Dynamic intelligent control method and system for improving efficiency of rectifying tower

Abstract

The invention discloses a dynamic intelligent control method and system for improving the efficiency of a rectifying tower, particularly relates to the technical field of chemical process control, and is used for solving the technical defect that an operation variable is mismatched with a physical state due to the fact that an existing distributed control system cannot recognize a hidden degradation state of equipment. The method comprises the following steps: acquiring temperature gradient distribution of each tower plate of a rectifying tower, thermal load of a reboiler and tower top condensation load parameters in real time, calculating a mass transfer efficiency offset by combining the temperature gradient distribution to generate a scaling trend index, and synchronously calculating a local entropy yield based on the thermal load and the condensation load to generate a thermodynamic irreversibility index. When the double indexes exceed a preset threshold value, it is judged that a recessive degradation state is entered, then a backflow ratio correction coefficient is reversely derived through a dynamic correlation model, and the correction coefficient is input into a distributed control system to dynamically adjust backflow ratio operation parameters; the product purity is maintained, the tower plate scaling process is obviously delayed, and long-term efficient operation of the rectifying tower is guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of chemical process control, and more particularly to a dynamic intelligent control method and system for improving the efficiency of a distillation tower. Background Art

[0002] In the fine chemical industry, distillation towers serve as core separation equipment, and their efficiency directly impacts production energy consumption and economic efficiency. Modern industry widely uses distributed control systems (DCS) to automate the distillation process. These systems monitor key parameters such as temperature, pressure, and flow in real time, and dynamically adjust operational variables using model-based optimization algorithms (such as real-time optimization (RTO)). However, DCS control decisions rely heavily on process variables directly measurable by sensors. Implicit parameters such as in-tower mass transfer efficiency and tray fouling, which are difficult to quantify, can only be roughly estimated through empirical assumptions or offline analysis.

[0003] Existing DCS control strategies have fundamental limitations: when implicit state parameters drift slowly, the control system will continuously adjust operating variables (such as increasing the reflux ratio) to maintain short-term goals (such as product purity). This leads to a self-reinforcing cycle between the actual operating state of the equipment and the degradation of implicit parameters. This mismatch between the control logic and the changes in the physical state of the equipment causes the distillation tower to operate at a local optimal condition for a long time, accelerating the degradation of equipment performance and forcing operators to frequently adopt conservative operating strategies to avoid risks. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, the present invention provides a dynamic intelligent control method and system for improving the efficiency of a distillation tower to solve the problems raised in the above-mentioned background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions: The dynamic intelligent control method for improving the efficiency of the distillation tower includes: S1. Real-time acquisition of the temperature gradient distribution of each tray of the distillation tower, the reboiler heat load parameters, and the tower top condensation load parameters; S2. Calculate the mass transfer efficiency offset between adjacent trays based on the temperature gradient distribution and generate a fouling trend index based on the reboiler heat load parameter; S3. Calculate the local entropy production rate of the distillation tower based on the reboiler heat load parameter and the tower top condensing load parameter to generate a thermodynamic irreversibility index; S4. When both the scaling trend index and the thermodynamic irreversibility index exceed the preset threshold, it is determined to enter the latent degradation state; S5. Under the latent degradation state, the optimal control compensation amount is reversely derived based on the dynamic correlation model to generate a reflux ratio correction coefficient that suppresses the scaling rate; S6. Input the reflux ratio correction coefficient into the distributed control system to dynamically adjust the reflux ratio operating parameters of the distillation tower.

[0006] Furthermore, the temperature gradient distribution of each tray of the distillation tower, the reboiler heat load parameters and the tower top condensation load parameters are obtained in real time, including: The temperature sensor array installed on each tray of the distillation tower collects raw temperature data, and the temperature gradient distribution is calculated based on the temperature difference between adjacent trays; The steam mass flow rate at the reboiler inlet is measured by a steam mass flow meter, and the reboiler heat load parameters are calculated based on the steam specific enthalpy value; The cooling water mass flow rate of the top condenser is measured by a cooling water mass flow meter, and the top condensing load parameters are calculated based on the cooling water inlet and outlet temperature difference.

[0007] Furthermore, the mass transfer efficiency offset between adjacent trays is calculated based on the temperature gradient distribution, and the fouling trend index is generated in combination with the reboiler heat load parameter, including: Extract the axial temperature difference value from the temperature gradient distribution. Based on the positive correlation between the mass transfer coefficient and the temperature gradient, multiply the ratio of the axial temperature difference values ​​of adjacent trays by the reference mass transfer coefficient to obtain the actual mass transfer coefficient. Subtract the theoretical mass transfer coefficient under the design conditions from the actual mass transfer coefficient to obtain the mass transfer efficiency offset; Read the real-time value of the reboiler heat load parameter, and calculate the heat load deviation percentage based on the theoretical heat load value under the design conditions; The mass transfer efficiency offset is multiplied by the heat load deviation percentage and then multiplied by the fouling influence weight coefficient to generate the fouling trend index.

[0008] Furthermore, the local entropy production rate of the distillation tower is calculated based on the reboiler heat load parameter and the tower top condensing load parameter to generate a thermodynamic irreversibility index, including: Obtain the real-time value of the reboiler heat load parameter and the reboiler operating temperature value, and calculate the ideal reboiler heat load value; Obtain the real-time value of the tower top condensing load parameter and the condenser operating temperature value, and calculate the ideal heat load value of the condenser; The reboiler entropy production rate is obtained by subtracting the ideal reboiler heat load value from the real-time value of the reboiler heat load parameter and dividing it by the absolute value of the ambient temperature. The condenser entropy production rate is obtained by subtracting the ideal heat load value of the condenser from the real-time value of the tower top condensing load parameter and dividing it by the absolute value of the ambient temperature. The entropy production rate of the reboiler is added to the entropy production rate of the condenser to obtain the local entropy production rate; The current total local entropy production rate is divided by the baseline total entropy production rate under the design conditions to generate the thermodynamic irreversibility index.

[0009] Furthermore, the ideal heat load value of the condenser is calculated by the Carnot efficiency formula; the ideal heat load value of the reboiler is calculated by the Carnot efficiency formula.

[0010] Furthermore, when both the scaling trend index and the thermodynamic irreversibility index exceed the preset threshold, it is determined to have entered a latent degradation state, including: Read the real-time value of scaling trend index and thermodynamic irreversibility index; Obtaining a preset threshold value of a scaling trend index and a preset threshold value of a thermodynamic irreversibility index; When the real-time value of the scaling trend index is greater than the preset threshold value of the scaling trend index and the real-time value of the thermodynamic irreversibility index is greater than the preset threshold value of the thermodynamic irreversibility index, a latent degradation state determination signal is generated.

