Preliminary separation method and system of liquid chromatography system

Abstract

The invention discloses a primary separation method and system of a liquid chromatography system.The method comprises the steps that a sample is pretreated according to physicochemical properties of a target sample, and a high-purity enriched sample is obtained; optimizing the composition proportion of the mobile phase according to the composition characteristics of the high-purity enriched sample to obtain the optimized mobile phase composition; selecting an optimal chromatographic column type according to the optimized mobile phase composition, and carrying out dynamic temperature control separation through a dynamic temperature control technology to obtain a high-selectivity separation condition; performing real-time detection on the separated chromatographic peak according to a high-selectivity separation condition to obtain a high-signal-to-noise-ratio detection signal; and evaluating the preliminary separation effect according to the high signal-to-noise ratio detection signal. By utilizing the embodiment of the invention, the enrichment efficiency and the separation selectivity of the sample can be improved, and the accuracy and the reliability of integral separation are further improved.

Description

Technical Field

[0001] This invention belongs to the field of liquid chromatography technology, and in particular to a preliminary separation method and system for a liquid chromatography system. Background Technology

[0002] Liquid chromatography (LC) is widely used in chemical analysis for its efficient separation and precise quantification capabilities. However, existing techniques still face numerous challenges in sample pretreatment, mobile phase optimization, column selection, and online detection. Traditional sample pretreatment methods often rely on conventional solid-phase extraction, which struggles to achieve high-purity enrichment and is insufficient for handling complex matrices. Furthermore, mobile phase composition optimization typically lacks a systematic approach and relies on experience, leading to unstable separation results. Column selection and temperature control techniques also fail to fully incorporate the characteristics of the target compounds, affecting separation selectivity and efficiency. Meanwhile, while online detection technology offers some signal monitoring capabilities, it remains insufficient in signal quality and real-time performance, and is limited by traditional data processing methods, making it difficult to provide high signal-to-noise ratio detection results. Summary of the Invention

[0003] The purpose of this invention is to provide a preliminary separation method and system for liquid chromatography to overcome the shortcomings of the prior art, improve the enrichment efficiency and separation selectivity of samples, and thus enhance the overall accuracy and reliability of separation.

[0004] One embodiment of this application provides a preliminary separation method for a liquid chromatography system, the method comprising: Based on the physicochemical properties of the target sample, a functionalized solid-phase extraction technology based on nanomaterials was used to pretreat the sample. Through porous nanomaterials and dynamic adsorption optimization algorithms, combined with the characteristic parameters of the target compound, a high-purity enriched sample was obtained. Based on the compositional characteristics of the high-purity enriched sample, a mobile phase composition optimization algorithm based on response surface methodology was adopted, combined with a retention behavior prediction model of the target compound, to optimize the composition ratio of the mobile phase. Through microfluidic mixing technology, the mixing uniformity and flow rate stability of the mobile phase were dynamically adjusted to obtain the optimized mobile phase composition. Based on the optimized mobile phase composition, a machine learning-based column selection model is used to select the optimal column type in combination with the separation requirements of the target compound. Dynamic temperature control separation is performed through dynamic temperature control technology. The column temperature is adjusted in real time in combination with the thermal stability of the target compound to obtain highly selective separation conditions. Based on the high-selectivity separation conditions, an online detection technology based on ultraviolet-visible spectroscopy and mass spectrometry is adopted, combined with a signal enhancement algorithm, to detect the separated chromatographic peaks in real time. Through multi-channel signal fusion technology, ultraviolet-visible spectroscopy and mass spectrometry data are dynamically integrated to obtain a high signal-to-noise ratio detection signal. Based on the high signal-to-noise ratio detection signal, a machine learning-based separation effect evaluation model is used, combined with the retention time, peak shape symmetry and separation degree index of the target compound, to evaluate the preliminary separation effect.

[0005] Optionally, based on the physicochemical properties of the target sample, a functionalized solid-phase extraction technique based on nanomaterials is used to pretreat the sample. Through porous nanomaterials and dynamic adsorption optimization algorithms, combined with the characteristic parameters of the target compound, a high-purity enriched sample is obtained, including: Based on the physicochemical properties of the target sample, a prediction model based on molecular descriptors is used to calculate its adsorption characteristic parameters. A preliminary range of adsorption conditions is generated through a multi-objective optimization algorithm. For the initial adsorption condition range, a functionalized solid-phase extraction technology based on nanomaterials was adopted. Porous nanomaterials were selected, and a preliminary functionalized solid-phase extraction material was generated through surface modification technology. Based on the preliminary functionalized solid-phase extraction materials, a dynamic adsorption optimization algorithm was used, combined with the characteristic parameters of the target compound, to adjust the adsorption conditions in real time. Through adsorption efficiency monitoring technology, a preliminary enriched sample was generated. For the initial enriched sample, an elution optimization algorithm is used to optimize the elution conditions by combining the desorption characteristics of the target compound. The final high-purity enriched sample is generated by using purity evaluation indicators.

[0006] Optionally, based on the compositional characteristics of the high-purity enriched sample, a mobile phase composition optimization algorithm based on response surface methodology is used, combined with a retention behavior prediction model of the target compound, to optimize the composition ratio of the mobile phase. Through microfluidic mixing technology, the mixing uniformity and flow rate stability of the mobile phase are dynamically adjusted to obtain the optimized mobile phase composition, including: Based on the compositional characteristics of the high-purity enriched sample, a mobile phase composition optimization algorithm based on response surface methodology is adopted, combined with a retention behavior prediction model of the target compound, to generate a preliminary range of mobile phase composition. For the initial range of mobile phase composition, the response surface methodology was used to optimize the mobile phase composition ratio by combining the retention time and peak shape symmetry of the target compound. Through a multi-objective optimization mechanism, the initial optimized mobile phase composition was generated. For the initial optimized mobile phase composition, microfluidic mixing technology is used, combined with dynamic mixing uniformity monitoring data, to adjust the mixing ratio and flow rate in real time, and generate an initial uniform mobile phase through a feedback control mechanism; For the initial homogeneous mobile phase, a stability evaluation index is used to optimize the stability of the mobile phase. Then, a dynamic adjustment algorithm is used to generate the final optimized mobile phase composition.

[0007] Optionally, based on the optimized mobile phase composition, a machine learning-based column selection model is used to select the optimal column type, combined with the separation requirements of the target compound. Dynamic temperature control separation is then performed using dynamic temperature control technology, and the column temperature is adjusted in real time based on the thermal stability of the target compound to obtain highly selective separation conditions. This includes: Based on the optimized mobile phase composition, a machine learning-based column selection model was used to analyze the separation potential of different columns in conjunction with the separation requirements of the target compounds. Preliminary column selection results were generated through column efficiency evaluation indicators. Based on the column selection results and the thermal stability of the target compound, the dynamic temperature control parameters are initialized, and the initial temperature control parameters are generated through the temperature control efficiency prediction model. Based on the initial temperature control parameters, the chromatographic separation process is started. An embedded temperature sensor is used to monitor the column temperature in real time and collect separation effect data simultaneously. Through feedback control algorithm, combined with real-time separation effect data, the temperature parameters are dynamically adjusted to generate preliminary optimized temperature control separation conditions. Based on the preliminary optimized temperature-controlled separation conditions, a comprehensive evaluation model for separation effect was adopted, combining separation degree, peak capacity and column efficiency indicators to verify the separation conditions. Through iterative optimization algorithm, the temperature parameters were further fine-tuned to generate the final high-selectivity separation conditions.

[0008] Optionally, based on the high-selectivity separation conditions, online detection technology based on ultraviolet-visible spectroscopy and mass spectrometry, combined with signal enhancement algorithms, is used to detect the separated chromatographic peaks in real time. Through multi-channel signal fusion technology, ultraviolet-visible spectroscopy and mass spectrometry data are dynamically integrated to obtain a high signal-to-noise ratio detection signal, including: Based on the high-selectivity separation conditions, the detection parameters of ultraviolet-visible spectroscopy and mass spectrometry are initialized, and preliminary detection conditions are generated through the detection efficiency prediction model. For the initial detection conditions, a signal enhancement algorithm is used, combined with real-time chromatographic peak shape monitoring data, to optimize the detection signal quality. Through noise filtering technology, a preliminary high-quality detection signal is generated. For the initial high-quality detection signal, multi-channel signal fusion technology is used to combine ultraviolet-visible spectroscopy and mass spectrometry data to dynamically integrate multi-source signals and generate an initial fused signal through a signal alignment algorithm; For the initial fused signal, the signal-to-noise ratio (SNR) evaluation index is used to optimize the signal quality, and a dynamic adjustment algorithm is used to generate the final high SNR detection signal.

[0009] Optionally, the step of evaluating the preliminary separation effect using a machine learning-based separation performance evaluation model based on the high signal-to-noise ratio detection signal, combined with the retention time, peak shape symmetry, and separation degree index of the target compound, includes: Based on the high signal-to-noise ratio detection signal, data preprocessing techniques are used, combined with the retention time, peak shape symmetry and separation index of the target compound, to generate a preliminary separation effect dataset; For the preliminary separation effect dataset, a machine learning-based separation effect evaluation model was used, combined with the standard curve of the target compound, to train the model and generate preliminary evaluation results through cross-validation. Based on the preliminary evaluation results, a dynamic adjustment algorithm is used, combined with the separation effect evaluation index, to optimize the separation conditions. Through a feedback control mechanism, a preliminary optimized separation effect is generated. For the initial optimized separation effect, an accuracy evaluation index is used to further optimize the separation effect. Through an error correction mechanism, the final separation effect evaluation result is generated.

[0010] Another embodiment of this application provides a preliminary separation system for a liquid chromatography system, the system comprising: The processing module is used to pre-treat the target sample based on the physicochemical properties of the sample using a functionalized solid-phase extraction technology based on nanomaterials. Through porous nanomaterials and dynamic adsorption optimization algorithms, combined with the characteristic parameters of the target compound, a high-purity enriched sample is obtained. The optimization module is used to optimize the composition ratio of the mobile phase based on the composition characteristics of the high-purity enriched sample by using a mobile phase composition optimization algorithm based on response surface methodology and combined with a retention behavior prediction model of the target compound. Through microfluidic mixing technology, the mixing uniformity and flow rate stability of the mobile phase are dynamically adjusted to obtain the optimized mobile phase composition. The separation module is used to select the optimal column type based on the optimized mobile phase composition and a machine learning-based column selection model, combined with the separation requirements of the target compound. It performs dynamic temperature-controlled separation using dynamic temperature control technology, and adjusts the column temperature in real time based on the thermal stability of the target compound to obtain highly selective separation conditions. The detection module is used to detect the separated chromatographic peaks in real time by using online detection technology based on ultraviolet-visible spectroscopy and mass spectrometry, combined with signal enhancement algorithms, according to the high-selectivity separation conditions. Through multi-channel signal fusion technology, ultraviolet-visible spectroscopy and mass spectrometry data are dynamically integrated to obtain a high signal-to-noise ratio detection signal. The evaluation module is used to evaluate the initial separation effect based on a machine learning-based separation effect evaluation model, combined with the retention time, peak shape symmetry and separation degree index of the target compound, based on the high signal-to-noise ratio detection signal.

