the baseline value. To comprehensively reflect the offset of all characteristic frequency bands, the energy offsets of each band need to be weighted and superimposed to obtain the packing state change index. The weight allocation is determined based on the sensitivity of each characteristic frequency band in relation to the packing state; the higher the sensitivity, the greater the weight of the frequency band. The total weight is 1. The formula for weighted superposition is:

[0129] ;

[0130] This formula is derived from a multi-index comprehensive evaluation logic, highlighting the influence of high-sensitivity frequency bands through weight allocation to ensure that the index accurately reflects changes in the packing condition; in the formula... The index represents the change in packing condition; n represents the number of locked characteristic frequency bands. The weight representing the i-th feature frequency band. The value of the packing state change index directly reflects the degree of packing degradation. The larger the index, the more severe the degradation. Those skilled in the art can determine the index threshold range corresponding to different degradation levels through prior calibration experiments.

[0131] Step 63: Based on the time series of the packing state change index, analyze its rate of change and acceleration to identify the dynamic mode of packing degradation, and extrapolate to estimate the remaining healthy life. The specific operation is as follows:

[0132] The packing condition change index is continuously collected according to a fixed evaluation cycle. The index values ​​obtained from each evaluation are arranged in chronological order to form a time series of the packing condition change index. The length of the time series must be no less than 5 evaluation cycles to ensure complete capture of the degradation trend. Based on this time series, the rate of change and acceleration of change of the index are calculated using linear or nonlinear fitting algorithms. The rate of change represents the average change of the index per unit time, and the acceleration of change represents the trend of the rate of change, i.e., whether the degradation rate is accelerating, slowing down, or remaining stable. Based on the combined characteristics of the rate of change and acceleration, the dynamic mode of packing degradation is identified. Common modes include linear degradation mode, accelerated degradation mode, and decelerated degradation mode, such as a constant rate of change. Furthermore, linear degradation occurs when the acceleration approaches zero, while accelerated degradation occurs when the rate of change gradually increases and the acceleration is positive. After the dynamic degradation pattern is identified, the remaining healthy life is estimated based on the time series data and the identified degradation pattern using a trend extrapolation algorithm. The extrapolation process terminates with a pre-stored degradation failure threshold. That is, when the packing state change index reaches this threshold, it is determined that the packing can no longer meet the separation process requirements and needs to be replaced or maintained. The impact of process fluctuations on index changes needs to be considered during extrapolation. Interference is offset by introducing a correction coefficient, which is determined based on the correlation data between previous process fluctuations and index changes. Those skilled in the art can use common algorithms such as linear extrapolation and exponential extrapolation to complete the life estimation, ensuring the reliability and reference value of the estimation results.

[0133] Step 64: Combining the packing deterioration dynamic mode with process pressure measurements and flow rate settings, assess the risk level of the micro-column based on the preset decision matrix and generate maintenance recommendations. The specific operation is as follows:

[0134] The dynamic degradation pattern of the packing identified in step 63, the real-time process pressure measurement value obtained in step 32, and the current process set flow rate value are combined to determine the operational risk of the microcolumn. The degradation dynamic pattern reflects the long-term performance degradation trend, the process pressure measurement value reflects the current pressure state, and the flow rate set value reflects the process load. The preset decision matrix is ​​constructed based on previous multi-condition calibration experiments. The rows and columns of the matrix correspond to the value ranges of different parameters, and the elements within the matrix represent the corresponding risk level and maintenance recommendations. During the construction of the decision matrix, each parameter needs to be divided into multiple intervals. For example, the process pressure measurement value is divided into normal, slightly exceeding, and severely exceeding intervals; the packing degradation dynamic pattern is divided into mild, moderate, and severe degradation intervals; and the flow rate set value... The speed setpoint is divided into low-load, normal-load, and high-load ranges. The three parameters acquired in real time are matched to the corresponding ranges in the decision matrix. By locating the corresponding element in the matrix, the current risk level of the micro-column can be obtained. The risk level is usually divided into three levels: low risk, medium risk, and high risk. At the same time, based on the maintenance suggestion item corresponding to the element, fine-tuning is made in combination with the current process operation to generate targeted maintenance suggestions. For example, it is recommended to increase the monitoring frequency when the risk is low, to reduce the flow rate or to perform online cleaning when the risk is medium, and to stop the process immediately and replace the packing when the risk is high. The construction of the decision matrix and the generation logic of maintenance suggestions can be completed by those skilled in the art through sufficient experimental data of the operating conditions in the early stage to ensure the accuracy of the risk assessment and the feasibility of the maintenance suggestions.