[0011] Furthermore, under the latent degradation state, the optimal control compensation is reversely derived based on the dynamic correlation model to generate a reflux ratio correction coefficient that suppresses the scaling rate, including: When a latent degradation state judgment signal is detected, the real-time value of the current scaling trend index and the real-time value of the thermodynamic irreversibility index are read; Calculating the scaling deviation value of the real-time value of the scaling trend index relative to the preset threshold value of the scaling trend index, and simultaneously calculating the irreversibility deviation value of the real-time value of the thermodynamic irreversibility index relative to the preset threshold value of the thermodynamic irreversibility index; The scaling deviation value and the irreversibility deviation value are input into the dynamic correlation model. The dynamic correlation model is a pre-trained proportional-integral controller. The optimal control compensation is output by multiplying the proportional term by the scaling deviation value and the integral term by the irreversibility deviation value. The optimal control compensation amount is multiplied by the reference reflux ratio under the design operating conditions of the distillation tower to generate a reflux ratio correction coefficient.

[0012] Furthermore, the reflux ratio correction coefficient is input into the distributed control system to dynamically adjust the reflux ratio operating parameters of the distillation column, including: Read the current effective reflux ratio correction coefficient; Transmitting the reflux ratio correction factor to the distillation column control unit of the distributed control system via an industry standard communication interface; In the distillation column control unit, the reflux ratio correction coefficient is multiplied by the current reflux ratio operating parameter of the distillation column to generate an updated reflux ratio operating parameter; Verify whether the updated reflux ratio operating parameters are within the safe operating range of the distillation column; When the verification passes, the updated reflux ratio operating parameter is written into the loop setting value register of the distributed control system at the beginning of the next control cycle.

[0013] In another aspect, the present invention provides a dynamic intelligent control system for improving the efficiency of a distillation column, comprising: Real-time acquisition module, used to obtain the temperature gradient distribution of each tray of the distillation tower, reboiler heat load parameters and tower top condensation load parameters in real time; The fouling prediction module is used to calculate the mass transfer efficiency offset between adjacent trays based on the temperature gradient distribution and generate a fouling trend index based on the reboiler heat load parameters; Thermal entropy analysis module, used to calculate the local entropy production rate of the distillation tower based on the reboiler heat load parameters and the tower top condensing load parameters, and generate the thermodynamic irreversibility index; The state diagnosis module is used to determine that the system has entered a latent degradation state when both the scaling trend index and the thermodynamic irreversibility index exceed preset thresholds; The compensation deduction module is used to reversely deduce the optimal control compensation amount based on the dynamic correlation model under the latent degradation state and generate the reflux ratio correction coefficient to suppress the scaling rate; The parameter control module is used to input the reflux ratio correction coefficient into the distributed control system and dynamically adjust the reflux ratio operating parameters of the distillation tower.

[0014] Compared with the prior art, the present invention has the following beneficial effects: 1. By establishing a closed-loop control chain for latent degradation state identification and dynamic compensation, early proactive intervention in distillation process performance degradation is achieved. Based on a dual collaborative diagnostic mechanism combining temperature gradient distribution and thermodynamic entropy production, this system accurately captures mass transfer efficiency degradation characteristics that are undetectable by traditional sensors. By coupling the thresholds of the fouling trend index and the thermodynamic irreversibility index, an early warning is triggered at the initial stage of latent equipment degradation, breaking through the limitations of existing control systems' reliance on explicit parameters. A dynamic correlation model establishes a quantitative mapping between thermodynamic state attenuation and control parameter compensation. The reversely derived reflux ratio correction coefficient acts directly on the distributed control system's execution layer, forming a technical closed loop that suppresses fouling rates.

[0015] 2. Through real-time entropy generation analysis of reboiler heat load parameters and overhead condensing load parameters, the reflux ratio operating parameters are dynamically corrected, directly interrupting the self-reinforcing cycle between operating variable adjustments and equipment degradation. While maintaining product purity, this significantly slows tray fouling and reduces the frequency of unplanned downtime. The deep integration of compensation deduction and distributed control systems enables online adaptive control strategies, allowing operators to maintain long-term efficient operation of the distillation column without relying on empirical decisions, significantly reducing energy losses caused by conservative operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a flow chart of the dynamic intelligent control method for improving the efficiency of a distillation column according to the present invention; Figure 2This is a schematic diagram of the structure of a dynamic intelligent control system for improving the efficiency of a distillation tower according to the present invention. DETAILED DESCRIPTION

[0017] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0018] Example 1: Figure 1 The present invention provides a dynamic intelligent control method for improving the efficiency of a distillation tower, comprising: S1. Real-time acquisition of the temperature gradient distribution of each tray of the distillation tower, the reboiler heat load parameters, and the tower top condensation load parameters; S2. Calculate the mass transfer efficiency offset between adjacent trays based on the temperature gradient distribution and generate a fouling trend index based on the reboiler heat load parameter; S3. Calculate the local entropy production rate of the distillation tower based on the reboiler heat load parameter and the tower top condensing load parameter to generate a thermodynamic irreversibility index; S4. When both the scaling trend index and the thermodynamic irreversibility index exceed the preset threshold, it is determined to enter the latent degradation state; S5. Under the latent degradation state, the optimal control compensation amount is reversely derived based on the dynamic correlation model to generate a reflux ratio correction coefficient that suppresses the scaling rate; S6. Input the reflux ratio correction coefficient into the distributed control system to dynamically adjust the reflux ratio operating parameters of the distillation tower.

[0019] S1. Real-time acquisition of the temperature gradient distribution of each tray of the distillation tower, the reboiler heat load parameter, and the tower top condensation load parameter. The specific implementation is as follows: A temperature sensor array is installed on each plate of the distillation tower. The array consists of platinum resistance temperature sensors distributed at equal intervals. The spacing between adjacent sensors is controlled within a specific ratio of the effective diameter of the plate, for example, 5% to 8%. The raw temperature data of each plate is collected in real time. The raw temperature data includes the temperature values ​​of the center point of the plate and four symmetrically distributed measurement points. The temperature gradient distribution is calculated based on the temperature difference between the corresponding measurement points between adjacent plates. The specific process is as follows: the temperature measurement value at the center point of the upper plate is subtracted from the temperature measurement value at the center point of the lower plate to obtain the axial temperature difference value; the difference between the temperature measurement values ​​of the two diagonal measurement points on the same plate is divided by the sensor installation spacing to obtain the radial temperature gradient value. The temperature gradient distribution is composed of the axial temperature difference value and the radial temperature gradient value. The data update frequency is set to a specific number of times per second, for example, once.

