Adaptive control method for condenser operation in silicone oil production lines

Abstract

This invention relates to the field of refrigeration equipment control technology, specifically to an adaptive control method for condenser operation in silicone oil production lines. The method includes: First, based on historical production data, determining a reference range for the condenser outlet vapor phase temperature to produce silicone oil with preset quality indicators, and determining a target temperature within this range based on historical operating efficiency. Second, during production, using a model predictive control algorithm to adjust the condenser's operating parameters in real time, bringing the predicted future outlet temperature closer to the target temperature. Finally, based on the recent changes in actual operating efficiency with temperature and the difference between historical and predicted trends in outlet temperature, the target temperature is dynamically readjusted. This invention achieves adaptive tracking and coordination of the condenser with the de-boiling vessel conditions, ensuring key indicators such as product viscosity and volatile matter content while further achieving energy saving and efficient recovery of low-boiling-point substances in the production process.

Description

Technical Field

[0001] This invention relates to the field of refrigeration equipment control technology, and more specifically to an adaptive control method for condenser operation applicable to silicone oil production lines. Background Technology

[0002] In silicone oil production, the condenser after the de-lowering process is a core piece of equipment for controlling key quality indicators such as product viscosity and volatile matter. Traditional control methods typically set a fixed target outlet gas phase temperature for the condenser and then adjust it using MPC or PID algorithms, or rely on manual experience, to produce a qualified product. However, silicone oil production is a complex dynamic process, and the load on the evaporation stage of the de-lowering reactor (often operated manually or by other independent control systems) varies independently. If the condenser operates at a fixed temperature target, its recovery capacity cannot adaptively match the real-time evaporation rate of the de-lowering reactor, leading to a mismatch between the two devices. This causes two drawbacks: first, when the evaporation rate of the de-lowering reactor increases dramatically, the condenser's recovery is not timely, resulting in raw material waste and emission pressure; second, when the evaporation rate of the de-lowering reactor is low, the condenser still operates at high power, leading to energy waste. While this rigid control mode may produce a qualified product, it is difficult to achieve energy saving and consumption reduction in the production process and efficient recovery of raw materials (unreacted low-boiling substances such as D4, D5, DMC, and MM), thus restricting the improvement of the overall energy efficiency and economic benefits of the production line. Summary of the Invention

[0003] To address the challenges of achieving energy conservation and efficient recovery of low-boiling-point raw materials in production processes when condensers are operated at a fixed temperature target, this invention provides an adaptive control method for condenser operation suitable for silicone oil production lines.

[0004] The adaptive control method for condenser operation in an organosilicon oil production line of the present invention adopts the following technical solution:

[0005] One embodiment of the present invention provides an adaptive control method for condenser operation in an organosilicon oil production line, the method comprising the following steps:

[0006] The reference range of the condenser outlet vapor temperature is determined based on the historical production process to produce silicone oil with preset specifications. The target temperature is then determined based on the condenser's operating efficiency within the reference range during the historical production process. The operating efficiency is determined by the evaporation rate of the de-evaporation reactor and the condensation recovery rate of the condenser. During the production of silicone oil with preset specifications, the MPC algorithm is used to control the condenser's operating parameters in real time, causing the outlet vapor temperature predicted by the MPC algorithm for several future moments to change towards the target temperature. Simultaneously, based on the change in condenser operating efficiency with outlet vapor temperature over several recent historical moments, and the difference in the changing trends of outlet vapor temperature between several recent historical moments and several future moments, the target temperature is re-determined within the reference range based on the condenser's operating efficiency during the historical production process, and the MPC algorithm is continued to be used to control the condenser's operating parameters in real time.

[0007] Preferably, the specific steps for determining the target temperature based on the condenser's operating efficiency within a reference range during historical production processes are as follows:

[0008] Each temperature within the reference range is assigned a historical operating efficiency. All temperatures and historical operating efficiencies within the reference range are clustered into several categories. The mean of all temperatures within each category is recorded as the center temperature. An evaluation index is calculated for each category. The evaluation index is positively correlated with the temperature distribution density and the average historical operating efficiency within each category. The center temperature of the category with the largest evaluation index is recorded as the target temperature.