[0011] Another embodiment of this application provides a storage medium storing a computer program, wherein the computer program is configured to execute the method described in any of the preceding claims when running.

[0012] Another embodiment of this application provides an electronic device including a memory and a processor, wherein the memory stores a computer program and the processor is configured to run the computer program to perform the method described in any of the preceding claims.

[0013] Compared with existing technologies, the present invention provides a preliminary separation method for a liquid chromatography system. Based on the physicochemical properties of the target sample, the sample is pretreated to obtain a high-purity enriched sample. Based on the compositional characteristics of the high-purity enriched sample, the composition ratio of the mobile phase is optimized to obtain an optimized mobile phase composition. Based on the optimized mobile phase composition, the optimal chromatographic column type is selected, and dynamic temperature control separation is performed using dynamic temperature control technology to obtain highly selective separation conditions. Based on the highly selective separation conditions, the separated chromatographic peaks are detected in real time to obtain a high signal-to-noise ratio detection signal. Based on the high signal-to-noise ratio detection signal, the preliminary separation effect is evaluated, thereby improving the enrichment efficiency and separation selectivity of the sample, and ultimately enhancing the overall accuracy and reliability of the separation. Attached Figure Description

[0014] Figure 1 Hardware structure block diagram of a computer terminal for a preliminary separation method of a liquid chromatography system provided in an embodiment of the present invention; Figure 2 A schematic flowchart of a preliminary separation method for a liquid chromatography system provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of a preliminary separation system of a liquid chromatography system provided in an embodiment of the present invention. Detailed Implementation

[0015] The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0016] This invention first provides a preliminary separation method for a liquid chromatography system, which can be applied to electronic devices, such as computer terminals, specifically ordinary computers.

[0017] The following detailed explanation uses a computer terminal as an example. Figure 1 This is a hardware block diagram of a computer terminal for a preliminary separation method of a liquid chromatography system provided in an embodiment of the present invention. (See diagram below.) Figure 1 As shown, the computer device includes a processor, memory, and network interface connected via a system bus, wherein the memory may include non-volatile storage media and internal memory.

[0018] Non-volatile storage media can store operating systems and computer programs. These computer programs include program instructions that, when executed, cause the processor to perform a preliminary separation method for any liquid chromatography system.

[0019] The processor provides computing and control capabilities, supporting the operation of the entire computer device.

[0020] Internal memory provides an environment for the execution of computer programs in non-volatile storage media. When executed by a processor, the computer program enables the processor to perform a preliminary separation method for any liquid chromatography system.

[0021] This network interface is used for network communication, such as sending assigned tasks. Those skilled in the art will understand that... Figure 1 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0022] It should be understood that the processor can be a Central Processing Unit (CPU), but it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among these, a general-purpose processor can be a microprocessor or any conventional processor.

[0023] See Figure 2 The present invention provides a preliminary separation method for a liquid chromatography system, which may include the following steps: S201. Based on the physicochemical properties of the target sample, a functionalized solid-phase extraction technology based on nanomaterials is used to pretreat the sample. Through porous nanomaterials and dynamic adsorption optimization algorithms, combined with the characteristic parameters of the target compound, a high-purity enriched sample is obtained. This step utilizes functionalized nanomaterials (such as mesoporous silica and metal-organic frameworks) as the solid-phase extraction medium, achieving selective adsorption of target compounds through surface modification with specific functional groups (such as amino and carboxyl groups). A dynamic adsorption optimization algorithm, based on the target analyte's logP value, molecular volume, and other physicochemical parameters, adjusts conditions such as pH and ionic strength in real time. This ensures high adsorption efficiency while minimizing matrix interference, significantly improving the recovery rate and purity of target compounds in complex matrices. This provides a high-quality sample basis for subsequent chromatographic separation and effectively avoids column contamination and detection interference.

[0024] Specifically, based on the physicochemical properties of the target sample, a prediction model based on molecular descriptors can be used to calculate its adsorption characteristic parameters, and a preliminary range of adsorption conditions can be generated through a multi-objective optimization algorithm. In this step, the physicochemical properties of the target sample are first analyzed using a molecular descriptor prediction model. Molecular descriptors are numerical indicators used to describe molecular structure and properties, and these indicators can be used to predict the adsorption behavior of the sample under different conditions. Next, a multi-objective optimization algorithm is applied, comprehensively considering multiple adsorption characteristic parameters, to generate a preliminary set of adsorption condition ranges. These condition ranges provide a reasonable starting point for subsequent experiments, ensuring the effective enrichment of the target compound during the experiment. The significance of this step lies in reducing the blind spots and trial-and-error costs of experiments through scientific prediction and optimization methods. Through calculation and optimization, a set of adsorption condition ranges most likely to succeed can be determined before the experiment begins, thereby improving experimental efficiency and success rate. This not only saves time and resources but also provides a reliable foundation for subsequent sample processing.

[0025] In the preliminary separation methods of liquid chromatography, the first step is to analyze the physicochemical properties of the target sample. First, basic information about the sample, such as molecular weight, polarity, and solubility, is collected. This information can be obtained through experimental measurements or literature review. Next, a predictive model based on molecular descriptors is used to calculate the sample's adsorption characteristic parameters. Molecular descriptors are numerical indicators used to quantify molecular structure and properties, such as molecular volume and surface charge distribution. These descriptors can be used to predict the behavior of the sample under different adsorption conditions.

[0026] To determine the optimal adsorption conditions, a multi-objective optimization algorithm was applied. This algorithm considers multiple adsorption characteristic parameters, such as adsorption capacity, selectivity, and stability, generating a preliminary range of adsorption conditions. These ranges provide a reasonable starting point for subsequent experiments, ensuring the effective enrichment of the target compound during the experimental process. For example, when processing a complex biological sample, it may be found to contain a variety of different compounds. Using a molecular descriptor prediction model, compounds with high adsorption potential can be identified. Then, using the multi-objective optimization algorithm, a set of adsorption conditions, such as pH, temperature, and solvent concentration, can be generated to maximize the adsorption efficiency of the target compound.

[0027] Suppose we are processing a complex environmental water sample containing various organic pollutants, such as polycyclic aromatic hydrocarbons (PAHs). First, we collect physicochemical data on these pollutants, including molecular weight, polarity, solubility, and melting point. Next, we use a molecular descriptor-based predictive model to calculate the adsorption characteristics of each pollutant. Molecular descriptors, such as molecular volume, surface charge distribution, and hydrophobicity index, are used to predict the behavior of these compounds under different adsorption conditions. For example, molecular descriptors can predict the diffusion capacity of molecules in porous materials by calculating molecular volume, or predict the interaction between molecules and charged surfaces by calculating surface charge distribution. Software tools, such as ChemAxon or Molecular Operating Environment (MOE), may be used to calculate these descriptors. A preliminary set of adsorption condition ranges is generated by using a multi-objective optimization algorithm, considering multiple adsorption characteristics such as adsorption capacity, selectivity, and stability. These conditions include pH, temperature, and solvent concentration. Multi-objective optimization algorithms, such as genetic algorithms or particle swarm optimization algorithms, can help find the optimal balance among multiple objectives. For example, genetic algorithms can improve the adsorption efficiency of target compounds by simulating the natural selection process and gradually optimizing adsorption conditions.

[0028] For the initial adsorption condition range, a functionalized solid-phase extraction technology based on nanomaterials was adopted. Porous nanomaterials were selected, and a preliminary functionalized solid-phase extraction material was generated through surface modification technology. In this step, suitable porous nanomaterials are selected as the base material for solid-phase extraction (SPE) based on the initially determined adsorption condition range. These nanomaterials are functionalized using surface modification techniques to enhance their selective adsorption capacity for target compounds. The functionalization process may involve introducing specific chemical groups onto the surface of the nanomaterials to improve their interaction with the target compounds, thereby increasing enrichment efficiency. The selection and preparation of functionalized SPE materials are crucial to ensuring successful sample pretreatment. By using functionalized nanomaterials, the enrichment efficiency and selectivity of target compounds can be significantly improved. This process not only improves sample purity but also provides a higher-quality sample basis for subsequent analytical steps, ensuring the accuracy and reliability of the final analytical results.

[0029] After determining the initial adsorption condition range, it is necessary to select suitable nanomaterials for functionalized solid-phase extraction. Porous nanomaterials are widely used due to their high specific surface area and good adsorption performance. First, several promising porous nanomaterials, such as silica nanoparticles and carbon nanotubes, were screened. Next, these materials were functionalized using surface modification techniques.

[0030] Surface modification techniques involve introducing specific chemical groups onto the surface of nanomaterials to enhance their selective adsorption capacity for target compounds. For example, amino, carboxyl, or sulfonic acid groups can be introduced onto the surface of nanomaterials to improve their adsorption capacity for charged compounds. In this way, preliminary functionalized solid-phase extraction materials are generated. Consider a water sample containing multiple metal ions. Silica nanoparticles are selected as the base material, and amino groups are introduced onto their surface. Amino groups can form coordination bonds with metal ions, thereby improving the material's adsorption capacity for metal ions. Through this functionalization treatment, a solid-phase extraction material suitable for metal ion enrichment is successfully prepared.

[0031] After determining the initial adsorption condition range, silica nanoparticles were selected as the base material due to their high specific surface area and good adsorption performance. To enhance their selective adsorption capacity for polycyclic aromatic hydrocarbons (PAHs), amino groups were introduced onto the surface of the silica nanoparticles using surface modification techniques. These amino groups can form weak hydrogen bonds and π-π interactions with PAHs, thereby improving the material's adsorption capacity for these compounds. For example, surface modification techniques can attach aminosilane coupling agents to the silica surface via chemical reactions. This chemical modification can be achieved through simple impregnation or sol-gel methods. Through this functionalization treatment, preliminary functionalized solid-phase extraction materials were generated. These materials were used in subsequent experiments to enrich PAH pollutants in water samples.

