High-purity dimethyl carbonate and propylene glycol synchronous refining and regulating system and method

Abstract

The present invention discloses a high-purity dimethyl carbonate and propylene glycol synchronous refining adjustment system and method, which relates to the technical field of high-purity dimethyl carbonate and propylene glycol synchronous refining, and comprises the following steps: at the beginning of the process, the reflux ratio control frequency of the initial distillation stage is first used to quickly remove low-boiling impurities and establish a thermal balance in the tower; when entering the switching process between the initial distillation stage and the refining stabilization stage, the thermal balance information of the system is collected in real time. The present invention introduces real-time monitoring and support vector machine models, intelligently identifies changes in thermal balance states, dynamically adjusts the reflux ratio control frequency, ensures gas-liquid mass transfer balance, avoids impurity entrainment, improves product purity, reduces the burden on downstream separation equipment, reduces energy consumption and costs, quickly restores system thermal balance, significantly improves production efficiency and product reliability, avoids the scrapping of a whole batch of products, and realizes an efficient and high-quality synchronous refining process.

Description

Technical Field

[0001] The invention relates to the technical field of synchronous refining of high-purity dimethyl carbonate and propylene glycol, and in particular to a synchronous refining regulating system and method for high-purity dimethyl carbonate and propylene glycol. Background Art

[0002] The simultaneous refining of high-purity dimethyl carbonate and propylene glycol refers to the simultaneous purification of dimethyl carbonate (DMC) and propylene glycol (PG) in the same process flow to improve the purity of both and meet the needs of industrial applications. This process usually uses processes such as azeotropic distillation, extractive distillation or reactive distillation to gradually separate impurities (such as water, alcohols or other by-products) in the mixture. Synchronous refining can optimize energy consumption and material consumption, and reduce production time and cost through integrated design. During the process, by controlling the temperature, pressure and reflux ratio in the distillation tower, DMC and PG are effectively separated at their respective boiling points or azeotropic points, thereby achieving efficient refining. This process is widely used in the production of lithium battery electrolytes and high-performance polyesters to ensure product purity and stability.

[0003] In the simultaneous refining of high-purity dimethyl carbonate (DMC) and propylene glycol (PG), the reflux ratio refers to the part of the condensed liquid that returns to the distillation tower as reflux liquid to carry out mass transfer and heat exchange with the rising steam, while the other part flows out of the tower as product discharge. The value of the reflux ratio is defined as the ratio of the amount of reflux liquid to the amount of discharge product. It plays a key role in the distillation process: a higher reflux ratio can improve the separation efficiency in the tower and make the separation of DMC and PG more thorough, but at the same time it will increase energy consumption; while a lower reflux ratio can save energy, but may lead to incomplete separation and reduce product purity. In the simultaneous refining process, precise control of the reflux ratio helps to achieve efficient separation at different stages, optimize energy consumption, and ensure that DMC and PG meet the required high purity requirements.

[0004] The prior art has the following deficiencies:

[0005] In the process of simultaneous refining of high-purity dimethyl carbonate and propylene glycol, the prior art usually uses the reflux ratio control frequency in the initial distillation stage to manage the switching between the initial distillation stage and the refining stabilization stage. The initial distillation stage requires high-frequency control (usually every 1-5 minutes) to quickly remove impurities and establish thermal balance in the tower. Therefore, in the switching stage, the high-frequency control strategy of the initial stage is usually continued to respond to system fluctuations in a timely manner. However, when the thermal balance stability of the system in the switching stage is poor, the reflux ratio control frequency in the initial distillation stage may not meet the demand. If the control frequency continues to be used and the reflux ratio is not adjusted in time, the impurities in the tower may not be completely separated, resulting in an increase in impurity entrainment. In particular, propylene glycol and other by-products may be distilled out with high-purity dimethyl carbonate, causing cross contamination. This will not only reduce the purity of high-purity dimethyl carbonate, but also increase the separation burden of the subsequent distillation tower, resulting in the accumulation of impurities in the system. Ultimately, this situation may undermine the quality of the entire production batch, increase process difficulty and cost, and have a significant impact on production efficiency.

[0006] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0007] The purpose of the present invention is to provide a high-purity dimethyl carbonate and propylene glycol synchronous refining regulation system and method. The present invention realizes intelligent division of switching stages and adaptive regulation of reflux ratio by introducing real-time monitoring and support vector machine model prediction. For the normal variable stage, the frequency of the initial distillation stage is continued to ensure process consistency and energy efficiency; for the variable stage, the reflux ratio is dynamically adjusted to maintain gas-liquid mass transfer balance, prevent impurity entrainment and cross contamination, and reduce the burden on downstream equipment, energy consumption and cost while improving the purity of DMC and PG products. Secondly, combined with intelligent prediction and dynamic feedback, the method accelerates thermal balance recovery, improves production efficiency and product reliability, avoids batch scrapping, and realizes high-quality and high-efficiency synchronous refining to solve the problems in the above-mentioned background technology.

[0008] In order to achieve the above object, the present invention provides the following technical solution: a method for synchronously refining and regulating high-purity dimethyl carbonate and propylene glycol, comprising the following steps:

[0009] At the beginning of the process, the reflux ratio of the initial distillation stage is used to adjust the frequency to quickly remove low-boiling impurities and establish heat balance in the tower;

[0010] When entering the switching process between the initial distillation stage and the refining stabilization stage, the thermal balance information of the system is collected in real time;

[0011] Extract key features representing the dynamic changes of the system from the collected thermal balance data, set a short-term monitoring window to conduct in-depth analysis of the extracted key features, and input the key feature data analyzed within the short-term monitoring window into the pre-trained support vector machine model to predict the thermal balance change trend in the current switching stage;

[0012] Based on the prediction results of the support vector machine model, the thermal equilibrium state of the system in the switching stage is divided into normal thermal equilibrium change type and thermal equilibrium change type.

[0013] For the switching stage classified as the normal change type of heat balance, the switching between the initial distillation stage and the refining stabilization stage is managed by continuing to control the reflux ratio frequency of the initial distillation stage;

[0014] For the switching stage classified as the thermal balance variable type, the actual reflux ratio control frequency of the switching stage is regulated based on the reflux ratio control frequency of the initial distillation stage and the thermal balance stability of the system in the switching stage to meet the dynamic change requirements of the thermal balance of the system in the switching stage.

[0015] Preferably, in the initial distillation stage of simultaneous refining of high-purity dimethyl carbonate and propylene glycol, the reflux ratio control frequency refers to the time interval for adjusting the reflux ratio in the system, that is, how often the reflux ratio is adjusted.