[0009] Preferably, the specific steps for re-determining the target temperature within a reference range based on the change in condenser efficiency with outlet gas temperature over several recent historical periods, and the difference in the trend of outlet gas temperature between several recent historical periods and several future periods, include the following:

[0010] The ratio of the first rate of change of the outlet gas phase temperature over time at several future moments to the second rate of change of the outlet gas phase temperature over time at several recent historical moments is denoted as the trend difference.

[0011] When the change is greater than 0, the target temperature remains unchanged;

[0012] When the change is less than or equal to 0, if the difference in the trend of change is greater than or equal to 1, the target temperature is updated within the reference range based on the working efficiency of the condenser in the historical production process, so that the change in the target temperature is positively correlated with the absolute value of the change and the difference in the trend of change, respectively; if the difference in the trend of change is less than 1, the target temperature is not changed.

[0013] Preferably, the specific steps for updating the target temperature within a reference range based on the historical condenser operating efficiency during the production process are as follows:

[0014] When both the first rate of change and the second rate of change are greater than or equal to 0, among all the categories, the category with a center temperature less than or equal to the target temperature is recorded as the reference category, and the target temperature minus the update change amount is used as the initial update result; when both the first rate of change and the second rate of change are less than 0, the category with a center temperature greater than or equal to the target temperature is recorded as the reference category, and the target temperature plus the update change amount is used as the initial update result.

[0015] The difference between the center temperature of each reference category and the initial update result is the selection index for each reference category; the selection index of each reference category is reduced, and the amount of reduction is positively correlated with the evaluation index of each reference category; the center temperature of the reference category with the smallest reduced selection index among all reference categories is taken as the updated target temperature.

[0016] Preferably, the specific steps for updating the change amount are as follows:

[0017] The update percentage is determined, and the update percentage is positively correlated with the absolute value of the change and the difference in the trend of change, respectively; the product of the first difference between the maximum and minimum center temperatures of all reference categories and the update percentage is recorded as the update change.

[0018] Preferably, the specific steps for determining the reference range of the condenser outlet vapor phase temperature when producing silicone oil with preset indicators based on historical production processes are as follows:

[0019] In the historical production process, after each batch of silicone oil is produced, the indicators of that batch of silicone oil are tested, including viscosity and volatile matter, and recorded as the historical indicators of that batch of silicone oil; the average value of the outlet vapor phase temperature of the condenser at all times during the production of that batch of silicone oil is obtained and recorded as the historical outlet temperature of that batch of silicone oil.

[0020] For all batches of silicone oil in all historical production processes, the historical indicators and historical outlet temperatures are used to obtain the most similar historical indicators to the preset indicators. The temperature range formed by the maximum and minimum values ​​of the historical outlet temperatures corresponding to the most similar historical indicators is recorded as the reference range.

[0021] Preferably, each temperature within the reference range is assigned a historical operating efficiency, including the following specific steps:

[0022] For any historical production process, the average outlet vapor temperature of the condenser at all times is recorded as the average operating efficiency at all times of the historical production process. For all historical production processes, the historical operating efficiency corresponding to any temperature within the reference range is interpolated using a linear interpolation algorithm based on the historical operating efficiency corresponding to all average outlet vapor temperatures.

[0023] Preferably, the specific steps for obtaining the work efficiency are as follows:

[0024] For any given moment in the production process, obtain the evaporation rate and the condensation recovery rate in the de-evaporation vessel at that moment; the ratio of the condensation recovery rate to the evaporation rate is taken as the working efficiency at that moment.

[0025] Preferably, the specific steps for obtaining the change in the condenser's operating efficiency with respect to the outlet gas phase temperature at the most recent historical moments are as follows:

[0026] The first temperature change is determined based on the differences in outlet gas phase temperature over several recent historical periods; the first efficiency change is determined based on the differences in operating efficiency over several recent historical periods, and the ratio of the first efficiency change to the first temperature change is taken as the change in condenser operating efficiency with outlet gas phase temperature.