[0135] Example 2

[0136] Please see Figure 2 Based on Example 1, this embodiment provides a pressure and temperature coupled monitoring system for radioactive column separation, including:

[0137] The signal extraction module is used to inject pressure oscillations with a predetermined frequency into the microcolumn; detect the deformation signals generated at the inlet and outlet of the microcolumn due to internal pressure changes; and extract the inlet response amplitude, outlet response amplitude, and phase difference between the inlet and outlet responses corresponding to the pressure oscillation frequency from the deformation signals.

[0138] The sound velocity calculation module calculates the propagation time of pressure oscillation from inlet to outlet based on phase difference; and determines the equivalent sound velocity of the fluid inside the micro-pillar based on the propagation time and the inherent structural parameters of the micro-pillar.

[0139] The model building module obtains the measured process pressure values ​​of the micropillar; by combining the equivalent sound velocity and the measured process pressure values, a mapping model is constructed with the equivalent sound velocity and the measured process pressure values ​​as inputs and the internal pressure of the micropillar as the output.

[0140] The pressure estimation module is used to input the real-time obtained equivalent sound velocity and process pressure measurement values ​​into the mapping model to obtain the pressure estimate inside the micropillar;

[0141] The frequency adjustment module is used to compare the pressure estimate with the target pressure value to obtain the pressure deviation; and to adjust the frequency of pressure oscillation based on the relationship between the characteristics of the pressure deviation and the oscillation response characteristics of the inlet response amplitude and the outlet response amplitude.

[0142] The state assessment module is used to obtain the energy attenuation ratio of pressure oscillation from inlet to outlet at multiple frequencies to form an attenuation spectrum; the attenuation spectrum is compared with the pre-stored benchmark attenuation spectrum, and the state of the microcolumn is assessed based on the comparison results.

[0143] The above description is merely a preferred embodiment of the present invention; however, the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and its improved concepts, should be covered within the scope of protection of the present invention.

Claims

1. A method for coupled pressure and temperature monitoring in radioactive column separation, characterized in that, include: Step 1: Inject pressure oscillations with a predetermined frequency into the micropillar; Deformation signals generated at the inlet and outlet of the microcolumn due to changes in internal pressure were detected. From the deformation signal, extract the inlet response amplitude, outlet response amplitude, and phase difference between the inlet and outlet responses corresponding to the pressure oscillation frequency, including: Step 11: Inject a pressure wave with a continuously varying frequency into the microcolumn, simultaneously detect the deformation signals at the inlet and outlet, analyze the frequency response spectrum based on the deformation signal at the outlet, and select a characteristic resonance frequency from the frequency response spectrum. Step 12: Inject phase-locked pressure oscillation into the micropillar at the characteristic resonant frequency. After reaching a steady state, detect the first deformation signal at the inlet and the second deformation signal at the outlet respectively. Subtract the first deformation signal from the second deformation signal to obtain the differential signal. Step 13: Perform zero-crossing detection on the differential signal to determine its real-time frequency, and perform analog multiplication operations on the differential signal and the reference sine wave and reference cosine wave of the same frequency, and then perform low-pass filtering. Calculate the amplitude and phase of the differential signal based on the filtering results to obtain the input response amplitude, the output response amplitude, and the phase difference between the two. Step 2: Calculate the propagation time of the pressure oscillation from the inlet to the outlet based on the phase difference; determine the equivalent sound velocity of the fluid inside the micro-pillar based on the propagation time and the inherent structural parameters of the micro-pillar, including: Step 21: Switch the fluid inside the micro-column to a reference fluid with a known sound velocity-temperature relationship and maintain it at a known temperature. Obtain the phase difference between the inlet and outlet based on the reference fluid. Calculate the propagation time based on the phase difference and oscillation frequency. Combine the known sound velocity of the reference fluid at a known temperature to calculate the equivalent sound propagation path length of the micro-column in reverse. Step 22: Switch the fluid inside the micro-column to the working fluid, obtain the real-time phase difference between the inlet and outlet based on the working fluid, calculate the real-time propagation time based on the phase difference, and determine the equivalent sound velocity of the working fluid using the equivalent sound propagation path length and the real-time propagation time. Step 3: Obtain the process pressure measurement value of the micro-column; combine the equivalent sound velocity and process pressure measurement value to construct a mapping model with the equivalent sound velocity and process pressure measurement value as input and the internal pressure of the micro-column as output. Step 4: Input the real-time equivalent sound velocity and process pressure measurement values ​​into the mapping model to obtain the pressure estimate inside the micropillar; Step 5: Compare the pressure estimate with the target pressure value to obtain the pressure deviation; adjust the frequency of pressure oscillation based on the relationship between the characteristics of the pressure deviation and the oscillation response characteristics of the inlet and outlet response amplitudes. Step 6: At multiple frequencies, obtain the energy attenuation ratio of the pressure oscillation from the inlet to the outlet to form an attenuation spectrum; compare the attenuation spectrum with the pre-stored reference attenuation spectrum, and evaluate the state of the microcolumn based on the comparison results.