[0020] The reboiler heat load parameter is obtained through the following process: A mass flow meter with a specified accuracy level, such as a Coriolis mass flow meter with an accuracy error of no more than ±0.5%, is installed at the reboiler steam inlet piping to measure the saturated steam mass flow rate in real time, with the mass flow rate expressed in kg / h. Simultaneously, a pressure transmitter is used to measure the absolute steam pressure, covering a specified range, such as 0 to 2.5 MPa, and a temperature transmitter is used to measure the steam temperature, with a specified accuracy, such as ±0.5°C. The measured steam pressure and temperature values ​​are used to determine the steam specific enthalpy according to the calculation method specified in the International Association for the Properties of Water and Steam (IAPWS-IF97). The reboiler heat load parameter is calculated by multiplying the steam mass flow rate by the steam specific enthalpy. The calculation relationship is expressed as follows: The reboiler heat load parameter equals the product of the steam mass flow rate and the steam specific enthalpy. The calculated result is expressed in kW. The calculation process of steam specific enthalpy distinguishes pressure conditions: when the steam pressure does not exceed a certain threshold, such as 1 MPa, the saturated steam table interpolation method is used; when the steam pressure exceeds this threshold, the superheated steam state equation is solved iteratively.

[0021] The tower top condensing load parameter is obtained through the following process: Flow meters of a specific accuracy level, such as electromagnetic flow meters with an accuracy error of no more than ±0.5%, are installed on the condenser cooling water inlet and outlet pipes to measure the cooling water mass flow rate in real time. The mass flow rate unit is kg / h. The cooling water inlet and outlet temperatures are measured using platinum resistance thermometers. The temperature measurement accuracy is controlled within a specific range, such as ±0.1°C. The cooling water inlet and outlet temperature difference is calculated by subtracting the inlet temperature from the outlet temperature. The tower top condensing load parameter is calculated by multiplying the cooling water mass flow rate by the cooling water specific heat capacity and then by the inlet and outlet temperature difference. The calculation relationship is expressed as: The tower top condensing load parameter equals the cooling water mass flow rate multiplied by the cooling water specific heat capacity multiplied by the inlet and outlet temperature difference. The specific heat capacity of cooling water is adjusted according to the temperature conditions: when the average value of the cooling water inlet and outlet temperatures is lower than a certain critical value, for example, 40°C, the specific heat capacity is fixed at 4.18 kJ / (kg·K); when the average value is higher than the critical value, a linear correction rule is adopted, for example, the specific heat capacity increases by 0.007 kJ / (kg·K) for every 1°C increase in temperature.

[0022] All collected data is transmitted to the distributed control system via current signals, for example, a 4 to 20 mA current signal is transmitted to the signal acquisition interface. The temperature gradient distribution, reboiler heat load parameters, and overhead condenser load parameters are stored in a structured format in the data storage area. The temperature gradient distribution is organized into a two-dimensional array by tray number and measurement time, while the reboiler heat load parameters and overhead condenser load parameters are organized into one-dimensional arrays by time series. Data preprocessing includes the following operations: noise removal using a time window sliding average method with a specific time window length, such as 10 seconds; and statistical discrimination of outlier measurement points in the temperature gradient distribution. When the temperature difference at a measurement point exceeds a specific multiple of the standard deviation of the average temperature difference of the same point on three adjacent trays, for example, three standard deviations, the point is identified as an outlier and removed. The preprocessed parameters are transmitted to a real-time database via an industrial communication protocol, such as OPC, for subsequent analysis.

[0023] S2. Calculate the mass transfer efficiency offset between adjacent trays based on the temperature gradient distribution, and generate a fouling trend index based on the reboiler heat load parameter. The specific implementation is as follows: Extract the axial temperature difference value from the temperature gradient distribution. This axial temperature difference value is derived from the calculation result obtained in step S1 by subtracting the temperature measurement at the center of the lower tray from the temperature measurement at the center of the upper tray. Based on the axiom that the mass transfer coefficient and the temperature gradient are positively correlated in chemical mass transfer processes, perform a ratio calculation on the axial temperature difference values ​​of two adjacent trays. Specifically, divide the axial temperature difference value of the lower tray by the axial temperature difference value of the upper tray to obtain the temperature gradient change ratio. Multiply the temperature gradient change ratio by the benchmark mass transfer coefficient. The benchmark mass transfer coefficient is derived from the calibration value under the design operating conditions of the distillation tower. This calibration value is recorded in the equipment's factory technical documentation and stored in the real-time database. It can be obtained through the data query interface. The unit of the benchmark mass transfer coefficient is kmol / (m²·s). The actual mass transfer coefficient is equal to the product of the temperature gradient change ratio and the benchmark mass transfer coefficient. The calculation result is rounded to three significant figures.

[0024] The actual mass transfer coefficient is subtracted from the theoretical mass transfer coefficient under the design conditions. The theoretical mass transfer coefficient is determined by the following process: the current feed component concentration value of the distillation tower is read from the real-time database, the tower operating pressure value is obtained through the pressure transmitter, and the gas-liquid equilibrium constant of the component concentration under the operating pressure conditions is combined with the theoretical plate number calculation formula to obtain the theoretical mass transfer coefficient under the design conditions. The theoretical mass transfer coefficient calculation process adopts the plate-by-plate iteration method. The iteration termination condition is that the difference between two consecutive calculation results is less than a specific threshold, such as 10 -3 kmol / (m²·s). The mass transfer efficiency offset is equal to the difference between the actual mass transfer coefficient and the theoretical mass transfer coefficient. The units of the difference are the same as the reference mass transfer coefficient.

[0025] Read the real-time value of the reboiler heat load parameter, which is derived from the calculation result of the product of the steam mass flow value and the steam specific enthalpy value in step S1. The theoretical heat load value under the design conditions is derived from the equipment thermal performance curve. This curve is stored in the real-time database as a function of the steam pressure as the horizontal axis and the heat load as the vertical axis. The theoretical heat load value is obtained by looking up the table and interpolating the current steam pressure value. The interpolation method uses linear interpolation. The heat load deviation percentage is obtained by the following calculation relationship: the heat load deviation percentage is equal to the difference between the real-time value of the reboiler heat load parameter and the theoretical heat load value, divided by the theoretical heat load value, and finally multiplied by 100%. The calculation result is expressed as a percentage.

[0026] The scaling influence weight coefficient is determined through historical data analysis. The specific process is as follows: the mass transfer efficiency offset data of the distillation tower over a specific time period in the past, such as 12 months, is collected along with the plate scale thickness data manually inspected during the same period. A linear regression model is then established with the mass transfer efficiency offset as the horizontal axis and the scale thickness as the vertical axis. The slope of the regression equation serves as the baseline value for the scaling influence weight coefficient. This baseline value is updated at a specific period, such as monthly, and stored in a real-time database. The scaling trend index is calculated by multiplying the mass transfer efficiency offset by the heat load deviation percentage, and then by the scaling influence weight coefficient. The result is a dimensionless value. When the scaling trend index exceeds a specific threshold, for example, a value greater than or equal to 1.0, a scaling warning signal is triggered.