[0027] Preferably, the operating parameters include the electric regulating valve on the cooling medium pipeline of the condenser and the operating frequency of the compressor of the variable frequency chiller in the condenser.

[0028] The beneficial effects of the technical solution of the present invention are:

[0029] This invention firstly achieves a shift from fixed settings to optimized settings based on historical experience, laying the foundation for efficient collaboration. Traditional methods set a fixed target outlet gas temperature for the condenser, often based on rough experience or single operating conditions, failing to reflect the complex and ever-changing actual production process. This invention, however, first determines the reference range of outlet temperatures capable of producing silicone oil with preset quality indicators (viscosity, volatile matter) based on historical production data, and further determines the initial target temperature within this range based on the condenser's operating efficiency (determined by the evaporation rate of the de-evaporation vessel and the condenser's recovery rate) from historical data. This step ensures that the initial target temperature is not arbitrarily set, but rather derived from the mining and quantification of historical high-efficiency and energy-saving production experience, thus providing an optimized starting point for the entire control process that ensures both product quality and energy efficiency.

[0030] Secondly, this invention achieves a fundamental breakthrough from static coordination to dynamic adaptive coordination. Under traditional fixed-temperature control, the condenser cannot sense or respond to independent changes in the evaporation load of the de-cooling vessel, resulting in a mismatch between the two and leading to an inability to balance energy consumption and recovery efficiency. This invention introduces a dynamic re-determination mechanism for the target temperature based on real-time operating status. This mechanism continuously monitors the change in operating efficiency with temperature at the most recent moment, as well as the difference between historical and predicted trends in outlet temperature. Based on this, it readjusts the target temperature in real time within a reference range determined by historical experience. This means that the target temperature is no longer a fixed value, but a variable that can adaptively adjust according to the current production situation (especially the efficiency trend of the condenser in responding to changes in the de-cooling vessel load). While ensuring the production of silicone oil with preset specifications, the system not only ensures that the control process based on the target temperature can take into account the working efficiency of the historical production process (i.e., take into account the historical energy saving and efficient recovery of low-boiling substances in the production process), but also can adaptively match the working condition of the de-low-boiling vessel in the current production process. This allows the active control process of the condenser to passively follow the control process of the de-low-boiling vessel, so that the combination of the evaporation of the de-low-boiling vessel and the recovery of the condenser ensures that the current production process is energy-saving and can efficiently recover low-boiling substances.

[0031] In summary, this invention integrates historical experience optimization, MPC predictive control, and real-time performance feedback dynamic optimization, enabling the condenser to not only stably produce the final silicone oil product with the required specifications, but also adaptively match the dynamic changes in the evaporation load of the front-end de-boiling kettle. Thus, in continuous production, it achieves the dual goals of energy saving and consumption reduction, as well as efficient recovery of low-boiling substances, thereby improving the energy efficiency and economy of the entire production line. Attached Figure Description

[0032] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0033] Figure 1 The flowchart illustrates the steps of an adaptive control method for condenser operation in an organosilicon oil production line, as provided in an embodiment of the present invention. Detailed Implementation

[0034] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of the adaptive control method for condenser operation in silicone oil production lines proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0035] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0036] The following description, in conjunction with the accompanying drawings, details the specific scheme of the adaptive control method for condenser operation in organosilicon oil production lines provided by this invention.

[0037] Please see Figure 1 The diagram illustrates a flowchart of an adaptive control method for condenser operation in an organosilicon oil production line, provided by an embodiment of the present invention. The method includes the following steps:

[0038] Step S101: Determine the reference range of the outlet vapor phase temperature of the condenser when producing the preset target silicone oil based on the historical production process.

[0039] In the silicone oil production line, the de-boiling reactor uses electric power to vaporize and evaporate low-boiling substances (unreacted D4, D5, DMC, MM, etc.). The condenser in the silicone oil production line is used to condense and recover the vaporized low-boiling substances from the de-boiling reactor. The condenser not only recovers raw materials and reduces raw material consumption, but also controls the final silicone oil viscosity, volatile matter, and other indicators by separating low-boiling substances. For example, by separating and recovering large quantities of monomers and small-molecule cyclosiloxanes (such as D4, D5, MM, etc.), it prevents them from mixing with long-chain polysiloxane molecules or from interfering with the entanglement and interaction of molecular chains, thereby improving the final silicone oil viscosity and volatile matter.