[0016] Preferably, key features representing dynamic changes of the system are extracted from the collected heat balance data, wherein the extracted key features include changes in the flow rate of rising steam and changes in the reflux liquid flow rate in the distillation tower. After acquisition, a short-term monitoring window is set to conduct in-depth analysis on the extracted changes in the flow rate of rising steam and changes in the reflux liquid flow rate in the distillation tower, and a steam flow rate change reference value and a reflux liquid fluctuation reference value are generated respectively. The steam flow rate change reference value is used to quantify the fluctuation amplitude and frequency of the rising steam flow rate in the distillation tower, and the stability of the gas phase mass transfer process in the reaction distillation tower is quantified. The reflux liquid fluctuation reference value is used to quantify the short-term fluctuation degree of the reflux liquid flow rate, which is used to evaluate the uniformity of liquid distribution in the distillation tower.

[0017] Preferably, after obtaining the steam flow rate change reference value and the reflux liquid fluctuation reference value generated by analyzing the key characteristics of the system's dynamic changes, the steam flow rate change reference value and the reflux liquid fluctuation reference value analyzed within the short-term monitoring window are input into a pre-trained support vector machine model to generate a thermal balance change evaluation index, and the thermal balance change trend in the current switching stage is predicted by the thermal balance change evaluation index.

[0018] Preferably, the thermal balance change evaluation index generated when predicting the thermal balance change trend of the system in the current switching stage within the short-term monitoring window by the support vector machine model is compared and analyzed with a preset reference threshold of the thermal balance change evaluation index, and the thermal balance state of the system in the switching stage is divided. The specific division steps are as follows:

[0019] If the thermal balance change evaluation index is greater than or equal to the thermal balance change evaluation index reference threshold, the thermal balance state of the system in the switching stage under the short-term monitoring window is classified as a thermal balance change type;

[0020] If the thermal balance change evaluation index is less than the thermal balance change evaluation index reference threshold, the thermal balance state of the system in the switching phase within the short-term monitoring window is classified as a thermal balance normal change type.

[0021] Preferably, the specific steps of setting a short-term monitoring window to conduct an in-depth analysis of the flow rate change of the rising steam in the extracted distillation tower to generate a reference value of the steam flow rate change are as follows:

[0022] Collect flow velocity change data and perform discrete processing. The expression of discrete processing is:

[0023] ,

[0024] In the formula, S i Indicates i The intensity of the steam velocity change during the period, V ( t i ) means in t i Steam flow rate at the time, represents the time interval between adjacent sampling points, ;

[0025] Using the steam flow rate variation intensity in each period S i Generate a nonlinear change cumulative curve to describe the overall steam flow rate fluctuation trend of the system. The expression of the nonlinear change cumulative curve is:

[0026] ,

[0027] In the formula, α is the nonlinear amplification factor, which controls the nonlinear amplification effect of flow velocity changes and highlights the impact of violent fluctuations. β is the time attenuation coefficient, which controls the rate at which the flow velocity changes with time and simulates the trend of the system tending to be stable. C ( t ) represents the cumulative value of flow velocity change at time t, It is an integral variable, an intermediate variable representing time, used to represent the state at each time point before time t. Indicated in The intensity of the steam flow rate change at each moment;

[0028] The reference value of steam flow rate change is calculated by calculating the difference of cumulative change and the integral of instantaneous fluctuation. The calculation expression is:

[0029] ,

[0030] In the formula, I v ( index ) is the reference value of steam flow rate change, indicating the intensity and frequency of steam flow rate fluctuation at the current moment. Indicates the cumulative change difference of steam flow rate between the current moment and the previous moment, which is used to capture instantaneous jump fluctuations. is a nonlinear weighting coefficient, which further amplifies the impact of violent fluctuations. T is the length of the short-term monitoring window.

[0031] Preferably, the specific steps of setting a short-term monitoring window to conduct in-depth analysis on the changes in the extracted reflux liquid flow rate to generate a reflux liquid fluctuation reference value are as follows:

[0032] In the short-term monitoring window, the data sequence of the reflux flow rate is collected in real time and recorded as F t ,in t =1,2,3,4,…, n , n is the total number of sampling time points within the short-term monitoring window, F t is the reflux flow rate at time t, and then the difference in reflux flow rate at each adjacent time point is calculated. The calculation expression is: ,

[0033] In the formula, It is the difference in flow rate change between adjacent sampling time points, reflecting the short-term fluctuation of the reflux liquid;

[0034] In order to further characterize the fluctuation characteristics of the reflux fluid, an exponential weighted fluctuation model is introduced to give higher weights to the most recent data. The definition of the weighted flow change is expressed as: ,

[0035] In the formula, W t is the weighted flow change at time t, is the weighting coefficient, between 0 and 1, W t-1 is the weighted flow change at the previous time point;

[0036] A nonlinear normalization formula is introduced to calculate the reference value of reflux liquid fluctuation. The calculation expression is:

[0037] ,

[0038] In the formula, RLI ( index ) is the reflux fluid fluctuation reference value, which indicates the degree of reflux fluid flow fluctuation within the short-term monitoring window. max ( W t ) is the maximum value of the weighted flow change in the short-term monitoring window, which is used for normalization processing. is the weighted traffic change in the short-term monitoring window W t The standard deviation is used to characterize the degree of dispersion of fluctuations. is a small constant used to avoid zero denominator.

[0039] Preferably, the actual reflux ratio control frequency in the switching stage is regulated based on the reflux ratio control frequency in the initial distillation stage and the thermal balance stability of the system in the switching stage. The specific steps are as follows:

[0040] S1. According to the difference between the thermal balance change evaluation index and the thermal balance change evaluation index reference threshold, a dynamic coefficient for adjusting the reflux ratio control frequency is calculated to cope with thermal balance fluctuations. The calculation expression is:

[0041] ,

[0042] In the formula, is the reflux ratio frequency control coefficient, which determines the frequency adjustment increase. Ther Balan is the currently calculated thermal balance change evaluation index, Ther Threshold The reference threshold value for the thermal balance change evaluation index, To adjust the sensitivity coefficient, determine the system's sensitivity to fluctuation response;

[0043] S2. Based on the reflux ratio control frequency and the reflux ratio frequency control coefficient in the initial distillation stage, the actual reflux ratio control frequency after adjustment is calculated. The calculation expression is:

[0044] ,

[0045] In the formula, f adj Indicates the actual reflux ratio control frequency after adjustment, f 0 is the reflux ratio control frequency in the initial distillation stage, The role of is to perform logarithmic transformation to capture the nonlinear effect of fluctuation amplitude;

[0046] S3. In actual operation, the calculation is performed again according to S1-S2 every short-term monitoring window, and the actual reflux ratio control frequency is dynamically adjusted according to the real-time data feedback. The adjusted actual reflux ratio control frequency is input into the control system to control the reflux liquid and steam flow rates in real time until the system restores thermal equilibrium. The expression of the real-time reflux ratio adjustment amount is: ,

[0047] In the formula, is the real-time reflux ratio adjustment, R target is the target reflux ratio, R actual is the current actual reflux ratio.