[0027] All calculation processes are completed within the data processing cycle, and the data processing cycle is set to be synchronized with the data update frequency of step S1, for example, it is executed once per second. The intermediate calculation results including the temperature gradient change ratio, the actual mass transfer coefficient, and the heat load deviation percentage are temporarily stored in the memory cache in floating point format, and the scaling trend index is written into the historical data record table of the real-time database for subsequent steps to call. The exception handling mechanism includes: when the iterative calculation of the theoretical mass transfer coefficient does not meet the convergence conditions, it is replaced by the sliding average of the most recent valid values, such as the arithmetic mean of the last three valid values; when the heat load deviation percentage exceeds the physical reasonable range, for example, greater than 50%, it is determined to be a sensor failure and the output result is frozen. The data post-processing process includes result range verification to ensure that the scaling trend index is within a specific range, such as between -5.0 and 5.0.

[0028] The mass transfer efficiency offset calculation utilizes a dynamic correlation between the axial temperature difference ratio and the benchmark mass transfer coefficient. Compared to traditional fixed threshold methods, this method can capture subtle changes in mass transfer between trays in real time. The axial temperature difference value within the temperature gradient distribution directly reflects the component transfer resistance. The gradient change is quantified by the axial temperature difference ratio between the lower and upper trays (for example, an increase in the lower temperature difference indicates a blocked tray). This is then multiplied by the factory-calibrated benchmark mass transfer coefficient, ensuring that the calculated result takes into account both equipment characteristics and real-time operating conditions. The fouling trend index generation incorporates a heat load deviation percentage as a correction factor to address the vulnerability of a single mass transfer parameter to feed fluctuations. The deviation between the real-time and theoretical reboiler heat load parameters (for example, changes in heat load due to steam pressure fluctuations) is calculated, combined with the mass transfer efficiency offset, and multiplied by a weighting factor derived from historical fouling data regression (for example, a 1% decrease in mass transfer efficiency corresponds to a 0.15mm fouling thickness). This creates a multi-parameter coupled fouling index. The advantages of implementation are: the axial temperature difference ratio method avoids the installation of additional mass transfer sensors, reducing implementation costs; the heat load deviation correction can distinguish between changes in energy input and real fouling (such as false efficiency drops caused by reboiler valve failure); the dynamic weight coefficient is updated regularly (for example, by fitting new data monthly) to adapt to changes in fouling sensitivity caused by equipment aging.

[0029] S3. Calculate the local entropy production rate of the distillation tower based on the reboiler heat load parameter and the tower top condensation load parameter to generate a thermodynamic irreversibility index, which is specifically implemented as follows: The real-time value of the reboiler heat load parameter and the reboiler operating temperature are obtained. The real-time value of the reboiler heat load parameter is derived from the product of the steam mass flow rate value and the steam specific enthalpy value calculated in step S1. This product calculation is completed within the data processing cycle of step S1. The reboiler operating temperature value is derived from the temperature sensor measurement value installed in the reboiler outlet pipe in step S1. The temperature sensor uses a platinum resistance thermometer with a specific measurement accuracy level, such as 0.5. The process of calculating the ideal reboiler heat load value is based on the Carnot efficiency formula: first, the absolute value of the ambient temperature is divided by the reboiler operating temperature value to obtain the Carnot efficiency factor; then, the difference between the Carnot efficiency factor and 1 is subtracted and multiplied by the real-time value of the reboiler heat load parameter to obtain the ideal reboiler heat load value. The absolute value of the ambient temperature is obtained in real time through the weather station data interface. When weather station data is unavailable, a default value of 298 Kelvin is used, which is set based on the standard thermodynamic reference state.

[0030] The real-time value of the tower top condensing load parameter and the condenser operating temperature value are obtained. The real-time value of the tower top condensing load parameter is derived from the calculation result of the product of the cooling water mass flow value and the cooling water inlet and outlet temperature difference in step S1, and then multiplied by the specific heat capacity of water. The specific heat capacity of water adopts the standard value of 4.18 kilojoules per kilogram per Kelvin. The condenser operating temperature value is derived from the measurement value of the temperature sensor installed in the condenser outlet pipe in step S1. This temperature sensor also uses a platinum resistance thermometer. The process of calculating the ideal heat load value of the condenser is also based on the Carnot efficiency formula: the absolute value of the ambient temperature is divided by the condenser operating temperature value to obtain the Carnot efficiency factor; the difference between 1 and the Carnot efficiency factor is multiplied by the real-time value of the tower top condensing load parameter to obtain the ideal heat load value of the condenser.

[0031] The reboiler entropy generation rate is calculated by subtracting the ideal reboiler heat load from the real-time reboiler heat load parameter and dividing it by the absolute value of the ambient temperature. This calculation requires consistent units: both the real-time reboiler heat load parameter and the ideal reboiler heat load are converted to kilowatts, the absolute value of the ambient temperature is in Kelvin, and the reboiler entropy generation rate is expressed in kilowatts per Kelvin. If the calculated difference is negative, the reboiler entropy generation rate is set to zero. This is based on the principle of non-negative entropy generation according to the second law of thermodynamics.

[0032] The condenser entropy generation rate is calculated by subtracting the ideal condenser heat load from the real-time overhead condensing load parameter and dividing it by the absolute value of the ambient temperature. The unit treatment is the same as for the reboiler entropy generation rate: both the real-time overhead condensing load parameter and the ideal condenser heat load are converted to kilowatts, the absolute value of the ambient temperature is in Kelvin, and the condenser entropy generation rate is expressed in kilowatts per Kelvin. Similarly, if the calculated difference is negative, the condenser entropy generation rate is set to zero.

[0033] The local entropy generation rate is calculated by adding the entropy generation rate of the reboiler to the entropy generation rate of the condenser. The local entropy generation rate is the sum of the entropy generation rates of the reboiler and the condenser, and the units are maintained in kilowatts per kelvin. If the distillation system contains multiple reboilers or condensers, the total local entropy generation rate is calculated by adding the entropy generation rates of all reboilers to the sum of the entropy generation rates of all condensers. This treatment covers the case of multiple units operating in parallel.