[0040] In this embodiment, the outlet gas phase temperature of the condenser is adaptively controlled, thereby indirectly changing the content of low-boiling-point substances, ultimately producing silicone oil with the viscosity and volatile matter required by the user. The preset silicone oil refers to the silicone oil with the viscosity and volatile matter required by the user. In this embodiment, the preset silicone oil is a silicone oil with 2% volatile matter and a viscosity of 1200 cSt. Other preset values ​​can be set in other examples; this embodiment does not impose specific limitations.

[0041] This embodiment determines the reference range of the condenser outlet vapor phase temperature when producing silicone oil with preset specifications based on historical production processes. The temperature within this reference range represents the outlet vapor phase temperature required by the condenser to ensure that the viscosity and volatile matter of the silicone oil are at or near the preset specifications. In this embodiment, adaptive control of the condenser outlet vapor phase temperature is performed to ensure that the outlet vapor phase temperature is within the reference range in order to produce silicone oil with preset specifications.

[0042] Step S102: Determine the target temperature within the reference range based on the working efficiency of the condenser in the historical production process.

[0043] The target temperature refers to the temperature at the outlet of the condenser, which needs to be controlled to ensure that the viscosity and volatile matter of the produced silicone oil are at or near the preset values.

[0044] The condenser's operating efficiency is determined by the evaporation rate of the de-lowering reactor and the condensation recovery rate of the condenser. It should be noted that this embodiment controls the condenser, not the de-lowering reactor (which can be controlled independently by other methods or manually). When the operating efficiency is high or low, it not only describes or quantifies the energy saving and efficient recovery in the production process, but also indicates whether the condenser can adapt to or follow the de-lowering reactor's operation in subsequent control processes. Furthermore, this embodiment, when controlling the condenser based on operating efficiency, can further balance energy saving and efficient recovery of low-boiling substances.

[0045] This step determines the target temperature based on the historical condenser efficiency during production, so that subsequent control based on the target temperature can refer to the energy saving and efficient recovery of low-boiling-point substances in historical production (i.e., the historical condenser efficiency during production).

[0046] Step S103: During the production of silicone oil products with preset indicators, the working parameters of the condenser are controlled in real time using the MPC algorithm, so that the outlet gas phase temperature predicted by the MPC algorithm at several future moments changes towards the target temperature.

[0047] In the process of producing silicone oil products with preset indicators, the MPC algorithm (model predictive control algorithm) is used to control the working parameters of the condenser in real time, so that the outlet gas phase temperature of the condenser changes towards the target temperature and gradually equals the target temperature or the dynamic temperature is near the target temperature.

[0048] The operating parameters mentioned include: the electric regulating valve on the cooling medium pipeline of the condenser, and the compressor operating frequency of the variable frequency chiller (or variable frequency chiller unit) of the condenser; wherein the electric regulating valve is used to regulate the cooling medium flow rate, and the compressor operating frequency is used to regulate the cooling medium temperature. The condenser described in this embodiment is a shell-and-tube condenser.

[0049] The MPC algorithm can predict the outlet gas phase temperature at several (e.g., 5) future moments after adjusting the condenser's operating parameters. Through convex optimization, the MPC algorithm makes the outlet gas phase temperature at the future moments change towards the target temperature, thereby making the outlet gas phase temperature of the condenser smoothly stabilize at the target temperature.

[0050] As an example, the state model of the MPC algorithm is set as c(i+1)=a×c(i)+[w(i),f(i)]×b. Here, a is a preset state parameter, b is a preset control parameter; c(i) represents the normalized value of the outlet gas phase temperature collected at time i, w(i) represents the valve opening percentage of the electric regulating valve at time i (dimensionless), f(i) represents the normalized value of the compressor operating frequency at time i, and c(i+1) represents the normalized value of the predicted outlet gas phase temperature at time i+1.