[0048] A high-purity dimethyl carbonate and propylene glycol synchronous refining and regulating system, comprising an initial distillation control module, a heat balance information acquisition module, a dynamic feature analysis module, a heat balance classification module, a normal change management module and a variable adaptive control module;

[0049] Initial distillation control module: At the beginning of the process, the reflux ratio control frequency of the initial distillation stage is used to quickly remove low-boiling impurities and establish thermal balance in the tower;

[0050] The heat balance information acquisition module collects the heat balance information of the system in real time when entering the switching process between the initial distillation stage and the refining stabilization stage;

[0051] The dynamic feature analysis module extracts key features representing the dynamic changes of the system from the collected thermal balance data, sets a short-term monitoring window to conduct in-depth analysis of the extracted key features, and inputs the key feature data analyzed within the short-term monitoring window into a pre-trained support vector machine model to predict the thermal balance change trend in the current switching stage;

[0052] The thermal balance classification module divides the thermal balance state of the system in the switching stage into thermal balance normal change type and thermal balance change type based on the prediction results of the support vector machine model;

[0053] The normal change management module, for the switching stage classified as the normal change type of heat balance, continues to manage the switching between the initial distillation stage and the refining stable stage by adjusting the reflux ratio frequency of the initial distillation stage;

[0054] The variable adaptive control module controls the actual reflux ratio control frequency of the switching stage for the switching stage classified as the heat balance variable type based on the reflux ratio control frequency of the initial distillation stage and the heat balance stability of the system in the switching stage, so as to meet the dynamic change requirements of the heat balance of the system in the switching stage.

[0055] The beneficial effects of the present invention are:

[0056] The present invention can intelligently identify the changing trend of the thermal equilibrium state by introducing real-time monitoring and support vector machine model prediction, and dynamically divide the switching stage into normal change type and variable type. For the normal change type stage, the reflux ratio control frequency of the initial distillation stage is continued to reduce unnecessary frequent adjustments, thereby ensuring the continuity and energy efficiency of the process; and for the variable type stage, the reflux ratio control frequency is adaptively adjusted through the analysis results of the support vector machine model, ensuring the balance of gas-liquid mass transfer, avoiding the occurrence of impurity entrainment and cross contamination, while improving the purity of high-purity dimethyl carbonate and propylene glycol products, effectively reducing the burden of downstream separation equipment, and reducing the energy consumption and production cost of the system. In addition, by combining intelligent prediction and dynamic feedback, the rapid recovery of the system's thermal balance under complex working conditions is ensured, the production efficiency and the reliability of the product batch are greatly improved, the risk of scrapping the entire batch of products is avoided, and finally a high-quality, high-efficiency synchronous refining process is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 This is a method flow chart of the method for synchronously refining and regulating high-purity dimethyl carbonate and propylene glycol of the present invention;

[0058] Figure 2 The module schematic diagram of the synchronous refining and regulating system of high-purity dimethyl carbonate and propylene glycol of the present invention. DETAILED DESCRIPTION

[0059] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of the present disclosure will be more comprehensive and complete, and the concept of the example embodiments will be fully conveyed to those skilled in the art.

[0060] The present invention provides Figure 1 The method for simultaneously refining and regulating high-purity dimethyl carbonate and propylene glycol comprises the following steps:

[0061] At the beginning of the process, the reflux ratio of the initial distillation stage is used to adjust the frequency to quickly remove low-boiling impurities and establish heat balance in the tower;

[0062] The reflux ratio in this stage is relatively high to minimize the accumulation of impurities in high-purity dimethyl carbonate (DMC) and propylene glycol (PG) and lay the foundation for switching to the stable stage. The purpose of the initial high-frequency control strategy is to establish the thermal dynamic balance of the system as quickly as possible and reduce the interference of uncertain factors on subsequent processes.

[0063] In the initial distillation stage of simultaneous refining of high-purity dimethyl carbonate (DMC) and propylene glycol (PG), the reflux ratio control frequency refers to the time interval for adjusting the reflux ratio in the system, that is, how often the reflux ratio is adjusted. This control frequency determines the efficiency of the exchange between the liquid and gas phases in the distillation tower so that low-boiling impurities (such as water, methanol, etc.) can be quickly removed. At this stage, the control frequency is high (for example, once every 1-5 minutes) to ensure that impurities are quickly removed to prevent them from entering the product flow and to help the system establish a thermal equilibrium state as soon as possible. This frequent regulation is to cope with the initial fluctuations in temperature and pressure in the tower and improve the stability of the distillation process.

[0064] When entering the switching process between the initial distillation stage and the refining stabilization stage, the thermal balance information of the system is collected in real time;

[0065] Extract key features representing the dynamic changes of the system from the collected thermal balance data, set a short-term monitoring window to conduct in-depth analysis of the extracted key features, and input the key feature data analyzed within the short-term monitoring window into the pre-trained support vector machine model to predict the thermal balance change trend in the current switching stage;

[0066] In the simultaneous refining process of high-purity dimethyl carbonate (DMC) and propylene glycol (PG), the system refers to the entire process system involved in the distillation and refining process.

[0067] The real-time acquisition of the thermal balance information of the system requires the coordinated work of multiple sensors and automated monitoring systems. This information includes the temperature, pressure, and liquid level at the top and bottom of the tower, as well as the flow rate (such as reflux liquid and feed flow), steam pressure, and component concentration at each key location. Common collection methods include: installing temperature and pressure sensors at the top, bottom, and middle of the tower to monitor the dynamic changes in thermal state; installing flow meters at the reflux line and feed port to monitor the flow of liquid and steam in real time; and using online gas chromatography (GC) or near-infrared spectroscopy (NIR) analyzers at key discharge points to detect changes in the components of the discharge material. In addition, the data from these sensors will be centrally collected through the PLC or DCS system and transmitted to the data processing and analysis module to achieve real-time monitoring of thermal balance information and subsequent predictive analysis.

[0068] The key features representing the dynamic changes of the system are extracted from the collected heat balance data, wherein the extracted key features include the changes in the flow rate of the rising steam and the changes in the reflux liquid flow rate in the distillation tower. After acquisition, a short-term monitoring window is set to conduct an in-depth analysis of the extracted changes in the flow rate of the rising steam and the changes in the reflux liquid flow rate in the distillation tower, and a steam flow rate change reference value and a reflux liquid fluctuation reference value are generated respectively. The steam flow rate change reference value is used to quantify the fluctuation amplitude and frequency of the rising steam flow rate in the distillation tower, and the stability of the gas phase mass transfer process in the reaction distillation tower. The reflux liquid fluctuation reference value is used to quantify the short-term fluctuation degree of the reflux liquid flow rate, which is used to evaluate the uniformity of the liquid phase distribution in the distillation tower.

[0069] After obtaining the steam flow rate change reference value and the reflux liquid fluctuation reference value generated by analyzing the key characteristics of the system's dynamic changes, the steam flow rate change reference value and the reflux liquid fluctuation reference value analyzed within the short-term monitoring window are input into a pre-trained support vector machine model to generate a thermal balance change evaluation index. The thermal balance change evaluation index is used to predict the thermal balance change trend in the current switching stage.