[0032] Based on the preliminary functionalized solid-phase extraction materials, a dynamic adsorption optimization algorithm was used, combined with the characteristic parameters of the target compound, to adjust the adsorption conditions in real time. Through adsorption efficiency monitoring technology, a preliminary enriched sample was generated. In this stage, dynamic adsorption optimization algorithms are used to adjust adsorption conditions in real time based on the characteristic parameters of the target compound (such as molecular size and polarity) to achieve the best adsorption effect. Adsorption efficiency monitoring technology is used to monitor the efficiency of the adsorption process in real time, ensuring that parameters can be adjusted in a timely manner to optimize the enrichment effect. Through this dynamic adjustment process, preliminary enriched samples can be obtained. The process of dynamically adjusting adsorption conditions ensures the flexibility and adaptability of sample processing. By monitoring and adjusting in real time, the enrichment efficiency of the target compound can be maximized during adsorption. This process not only improves the purity of the sample but also reduces sample loss, ensuring the accuracy and reliability of subsequent analyses.

[0033] After preparing the functionalized solid-phase extraction material, adsorption experiments are required to enrich the target compound. Dynamic adsorption optimization algorithms play a crucial role in this process. These algorithms can dynamically adjust adsorption conditions based on real-time monitoring data to achieve optimal results. First, the sample is brought into contact with the functionalized solid-phase extraction material, and the adsorption process is monitored in real-time using adsorption efficiency monitoring technology.

[0034] Adsorption efficiency monitoring technology can include online monitoring of parameters such as the adsorption capacity, selectivity, and stability of the adsorbent. Using this data, dynamic adsorption optimization algorithms can adjust adsorption conditions in real time, such as flow rate, temperature, and pH, to maximize the enrichment efficiency of the target compound. Ultimately, a preliminary enriched sample is obtained. For example, in treating a water sample containing multiple organic pollutants, functionalized carbon nanotubes are used as the adsorbent material. Through dynamic adsorption optimization algorithms, the flow rate and pH of the water sample can be adjusted in real time to improve the adsorption efficiency of carbon nanotubes for specific organic pollutants. This dynamic adjustment successfully enriches the target pollutant.

[0035] After preparing functionalized silica nanoparticles, they were contacted with environmental water samples to enrich polycyclic aromatic hydrocarbons (PAHs). A dynamic adsorption optimization algorithm played a crucial role in this process. Adsorption conditions, such as flow rate, temperature, and pH, were adjusted in real time by monitoring parameters like the adsorbent's adsorption capacity and selectivity online. For example, the dynamic adsorption optimization algorithm could use an adaptive control system to automatically adjust the flow rate and temperature based on real-time monitoring data. Adsorption efficiency monitoring techniques might include using UV-Vis spectroscopy to measure the PAH concentration in the effluent in real time. It was found that the highest adsorption efficiency was achieved at pH 7.0 and a temperature of 25°C. Therefore, the dynamic adsorption optimization algorithm was used to adjust these conditions in real time to maximize the PAH enrichment efficiency. Ultimately, a preliminary enriched sample was obtained, with a significantly increased PAH concentration.

[0036] For the initial enriched sample, an elution optimization algorithm is used to optimize the elution conditions by combining the desorption characteristics of the target compound. The final high-purity enriched sample is generated by using purity evaluation indicators.

[0037] In the final step, the pre-enriched sample undergoes elution optimization. By employing elution optimization algorithms, combined with the desorption characteristics of the target compound (such as solubility and polarity), the optimal elution conditions can be determined to maximize the recovery and purity of the target compound. Purity evaluation metrics are used to assess the purity of the eluted sample, ensuring that the final sample meets high purity requirements. Elution optimization is a crucial step in ensuring sample purity and recovery. By optimizing elution conditions, the target compound can be effectively separated and recovered, reducing interference from impurities. This process not only improves the analytical quality of the sample but also provides a high-purity sample basis for subsequent analyses, ensuring the accuracy and reliability of the analytical results.

[0038] After obtaining the initial enriched sample, elution is required to recover the target compound. Elution optimization algorithms play a crucial role in this process. First, the desorption characteristics of the target compound are analyzed, such as solubility, polarity, and stability. Based on these characteristics, the elution optimization algorithm can determine the optimal elution conditions, such as the type, concentration, and flow rate of the eluent.

[0039] During the elution process, the elution effect was monitored in real time using purity evaluation indicators. These indicators could include the purity, recovery rate, and selectivity of the target compound. By continuously adjusting the elution conditions, a high-purity enriched sample was ultimately obtained. Assuming a sample containing multiple organic acids is being processed, acetonitrile was selected as the eluent, and the optimal acetonitrile concentration and flow rate were determined using an elution optimization algorithm. The purity evaluation indicators allowed for real-time monitoring of the purity and recovery rate of the organic acids during elution. Ultimately, high-purity organic acids were successfully recovered.

[0040] After obtaining the initial enriched sample, elution is required to recover polycyclic aromatic hydrocarbons (PAHs). Elution optimization algorithms play a crucial role in this process. First, the desorption characteristics of PAHs, such as solubility and polarity, are analyzed. Based on these characteristics, acetonitrile is selected as the eluent. For example, elution optimization algorithms can use response surface methodology (RSM) to optimize the eluent concentration and flow rate. The elution effect is monitored in real time using purity evaluation metrics, such as high-performance liquid chromatography (HPLC). By continuously adjusting the elution conditions, a high-purity enriched sample is finally obtained, where the purity and recovery rate of PAHs both meet the expected standards. This optimization effectively separates and recovers the target compounds, ensuring high sample purity and high recovery rates.

[0041] S202. Based on the compositional characteristics of the high-purity enriched sample, a mobile phase composition optimization algorithm based on response surface methodology is adopted, combined with a retention behavior prediction model of the target compound, to optimize the composition ratio of the mobile phase. Through microfluidic mixing technology, the mixing uniformity and flow rate stability of the mobile phase are dynamically adjusted to obtain the optimized mobile phase composition. A mathematical model of mobile phase composition (such as acetonitrile / water ratio, buffer salt concentration, etc.) and retention time was established using response surface methodology. This model was then combined with machine learning to predict peak shape and separation under different compositions. The microfluidic mixer employs a multi-layered staggered channel design to achieve precise mixing at the nanoliter level. Online conductivity monitoring monitors mixing uniformity, ensuring optimal separation selectivity while maintaining mobile phase stability, reducing baseline fluctuations and retention time drift, and improving analytical reproducibility.

[0042] Specifically, based on the compositional characteristics of the high-purity enriched sample, a mobile phase composition optimization algorithm based on response surface methodology can be used, combined with a retention behavior prediction model of the target compound, to generate a preliminary range of mobile phase composition. In this step, the composition of the mobile phase is optimized using response surface methodology (RSM) based on the compositional characteristics of the high-purity enriched sample. RSM is a statistical technique used to optimize and predict the system's response. Combined with a retention behavior prediction model of the target compound, a preliminary set of mobile phase composition ranges can be generated, providing a foundation for subsequent mobile phase optimization. Generating these preliminary mobile phase composition ranges allows for the identification of a set of potentially optimal mobile phase conditions before the experiment begins. This process reduces experimental uncertainty and improves experimental efficiency and success rate. By optimizing the mobile phase composition, the retention behavior of the target compound can be better controlled, thereby improving separation performance.

[0043] After obtaining a high-purity enriched sample, the composition of the mobile phase needs to be optimized to achieve the best separation effect in liquid chromatography. First, the compositional characteristics of the enriched sample are analyzed, including the polarity, molecular weight, and solubility of the target compound. These characteristics will affect the retention behavior of the compound in the chromatographic column. A retention behavior prediction model is used to estimate the retention time of the target compound under different mobile phase conditions. This model may be trained using machine learning algorithms based on historical data and the physicochemical properties of the compound.

[0044] Next, response surface methodology (RSM) was used to optimize the mobile phase composition. RSM is a statistical technique used to study the effects of multiple variables on the response and to find optimal conditions. Several commonly used mobile phase components, such as water, acetonitrile, and methanol, were selected, and a series of experiments were designed to evaluate the effects of different combinations on retention time. These experiments generated a preliminary range of mobile phase compositions. For example, it might be found that the retention time of the target compound is closest to the ideal value at a specific ratio of water to acetonitrile. RSM allows for the plotting of the mobile phase composition, showing the effect of different component ratios on retention time. This process helps determine the preliminary range of mobile phase compositions, providing a foundation for subsequent optimization.

[0045] For example, a batch of high-purity enriched samples containing various target compounds, such as polycyclic aromatic hydrocarbons (PAHs), was obtained. To optimize the separation effect of liquid chromatography, the physicochemical properties of these compounds, including polarity, molecular weight, and solubility, were first analyzed. These properties will affect the retention behavior of the compounds in the chromatographic column. A retention behavior prediction model was used to estimate the retention time of the target compounds under different mobile phase conditions. This model may be trained using machine learning algorithms based on historical data and the physicochemical properties of the compounds. Through model prediction, a preliminary understanding of the impact of different mobile phase compositions on retention time can be obtained.

[0046] For the initial range of mobile phase composition, the response surface methodology was used to optimize the mobile phase composition ratio by combining the retention time and peak shape symmetry of the target compound. Through a multi-objective optimization mechanism, the initial optimized mobile phase composition was generated. In this step, the initial mobile phase composition range is further optimized. Response surface methodology (RSM) optimization algorithms, combined with the retention time and peak shape symmetry of the target compounds, allow for the optimization of the mobile phase composition ratio. A multi-objective optimization mechanism is used to balance different optimization objectives, generating an initial optimized mobile phase composition. Optimizing the mobile phase composition ratio is crucial for ensuring effective chromatographic separation. Through optimization, the resolution and peak shape symmetry of the target compounds can be improved, thereby enhancing the accuracy and reliability of the analysis. This process not only improves separation efficiency but also provides a better foundation for subsequent analyses.

[0047] After determining the initial range of mobile phase composition, the composition ratio of the mobile phase was further optimized to achieve the best chromatographic separation effect. Response surface methodology (RSM) optimization algorithms played a crucial role in this process. A series of experiments were designed to systematically change the proportions of the mobile phase components and measure the retention time and peak shape symmetry of the target compound. Peak shape symmetry is an important indicator for evaluating the quality of chromatographic peaks; an ideal peak shape should be close to symmetry to facilitate quantitative analysis. Through a multi-objective optimization mechanism, considering both retention time and peak shape symmetry, the optimal mobile phase composition ratio was found. For example, at a specific water-acetonitrile ratio, the retention time of the target compound was moderate, and the peak shape symmetry was good. RSM helped to plot the optimization surface, showing the combined effect of different mobile phase ratios on multiple response variables.