[0051] The normalized value of the outlet gas phase temperature refers to the ratio of the outlet gas phase temperature to 20°C; the normalized value of the compressor operating frequency refers to the ratio of the compressor operating frequency to 130Hz. The purpose of normalization by calculating the ratio is to remove dimensions and orders of magnitude.

[0052] As an example, a is set to 0.6, and b is a 2×1 matrix with elements of 0.2 and 0.4 respectively.

[0053] Since the control vector [w(i), f(i)] is a two-dimensional vector composed of the operating frequencies of the electric regulating valve and the compressor, the control input parameter R in the cost function used for convex optimization in the MPC algorithm is a 2×2 diagonal matrix. Furthermore, since the MPC algorithm predicts the outlet gas phase temperature, which is a scalar, the state deviation parameter Q in the cost function used for convex optimization in the MPC algorithm is also a scalar. In one example, the diagonal elements of this diagonal matrix R are 0.1 and 0.5, and the scalar Q is set to 1.0.

[0054] In this embodiment, the target value of the cost function is equal to the normalized value of the target temperature (i.e., the ratio of the target temperature to 20°C). In the MPC algorithm, by performing convex optimization on the cost function, the normalized value of the predicted outlet gas phase temperature can be made to approach the normalized value of the target temperature, thereby causing the outlet gas phase temperature at future times to change towards the target temperature.

[0055] It should be noted that the compressor operating frequency after MPC algorithm regulation needs to be multiplied by 130Hz (that is, after restoring the dimensions and order of magnitude) before it can be used as the compressor operating parameter. The outlet gas phase temperature predicted by MPC algorithm also needs to be multiplied by 20℃ (that is, after restoring the dimensions and order of magnitude) to obtain the outlet gas phase temperature at several future moments as described later.

[0056] In some other examples, during convex optimization in the MPC algorithm, the predicted outlet gas phase temperature is set to a range of 0~20℃, the compressor operating frequency is set to a range of 50~130Hz, and the valve opening percentage is set to a range of 0~1.

[0057] In other embodiments, the relevant parameters in the MPC algorithm can be set to other values. This embodiment does not impose specific limitations. In addition, the MPC algorithm is a well-known technology, and this embodiment will not elaborate on its specific principles.

[0058] In this embodiment, one second is considered as one moment. In other embodiments, considering that the MPC algorithm requires a long computation time, a longer time interval is used as one moment, such as 2 to 5 seconds.

[0059] Step S104: Based on the changes in the condenser's operating efficiency with the outlet gas phase temperature over several recent historical periods, and the differences in the changing trends of the outlet gas phase temperature between several recent historical periods and several future periods, the target temperature is re-determined within the reference range based on the condenser's operating efficiency in the historical production process, and the MPC algorithm is continued to be used to control the condenser's operating parameters in real time.

[0060] In the above process, after the target temperature is determined, the MPC algorithm is used to automatically control the working parameters of the condenser, so that the outlet gas phase temperature of the condenser approaches the target temperature and stabilizes at the target temperature, so that the produced silicone oil meets the preset indicators.

[0061] However, this embodiment takes into account that silicone oil production is a complex chemical reaction process involving multiple stages, each with its own independent control process (e.g., manual control). When producing silicone oil with preset specifications, the target temperature that the condenser outlet gas phase temperature needs to approach is not fixed and unique. When producing silicone oil at the fixed target temperature obtained in step S102, if the evaporation stage of the de-boiling vessel is controlled by other methods (e.g., manually and dynamically adjusting the heating power of the de-boiling vessel), there may be a problem that the condenser control process cannot adapt to or follow the working conditions of the de-boiling vessel. This results in the de-boiling vessel and condenser not working efficiently together. For example, when the de-boiling vessel is under manual control and evaporates a large amount of low-boiling substances with high energy consumption, the condenser may not recover the evaporation in time. Or, when evaporating a small amount of low-boiling substances with low energy consumption, the condenser may recover a large amount of evaporation with high power. This results in the production of silicone oil with the preset target, but it is impossible to achieve both energy saving and efficient recovery of low-boiling substances (i.e., low working efficiency).