[0070] The support vector machine model (SVM) is a powerful supervised learning algorithm that is often used for classification and regression analysis. In this solution, the support vector machine is used to process complex nonlinear relationships in time series data, such as the dynamic association between the reference value of the steam flow rate change and the reference value of the reflux liquid fluctuation. SVM classifies or predicts data points by finding the best hyperplane, and can use kernel functions (such as radial basis function kernel (RBF)) to map low-dimensional data to high-dimensional space, so that it can also find potential laws in complex nonlinear systems. In the synchronous refining process, the thermal equilibrium state at different stages may show complex changing trends, and the SVM model can learn the association between these trends and fluctuations, providing accurate results for subsequent thermal equilibrium state predictions.

[0071] In order to meet the needs of this solution, the SVM model is trained with a large amount of historical data before the system is deployed to learn the normal state and fluctuation pattern of the distillation tower system. For example, the model will learn how to quantify the stability and change trend of the heat balance under different flow rates and liquid reflux fluctuation conditions. In the short-term monitoring window, when the system generates a reference value for the change in steam flow rate and a reference value for the fluctuation in reflux liquid, these data will be used as input and sent to the SVM model for real-time calculation. The model will generate a heat balance change evaluation index based on the distribution and fluctuation amplitude of the input features to quantify the current stability of the system. Finally, the model predicts whether the system will experience severe fluctuations or instability during the switching stage by judging the change trend. In this way, the SVM model can provide efficient and accurate heat balance predictions, helping operators or automatic control systems to adjust the reflux ratio regulation frequency in a timely manner to avoid a decrease in product purity or an increase in energy consumption.

[0072] When the flow rate of the rising steam in the distillation tower fluctuates frequently, it usually indicates that the trend of thermal balance changes in the current switching stage is extremely unstable. The specific reason is that during the distillation process, the flow rate of the rising steam directly affects the gas-liquid mass transfer process and the temperature gradient distribution in the tower. If the flow rate fluctuates frequently, it means that the evaporation and condensation processes of the system are unbalanced, which may be caused by fluctuations in heat supply, unstable feed flow, or uncontrolled reflux liquid volume. This fluctuation will reduce the efficiency of gas-liquid phase separation in the tower, make material exchange unstable, and make it difficult to maintain a dynamic balance of temperature and pressure, which will eventually lead to a decrease in product purity. In addition, frequent fluctuations in flow rate also mean that the mass transfer process of the distillation tower is disturbed, and a stable steam rising path cannot be formed, so that some impurities may be entrained into the product, further exacerbating the instability of the system. Therefore, during the switching stage, the fluctuation of steam flow rate is a key signal of thermal balance out of control.

[0073] The specific steps of setting a short-term monitoring window to conduct an in-depth analysis of the flow rate change of the rising steam in the extracted distillation tower to generate a reference value of the steam flow rate change are as follows:

[0074] Collect flow velocity change data and perform discrete processing. The expression of discrete processing is:

[0075] ,

[0076] In the formula, S i Indicates i The intensity of the steam velocity change during the period, V ( t i ) means in t i Steam flow rate at the moment (unit: m / s), Indicates the time interval between adjacent sampling points (unit: seconds). ;

[0077] This step extracts the instantaneous change in flow velocity from real-time data and generates the discretized change intensity. S i ,This step ensures that the flow rate fluctuation of the system is captured in each monitoring period.

[0078] Using the steam flow rate variation intensity in each period S i Generate a nonlinear change cumulative curve to describe the overall steam flow rate fluctuation trend of the system. The expression of the nonlinear change cumulative curve is:

[0079] ,

[0080] In the formula, αis the nonlinear amplification factor, which controls the nonlinear amplification effect of flow velocity changes and highlights the impact of violent fluctuations. β is the time attenuation coefficient, which controls the rate at which the flow velocity changes with time and simulates the trend of the system tending to be stable. C ( t ) represents the cumulative value of flow velocity change at time t, It is an integral variable, an intermediate variable representing time, used to represent the state at each time point before time t. Indicated in The intensity of the steam flow rate change at each moment;

[0081] This step generates a dynamic change curve by accumulating nonlinear changes in flow rate fluctuations. C ( t ). This step can capture the complex fluctuation patterns within the system and adjust α and β Control for varying degrees of volatility and time effects.

[0082] The reference value of steam flow rate change is calculated by calculating the difference of cumulative change and the integral of instantaneous fluctuation. The calculation expression is:

[0083] ,

[0084] In the formula, I v ( index ) is the reference value of steam flow rate change, indicating the intensity and frequency of steam flow rate fluctuation at the current moment. Indicates the cumulative change difference of steam flow rate between the current moment and the previous moment, which is used to capture instantaneous jump fluctuations. is a nonlinear weighting coefficient, which further amplifies the impact of violent fluctuations. T is the length of the short-term monitoring window;

[0085] The nonlinear weighting coefficient is used to amplify the impact of violent fluctuations. Its purpose is to give higher weights to large fluctuations or violent changes within a short monitoring window, so that these changes occupy a larger proportion in the calculation results of the index. This is because in a dynamic system, violent fluctuations often indicate that the thermal equilibrium state of the system is about to become unstable, so it needs to be paid special attention. Usually, The value range is , but it can be adjusted according to the actual working conditions, and no specific limitation is given here. Values ​​(close to 1.5) focus more on the overall trend, while higher Smaller values ​​(close to 3.0 or above) will emphasize the impact of large fluctuations. This flexible nonlinear weighting allows the model to keenly capture abnormal fluctuations in the system while avoiding over-response to minor fluctuations.

[0086] It can be seen from the reference value of the steam flow rate change that the larger the performance value of the reference value of the steam flow rate change generated after an in-depth analysis of the flow rate change of the rising steam in the extracted distillation tower within the short-term monitoring window, it usually indicates that the flow rate fluctuation of the rising steam in the distillation tower within the short-term monitoring window is more drastic, which means that the thermal balance stability of the system in the current switching stage is worse. Drastic flow rate fluctuations usually indicate that the mass transfer process between the gas phase and the liquid phase is unstable, resulting in large fluctuations in key parameters such as temperature, pressure, and reflux in the tower, affecting the separation efficiency and product purity. Therefore, an increase in the reference value of the steam flow rate change indicates that the system is about to be unbalanced. On the contrary, the smaller the reference value of the steam flow rate change, the more stable the flow rate change, the more stable the system mass transfer process, and the better the thermal balance, indicating that the process operation at the current stage is relatively smooth.

[0087] When entering the switching process between the initial distillation stage and the refining stabilization stage, if the reflux liquid flow rate changes dramatically, it usually indicates that the thermal balance change trend in the current stage is extremely unstable. The reflux liquid plays a key role in liquid phase mass transfer and heat recovery in the distillation tower, and its stability is essential to maintaining the gas-liquid balance and temperature gradient in the tower. Drastic fluctuations in the reflux liquid flow rate may mean insufficient or excessive reflux, resulting in a decrease in the heat transfer efficiency and separation efficiency of different sections in the distillation tower, which in turn disrupts the stable exchange between the gas phase and the liquid phase. At the same time, this may also be caused by fluctuations in the top condensation system, sudden changes in feed composition, or improper adjustment of operating parameters. These problems will break the original thermal equilibrium state and cause the temperature and pressure of the system to fluctuate continuously. Therefore, the drastic changes in the reflux liquid flow rate reflect that the system cannot achieve stable heat exchange and material separation, resulting in an imbalance in the flow of gas and liquid phases in the tower, indicating that the thermal balance is extremely unstable.