[0034] The current total local entropy production rate is divided by the baseline total entropy production rate under design conditions to generate a thermodynamic irreversibility index. The baseline total entropy production rate is derived from the equipment's factory performance test report, which records the total local entropy production rate values ​​calculated using the same method under design conditions (feed flow rate, component concentration, and operating pressure are all rated values). The thermodynamic irreversibility index is a dimensionless ratio. When the index value is greater than 1, it indicates that the system's irreversibility exceeds the design baseline. The index value is written to the thermodynamic performance table in a real-time database. The storage period is synchronized with the data acquisition frequency in step S1, for example, storing one data record every 1 second.

[0035] The exception handling mechanism includes the following: When the reboiler operating temperature or condenser operating temperature is lower than the ambient temperature, the Carnot efficiency factor calculation is forced to set the ratio of the absolute value of the ambient temperature divided by the operating temperature to 1, resulting in an ideal heat load of zero. When the baseline total entropy production rate is not initialized, the average value of the local entropy production rate within a specific time period is used as a temporary baseline, such as the arithmetic mean within the last 24 hours. When the absolute value of the ambient temperature falls below a specific threshold, such as 250 Kelvin, the data is considered abnormal and the backup weather station data source is activated. All intermediate calculation results are rounded to three significant digits, and the final thermodynamic irreversibility index outputs a sliding average every specified time interval to smooth fluctuations, such as the arithmetic mean of the past 300 data points every 5 minutes.

[0036] The entropy production rate calculation uses the Carnot efficiency formula to link the operating temperature and ambient temperature, which more accurately quantifies the effective energy loss than the traditional heat balance method. The ideal heat load is calculated by combining the reboiler operating temperature value with the absolute value of the ambient temperature (for example, the Carnot efficiency factor is 298 / 393≈0.76 when the operating temperature is 393K). The difference between this and the actual heat load is then divided by the ambient temperature, strictly adhering to the entropy increase principle of the second law of thermodynamics. Multi-source data collaboration fully utilizes the existing parameters (heat load / operating temperature) in step S1, eliminating the need for new sensors. The ambient temperature is obtained through a dual channel of real-time data from the weather station and default values ​​(for example, the default of 298K corresponds to standard operating conditions), ensuring calculation continuity under extreme weather conditions. The advantages of this implementation are as follows: the calculation of local entropy production rates accounts for both equipment characteristics (such as reboiler outlet temperature) and environmental influences (such as low temperatures in winter), making it more sensitive to energy efficiency degradation than fixed-baseline methods. The thermodynamic irreversibility index directly quantifies the degree to which the system deviates from optimal operating conditions by measuring the ratio of the current total local entropy production rate to the design baseline (for example, 1.15 indicates a 15% increase in irreversible losses). Anomaly handling mechanisms (such as forcing the Carnot factor to 1 when the operating temperature is below ambient) prevent counter-physical calculation results. By integrating thermodynamic theory with real-time data, the false alarm rate is reduced under fluctuating steam pressure conditions compared to conventional temperature differential monitoring methods.

[0037] S4. When both the scaling trend index and the thermodynamic irreversibility index exceed the preset threshold, it is determined to enter the latent degradation state, which is specifically implemented as follows: The real-time value of the scaling trend index and the real-time value of the thermodynamic irreversibility index are read from the historical performance record table of the real-time database. The historical performance record table contains the scaling trend index value generated in step S2 and the thermodynamic irreversibility index value generated in step S3. Both parameters are stored in timestamp order, and the storage frequency is 1 time per second. The reading process obtains the latest data point within the most recent specific time window, such as the latest record in the last 5 minutes. The real-time value of the scaling trend index and the real-time value of the thermodynamic irreversibility index are both stored in floating-point format, and the numerical precision is retained to 3 decimal places. When there is no valid data in the historical performance record table, the system waits for the next data collection cycle before trying to read again. The waiting time shall not exceed the data collection interval, such as 1 second.

[0038] Obtain the preset thresholds for the scaling trend index and the thermodynamic irreversibility index. The preset thresholds are derived from the maximum allowable values ​​for the corresponding indicators during the normal operation of the equipment in the historical performance record table. The normal operation period of the equipment is defined as a continuous period with no alarm records within a specific number of days, such as the last 30 days. The maximum allowable value is determined through the following statistical process: First, extract all scaling trend index values ​​within the time period to form a first data set, and all thermodynamic irreversibility index values ​​to form a second data set. Then, sort the first data set in ascending order, and take the value at the 99th percentile after sorting as the preset threshold for the scaling trend index. Then, sort the second data set in ascending order, and take the value at the 98th percentile after sorting as the preset threshold for the thermodynamic irreversibility index. The thresholds are automatically updated once a week at a specific time every Monday, such as 2:00 AM.

[0039] A latent degradation state determination signal is generated when both the real-time value of the fouling trend index and the real-time value of the thermodynamic irreversibility index exceed the preset threshold. This determination utilizes a logical AND operation, triggering signal generation only when both comparison conditions are met. The determination frequency is synchronized with the data acquisition frequency, for example, once per second. When a latent degradation state determination signal is generated, a millisecond-accurate timestamp, the actual value of the exceeded parameter, and its corresponding threshold value are recorded simultaneously. The signal is a Boolean true value.

[0040] The latent degradation state determination signal is written to the abnormal state record table in the real-time database, simultaneously triggering the alarm logging function. The abnormal state record table contains three fields: a timestamp field that stores time information in ISO8601 format, a status type field that stores the preset status code, and a parameter details field that stores out-of-limit data in JSON format. During writing, the status type field is assigned a specific code, for example, the value 3 represents a latent degradation state. The alarm logging function calls the system's standard log interface to generate structured text containing a timestamp, device number, and status description, such as "2023-06-15T14:30:25.123Z, Device T-101, Code 3: Hidden degradation state." Database write operations utilize a transaction mechanism to ensure atomic updates to the abnormal state record table and alarm log.

[0041] If a latent degradation status determination signal is generated for multiple consecutive determination cycles, for example, ten consecutive true determinations, the latent degradation status determination signal is upgraded to a confirmed status signal. The confirmed status signal is also written to the preventive maintenance queue of the equipment maintenance work order system. The work order contains the equipment number, failure mode code, and first anomaly timestamp. If the latent degradation status determination signal changes from true to false within a specific time period, such as 24 hours, the status type field in the abnormal status record table is marked with a specific code, such as a value of 7 to indicate an intermittent abnormal event.

[0042] The threshold calibration mechanism includes two trigger conditions: When the device is put back into operation after maintenance, or the operator manually clicks the Maintenance Complete button on the console, the system resets the threshold calculation cycle; and when the ambient temperature deviates from the design value by more than a certain amount for a specific number of consecutive days, for example, by more than 5 degrees Celsius for three consecutive days, the system automatically shortens the threshold update cycle to once a day. All threshold values ​​are rounded to three significant digits, and the abnormal status record table stores data for the most recent specific number of days, for example, within 180 days.