[0048] This multi-objective optimization mechanism allows for finding the optimal balance among multiple objectives, generating a preliminary optimized mobile phase composition. This process ensures that the mobile phase composition provides the best separation performance and analytical accuracy in chromatographic separation.

[0049] For the initial optimized mobile phase composition, microfluidic mixing technology is used, combined with dynamic mixing uniformity monitoring data, to adjust the mixing ratio and flow rate in real time, and generate an initial uniform mobile phase through a feedback control mechanism; In this stage, microfluidic mixing technology is used to further homogenize the initially optimized mobile phase composition. By dynamically monitoring the mixing homogeneity, the mixing ratio and flow rate of the mobile phase can be adjusted in real time. A feedback control mechanism is used to ensure the homogeneity and stability of the mobile phase, generating a preliminary homogeneous mobile phase. A homogeneous mobile phase is crucial for ensuring the stability and reproducibility of chromatographic separation. Microfluidic mixing technology allows for precise control of the mobile phase mixing process, improving its homogeneity and stability. This process not only improves separation efficiency but also reduces experimental errors, ensuring the reliability of analytical results.

[0050] After determining the initial optimized mobile phase composition, microfluidic mixing technology was used to achieve precise mixing of the mobile phase. Microfluidic technology allows for the control of fluid flow and mixing at the micrometer scale, characterized by high precision and efficiency. A microfluidic mixing device was designed to precisely control the proportions and flow rates of the mobile phase components. To ensure the homogeneity of the mobile phase, dynamic mixing uniformity monitoring data was used to monitor the mixing effect of the mobile phase in real time. Online sensors can detect the component proportions and flow rates of the mobile phase, and adjustments are made in real time through a feedback control mechanism. For example, if non-uniform mixing of the mobile phase is detected, the system automatically adjusts the flow rate or component proportions to achieve the desired uniformity. This real-time adjustment mechanism ensures the homogeneity and stability of the mobile phase, generating an initially homogeneous mobile phase. This process is crucial for successful separation in liquid chromatography, as the homogeneity of the mobile phase directly affects the shape of the chromatographic peaks and the separation effect.

[0051] For the initial homogeneous mobile phase, a stability evaluation index is used to optimize the stability of the mobile phase. Then, a dynamic adjustment algorithm is used to generate the final optimized mobile phase composition.

[0052] In the final step, the stability of the initial homogeneous mobile phase is assessed. Stability assessment indicators identify unstable factors in the mobile phase, which are then optimized using a dynamic adjustment algorithm. Ultimately, this results in an optimized mobile phase composition with good stability. Optimizing the stability of the mobile phase is crucial for ensuring chromatographic separation and the reliability of analytical results. Through stability assessment and dynamic adjustment, the consistency and stability of the mobile phase are ensured throughout the analytical process. This process not only improves separation efficiency but also provides a reliable foundation for subsequent analyses.

[0053] After obtaining a preliminary homogeneous mobile phase, it is necessary to ensure its long-term stability to achieve consistent separation in liquid chromatography. Stability assessment metrics are used to evaluate the stability of the mobile phase, which may include changes in pH, ionic strength, and component ratios. A dynamic adjustment algorithm is used to optimize the stability of the mobile phase. This algorithm automatically adjusts the component ratios and flow rates of the mobile phase based on real-time monitoring data to cope with potential changes. For example, if the pH of the mobile phase deviates from the expected value, the system automatically adjusts the amount of acid or base added to restore it to the target pH. Through this dynamic adjustment mechanism, a final optimized mobile phase composition can be generated, ensuring that the mobile phase remains stable throughout the chromatographic analysis. This process is crucial for obtaining reliable chromatographic data, as the stability of the mobile phase directly affects the reproducibility of chromatographic peaks and the accuracy of analytical results.

[0054] S203, based on the optimized mobile phase composition, adopts a machine learning-based column selection model, combined with the separation requirements of the target compound, to select the optimal column type, and performs dynamic temperature control separation through dynamic temperature control technology. Combined with the thermal stability of the target compound, the column temperature is adjusted in real time to obtain highly selective separation conditions. Based on the target compound's polarity, molecular size, and other characteristics, a trained random forest model recommends the optimal column type from C18, HILIC, and other chromatographic columns. The dynamic temperature control system adjusts the column temperature with an accuracy of 0.1℃ during separation based on the compound's thermal stability data (such as thermogravimetric analysis results), optimizing retention behavior and peak shape. This achieves precise control of separation selectivity, which is particularly beneficial for separating thermally unstable compounds and difficult-to-separate substance pairs, improving peak capacity and separation efficiency.

[0055] Specifically, based on the optimized mobile phase composition, a machine learning-based column selection model can be used to analyze the separation potential of different columns in conjunction with the separation requirements of the target compound. Preliminary column selection results can be generated through column efficiency evaluation indicators. In this step, machine learning models are used to analyze the separation potential of different chromatographic columns. Combined with the separation requirements of the target compound, the model evaluates the performance of different columns under optimized mobile phase conditions. Column efficiency evaluation metrics are used to quantify the separation capability of the columns, helping to generate preliminary column selection results. Selecting the appropriate column is crucial to ensuring effective chromatographic separation. Machine learning models can quickly screen for the most suitable column type, improving separation efficiency and effectiveness. This process not only improves experimental efficiency but also provides a reliable foundation for subsequent separations.

[0056] In liquid chromatography (HPLC) systems, selecting the appropriate column is a crucial step in achieving efficient separation. First, the separation requirements of the target compound are determined based on the optimized mobile phase composition. These requirements may include the compound's polarity, molecular weight, and retention behavior under specific mobile phase conditions. To select the optimal column type, a machine learning-based column selection model is employed. This model is trained using a large amount of historical data, including the separation performance of different columns under various conditions. The model can predict the separation potential of different columns based on the input compound characteristics and mobile phase composition. By inputting characteristic parameters of the target compound, such as molecular structure and polarity index, the model outputs a set of recommended column types.

[0057] To validate the model's recommendations, column efficiency evaluation metrics were used to analyze the separation potential of each column. These metrics might include column efficiency (theoretical plate number), resolution, peak capacity, etc. By comparing the column efficiency evaluation metrics of different columns, preliminary column selection results can be generated, ensuring that the selected columns meet the separation requirements. For example, a certain C18 reversed-phase column might be found to exhibit the best separation performance and peak shape symmetry under the current mobile phase conditions, and therefore be selected as a preliminary column.

[0058] Machine learning model: The Random Forest algorithm could be used to build a column selection model. Random Forest is an ensemble learning method that improves the accuracy and robustness of predictions by constructing multiple decision trees and combining their outputs. The model input includes the molecular descriptor of the compound and the composition of the mobile phase, and the output is a recommended column type.

[0059] Column performance evaluation metrics: When selecting chromatographic columns, column performance evaluation metrics are used to analyze the separation potential of each column. These metrics may include theoretical plate number (N), resolution (Rs), peak capacity (Pc), etc. By comparing the column performance evaluation metrics of different columns, preliminary column selection results can be generated.

[0060] Based on the column selection results and the thermal stability of the target compound, the dynamic temperature control parameters are initialized, and the initial temperature control parameters are generated through the temperature control efficiency prediction model. In this step, dynamic temperature control parameters are initialized based on the column selection and the thermal stability of the target compound. A temperature control efficiency prediction model is used to predict the impact of different temperature control parameters on separation, helping to generate the initial temperature control parameters. Initialization of dynamic temperature control parameters is crucial for ensuring effective chromatographic separation. By setting appropriate temperature control parameters, separation efficiency and effectiveness can be improved, ensuring the stability and resolution of the target compound within the column. This process not only improves separation performance but also provides a reliable foundation for subsequent separation processes.

[0061] After selecting a suitable chromatographic column, the thermal stability of the target compound needs to be considered to ensure that it does not degrade or denature during separation. Based on the column selection results, the stability of the target compound at different temperatures was evaluated, and a suitable temperature range was determined. A temperature control efficiency prediction model was used to generate initial temperature control parameters. This model is trained using a machine learning algorithm based on historical experimental data and the thermal stability of the compound. The model can predict separation efficiency and compound stability under different temperature conditions. Inputting the column type and the thermal stability parameters of the target compound, the model outputs a set of initial temperature control parameters, including the starting temperature and heating rate. These initial temperature control parameters enable the chromatographic separation process to be initiated, ensuring optimal temperature conditions during separation to achieve efficient separation and compound stability.

[0062] After selecting a suitable chromatographic column, the thermal stability of the target compound needs to be considered to ensure that the compound does not degrade or denature during separation. Suppose a group of thermosensitive compounds is being processed, which may degrade at high temperatures. Based on the column selection, the stability of these compounds at different temperatures is evaluated, and a suitable temperature range is determined. A temperature control efficiency prediction model is used to generate initial temperature control parameters. This model is trained using a machine learning algorithm based on historical experimental data and the thermal stability of the compounds. The model can predict separation efficiency and compound stability under different temperature conditions. Inputting the column type and the thermal stability parameters of the target compound, the model outputs a set of initial temperature control parameters, including the starting temperature and heating rate. These initial temperature control parameters enable the chromatographic separation process to be initiated, ensuring optimal temperature conditions during separation for efficient separation and compound stability. For example, the model might suggest setting the starting temperature to 30°C and increasing it to 50°C at a rate of 1°C per minute to ensure compound stability and separation effectiveness.

[0063] Temperature control efficiency prediction model: A support vector machine (SVM) algorithm may be used to construct the temperature control efficiency prediction model. SVM is a supervised learning model suitable for classification and regression problems. The model input includes the thermal stability parameters of the compound and the chromatographic column type, and the output is the initial temperature control parameters.

[0064] Thermal stability assessment: The thermal stability of compounds is assessed using techniques such as thermogravimetric analysis (TGA) or differential scanning calorimetry (DSC). These techniques provide stability data for compounds at different temperatures, helping to determine suitable temperature ranges.