[0088] The specific steps for setting a short-term monitoring window to conduct in-depth analysis of the changes in the extracted reflux flow rate and generate a reflux fluctuation reference value are as follows:

[0089] In the short-term monitoring window, the data sequence of the reflux flow rate is collected in real time and recorded as F t ,in t =1,2,3,4,…, n , n is the total number of sampling time points within the short-term monitoring window, F t is the reflux flow rate at time t, and then the difference in reflux flow rate at each adjacent time point is calculated. The calculation expression is: ,

[0090] In the formula, It is the difference in flow rate change between adjacent sampling time points, reflecting the short-term fluctuation of the reflux liquid;

[0091] This step captures the fluctuation trend of the reflux liquid flow rate by calculating the reflux liquid flow rate difference, laying the foundation for the subsequent fluctuation characteristic analysis.

[0092] In order to further characterize the fluctuation characteristics of the reflux fluid, an exponential weighted fluctuation model is introduced. By assigning different weights to the change differences of each sampling, the most recent data is given a higher weight. The definition of the weighted flow change is expressed as: ,

[0093] In the formula, W t is the weighted flow change at time t, is a weighting coefficient between 0 and 1, which controls the weight of new and old data. It is usually set to 0.8, so that the recent fluctuations have a greater impact on the results. W t-1 is the weighted flow change at the previous time point;

[0094] This step captures the rapid fluctuation trend of the reflux flow rate within a short time window through an exponential weighted formula.

[0095] In order to express the degree of reflux liquid fluctuation in a quantitative exponential form, a nonlinear normalization formula is introduced to calculate the reflux liquid fluctuation reference value. The calculation expression is:

[0096] ,

[0097] In the formula, RLI ( index ) is the reflux fluid fluctuation reference value, which indicates the degree of reflux fluid flow fluctuation within the short-term monitoring window. max ( W t ) is the maximum value of the weighted flow change in the short-term monitoring window, which is used for normalization processing. is the weighted traffic change in the short-term monitoring window W t The standard deviation is used to characterize the degree of dispersion of fluctuations. is a small constant used to avoid zero denominator (such as );

[0098] This step combines the amplitude of flow rate change with the degree of dispersion, and amplifies significant fluctuations through logarithmic function and square normalization to generate the final reflux liquid fluctuation reference value.

[0099] It can be seen from the reflux liquid fluctuation reference value that the larger the performance value of the reflux liquid fluctuation reference value generated after in-depth analysis of the changes in the extracted reflux liquid flow rate in the short-term monitoring window, the worse the thermal balance stability of the system in the current switching stage, and vice versa. The better the stability of the system. In the short-term monitoring window, the reflux liquid fluctuation reference value reflects the stability of the liquid phase mass transfer and heat exchange process in the tower by quantifying the short-term fluctuation amplitude and frequency of the reflux liquid flow rate. When the reflux liquid fluctuation reference value is large, it means that the reflux liquid flow rate has violent and frequent fluctuations, which will destroy the gas-liquid balance in the tower, resulting in a decrease in separation efficiency, an increase in impurity entrainment, a decrease in product purity, and an increase in energy consumption. When the reflux liquid fluctuation reference value is small, it indicates that the reflux liquid flow rate tends to be stable, and the mass transfer and thermal balance process in the tower proceeds smoothly, ensuring efficient separation effect and reliable operation of the system. Therefore, the reflux liquid fluctuation reference value is an important evaluation indicator of system stability, and its size directly reflects the dynamic change state of thermal balance.

[0100] The above support vector machine model is not specifically limited here, and can realize the steam flow rate change reference value I v ( index ) and reflux fluid fluctuation reference value RLI ( index ) to generate a comprehensive analysis of the thermal balance change evaluation index Ther Balan The support vector machine model can be used. In order to realize the technical solution of the present invention, the present invention provides a specific implementation method; thermal balance change evaluation index Ther Balan The resulting calculation formula is:

[0101] ,

[0102] In the formula, l 1. l 2 are the reference values ​​of steam flow rate change I v ( index ) and reflux fluid fluctuation reference value RLI ( index ), and l 1. l 2 are both greater than 0.

[0103] The preset proportionality coefficient is used to control the weight ratio of the steam flow rate change reference value and the reflux liquid fluctuation reference value in generating the heat balance change evaluation index. The size of the preset proportionality coefficient determines the influence of the steam flow rate change reference value and the reflux liquid fluctuation reference value in calculating the total heat balance change evaluation index. For example, when l 1 is larger, the influence of steam flow rate change on the heat balance change evaluation index will be more significant; l2 is larger, the impact of reflux liquid fluctuation will be greater. The selection of the preset proportional coefficient needs to be set according to the actual working conditions and operating characteristics of the system to reflect the degree of influence of different dynamic characteristics on the stability of thermal balance and make the model more in line with actual operating needs. In addition, the preset proportional coefficient ensures the balanced calculation between the reference value of steam flow rate change and the reference value of reflux liquid fluctuation, preventing the fluctuation of a certain parameter from excessively affecting the overall result.

[0104] It can be seen from the heat balance change evaluation index that the greater the performance value of the steam flow rate change reference value generated after in-depth analysis of the flow rate change of the rising steam in the extracted distillation tower within the short-time monitoring window, the greater the performance value of the reflux liquid fluctuation reference value generated after in-depth analysis of the change in the reflux liquid flow extracted within the short-time monitoring window. That is, the greater the performance value of the heat balance change evaluation index generated when predicting the heat balance change trend of the system in the current switching stage within the short-time monitoring window through the support vector machine model, the worse the heat balance stability of the system in the current switching stage, and vice versa.

[0105] Based on the prediction results of the support vector machine model, the thermal equilibrium state of the system in the switching stage is divided into normal thermal equilibrium change type and thermal equilibrium change type.

[0106] The thermal balance change evaluation index generated by the support vector machine model when predicting the thermal balance change trend of the system in the current switching stage within the short-term monitoring window is compared and analyzed with the pre-set thermal balance change evaluation index reference threshold, and the thermal balance state of the system in the switching stage is divided. The specific division steps are as follows:

[0107] If the thermal balance change evaluation index is greater than or equal to the thermal balance change evaluation index reference threshold, the thermal balance state of the system in the switching stage under the short-term monitoring window is classified as a thermal balance change type;

[0108] Thermal balance fluctuation type: Indicates that the system has large thermal balance fluctuations, and the reflux ratio control frequency needs to be adjusted according to dynamic changes to avoid impurity entrainment and system imbalance.