[0043] S5. Under the implicit degradation state, the optimal control compensation is reversely derived based on the dynamic correlation model to generate a reflux ratio correction coefficient that suppresses the scaling rate. The specific implementation is as follows: When a latent degradation state judgment signal is detected, the current real-time value of the scaling trend index and the real-time value of the thermodynamic irreversibility index are read from the abnormal state record table of the real-time database. The latent degradation state judgment signal is generated by step S4 and written into the state type field of the abnormal state record table. When the value of this field is a specific state code, the reading operation is triggered. For example, when the state code is 3, it is determined to be a latent degradation state. The reading process obtains the real-time value of the scaling trend index and the real-time value of the thermodynamic irreversibility index in the latest valid record. Both parameters are stored in the value field in floating point format. If there are multiple unprocessed signal records at the same time, only the record with the latest timestamp is selected for data reading. A retry mechanism is executed when the reading fails, for example, a maximum of 3 retries are made, and each retry is separated by a specific time, such as 1 second.

[0044] Calculate the scaling deviation value of the real-time value of the scaling trend index relative to the preset threshold value of the scaling trend index. Simultaneously, calculate the irreversibility deviation value of the real-time value of the thermodynamic irreversibility index relative to the preset threshold value of the thermodynamic irreversibility index. The scaling deviation value is equal to the real-time value of the scaling trend index minus the preset threshold value of the scaling trend index; the irreversibility deviation value is equal to the real-time value of the thermodynamic irreversibility index minus the preset threshold value of the thermodynamic irreversibility index. The preset thresholds are derived from historical statistical values ​​updated weekly and stored in the threshold configuration table in the real-time database. The deviation value calculation result is rounded to 4 decimal places. If the calculated result is a negative number, the deviation value is forcibly set to zero. For example, if the real-time value of the thermodynamic irreversibility index is 5.2 and the corresponding preset threshold is 5.3, the irreversibility deviation value is reset to zero.

[0045] The scaling deviation and irreversibility deviation values ​​are input into the dynamic correlation model, which is a pre-trained proportional-integral controller. The optimal control compensation is output by multiplying the proportional term by the scaling deviation value and the integral term by the irreversibility deviation value. The proportional term is calculated by multiplying the proportional coefficient by the scaling deviation value; the integral term is calculated by multiplying the integral coefficient by the irreversibility deviation value and then by the time integral factor. The time integral factor is the arithmetic mean of all irreversibility deviation values ​​within a specific recent time period, such as the average deviation value recorded every minute for the last 30 minutes. The proportional coefficient and integral coefficient are set as fixed constants based on the historical operating data of the equipment, for example, the proportional coefficient is 0.8 and the integral coefficient is 0.05. The output range of the optimal control compensation is limited to a specific numerical range, such as between 0.5 and 1.5. If it exceeds this range, it is automatically truncated to the nearest boundary value.

[0046] The optimal control compensation amount is multiplied by the baseline reflux ratio under the distillation tower's design conditions to generate a reflux ratio correction factor. The baseline reflux ratio is derived from the equipment's factory technical parameter sheet, which is stored in the static parameter table of the real-time database. The multiplication result is rounded to three decimal places. If the calculated result exceeds the distillation tower's allowable operating range, for example, if it is less than the equipment's minimum reflux ratio limit or greater than its maximum reflux ratio limit, the correction factor is set to the corresponding limit. The generated reflux ratio correction factor is accompanied by a millisecond-accurate timestamp and version number.

[0047] The reflux ratio correction factor is written to the control parameter table in the real-time database and transmitted to the distillation tower reflux ratio controller for compensation. The control parameter table contains a device number field, a parameter type code field, a correction factor field, and an effective time field. During the write operation, the parameter type code field is assigned a specific code, such as the code R for reflux ratio correction. Data transmission uses industry-standard communication protocols, such as OPC UA, with data synchronization cycles matching the controller scan cycle, for example, data synchronization every 5 seconds. After receiving the new correction factor, the reflux ratio controller replaces the original parameter value at the beginning of the next control cycle.

[0048] The compensation effectiveness verification mechanism includes retrieving the current real-time value of the fouling trend index generated in step S2 within a specified period of time (e.g., 30 minutes) after the reflux ratio correction factor takes effect. If this value does not decrease by a specified percentage compared to the real-time value of the fouling trend index before correction (e.g., a decrease of less than 5%), the control parameter adaptive adjustment process is initiated. The adaptive adjustment process recalculates the proportional and integral coefficients based on the rate of change of the deviation and writes the updated values ​​to the configuration parameter storage area of ​​the proportional-integral controller.

[0049] All correction operations are recorded in a maintenance log system. The log includes the original deviation value, the compensation calculation process, the final correction coefficient, and the execution results. Log data is stored in chronological order and retained for a specific number of days, such as 365 days, for subsequent control strategy optimization analysis.

[0050] S6. Input the reflux ratio correction coefficient into the distributed control system to dynamically adjust the reflux ratio operating parameters of the distillation tower. The specific implementation is as follows: The currently effective reflux ratio correction coefficient is read from the control parameter table of the real-time database, and the reflux ratio correction coefficient is generated and stored by step S5. The reading operation is triggered when it is detected that the parameter type code field of the control parameter table is a specific state code, for example, the state code is R representing the reflux ratio correction. The control parameter table contains an effective time field and a timestamp field. When reading, the valid record with the latest timestamp is given priority. If there are multiple valid records at the same time, only the record data with the largest timestamp value is read. When the reading fails, a retry mechanism is executed, for example, a maximum of 5 retries are made, and each retry is separated by a specific time, for example, 500 milliseconds. The read reflux ratio correction coefficient is stored in the correction coefficient field in floating point format, and the numerical precision is retained to 3 decimal places.

[0051] The reflux ratio correction factor is transmitted to the distillation column control unit of the distributed control system via an industry-standard communication interface. This industry-standard communication interface utilizes an Ethernet-based industrial communication protocol, such as OPC UA. Before transmission, the reflux ratio correction factor is converted into a standardized data format recognizable by the distributed control system, such as a 32-bit single-precision floating-point number compliant with the IEEE 754 standard. The transmission cycle is synchronized with the distributed control system's scan cycle, for example, transmitting data every 500 milliseconds. Bidirectional authentication is performed when establishing the transmission connection, for example, by exchanging digital certificates to verify device legitimacy.