2. The method for pressure and temperature coupling monitoring in radioactive column separation according to claim 1, characterized in that, Obtain process pressure measurements for the micropillars, including: Step 31: Control the pressure of the micro-column at multiple known reference pressure points, and under constant temperature conditions, obtain the equivalent sound velocity of the working fluid at each reference pressure point. Based on the correspondence between the equivalent sound velocity and the reference pressure point, generate a sound velocity-pressure correlation calibration curve. Step 32: Input the equivalent sound velocity of the working fluid obtained in real time into the sound velocity-pressure correlation calibration curve to obtain a preliminary pressure estimate, and obtain the real-time temperature observation value. Use the real-time temperature observation value and the pre-stored sound velocity-temperature coefficient of the working fluid to perform temperature compensation calculation on the preliminary pressure estimate value to obtain the process pressure measurement value.

3. The method for pressure and temperature coupling monitoring in radioactive column separation according to claim 2, characterized in that, The construction of the mapping model includes: Step 33: Under multiple preset working conditions, measure the local pressure at multiple different axial positions inside the micro-pillar, and simultaneously acquire the equivalent sound velocity and process pressure measurement values ​​at each working condition. Use the equivalent sound velocity and process pressure measurement values ​​at each working condition as coordinates, and the corresponding multiple local pressures as pressure field vectors to form a set of probe points. Step 34: Triangulate all points in the probe point set on a two-dimensional plane with equivalent sound velocity and process pressure measurement values ​​as axes to form a triangular mesh. Determine the target triangle to which the point belongs in the triangular mesh based on the real-time input equivalent sound velocity and process pressure measurement values. Calculate the centroid coordinates of the real-time input point relative to the three vertices of the target triangle. Based on the centroid coordinates, perform a weighted superposition of the pressure field vectors stored at the three vertices to obtain the predicted value of the internal pressure of the micropillar.

4. The method for pressure and temperature coupling monitoring in radioactive column separation according to claim 3, characterized in that, The real-time equivalent sound velocity and process pressure measurements are input into the mapping model to obtain the pressure estimate inside the micropillar, including: Step 41: Analyze the geometric position of the real-time input equivalent sound velocity and process pressure measurement values ​​in the historical calibration data space composed of probe point sets. Based on its distance from the calibration point and its relative positional relationship with the convex hull formed by the calibration point, select the corresponding solution strategy. Step 42: Based on the selected solution strategy, call the corresponding solver to process the real-time input equivalent sound velocity and process pressure measurement values ​​to obtain the pressure estimate inside the micro-column.