[0109] If the thermal balance change evaluation index is less than the thermal balance change evaluation index reference threshold, the thermal balance state of the system in the switching phase within the short-term monitoring window is classified as a thermal balance normal change type.

[0110] Normal variation of heat balance: indicates that the system fluctuation is within the normal range and the reflux ratio can maintain the control frequency of the initial distillation stage without additional adjustment.

[0111] For the switching stage classified as the normal change type of heat balance, the switching between the initial distillation stage and the refining stabilization stage is managed by continuing to control the reflux ratio frequency of the initial distillation stage;

[0112] For the switching stage classified as the normal change type of heat balance, it means that the heat balance fluctuation of the system is within the acceptable range during the switching stage, and the gas-liquid mass transfer process and liquid distribution remain stable. Therefore, the system does not need to make additional adjustments, and the reflux ratio control frequency of the initial distillation stage can continue to be used to manage the switch from the initial distillation stage to the refining stabilization stage. This means that the system will adjust the reflux ratio at a preset higher frequency (usually every 1-5 minutes) to ensure the continuous stability of the heat balance and separation effect in the tower. The purpose of this is to avoid unnecessary operational changes, maintain the continuity and stability of the system operation, thereby optimizing the separation efficiency and energy consumption, while ensuring a smooth transition from the switching process to the refining stabilization stage.

[0113] For the switching stage classified as the thermal balance variable type, the actual reflux ratio control frequency of the switching stage is regulated based on the reflux ratio control frequency of the initial distillation stage and the thermal balance stability of the system in the switching stage to meet the dynamic change requirements of the thermal balance of the system in the switching stage;

[0114] Based on the reflux ratio control frequency in the initial distillation stage and the thermal balance stability of the system in the switching stage, the actual reflux ratio control frequency in the switching stage is regulated. The specific steps are as follows:

[0115] S1. According to the difference between the thermal balance change evaluation index and the thermal balance change evaluation index reference threshold, a dynamic coefficient for adjusting the reflux ratio control frequency is calculated to cope with thermal balance fluctuations. The calculation expression is:

[0116] ,

[0117] In the formula, is the reflux ratio frequency control coefficient, which determines the frequency adjustment increase. Ther Balan is the currently calculated thermal balance change evaluation index, Ther Threshold It is the reference threshold of the thermal balance change evaluation index, representing the normal fluctuation range. To adjust the sensitivity coefficient, determine the sensitivity of the system to fluctuation response, usually >0, usually set according to the actual application scenario and requirements, no specific limitation is given here, 10 -4 This is to avoid the denominator being zero;

[0118] This formula quantifies the severity of thermal equilibrium fluctuations when Ther Balan > Ther Threshold When the reflux ratio frequency control coefficient will increase, indicating that the system requires a higher frequency of reflux ratio regulation.

[0119] S2. Based on the reflux ratio control frequency and the reflux ratio frequency control coefficient in the initial distillation stage, the actual reflux ratio control frequency after adjustment is calculated. The calculation expression is:

[0120] ,

[0121] In the formula, f adj Indicates the actual reflux ratio control frequency after adjustment (unit: times / minute). f 0 is the reflux ratio control frequency in the initial distillation stage, usually 1-5 times / minute, The role of is to perform logarithmic transformation to capture the nonlinear effect of fluctuation amplitude;

[0122] This step ensures that when the thermal balance fluctuation intensifies, the frequency of reflux ratio regulation increases exponentially to quickly respond to system changes and avoid imbalance.

[0123] S3. In actual operation, the calculation is repeated according to S1-S2 every short-term monitoring window, and the actual reflux ratio control frequency is dynamically adjusted according to the real-time data feedback. The adjusted actual reflux ratio control frequency is input into the control system to control the reflux liquid and steam flow rates in real time until the system restores thermal equilibrium (i.e. Ther Balan < Ther Threshold ), the expression of real-time reflux ratio adjustment is: ,

[0124] In the formula, is the real-time reflux ratio adjustment, R target is the target reflux ratio, R actual is the current actual reflux ratio;

[0125] The target reflux ratio refers to the ratio of reflux liquid to product discharge, which is pre-set in the distillation process in order to achieve the best separation effect and system stability. This ratio reflects the relationship between the amount of liquid returned to the distillation tower after condensation and the amount of product flowing out of the tower in each cycle. The setting of the target reflux ratio is usually based on the system's process requirements, feed components, temperature, pressure and other parameters to ensure the balance of the gas-liquid mass transfer process in the tower, the optimal separation efficiency, and to avoid impurity entrainment and cross contamination. In the switching stage, by dynamically adjusting the actual reflux ratio to approach the target reflux ratio, the product purity can be improved, the burden of subsequent distillation can be reduced, and energy efficiency can be maximized;

[0126] This adjustment process ensures the balance between the reflux liquid volume and the steam flow rate, and gradually achieves system stability within each monitoring window.

[0127] To illustrate the optimization effect of the above-mentioned high-purity dimethyl carbonate (DMC) and propylene glycol (PG) simultaneous refining method, the following experimental data comparison table is provided. The table lists the changes in various process indicators under different heat balance control methods, indicating the significant advantages of introducing real-time monitoring, support vector machine prediction model and dynamic reflux ratio control.

[0128] The experimental data are divided into two groups: the traditional reflux ratio control frequency (adjusted every 5 minutes, fixed frequency) and the intelligent control method of the present invention. The specific table is as follows:

[0129]

[0130] Data analysis and optimization effect summary:

[0131] Product purity and impurity control: The intelligent control method increased the purity of DMC to 99.8%, reduced the impurity content by 80%, and reduced the amount of propylene glycol entrainment by 88.9%, indicating that dynamic control is effective in improving product quality and reducing cross contamination;

[0132] Energy consumption and efficiency: The intelligent control method reduces system energy consumption by 13.9% and shortens the refining cycle by 15% by adaptively adjusting the reflux ratio frequency, significantly improving production efficiency;

[0133] Qualified rate and scrap rate: The qualified rate of product batches increased to 99%, and the scrap rate dropped from 3.5% to 0.5%, ensuring higher production reliability and stability and avoiding the loss of the entire batch of products;

[0134] Downstream distillation burden: By reducing impurity carryover and propylene glycol residue, the burden on downstream distillation equipment is reduced, the processing capacity is reduced by 26.7%, and separation energy consumption and resources are saved;

[0135] System response and thermal balance recovery: The intelligent control method shortened the reflux ratio control lag time by 75% and the system thermal balance recovery time by 66.7%, indicating that dynamic control can respond to system fluctuations more quickly and achieve rapid balance.