[0052] Within the distillation column control unit, the reflux ratio correction factor is multiplied by the current reflux ratio operating parameter of the distillation column to generate the updated reflux ratio operating parameter. The current reflux ratio operating parameter of the distillation column is read from the process variable real-time data register of the distributed control system, with the reading frequency consistent with the transmission cycle. The multiplication result is rounded to three decimal places. If the calculated result exceeds the valid floating-point representation range, a specific error code is returned and the process terminates. For example, if the reflux ratio correction factor is 1.25 and the current reflux ratio operating parameter is 4.8, the updated reflux ratio operating parameter is calculated to be 6.0.

[0053] Verify that the updated reflux ratio operating parameter is within the distillation column's safe operating range. This safe operating range is derived from the equipment's factory technical documentation and includes minimum and maximum reflux ratio operating parameter limits, stored in the distributed control system's safety parameter database. The verification process performs a double comparison: first, determining whether the updated reflux ratio operating parameter is greater than or equal to the minimum reflux ratio operating parameter limit, and then determining whether it is less than or equal to the maximum reflux ratio operating parameter limit. If either comparison condition is not met, a parameter out-of-limit alarm event is generated and recorded in the alarm log.

[0054] If verification passes, the updated reflux ratio operating parameters are written to the DCS's loop setpoint register at the start of the next control cycle. The control cycle length is determined by the DCS configuration parameters, for example, 500 milliseconds. The write operation is performed within a specific time window at the beginning of the control cycle, for example, within the first 50 milliseconds after the start of the scan cycle. Before writing, the write enable flag in the loop setpoint register is checked and the write operation is performed only if the flag is true.

[0055] Changes in the scaling trend index after the reflux ratio operating parameter adjustment are monitored in real time, and the operation record is written to the process log table of the real-time database. The monitoring process is initiated at a specific time period after the parameter adjustment, for example, 15 minutes later. The current real-time value of the scaling trend index generated in step S2 is read from the real-time database. The operation record includes a field for the equipment number, a field for the reflux ratio operating parameter before adjustment, a field for the reflux ratio operating parameter after adjustment, a field for the correction factor, and a field for the execution time. The operator identification and operation verification code are appended when writing to the process log table.

[0056] The parameter rollback mechanism automatically restores the reflux ratio operating parameters to their original state if the scaling trend index fails to improve within a specified timeframe—for example, if the scaling trend index decreases by less than 2% within one hour of adjustment. This rollback process retrieves the original parameter values ​​from the process log's historical operation records and rewrites them into the loop setpoint registers. All parameter changes are recorded in the process log for a specified number of days, such as 90 days, for operational audit analysis.

[0057] The technical solution of this embodiment shows a technical synergy effect in the field of scaling control in the distillation process. Existing technologies usually only take cleaning or shutdown measures in the stage of obvious scaling, while this solution builds a closed-loop control chain for latent degradation state identification and dynamic compensation: by tracking the coupling relationship between thermodynamic irreversibility and scaling trend in real time, an early warning is triggered at the early stage of implicit degradation of equipment performance; then a pre-trained proportional-integral controller is used to analyze the deep correlation of the dual-parameter deviations to generate a reflux ratio correction that accurately matches the current operating conditions; a specially designed parameter rollback mechanism is seamlessly integrated with the distributed control system to achieve dynamic optimization of operating parameters while ensuring process safety. This technical path of deep integration of state recognition models, dynamic association algorithms and industrial control systems effectively solves the problem that traditional methods cannot suppress early scaling without shutting down the machine.

[0058] Example 2: Figure 2 A schematic structural diagram of a dynamic intelligent control system for improving distillation tower efficiency according to the present invention is provided. The dynamic intelligent control system for improving distillation tower efficiency comprises: Real-time acquisition module, used to obtain the temperature gradient distribution of each tray of the distillation tower, reboiler heat load parameters and tower top condensation load parameters in real time; The fouling prediction module is used to calculate the mass transfer efficiency offset between adjacent trays based on the temperature gradient distribution and generate a fouling trend index based on the reboiler heat load parameters; Thermal entropy analysis module, used to calculate the local entropy production rate of the distillation tower based on the reboiler heat load parameters and the tower top condensing load parameters, and generate the thermodynamic irreversibility index; The state diagnosis module is used to determine that the system has entered a latent degradation state when both the scaling trend index and the thermodynamic irreversibility index exceed preset thresholds; The compensation deduction module is used to reversely deduce the optimal control compensation amount based on the dynamic correlation model under the latent degradation state and generate the reflux ratio correction coefficient to suppress the scaling rate; The parameter control module is used to input the reflux ratio correction coefficient into the distributed control system and dynamically adjust the reflux ratio operating parameters of the distillation tower.

[0059] The calculations involved in the embodiments are all dimensionless numerical calculations, and the preset parameters and thresholds in the calculations are set by those skilled in the art according to actual conditions.

[0060] It should be noted that the present invention can be deployed on the device itself to implement embedded applications, and can also be run on a PC or other terminal with a user interface, thereby meeting various hardware environments and usage requirements.

[0061] The above embodiments can be implemented in whole or in part via software, hardware, firmware, or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product comprises one or more computer instructions or computer programs. When loaded or executed on a computer, the processes or functions described in the embodiments of this application are fully or partially performed. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wireless or wired transmission. Wired transmission methods include optical fiber, twisted pair, coaxial cable, etc.; wireless transmission methods include infrared, microwave, etc. The computer-readable storage medium can be any available medium accessible by a computer, or a data storage device such as a server or data center that contains a collection of one or more available media. The available medium can be magnetic media (e.g., floppy disks, hard disks, tapes), optical media (e.g., DVDs), or semiconductor media. Semiconductor media can be solid-state drives.

[0062] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and modules described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0063] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the modules is only a logical function division. In actual implementation, there may be other division methods, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.

[0064] The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical modules, and may be located in one place or distributed across multiple network modules. Some or all of the modules may be selected to achieve the purpose of this embodiment according to actual needs.

[0065] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0066] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0067] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

[0068] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A dynamic intelligent control method for improving distillation tower efficiency, characterized in that: include: S1. Real-time acquisition of the temperature gradient distribution of each tray of the distillation tower, the reboiler heat load parameters, and the tower top condensation load parameters; S2. Calculate the mass transfer efficiency offset between adjacent trays based on the temperature gradient distribution and generate a fouling trend index based on the reboiler heat load parameter; S3. Calculate the local entropy production rate of the distillation tower based on the reboiler heat load parameter and the tower top condensing load parameter to generate a thermodynamic irreversibility index; S4. When both the scaling trend index and the thermodynamic irreversibility index exceed the preset threshold, it is determined to enter the latent degradation state; S5. Under the latent degradation state, the optimal control compensation amount is reversely derived based on the dynamic correlation model to generate a reflux ratio correction coefficient that suppresses the scaling rate; S6. Input the reflux ratio correction coefficient into the distributed control system to dynamically adjust the reflux ratio operating parameters of the distillation tower.