[0065] Based on the initial temperature control parameters, the chromatographic separation process is started. An embedded temperature sensor is used to monitor the column temperature in real time and collect separation effect data simultaneously. Through feedback control algorithm, combined with real-time separation effect data, the temperature parameters are dynamically adjusted to generate preliminary optimized temperature control separation conditions. In this stage, the chromatographic separation process is initiated, and the column temperature is monitored in real time using an embedded temperature sensor. Synchronous acquisition of separation data and the application of feedback control algorithms help dynamically adjust temperature parameters to optimize separation conditions. Real-time temperature monitoring and dynamic temperature control adjustment are crucial for ensuring effective chromatographic separation. Through real-time monitoring and adjustment, separation efficiency and effectiveness can be improved, ensuring the stability and resolution of the target compound within the column. This process not only enhances separation performance but also provides a reliable foundation for subsequent separation processes.

[0066] Temperature is a crucial factor affecting separation efficiency during chromatographic separation. The chromatographic separation process is initiated based on initial temperature control parameters, and embedded temperature sensors are installed in the chromatographic column. These sensors monitor the column temperature in real time, ensuring it remains within the preset range. To optimize separation, a feedback control algorithm dynamically adjusts the temperature parameters. This algorithm automatically adjusts the temperature settings based on real-time separation data (such as retention time and peak shape symmetry). For example, if the separation effect at a certain temperature is detected as unsatisfactory, the system automatically adjusts the temperature to improve separation. This dynamic adjustment mechanism generates preliminary optimized temperature control separation conditions. This process ensures that the temperature conditions remain optimal throughout the separation process, thereby achieving efficient separation. For instance, the system might adjust the temperature from 50°C to 55°C during separation to improve separation efficiency and peak shape symmetry.

[0067] Feedback control algorithm: Proportional-integral-derivative (PID) control algorithm may be used to achieve dynamic temperature adjustment. PID control is a classic feedback control algorithm that adjusts the control variable (such as temperature) to achieve the desired system behavior by calculating the error value (the difference between the setpoint and the measured value).

[0068] Embedded temperature sensors: High-precision embedded temperature sensors, such as thermocouples or RTDs (resistance temperature detectors), are used to monitor the temperature of the chromatographic column in real time. These sensors provide fast and accurate temperature data, supporting the operation of feedback control systems.

[0069] Based on the preliminary optimized temperature-controlled separation conditions, a comprehensive evaluation model for separation effect was adopted, combining separation degree, peak capacity and column efficiency indicators to verify the separation conditions. Through iterative optimization algorithm, the temperature parameters were further fine-tuned to generate the final high-selectivity separation conditions.

[0070] In the final step, the preliminary optimized temperature-controlled separation conditions are validated using a comprehensive separation performance evaluation model. Combining resolution, peak capacity, and column efficiency metrics, an iterative optimization algorithm is used to further fine-tune the temperature parameters, generating the final highly selective separation conditions. Through comprehensive evaluation and iterative optimization, the high selectivity and efficiency of the separation conditions are ensured. This process not only improves the separation performance but also provides a reliable foundation for subsequent analyses, ensuring the accuracy and reliability of the analytical results.

[0071] After obtaining the initial optimized temperature-controlled separation conditions, it is necessary to verify whether these conditions can achieve the expected separation effect. A comprehensive separation effect evaluation model is used, which combines multiple indicators such as resolution, peak capacity, and column efficiency to comprehensively evaluate the separation conditions. Inputting data collected during the separation process, including retention time, peak shape, and resolution, the model outputs a comprehensive evaluation result of the current separation conditions. If the evaluation results show that the separation effect is not ideal, an iterative optimization algorithm is used to further fine-tune the temperature parameters. The iterative optimization algorithm gradually optimizes the temperature parameters through multiple adjustments and verifications to ensure that the separation conditions reach the optimal state. This may involve adjusting the rate of temperature increase and decrease, the maximum and minimum temperature range, etc., until the separation effect reaches the expected level. For example, through multiple iterations, it may be found that raising the temperature to 60°C and keeping it constant can significantly improve resolution and peak capacity. Through this series of optimizations and verifications, highly selective separation conditions are finally generated, ensuring efficient and stable separation in the liquid chromatography system. This process not only improves separation efficiency but also ensures the integrity of the target compounds and the accuracy of the analytical results.

[0072] A comprehensive evaluation model for separation performance may employ multi-objective optimization algorithms, such as genetic algorithms (GA), to evaluate and optimize separation conditions. Genetic algorithms are optimization algorithms based on natural selection and genetic mechanisms, suitable for multi-objective optimization problems. The model input includes separation performance data, and the output is the optimized temperature control parameters.

[0073] Iterative optimization algorithm: Through multiple adjustments and verifications, temperature parameters are progressively optimized to ensure optimal separation conditions. The iterative process may involve adjusting the rate of temperature increase and decrease, the maximum and minimum temperature ranges, etc., until the separation effect meets expectations. Through this series of optimizations and verifications, highly selective separation conditions are ultimately generated, ensuring efficient and stable separation in the liquid chromatography system. This process not only improves separation efficiency but also ensures the integrity of the target compounds and the accuracy of the analytical results.

[0074] S204, based on the high-selectivity separation conditions, adopts online detection technology based on ultraviolet-visible spectroscopy and mass spectrometry, combined with signal enhancement algorithm, to detect the separated chromatographic peaks in real time. Through multi-channel signal fusion technology, ultraviolet-visible spectroscopy and mass spectrometry data are dynamically integrated to obtain a high signal-to-noise ratio detection signal; The ultraviolet detector employs diode array detection (DAD) to acquire full-spectral information, while mass spectrometry utilizes an electrospray ionization (ESI) source for high-sensitivity detection. The signal enhancement algorithm uses wavelet transform to remove high-frequency noise, and multi-channel data fusion integrates complementary information through time alignment and weighted averaging, simultaneously obtaining the spectral characteristics and molecular weight information of compounds. This improves the detection capability of low-abundance compounds and provides more comprehensive data support for substance identification.

[0075] Specifically, based on the high-selectivity separation conditions, the detection parameters of ultraviolet-visible spectroscopy and mass spectrometry can be initialized, and preliminary detection conditions can be generated through the detection efficiency prediction model; In this step, initial detection parameters for UV-Vis spectroscopy and mass spectrometry are set based on the high-selectivity separation conditions. These parameters may include wavelength, scan rate, and sensitivity. A detection efficiency prediction model can be used to predict the impact of different parameter settings on detection efficiency, thereby generating a preliminary set of detection conditions. This process ensures that the detection system is in optimal condition from the outset to accurately capture the separated compound signal. Initializing detection parameters is crucial for ensuring the efficient operation of the detection system. Appropriate parameter settings can improve detection sensitivity and accuracy, ensuring high-quality detection signals in subsequent steps. This process not only improves detection efficiency but also provides a reliable foundation for subsequent signal processing and analysis.

[0076] In liquid chromatography systems, initializing detection parameters is a crucial step in ensuring detection accuracy and sensitivity. First, the detection requirements for the target compound are determined based on high-selectivity separation conditions. These requirements may include detection sensitivity, resolution, and linear range. To meet these requirements, the detection parameters for UV-Vis spectroscopy and mass spectrometry need to be initialized. UV-Vis spectroscopy detection parameters may include detection wavelength, bandwidth, and integration time. Preliminary detection conditions can be generated using a detection efficiency prediction model combined with the absorbance characteristics of the target compound. Mass spectrometry detection parameters may include ionization mode, scan rate, and mass range. By analyzing the molecular weight and structural characteristics of the target compound, the mass spectrometry detection parameters can be optimized.

[0077] Detection efficiency prediction model: Machine learning algorithms such as Support Vector Machine (SVM) or Random Forest may be used to construct the detection efficiency prediction model. The model input includes the feature parameters of the target compound and the separation conditions, and the output is the optimized detection parameters.

[0078] Optimization of UV-Vis spectral parameters: Response surface methodology (RSM) can be used to optimize the detection parameters of UV-Vis spectroscopy. RSM is a statistical technique that can optimize the response of multivariate systems through experimental design and regression analysis.

[0079] In a liquid chromatography experiment, a complex mixture of organic compounds, including various polycyclic aromatic hydrocarbons (PAHs), needs to be detected. The separation of these compounds has been achieved through optimized mobile phase and column conditions. To ensure detection accuracy, the detection parameters for UV-Vis spectroscopy and mass spectrometry were first initialized based on highly selective separation conditions. A detection efficiency prediction model, trained on historical experimental data and the spectral characteristics of the compounds, was used. Model inputs included the molecular structure, polarity, and known absorbance properties of the target compounds. Model prediction determined the optimal detection wavelength for UV-Vis spectroscopy to be 254 nm, which is the characteristic absorption wavelength of PAHs. Electron spray ionization (ESI) mode was selected for mass spectrometry, suitable for analyzing highly polar compounds. These initialization steps generated preliminary detection conditions, ensuring efficient capture of the characteristic signals of the target compounds in subsequent detection processes.

[0080] For the initial detection conditions, a signal enhancement algorithm is used, combined with real-time chromatographic peak shape monitoring data, to optimize the detection signal quality. Through noise filtering technology, a preliminary high-quality detection signal is generated. In this step, signal enhancement algorithms and real-time chromatographic peak shape monitoring data are used to optimize the quality of the detection signal. Signal enhancement algorithms improve signal intensity and clarity, while noise filtering techniques remove background noise and interference from the detection signal. These techniques generate a preliminary high-quality detection signal, laying the foundation for subsequent signal fusion and analysis. Optimizing the detection signal quality is crucial to ensuring the accuracy of the analytical results. Through signal enhancement and noise filtering, the signal-to-noise ratio can be improved, reducing errors and interference. This process not only improves the sensitivity and accuracy of detection but also provides a high-quality foundation for subsequent signal fusion.

[0081] After initializing the detection parameters, it is crucial to ensure the detection signal quality is high enough for accurate identification and quantification of the target compound. Signal enhancement algorithms play a key role in this process. Signal processing techniques such as wavelet transform can be used to enhance the characteristics of the detection signal and remove background noise and interference signals. Real-time chromatographic peak shape monitoring data provides important information about the detection signal quality. By analyzing this data, detection parameters can be dynamically adjusted to optimize signal quality. Noise filtering techniques, such as Kalman filtering or adaptive filtering, can further remove noise from the detection signal, generating a preliminary high-quality detection signal.

[0082] Signal enhancement algorithm: Wavelet transform is a commonly used signal enhancement technique that can analyze signals simultaneously in the time and frequency domains, and is suitable for removing noise and enhancing signal features.

[0083] Noise filtering techniques: Kalman filtering is a recursive filtering algorithm that can estimate the true state of a signal in a dynamic system and is suitable for real-time signal processing.