5. The method for pressure and temperature coupling monitoring in radioactive column separation according to claim 1, characterized in that, The pressure estimate is compared with the target pressure value to obtain the pressure deviation, including: Step 51: Based on the continuously acquired pressure estimate and target pressure value, generate a time series of pressure deviation, and extract morphological feature values ​​including the rising slope, falling slope, duration of steady-state offset plateau, and number of zero crossings from the time series. Step 52: Perform a linkage analysis between the morphological characteristic value and the oscillation response attenuation mode characteristics from the inlet response amplitude and the outlet response amplitude, and separate the pressure deviation into deviation components corresponding to different physical sources according to the pre-stored judgment logic.

6. The method for pressure and temperature coupling monitoring in radioactive column separation according to claim 5, characterized in that, Based on the relationship between the characteristics of the pressure deviation and the oscillation response characteristics of the inlet and outlet response amplitudes, the frequency of pressure oscillation is adjusted, including: Step 53: Based on the identification of the dominant component of the deviation and the fluctuation characteristics of the outlet response amplitude under the current pressure oscillation frequency, evaluate the suppression efficacy of the current frequency on the dominant component. Step 54: Based on the evaluation results of the suppression effectiveness, start and execute a directional frequency scan starting from the current frequency. During the scan, calculate the instantaneous suppression effectiveness at each trial frequency in real time, and lock the frequency that makes the instantaneous suppression effectiveness reach the optimal condition as the new pressure oscillation frequency.

7. The method for pressure and temperature coupling monitoring in radioactive column separation according to claim 1, characterized in that, The condition of the micropillars was evaluated based on the comparison results, including: Step 61: Analyze the morphological characteristics of the pre-stored reference attenuation spectrum to identify a set of characteristic frequency bands associated with the filler state; Step 62: For the attenuation spectrum acquired in real time, calculate the offset of its energy relative to the reference value based on the characteristic frequency band locked in step 61, and obtain the filler state change index by combining the offsets of each frequency band. Step 63: Based on the time series of the packing state change index, analyze its change rate and acceleration to identify the dynamic mode of packing degradation, and extrapolate to predict the remaining healthy life. Step 64: Combining the packing deterioration dynamic mode with process pressure measurements and flow rate settings, assess the risk level of the microcolumn based on the preset decision matrix and generate maintenance recommendations.

8. A pressure-temperature coupled monitoring system for radioactive column separation, applied to the pressure-temperature coupled monitoring method for radioactive column separation as described in any one of claims 1-7, characterized in that, include: The signal extraction module is used to inject pressure oscillations with a predetermined frequency into the micropillar; Deformation signals generated at the inlet and outlet of the microcolumn due to changes in internal pressure were detected. From the deformation signal, extract the inlet response amplitude, outlet response amplitude, and phase difference between the inlet and outlet responses corresponding to the pressure oscillation frequency; The sound velocity calculation module calculates the propagation time of pressure oscillations from the inlet to the outlet based on the phase difference; The equivalent sound velocity of the fluid inside the micropillar is determined based on the propagation time and the inherent structural parameters of the micropillar. The model building module obtains the measured values ​​of the process pressure of the micropillar; By combining the equivalent sound velocity and process pressure measurements, a mapping model is constructed with the equivalent sound velocity and process pressure measurements as inputs and the internal pressure of the micropillar as the output. The pressure estimation module is used to input the real-time obtained equivalent sound velocity and process pressure measurement values ​​into the mapping model to obtain the pressure estimate inside the micropillar; The frequency adjustment module is used to compare the pressure estimate with the target pressure value to obtain the pressure deviation; and to adjust the frequency of pressure oscillation based on the relationship between the characteristics of the pressure deviation and the oscillation response characteristics of the inlet response amplitude and the outlet response amplitude. The condition assessment module is used to obtain the energy attenuation ratio of pressure oscillations from inlet to outlet at multiple frequencies, forming an attenuation spectrum; The attenuation spectrum is compared with a pre-stored reference attenuation spectrum, and the state of the microcolumn is evaluated based on the comparison results.

Images