2. The dynamic intelligent control method for improving distillation tower efficiency according to claim 1, characterized in that: Real-time acquisition of the temperature gradient distribution of each tray of the distillation tower, reboiler heat load parameters and tower top condenser load parameters, including: The temperature sensor array installed on each tray of the distillation tower collects raw temperature data, and the temperature gradient distribution is calculated based on the temperature difference between adjacent trays; The steam mass flow rate at the reboiler inlet is measured by a steam mass flow meter, and the reboiler heat load parameters are calculated based on the steam specific enthalpy value; The cooling water mass flow rate of the top condenser is measured by a cooling water mass flow meter, and the top condensing load parameters are calculated based on the cooling water inlet and outlet temperature difference.

3. The dynamic intelligent control method for improving distillation tower efficiency according to claim 1, characterized in that: The mass transfer efficiency offset between adjacent trays is calculated based on the temperature gradient distribution, and the fouling trend index is generated in combination with the reboiler heat load parameters, including: Extract the axial temperature difference value from the temperature gradient distribution. Based on the positive correlation between the mass transfer coefficient and the temperature gradient, multiply the ratio of the axial temperature difference values ​​of adjacent trays by the reference mass transfer coefficient to obtain the actual mass transfer coefficient. Subtract the theoretical mass transfer coefficient under the design conditions from the actual mass transfer coefficient to obtain the mass transfer efficiency offset; Read the real-time value of the reboiler heat load parameter, and calculate the heat load deviation percentage based on the theoretical heat load value under the design conditions; The mass transfer efficiency offset is multiplied by the heat load deviation percentage and then multiplied by the fouling influence weight coefficient to generate the fouling trend index.

4. The dynamic intelligent control method for improving distillation tower efficiency according to claim 1, characterized in that: The local entropy production rate of the distillation tower is calculated based on the reboiler heat load parameters and the tower top condensing load parameters, and the thermodynamic irreversibility index is generated, including: Obtain the real-time value of the reboiler heat load parameter and the reboiler operating temperature value, and calculate the ideal reboiler heat load value; Obtain the real-time value of the tower top condensing load parameter and the condenser operating temperature value, and calculate the ideal heat load value of the condenser; The reboiler entropy production rate is obtained by subtracting the ideal reboiler heat load value from the real-time value of the reboiler heat load parameter and dividing it by the absolute value of the ambient temperature. The condenser entropy production rate is obtained by subtracting the ideal heat load value of the condenser from the real-time value of the tower top condensing load parameter and dividing it by the absolute value of the ambient temperature. The entropy production rate of the reboiler is added to the entropy production rate of the condenser to obtain the local entropy production rate; The current total local entropy production rate is divided by the baseline total entropy production rate under the design conditions to generate the thermodynamic irreversibility index.

5. The dynamic intelligent control method for improving distillation tower efficiency according to claim 4, characterized in that: The ideal heat load value of the condenser is calculated using the Carnot efficiency formula; the ideal heat load value of the reboiler is calculated using the Carnot efficiency formula.

6. The dynamic intelligent control method for improving distillation tower efficiency according to claim 1, characterized in that: When both the scaling trend index and the thermodynamic irreversibility index exceed the preset threshold, it is determined to have entered a latent degradation state, including: Read the real-time value of scaling trend index and thermodynamic irreversibility index; Obtaining a preset threshold value of a scaling trend index and a preset threshold value of a thermodynamic irreversibility index; When the real-time value of the scaling trend index is greater than the preset threshold value of the scaling trend index and the real-time value of the thermodynamic irreversibility index is greater than the preset threshold value of the thermodynamic irreversibility index, a latent degradation state determination signal is generated.

7. The dynamic intelligent control method for improving distillation tower efficiency according to claim 1, characterized in that: In the latent degradation state, the optimal control compensation is reversely derived based on the dynamic correlation model to generate the reflux ratio correction coefficient that suppresses the scaling rate, including: When a latent degradation state judgment signal is detected, the real-time value of the current scaling trend index and the real-time value of the thermodynamic irreversibility index are read; Calculating the scaling deviation value of the real-time value of the scaling trend index relative to the preset threshold value of the scaling trend index, and simultaneously calculating the irreversibility deviation value of the real-time value of the thermodynamic irreversibility index relative to the preset threshold value of the thermodynamic irreversibility index; The scaling deviation value and the irreversibility deviation value are input into the dynamic correlation model. The dynamic correlation model is a pre-trained proportional-integral controller. The optimal control compensation is output by multiplying the proportional term by the scaling deviation value and the integral term by the irreversibility deviation value. The optimal control compensation amount is multiplied by the reference reflux ratio under the design operating conditions of the distillation tower to generate a reflux ratio correction coefficient.

8. The dynamic intelligent control method for improving distillation tower efficiency according to claim 1, characterized in that: The reflux ratio correction coefficient is input into the distributed control system to dynamically adjust the reflux ratio operating parameters of the distillation column, including: Read the current effective reflux ratio correction coefficient; Transmitting the reflux ratio correction factor to the distillation column control unit of the distributed control system via an industry standard communication interface; In the distillation column control unit, the reflux ratio correction coefficient is multiplied by the current reflux ratio operating parameter of the distillation column to generate an updated reflux ratio operating parameter; Verify whether the updated reflux ratio operating parameters are within the safe operating range of the distillation column; When the verification passes, the updated reflux ratio operating parameter is written into the loop setting value register of the distributed control system at the beginning of the next control cycle.

9. A dynamic intelligent control system for improving distillation tower efficiency, used to implement the dynamic intelligent control method for improving distillation tower efficiency according to any one of claims 1 to 8, characterized in that: include: Real-time acquisition module, used to obtain the temperature gradient distribution of each tray of the distillation tower, reboiler heat load parameters and tower top condensation load parameters in real time; The fouling prediction module is used to calculate the mass transfer efficiency offset between adjacent trays based on the temperature gradient distribution and generate a fouling trend index based on the reboiler heat load parameters; Thermal entropy analysis module, used to calculate the local entropy production rate of the distillation tower based on the reboiler heat load parameters and the tower top condensing load parameters, and generate the thermodynamic irreversibility index; The state diagnosis module is used to determine that the system has entered a latent degradation state when both the scaling trend index and the thermodynamic irreversibility index exceed preset thresholds; The compensation deduction module is used to reversely deduce the optimal control compensation amount based on the dynamic correlation model under the latent degradation state and generate the reflux ratio correction coefficient to suppress the scaling rate; The parameter control module is used to input the reflux ratio correction coefficient into the distributed control system and dynamically adjust the reflux ratio operating parameters of the distillation tower.

Images