[0084] During the detection process, significant background noise was found in the detection signal, affecting the identification of the target compound. To improve signal quality, wavelet transform was employed as a signal enhancement algorithm. Wavelet transform can analyze signals simultaneously in the time and frequency domains, making it suitable for noise removal and signal feature enhancement. Real-time acquired chromatographic peak shape monitoring data was input into the wavelet transform algorithm, removing most of the background noise. Subsequently, Kalman filtering was used to further optimize the signal quality. Kalman filtering is a recursive filtering algorithm capable of estimating the true state of a signal in dynamic systems, making it suitable for real-time signal processing. Through these steps, a preliminary high-quality detection signal was generated, ensuring accurate identification of the characteristic peaks of the target compound.

[0085] For the initial high-quality detection signal, multi-channel signal fusion technology is used to combine ultraviolet-visible spectroscopy and mass spectrometry data to dynamically integrate multi-source signals and generate an initial fused signal through a signal alignment algorithm; In this stage, multi-channel signal fusion technology is used to integrate detection signals from UV-Vis spectroscopy and mass spectrometry. Signal alignment algorithms are used to ensure the consistency of data from different signal sources in terms of time and intensity, thereby generating a preliminary fused signal. This process enables the acquisition of more comprehensive and accurate compound information. Multi-channel signal fusion is crucial for improving the completeness and accuracy of detection information. By integrating data from different signal sources, richer compound information can be obtained, increasing the depth and breadth of the analysis. This process not only improves detection accuracy but also provides a reliable foundation for subsequent signal optimization.

[0086] After obtaining high-quality detection signals, it is necessary to integrate data from different detection channels. UV-Vis spectroscopy and mass spectrometry provide different information about compounds; the former is mainly used to detect the absorbance properties of compounds, while the latter is used to analyze the molecular weight and structure of compounds. Multichannel signal fusion technology can integrate this information to generate more comprehensive compound characterization data. Signal alignment algorithms play a crucial role in this process. Since the data acquisition rates and timelines of different detection channels may be inconsistent, signal alignment algorithms are needed to synchronize these data. Through techniques such as cross-correlation or dynamic time warping (DTW), precise signal alignment can be achieved, generating a preliminary fused signal.

[0087] Multichannel signal fusion techniques: Principal component analysis (PCA) may be used for signal fusion. PCA is a dimensionality reduction technique that can extract the main features from multidimensional data and is suitable for integrating multichannel signals.

[0088] Signal alignment algorithm: Dynamic Time Warping (DTW) is a commonly used signal alignment algorithm that can handle nonlinear alignment problems between different time series and is suitable for the synchronization of multi-channel signals.

[0089] After obtaining a high-quality detection signal, the UV-Vis spectroscopy and mass spectrometry data need to be integrated to obtain more comprehensive compound information. UV-Vis spectroscopy provides the absorption characteristics of the compound, while mass spectrometry provides molecular mass and structural information. Principal component analysis (PCA) was used as the multi-channel signal fusion technique. PCA can extract the main features from multi-dimensional data and is suitable for integrating multi-channel signals. Through PCA, the UV-Vis spectroscopy and mass spectrometry data were integrated to generate a fused signal containing the main features. To ensure data synchronization, the Dynamic Time Warping (DTW) algorithm was used for signal alignment. DTW can handle the nonlinear alignment problem between different time series, ensuring that the time axes of the UV and mass spectrometry data are consistent. Through these steps, a preliminary fused signal was generated, providing a foundation for subsequent signal quality optimization.

[0090] For the initial fused signal, the signal-to-noise ratio (SNR) evaluation index is used to optimize the signal quality, and a dynamic adjustment algorithm is used to generate the final high SNR detection signal.

[0091] In the final step, the signal-to-noise ratio (SNR) of the preliminary fused signal is evaluated. Through dynamic adjustment algorithms, signal quality can be further optimized to ensure a high SNR in the final detection signal. This process ensures the clarity and accuracy of the detection signal, providing reliable data support for subsequent analysis and decision-making. Optimizing signal quality is crucial for ensuring the reliability of analysis results. SNR evaluation and dynamic adjustment improve signal clarity and accuracy while reducing errors and interference. This process not only enhances detection sensitivity and accuracy but also provides a high-quality foundation for subsequent analysis.

[0092] After generating the initial fused signal, its quality needs to be evaluated to ensure a sufficiently high signal-to-noise ratio (SNR). SNR is a crucial indicator of signal quality; a higher SNR means a higher proportion of useful information and less noise interference. The quality of the fused signal is analyzed using the SNR evaluation metric. If the SNR is unsatisfactory, dynamic adjustment algorithms are used to optimize the signal quality. This may involve adjusting detection parameters, recalibrating the detection equipment, or applying more advanced signal processing techniques. This process generates a final high SNR detection signal, ensuring the accuracy and reliability of subsequent analyses.

[0093] Signal-to-noise ratio (SNR) evaluation metrics: Signal quality may be evaluated using the SNR formula, which is typically defined as the ratio of signal power to noise power.

[0094] Dynamic adjustment algorithm: Adaptive filtering is a dynamic adjustment algorithm that can automatically adjust filter parameters according to changes in the input signal, suitable for real-time signal quality optimization. Through these steps, efficient online detection can be achieved in liquid chromatography systems, ensuring that the separated compounds can be accurately identified and quantified. This process not only improves the sensitivity and accuracy of detection but also provides reliable data support for subsequent separation effect evaluation.

[0095] S205. Based on the high signal-to-noise ratio detection signal, a machine learning-based separation effect evaluation model is used to evaluate the preliminary separation effect by combining the retention time, peak shape symmetry, and separation degree index of the target compound.

[0096] A deep neural network model is constructed, taking 12-dimensional features as input, including retention time, peak symmetry factor, and separation degree, and outputting a separation performance score (0-100). The model is trained on tens of thousands of historical chromatograms, enabling it to identify common problems such as peak overlap and tailing, achieving an objective quantitative evaluation of separation quality, providing clear direction for method optimization, and reducing the subjectivity of human judgment.

[0097] Specifically, based on the high signal-to-noise ratio detection signal, data preprocessing techniques can be used, combined with the retention time, peak symmetry, and separation index of the target compound, to generate a preliminary separation effect dataset; In this step, high signal-to-noise ratio detection signals are used for data preprocessing. Data preprocessing techniques include denoising, baseline correction, and signal normalization to ensure data accuracy and consistency. By combining the retention time, peak symmetry, and separation index of the target compound, a preliminary separation performance dataset is generated, providing a foundation for subsequent performance evaluation. Data preprocessing is crucial for ensuring the accuracy and consistency of analytical results. By removing noise and correcting signals, the quality and reliability of the data can be improved. This process not only improves the accuracy of the analysis but also provides a high-quality data foundation for subsequent performance evaluation.

[0098] In liquid chromatography (LC), the first step is to acquire high signal-to-noise ratio (SNR) detection signals containing the chromatographic peak information of the target compounds. To accurately assess separation performance, these signals require data preprocessing. The first step in preprocessing is baseline correction, which aims to eliminate baseline shifts caused by instrument drift or background noise. Moving averages or polynomial fitting methods can be used to correct the baseline, ensuring accurate start and end points for each peak. Next, peak detection and denoising are performed. Peak detection identifies the chromatographic peaks of each compound in the chromatogram and extracts their characteristic parameters, such as retention time, peak height, and peak area. Gaussian or Lorentz fitting methods can be used to accurately fit the peak shapes, thereby obtaining indicators such as peak symmetry and resolution. Denoising removes random noise from the signal through filtering techniques, improving signal clarity and accuracy.

[0099] These data preprocessing steps generated a preliminary separation performance dataset containing retention times, peak shape symmetry, and separation indexes of the target compounds. This dataset provides a foundation for subsequent separation performance evaluation and ensures the reliability and accuracy of the evaluation results.

[0100] For the preliminary separation effect dataset, a machine learning-based separation effect evaluation model was used, combined with the standard curve of the target compound, to train the model and generate preliminary evaluation results through cross-validation. In this step, a machine learning model is used to evaluate the initial separation performance dataset. By combining the standard curves of the target compounds, the model can identify and quantify various aspects of the separation performance. Cross-validation is used to verify the model's accuracy and robustness, ensuring the reliability of the evaluation results. Separation performance evaluation is crucial for ensuring the reliability of the analytical results. Through machine learning models, problems in the separation process can be quickly identified and optimized accordingly. This process not only improves the accuracy of the separation performance but also provides a reliable foundation for subsequent optimization.

[0101] After generating the initial separation performance dataset, it is necessary to evaluate the quality of the separation results. For this purpose, a machine learning-based separation performance evaluation model is employed. This model predicts the quality of the separation performance by learning the standard curve of the target compound. The standard curve is obtained by measuring compound samples with known concentrations and reflects the relationship between concentration and detection signal. The initial separation performance dataset is then input into the machine learning model for training. The model employs the random forest algorithm, an ensemble learning method capable of handling high-dimensional data and providing good predictive performance. To improve the model's generalization ability, cross-validation is used to validate the model. Cross-validation involves dividing the dataset into training and validation sets, repeatedly training and validating the model to ensure its stability and accuracy.

[0102] Preliminary evaluation results were generated through model training and cross-validation. These results include predicted retention times, peak shape symmetry, and separation indices for each target compound, providing a reference for subsequent optimization of separation conditions. This process not only improved the accuracy of the evaluation but also provided data support for further optimization of separation conditions.

[0103] Based on the preliminary evaluation results, a dynamic adjustment algorithm is used, combined with the separation effect evaluation index, to optimize the separation conditions. Through a feedback control mechanism, a preliminary optimized separation effect is generated. In this stage, based on the preliminary evaluation results, the separation conditions are optimized using a dynamic adjustment algorithm. Combined with separation performance evaluation indicators, aspects requiring improvement can be identified and adjusted through a feedback control mechanism, generating preliminary optimized separation results. Optimizing separation conditions is key to improving separation efficiency and analytical accuracy. Through dynamic adjustment and feedback control, separation efficiency and effectiveness can be improved, ensuring accurate separation of the target compound. This process not only improves separation performance but also provides a reliable foundation for subsequent analysis.