[0136] The present invention can intelligently identify the changing trend of the thermal equilibrium state by introducing real-time monitoring and support vector machine (SVM) model prediction, and dynamically divide the switching stage into normal change type and variable type. For the normal change type stage, the reflux ratio control frequency of the initial distillation stage is continued to reduce unnecessary frequent adjustments, thereby ensuring the continuity and energy efficiency of the process; while for the variable type stage, the reflux ratio control frequency is adaptively adjusted through the analysis results of the support vector machine model to ensure the balance of gas-liquid mass transfer and avoid the occurrence of impurity entrainment and cross contamination. This scheme effectively reduces the burden of downstream separation equipment and reduces the energy consumption and production cost of the system while improving the purity of high-purity dimethyl carbonate (DMC) and propylene glycol (PG) products. In addition, by combining intelligent prediction and dynamic feedback, the scheme ensures the rapid recovery of the system's thermal balance under complex working conditions, greatly improves production efficiency and the reliability of product batches, avoids the risk of scrapping the entire batch of products, and ultimately achieves a high-quality, high-efficiency synchronous refining process.

[0137] The present invention provides Figure 2 The high-purity dimethyl carbonate and propylene glycol synchronous refining and regulating system shown includes an initial distillation control module, a heat balance information acquisition module, a dynamic feature analysis module, a heat balance classification module, a normal change management module and a variable adaptive control module;

[0138] Initial distillation control module: At the beginning of the process, the reflux ratio control frequency of the initial distillation stage is used to quickly remove low-boiling impurities and establish thermal balance in the tower;

[0139] The heat balance information acquisition module collects the heat balance information of the system in real time when entering the switching process between the initial distillation stage and the refining stabilization stage;

[0140] The dynamic feature analysis module extracts key features representing the dynamic changes of the system from the collected thermal balance data, sets a short-term monitoring window to conduct in-depth analysis of the extracted key features, and inputs the key feature data analyzed within the short-term monitoring window into a pre-trained support vector machine model to predict the thermal balance change trend in the current switching stage;

[0141] The thermal balance classification module divides the thermal balance state of the system in the switching stage into thermal balance normal change type and thermal balance change type based on the prediction results of the support vector machine model;

[0142] The normal change management module, for the switching stage classified as the normal change type of heat balance, continues to manage the switching between the initial distillation stage and the refining stable stage by adjusting the reflux ratio frequency of the initial distillation stage;

[0143] The variable adaptive control module controls the actual reflux ratio control frequency of the switching stage for the switching stage classified as the heat balance variable type based on the reflux ratio control frequency of the initial distillation stage and the heat balance stability of the system in the switching stage, so as to meet the dynamic change requirements of the heat balance of the system in the switching stage.

[0144] The method for synchronously refining and regulating high-purity dimethyl carbonate and propylene glycol provided in an embodiment of the present invention is realized by the above-mentioned system for synchronously refining and regulating high-purity dimethyl carbonate and propylene glycol. The specific method and process of the system for synchronously refining and regulating high-purity dimethyl carbonate and propylene glycol are detailed in the embodiment of the method for synchronously refining and regulating high-purity dimethyl carbonate and propylene glycol, which will not be repeated here.

[0145] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0146] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0147] The above description is only by way of illustration of certain exemplary embodiments of the present invention. It is undoubted that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A method for simultaneously refining and regulating high-purity dimethyl carbonate and propylene glycol, characterized in that: The following steps are involved: At the beginning of the process, the reflux ratio of the initial distillation stage is used to adjust the frequency to quickly remove low-boiling impurities and establish heat balance in the tower; In the initial distillation stage of simultaneous refining of high-purity dimethyl carbonate and propylene glycol, the reflux ratio control frequency refers to the time interval for adjusting the reflux ratio in the system, that is, how often the reflux ratio is adjusted; When entering the switching process between the initial distillation stage and the refining stabilization stage, the thermal balance information of the system is collected in real time; Extract key features representing the dynamic changes of the system from the collected thermal balance data, set a short-term monitoring window to conduct in-depth analysis of the extracted key features, and input the key feature data analyzed within the short-term monitoring window into the pre-trained support vector machine model to predict the thermal balance change trend in the current switching stage; Among them, the key features extracted include the flow rate change of the rising steam in the distillation tower and the change of the reflux liquid flow rate. After acquisition, a short-term monitoring window is set to conduct in-depth analysis on the extracted flow rate change of the rising steam in the distillation tower and the change of the reflux liquid flow rate, and a steam flow rate change reference value and a reflux liquid fluctuation reference value are generated respectively. The fluctuation amplitude and frequency of the rising steam flow rate in the distillation tower and the stability of the gas phase mass transfer process in the reactive distillation tower are quantified by the steam flow rate change reference value. The short-term fluctuation degree of the reflux liquid flow rate is quantified by the reflux liquid fluctuation reference value, which is used to evaluate the uniformity of the liquid phase distribution in the distillation tower. Based on the prediction results of the support vector machine model, the thermal equilibrium state of the system in the switching stage is divided into normal thermal equilibrium change type and thermal equilibrium change type. For the switching stage classified as the normal change type of heat balance, the switching between the initial distillation stage and the refining stable stage is managed by continuing to control the reflux ratio frequency of the initial distillation stage; For the switching stage classified as the thermal balance variable type, the actual reflux ratio control frequency of the switching stage is regulated based on the reflux ratio control frequency of the initial distillation stage and the thermal balance stability of the system in the switching stage to meet the dynamic change requirements of the thermal balance of the system in the switching stage.

2. The method for synchronously refining and regulating high-purity dimethyl carbonate and propylene glycol according to claim 1, characterized in that: After obtaining the steam flow rate change reference value and the reflux liquid fluctuation reference value generated by analyzing the key characteristics of the system's dynamic changes, the steam flow rate change reference value and the reflux liquid fluctuation reference value analyzed within the short-term monitoring window are input into a pre-trained support vector machine model to generate a thermal balance change evaluation index. The thermal balance change evaluation index is used to predict the thermal balance change trend in the current switching stage.

3. The method for synchronously refining and regulating high-purity dimethyl carbonate and propylene glycol according to claim 2, characterized in that: The thermal balance change evaluation index generated by the support vector machine model when predicting the thermal balance change trend of the system in the current switching stage within the short-term monitoring window is compared and analyzed with the pre-set thermal balance change evaluation index reference threshold, and the thermal balance state of the system in the switching stage is divided. The specific division steps are as follows: If the thermal balance change evaluation index is greater than or equal to the thermal balance change evaluation index reference threshold, the thermal balance state of the system in the switching stage under the short-term monitoring window is classified as a thermal balance change type; If the thermal balance change evaluation index is less than the thermal balance change evaluation index reference threshold, the thermal balance state of the system in the switching phase within the short-term monitoring window is classified as a thermal balance normal change type.