[0104] After obtaining preliminary evaluation results, it was found that the separation effect of some target compounds was not ideal, possibly due to insufficient optimization of chromatographic conditions. To further improve the separation effect, a dynamic adjustment algorithm was used to optimize the separation conditions. The dynamic adjustment algorithm can adjust chromatographic parameters, such as mobile phase composition, flow rate, and temperature, in real time based on the evaluation results. Combined with separation effect evaluation indicators, such as resolution and peak capacity, the chromatographic conditions are optimized. Through a feedback control mechanism, the separation effect can be monitored in real time, and chromatographic parameters can be dynamically adjusted based on the monitoring data. The core of the feedback control mechanism is to adjust experimental conditions based on real-time data to achieve the best separation effect. This goal can be achieved using a PID controller or an adaptive control algorithm. After multiple iterations of optimization, preliminary optimized separation results were generated. This process not only improved the efficiency and accuracy of separation but also provided a better foundation for subsequent precision evaluation.

[0105] For the initial optimized separation effect, an accuracy evaluation index is used to further optimize the separation effect. Through an error correction mechanism, the final separation effect evaluation result is generated.

[0106] The final step involves an accuracy assessment of the preliminary optimized separation results. An error correction mechanism identifies and corrects errors in the separation process, ensuring high accuracy and reliability of the final separation performance evaluation. Accuracy assessment and error correction are crucial for ensuring the reliability of separation results and analytical outcomes. By identifying and correcting errors, the accuracy and consistency of the separation results can be improved. This process not only enhances the separation performance but also provides a high-quality foundation for subsequent analysis and decision-making.

[0107] To ensure the accuracy of the separation, precision evaluation metrics were used to validate the optimized results. These metrics included error analysis and repeatability testing to ensure the stability and reliability of the separation. Error analysis identified systematic errors by comparing experimental results with a standard curve and made corresponding adjustments. Repeatability testing verified the stability of the separation through multiple repeated experiments. The optimized separation was then evaluated for precision, and an error correction mechanism was used to adjust for any deviations. This mechanism could be implemented using linear regression or a deviation compensation algorithm to ensure the accuracy of the separation. By comparing experimental results with a standard curve, any systematic errors could be identified and adjusted accordingly. Ultimately, high-precision separation evaluation results were generated, providing reliable data support for the quantitative analysis of target compounds and subsequent research. This process not only improved the efficiency and accuracy of the separation but also provided important reference for the optimization of liquid chromatography systems.

[0108] As can be seen, by pretreating the sample according to its physicochemical properties, a high-purity enriched sample is obtained; by optimizing the composition ratio of the mobile phase according to the composition characteristics of the high-purity enriched sample, an optimized mobile phase composition is obtained; based on the optimized mobile phase composition, the optimal chromatographic column type is selected, and dynamic temperature control separation is performed using dynamic temperature control technology to obtain highly selective separation conditions; based on the highly selective separation conditions, the separated chromatographic peaks are detected in real time to obtain a high signal-to-noise ratio detection signal; based on the high signal-to-noise ratio detection signal, the preliminary separation effect is evaluated, thereby improving the enrichment efficiency and separation selectivity of the sample, and thus enhancing the overall separation accuracy and reliability.

[0109] Another embodiment of the present invention provides a preliminary separation system for a liquid chromatography system, see [link to relevant documentation]. Figure 3 The system may include: The processing module 301 is used to pre-treat the target sample according to its physicochemical properties using a functionalized solid-phase extraction technology based on nanomaterials. By combining porous nanomaterials and dynamic adsorption optimization algorithms with the characteristic parameters of the target compound, a high-purity enriched sample is obtained. The optimization module 302 is used to optimize the composition ratio of the mobile phase based on the composition characteristics of the high-purity enriched sample by using a mobile phase composition optimization algorithm based on response surface methodology and combined with a retention behavior prediction model of the target compound. Through microfluidic mixing technology, the mixing uniformity and flow rate stability of the mobile phase are dynamically adjusted to obtain the optimized mobile phase composition. The separation module 303 is used to select the optimal column type based on the optimized mobile phase composition and a machine learning-based column selection model, combined with the separation requirements of the target compound. It performs dynamic temperature control separation through dynamic temperature control technology, and adjusts the column temperature in real time based on the thermal stability of the target compound to obtain highly selective separation conditions. The detection module 304 is used to detect the separated chromatographic peaks in real time by using online detection technology based on ultraviolet-visible spectroscopy and mass spectrometry, combined with signal enhancement algorithms, according to the high-selectivity separation conditions. Through multi-channel signal fusion technology, it dynamically integrates ultraviolet-visible spectroscopy and mass spectrometry data to obtain a high signal-to-noise ratio detection signal. Evaluation module 305 is used to evaluate the initial separation effect based on the high signal-to-noise ratio detection signal using a machine learning-based separation effect evaluation model, combined with the retention time, peak shape symmetry and separation degree index of the target compound.

[0110] As can be seen, by pretreating the sample according to its physicochemical properties, a high-purity enriched sample is obtained; by optimizing the composition ratio of the mobile phase according to the composition characteristics of the high-purity enriched sample, an optimized mobile phase composition is obtained; based on the optimized mobile phase composition, the optimal chromatographic column type is selected, and dynamic temperature control separation is performed using dynamic temperature control technology to obtain highly selective separation conditions; based on the highly selective separation conditions, the separated chromatographic peaks are detected in real time to obtain a high signal-to-noise ratio detection signal; based on the high signal-to-noise ratio detection signal, the preliminary separation effect is evaluated, thereby improving the enrichment efficiency and separation selectivity of the sample, and thus enhancing the overall separation accuracy and reliability.

[0111] This invention also provides a storage medium storing a computer program, wherein the computer program is configured to execute the steps in any of the above method embodiments when running.

[0112] Specifically, in this embodiment, the storage medium can be configured to store a computer program for performing the following steps: S201. Based on the physicochemical properties of the target sample, a functionalized solid-phase extraction technology based on nanomaterials is used to pretreat the sample. Through porous nanomaterials and dynamic adsorption optimization algorithms, combined with the characteristic parameters of the target compound, a high-purity enriched sample is obtained. S202. Based on the compositional characteristics of the high-purity enriched sample, a mobile phase composition optimization algorithm based on response surface methodology is adopted, combined with a retention behavior prediction model of the target compound, to optimize the composition ratio of the mobile phase. Through microfluidic mixing technology, the mixing uniformity and flow rate stability of the mobile phase are dynamically adjusted to obtain the optimized mobile phase composition. S203, based on the optimized mobile phase composition, adopts a machine learning-based column selection model, combined with the separation requirements of the target compound, to select the optimal column type, and performs dynamic temperature control separation through dynamic temperature control technology. Combined with the thermal stability of the target compound, the column temperature is adjusted in real time to obtain highly selective separation conditions. S204, based on the high-selectivity separation conditions, adopts online detection technology based on ultraviolet-visible spectroscopy and mass spectrometry, combined with signal enhancement algorithm, to detect the separated chromatographic peaks in real time. Through multi-channel signal fusion technology, ultraviolet-visible spectroscopy and mass spectrometry data are dynamically integrated to obtain a high signal-to-noise ratio detection signal; S205. Based on the high signal-to-noise ratio detection signal, a machine learning-based separation effect evaluation model is used to evaluate the preliminary separation effect by combining the retention time, peak shape symmetry, and separation degree index of the target compound.

[0113] As can be seen, by pretreating the sample according to its physicochemical properties, a high-purity enriched sample is obtained; by optimizing the composition ratio of the mobile phase according to the composition characteristics of the high-purity enriched sample, an optimized mobile phase composition is obtained; based on the optimized mobile phase composition, the optimal chromatographic column type is selected, and dynamic temperature control separation is performed using dynamic temperature control technology to obtain highly selective separation conditions; based on the highly selective separation conditions, the separated chromatographic peaks are detected in real time to obtain a high signal-to-noise ratio detection signal; based on the high signal-to-noise ratio detection signal, the preliminary separation effect is evaluated, thereby improving the enrichment efficiency and separation selectivity of the sample, and thus enhancing the overall separation accuracy and reliability.

[0114] This invention also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to perform the steps in any of the above method embodiments.

[0115] Specifically, the aforementioned electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the aforementioned processor, and the input / output device is connected to the aforementioned processor.

[0116] Specifically, in this embodiment, the processor can be configured to perform the following steps via a computer program: S201. Based on the physicochemical properties of the target sample, a functionalized solid-phase extraction technology based on nanomaterials is used to pretreat the sample. Through porous nanomaterials and dynamic adsorption optimization algorithms, combined with the characteristic parameters of the target compound, a high-purity enriched sample is obtained. S202. Based on the compositional characteristics of the high-purity enriched sample, a mobile phase composition optimization algorithm based on response surface methodology is adopted, combined with a retention behavior prediction model of the target compound, to optimize the composition ratio of the mobile phase. Through microfluidic mixing technology, the mixing uniformity and flow rate stability of the mobile phase are dynamically adjusted to obtain the optimized mobile phase composition. S203, based on the optimized mobile phase composition, adopts a machine learning-based column selection model, combined with the separation requirements of the target compound, to select the optimal column type, and performs dynamic temperature control separation through dynamic temperature control technology. Combined with the thermal stability of the target compound, the column temperature is adjusted in real time to obtain highly selective separation conditions. S204, based on the high-selectivity separation conditions, adopts online detection technology based on ultraviolet-visible spectroscopy and mass spectrometry, combined with signal enhancement algorithm, to detect the separated chromatographic peaks in real time. Through multi-channel signal fusion technology, ultraviolet-visible spectroscopy and mass spectrometry data are dynamically integrated to obtain a high signal-to-noise ratio detection signal; S205. Based on the high signal-to-noise ratio detection signal, a machine learning-based separation effect evaluation model is used to evaluate the preliminary separation effect by combining the retention time, peak shape symmetry, and separation degree index of the target compound.

[0117] As can be seen, by pretreating the sample according to its physicochemical properties, a high-purity enriched sample is obtained; by optimizing the composition ratio of the mobile phase according to the composition characteristics of the high-purity enriched sample, an optimized mobile phase composition is obtained; based on the optimized mobile phase composition, the optimal chromatographic column type is selected, and dynamic temperature control separation is performed using dynamic temperature control technology to obtain highly selective separation conditions; based on the highly selective separation conditions, the separated chromatographic peaks are detected in real time to obtain a high signal-to-noise ratio detection signal; based on the high signal-to-noise ratio detection signal, the preliminary separation effect is evaluated, thereby improving the enrichment efficiency and separation selectivity of the sample, and thus enhancing the overall separation accuracy and reliability.

[0118] The above description, based on the embodiments shown in the figures, details the structure, features, and effects of the present invention. The above description is only a preferred embodiment of the present invention, but the present invention is not limited to the scope of implementation shown in the figures. Any changes made in accordance with the concept of the present invention, or equivalent embodiments modified to have equivalent changes, that do not exceed the spirit covered by the specification and figures, should be within the protection scope of the present invention.