4. The method for synchronously refining and regulating high-purity dimethyl carbonate and propylene glycol according to claim 1, characterized in that: The specific steps of setting a short-term monitoring window to conduct an in-depth analysis of the flow rate change of the rising steam in the extracted distillation tower to generate a reference value of the steam flow rate change are as follows: Collect flow velocity change data and perform discrete processing. The expression of discrete processing is: ， In the formula, S i Indicates i The intensity of the steam velocity change during the period, V ( t i ) means in t i Steam flow rate at the time, represents the time interval between adjacent sampling points, ; Using the steam flow rate variation intensity in each period S i Generate a nonlinear change cumulative curve to describe the overall steam flow rate fluctuation trend of the system. The expression of the nonlinear change cumulative curve is: ， In the formula, α is the nonlinear amplification coefficient, which controls the nonlinear amplification effect of flow velocity changes and highlights the impact of violent fluctuations. β is the time attenuation coefficient, which controls the rate at which the flow velocity changes with time and simulates the trend of the system tending to be stable. C ( t ) represents the cumulative value of flow velocity change at time t, It is an integral variable, an intermediate variable representing time, used to represent the state at each time point before time t. Indicated in The intensity of the steam flow rate change at each moment; The reference value of steam flow rate change is calculated by calculating the difference of cumulative change and the integral of instantaneous fluctuation. The calculation expression is: ， In the formula, I v ( index ) is the reference value of steam flow rate change, indicating the intensity and frequency of steam flow rate fluctuation at the current moment. Indicates the cumulative change difference of steam flow rate between the current moment and the previous moment, which is used to capture instantaneous jump fluctuations. is a nonlinear weighting coefficient, which further amplifies the impact of violent fluctuations. T is the length of the short-term monitoring window.

5. The method for synchronously refining and regulating high-purity dimethyl carbonate and propylene glycol according to claim 1, characterized in that: The specific steps for setting a short-term monitoring window to conduct in-depth analysis of the changes in the extracted reflux flow rate and generate a reflux fluctuation reference value are as follows: In the short-term monitoring window, the data sequence of the reflux flow rate is collected in real time and recorded as F t ,in t =1,2,3,4,…, n , n is the total number of sampling time points within the short-term monitoring window, F t is the reflux flow rate at time t, and then the difference in reflux flow rate at each adjacent time point is calculated. The calculation expression is: , In the formula, It is the difference in flow rate change between adjacent sampling time points, reflecting the short-term fluctuation of the reflux liquid; In order to further characterize the fluctuation characteristics of the reflux fluid, an exponential weighted fluctuation model is introduced to give higher weights to the most recent data. The definition of the weighted flow change is expressed as: , In the formula, W t is the weighted flow change at time t, is the weighting coefficient, between 0 and 1, W t-1 is the weighted flow change at the previous time point; A nonlinear normalization formula is introduced to calculate the reference value of reflux liquid fluctuation. The calculation expression is: ， In the formula, RLI ( index ) is the reflux fluid fluctuation reference value, which indicates the degree of reflux fluid flow fluctuation within the short-term monitoring window. max ( W t ) is the maximum value of the weighted flow change in the short-term monitoring window, which is used for normalization processing. is the weighted traffic change in the short-term monitoring window W t The standard deviation is used to characterize the degree of dispersion of fluctuations. is a small constant used to avoid zero denominator.

6. The method for synchronously refining and regulating high-purity dimethyl carbonate and propylene glycol according to claim 3, characterized in that: Based on the reflux ratio control frequency in the initial distillation stage and the thermal balance stability of the system in the switching stage, the actual reflux ratio control frequency in the switching stage is regulated. The specific steps are as follows: S1. According to the difference between the thermal balance change evaluation index and the thermal balance change evaluation index reference threshold, a dynamic coefficient for adjusting the reflux ratio control frequency is calculated to cope with thermal balance fluctuations. The calculation expression is: ， In the formula, is the reflux ratio frequency control coefficient, which determines the frequency adjustment increase. Ther Balan is the currently calculated thermal balance change evaluation index, Ther Threshold The reference threshold value for the thermal balance change evaluation index, To adjust the sensitivity coefficient, determine the system's sensitivity to fluctuation response; S2. Based on the reflux ratio control frequency and the reflux ratio frequency control coefficient in the initial distillation stage, the actual reflux ratio control frequency after adjustment is calculated. The calculation expression is: ， In the formula, f adj Indicates the actual reflux ratio control frequency after adjustment, f 0 is the reflux ratio control frequency in the initial distillation stage, The role of is to perform logarithmic transformation to capture the nonlinear effect of fluctuation amplitude; S3. In actual operation, the calculation is performed again according to S1-S2 every short-term monitoring window, and the actual reflux ratio control frequency is dynamically adjusted according to the real-time data feedback. The adjusted actual reflux ratio control frequency is input into the control system to control the reflux liquid and steam flow rates in real time until the system restores thermal equilibrium. The expression of the real-time reflux ratio adjustment amount is: , In the formula, is the real-time reflux ratio adjustment, R target is the target reflux ratio, R actual is the current actual reflux ratio.

7. A regulating system for realizing the method for synchronously refining and regulating high-purity dimethyl carbonate and propylene glycol according to any one of claims 1 to 6, characterized in that: It includes an initial distillation control module, a thermal balance information acquisition module, a dynamic feature analysis module, a thermal balance classification module, a normal change management module, and a variable adaptive control module; Initial distillation control module: At the beginning of the process, the reflux ratio control frequency of the initial distillation stage is first used to quickly remove low-boiling impurities and establish thermal balance in the tower; in the initial distillation stage of simultaneous refining of high-purity dimethyl carbonate and propylene glycol, the reflux ratio control frequency refers to the time interval for adjusting the reflux ratio in the system, that is, how often the reflux ratio is adjusted; The heat balance information acquisition module collects the heat balance information of the system in real time when entering the switching process between the initial distillation stage and the refining stabilization stage; The dynamic feature analysis module extracts key features representing the dynamic changes of the system from the collected thermal balance data, sets a short-term monitoring window to conduct in-depth analysis of the extracted key features, and inputs the key feature data analyzed within the short-term monitoring window into a pre-trained support vector machine model to predict the thermal balance change trend in the current switching stage; Among them, the key features extracted include the flow rate change of the rising steam in the distillation tower and the change of the reflux liquid flow rate. After acquisition, a short-term monitoring window is set to conduct in-depth analysis on the extracted flow rate change of the rising steam in the distillation tower and the change of the reflux liquid flow rate, and a steam flow rate change reference value and a reflux liquid fluctuation reference value are generated respectively. The fluctuation amplitude and frequency of the rising steam flow rate in the distillation tower and the stability of the gas phase mass transfer process in the reactive distillation tower are quantified by the steam flow rate change reference value. The short-term fluctuation degree of the reflux liquid flow rate is quantified by the reflux liquid fluctuation reference value, which is used to evaluate the uniformity of the liquid phase distribution in the distillation tower. The thermal balance classification module divides the thermal balance state of the system in the switching stage into thermal balance normal change type and thermal balance change type based on the prediction results of the support vector machine model; The normal change management module, for the switching stage classified as the normal change type of heat balance, continues to manage the switching between the initial distillation stage and the refining stable stage by adjusting the reflux ratio frequency of the initial distillation stage; The variable adaptive control module controls the actual reflux ratio control frequency of the switching stage for the switching stage classified as the heat balance variable type based on the reflux ratio control frequency of the initial distillation stage and the heat balance stability of the system in the switching stage, so as to meet the dynamic change requirements of the heat balance of the system in the switching stage.

Images