Claims

1. A preliminary separation method for a liquid chromatography system, characterized in that, The method includes: Based on the physicochemical properties of the target sample, a functionalized solid-phase extraction technology based on nanomaterials was used to pretreat the sample. Through porous nanomaterials and dynamic adsorption optimization algorithms, combined with the characteristic parameters of the target compound, a high-purity enriched sample was obtained. Based on the compositional characteristics of the high-purity enriched sample, a mobile phase composition optimization algorithm based on response surface methodology was adopted, combined with a retention behavior prediction model of the target compound, to optimize the composition ratio of the mobile phase. Through microfluidic mixing technology, the mixing uniformity and flow rate stability of the mobile phase were dynamically adjusted to obtain the optimized mobile phase composition. Based on the optimized mobile phase composition, a machine learning-based column selection model is used to select the optimal column type in combination with the separation requirements of the target compound. Dynamic temperature control separation is performed through dynamic temperature control technology. The column temperature is adjusted in real time in combination with the thermal stability of the target compound to obtain highly selective separation conditions. Based on the high-selectivity separation conditions, an online detection technology based on ultraviolet-visible spectroscopy and mass spectrometry is adopted, combined with a signal enhancement algorithm, to detect the separated chromatographic peaks in real time. Through multi-channel signal fusion technology, ultraviolet-visible spectroscopy and mass spectrometry data are dynamically integrated to obtain a high signal-to-noise ratio detection signal. Based on the high signal-to-noise ratio detection signal, a machine learning-based separation effect evaluation model is used, combined with the retention time, peak shape symmetry and separation degree index of the target compound, to evaluate the preliminary separation effect.

2. The method according to claim 1, characterized in that, Based on the physicochemical properties of the target sample, a functionalized solid-phase extraction technique based on nanomaterials is used to pretreat the sample. Through porous nanomaterials and dynamic adsorption optimization algorithms, combined with the characteristic parameters of the target compound, a high-purity enriched sample is obtained, including: Based on the physicochemical properties of the target sample, a prediction model based on molecular descriptors is used to calculate its adsorption characteristic parameters. A preliminary range of adsorption conditions is generated through a multi-objective optimization algorithm. For the initial adsorption condition range, a functionalized solid-phase extraction technology based on nanomaterials was adopted. Porous nanomaterials were selected, and a preliminary functionalized solid-phase extraction material was generated through surface modification technology. Based on the preliminary functionalized solid-phase extraction materials, a dynamic adsorption optimization algorithm was used, combined with the characteristic parameters of the target compound, to adjust the adsorption conditions in real time. Through adsorption efficiency monitoring technology, a preliminary enriched sample was generated. For the initial enriched sample, an elution optimization algorithm is used to optimize the elution conditions by combining the desorption characteristics of the target compound. The final high-purity enriched sample is generated by using purity evaluation indicators.

3. The method according to claim 2, characterized in that, Based on the compositional characteristics of the high-purity enriched sample, a mobile phase composition optimization algorithm based on response surface methodology is employed, combined with a retention behavior prediction model of the target compound, to optimize the composition ratio of the mobile phase. Microfluidic mixing technology is used to dynamically adjust the mixing uniformity and flow rate stability of the mobile phase, resulting in the optimized mobile phase composition, including: Based on the compositional characteristics of the high-purity enriched sample, a mobile phase composition optimization algorithm based on response surface methodology is adopted, combined with a retention behavior prediction model of the target compound, to generate a preliminary range of mobile phase composition. For the initial range of mobile phase composition, the response surface methodology was used to optimize the mobile phase composition ratio by combining the retention time and peak shape symmetry of the target compound. Through a multi-objective optimization mechanism, the initial optimized mobile phase composition was generated. For the initial optimized mobile phase composition, microfluidic mixing technology is used, combined with dynamic mixing uniformity monitoring data, to adjust the mixing ratio and flow rate in real time, and generate an initial uniform mobile phase through a feedback control mechanism; For the initial homogeneous mobile phase, a stability evaluation index is used to optimize the stability of the mobile phase. Then, a dynamic adjustment algorithm is used to generate the final optimized mobile phase composition.

4. The method according to claim 3, characterized in that, Based on the optimized mobile phase composition, a machine learning-based column selection model is used to select the optimal column type, taking into account the separation requirements of the target compound. Dynamic temperature control separation is then performed using dynamic temperature control technology, adjusting the column temperature in real time based on the thermal stability of the target compound to obtain highly selective separation conditions, including: Based on the optimized mobile phase composition, a machine learning-based column selection model was used to analyze the separation potential of different columns in conjunction with the separation requirements of the target compounds. Preliminary column selection results were generated through column efficiency evaluation indicators. Based on the column selection results and the thermal stability of the target compound, the dynamic temperature control parameters are initialized, and the initial temperature control parameters are generated through the temperature control efficiency prediction model. Based on the initial temperature control parameters, the chromatographic separation process is started. An embedded temperature sensor is used to monitor the column temperature in real time and collect separation effect data simultaneously. Through feedback control algorithm, combined with real-time separation effect data, the temperature parameters are dynamically adjusted to generate preliminary optimized temperature control separation conditions. Based on the preliminary optimized temperature-controlled separation conditions, a comprehensive evaluation model for separation effect was adopted, combining separation degree, peak capacity and column efficiency indicators to verify the separation conditions. Through iterative optimization algorithm, the temperature parameters were further fine-tuned to generate the final high-selectivity separation conditions.

5. The method according to claim 4, characterized in that, The method employs online detection technology based on ultraviolet-visible spectroscopy and mass spectrometry, combined with signal enhancement algorithms, to detect the separated chromatographic peaks in real time under highly selective separation conditions. Through multi-channel signal fusion technology, ultraviolet-visible spectroscopy and mass spectrometry data are dynamically integrated to obtain a high signal-to-noise ratio detection signal, including: Based on the high-selectivity separation conditions, the detection parameters of ultraviolet-visible spectroscopy and mass spectrometry are initialized, and preliminary detection conditions are generated through the detection efficiency prediction model. For the initial detection conditions, a signal enhancement algorithm is used, combined with real-time chromatographic peak shape monitoring data, to optimize the detection signal quality. Through noise filtering technology, a preliminary high-quality detection signal is generated. For the initial high-quality detection signal, multi-channel signal fusion technology is used to combine ultraviolet-visible spectroscopy and mass spectrometry data to dynamically integrate multi-source signals and generate an initial fused signal through a signal alignment algorithm; For the initial fused signal, the signal-to-noise ratio (SNR) evaluation index is used to optimize the signal quality, and a dynamic adjustment algorithm is used to generate the final high SNR detection signal.

6. The method according to claim 5, characterized in that, The preliminary separation effect is evaluated using a machine learning-based separation performance evaluation model based on the high signal-to-noise ratio detection signal, combined with the retention time, peak shape symmetry, and separation degree index of the target compound. This evaluation includes: Based on the high signal-to-noise ratio detection signal, data preprocessing techniques are used, combined with the retention time, peak shape symmetry and separation index of the target compound, to generate a preliminary separation effect dataset; For the preliminary separation effect dataset, a machine learning-based separation effect evaluation model was used, combined with the standard curve of the target compound, to train the model and generate preliminary evaluation results through cross-validation. Based on the preliminary evaluation results, a dynamic adjustment algorithm is used, combined with the separation effect evaluation index, to optimize the separation conditions. Through a feedback control mechanism, a preliminary optimized separation effect is generated. For the initial optimized separation effect, an accuracy evaluation index is used to further optimize the separation effect. Through an error correction mechanism, the final separation effect evaluation result is generated.

7. A preliminary separation system for a liquid chromatography system, characterized in that, The system includes: The processing module is used to pre-treat the target sample based on the physicochemical properties of the sample using a functionalized solid-phase extraction technology based on nanomaterials. Through porous nanomaterials and dynamic adsorption optimization algorithms, combined with the characteristic parameters of the target compound, a high-purity enriched sample is obtained. The optimization module is used to optimize the composition ratio of the mobile phase based on the composition characteristics of the high-purity enriched sample by using a mobile phase composition optimization algorithm based on response surface methodology and combined with a retention behavior prediction model of the target compound. Through microfluidic mixing technology, the mixing uniformity and flow rate stability of the mobile phase are dynamically adjusted to obtain the optimized mobile phase composition. The separation module is used to select the optimal column type based on the optimized mobile phase composition and a machine learning-based column selection model, combined with the separation requirements of the target compound. It performs dynamic temperature-controlled separation using dynamic temperature control technology, and adjusts the column temperature in real time based on the thermal stability of the target compound to obtain highly selective separation conditions. The detection module is used to detect the separated chromatographic peaks in real time by using online detection technology based on ultraviolet-visible spectroscopy and mass spectrometry, combined with signal enhancement algorithms, according to the high-selectivity separation conditions. Through multi-channel signal fusion technology, ultraviolet-visible spectroscopy and mass spectrometry data are dynamically integrated to obtain a high signal-to-noise ratio detection signal. The evaluation module is used to evaluate the initial separation effect based on a machine learning-based separation effect evaluation model, combined with the retention time, peak shape symmetry and separation degree index of the target compound, based on the high signal-to-noise ratio detection signal.

8. The system according to claim 7, characterized in that, The processing module is specifically used for: Based on the physicochemical properties of the target sample, a prediction model based on molecular descriptors is used to calculate its adsorption characteristic parameters. A preliminary range of adsorption conditions is generated through a multi-objective optimization algorithm. For the initial adsorption condition range, a functionalized solid-phase extraction technology based on nanomaterials was adopted. Porous nanomaterials were selected, and a preliminary functionalized solid-phase extraction material was generated through surface modification technology. Based on the preliminary functionalized solid-phase extraction materials, a dynamic adsorption optimization algorithm was used, combined with the characteristic parameters of the target compound, to adjust the adsorption conditions in real time. Through adsorption efficiency monitoring technology, a preliminary enriched sample was generated. For the initial enriched sample, an elution optimization algorithm is used to optimize the elution conditions by combining the desorption characteristics of the target compound. The final high-purity enriched sample is generated by using purity evaluation indicators.

9. A storage medium, characterized in that, The storage medium stores a computer program, wherein the computer program is configured to execute the method of any one of claims 1-6 when it is run.

10. An electronic device comprising a memory and a processor, characterized in that, The memory stores a computer program, and the processor is configured to run the computer program to perform the method of any one of claims 1-6.

Images