[0062] As described above, the working efficiency in this embodiment describes whether the condenser can adapt to or follow the working conditions of the de-boiling vessel under control. Based on this, in the MPC algorithm control process, for the most recent (e.g., 5) historical moments before the current moment, this embodiment re-determines the target temperature within the reference range based on the change in the condenser's working efficiency with the outlet gas phase temperature at the most recent historical moments, and the difference in the change trend of the outlet gas phase temperature at the most recent historical moments and at the outlet gas phase temperature at the most future moments (e.g., 5 future moments). This process realizes dynamic updating of the target temperature, ensuring the production of silicone oil with preset indicators, while not only enabling the control process based on the target temperature to take into account the working efficiency in the historical production process (i.e., taking into account the historical energy saving and efficient recovery of low-boiling substances in the production process), but also adaptively matching the working conditions of the de-boiling vessel in the current production process. This allows the active control process of the condenser to passively follow the control process of the de-boiling vessel, ensuring that the evaporation of the de-boiling vessel and the recovery of the condenser combine to guarantee energy saving and efficient recovery of low-boiling substances in the current production process.

[0063] The variation of condenser efficiency with outlet gas temperature at historical moments describes the increase or decrease in efficiency when the controlled outlet gas temperature changes during the current production process. The difference in the trend between the outlet gas temperature at several historical moments and the outlet gas temperature (predicted by the MPC algorithm) at several future moments describes the deviation from the predicted outlet gas temperature when the controlled outlet gas temperature changes, if the condenser continues to be controlled according to the target temperature (i.e., the old target temperature before updating or re-determining). For example, the difference in the trend is used to describe whether the outlet gas temperature continues to increase or decrease. This embodiment redefines or updates the target temperature based on the increase or decrease in working efficiency and the continued increase or decrease in outlet gas phase temperature. This ensures that when the condenser is subsequently controlled according to the updated target temperature, the subsequent controlled process of the condenser can adapt dynamically to or follow the working conditions of the de-cooling reactor in a timely manner. This guarantees that the current production process has appropriate working efficiency, allowing the condenser to cooperate with the de-cooling reactor in the current production process to achieve the goals of energy saving and efficient recovery. For example, by adapting to or following the working conditions of the de-cooling reactor, the decrease in working efficiency can be suppressed, thereby achieving the goals of energy saving and efficient recovery.

[0064] In special cases, if there are fewer than a certain number of moments (e.g., 5) prior to the current moment, the target temperature will not be re-determined.

[0065] After resetting the target temperature at the current moment, the operating parameters of the condenser are controlled in real time using the MPC algorithm according to step S103. At the next moment or after a preset time period (e.g., after 5 seconds), step S104 is executed again to reset the target temperature, and so on, until the silicone oil production process is completed. When silicone oil is produced again, all the above steps in this embodiment are repeated in real time.

[0066] As silicone oil production continues, the amount of historical production processes increases, and the reference range determined based on these processes becomes increasingly reliable. This leads to more accurate and stable control of the condenser (i.e., consistently producing silicone oil with the required viscosity and volatile content). When there are few historical production processes, although the former is less stable and reliable than the latter when there are many, compared to existing technologies such as manual experience-based condenser control or condenser control using a fixed target temperature combined with MPC algorithms (or PID algorithms), this embodiment still achieves energy saving and efficient recovery of low-boiling-point substances while producing silicone oil with preset specifications.

[0067] Specifically, when there are few historical production processes, such as less than 50 times, this embodiment is not run. Alternatively, in an experimental production environment, multiple batches of silicone oil with different viscosities and volatile contents are produced through artificial intervention and control methods to accumulate a large number of historical production processes (e.g., more than 50 times), and then this embodiment is run.

[0068] In summary, this embodiment integrates historical experience optimization, MPC predictive control, and real-time performance feedback dynamic optimization, enabling the condenser to not only stably produce the final silicone oil product with the required specifications, but also adaptively match the dynamic changes in the evaporation load of the front-end de-boiling kettle. Thus, in continuous production, it achieves the dual goals of energy saving and consumption reduction, as well as efficient recovery of low-boiling substances, thereby improving the energy efficiency and economy of the entire production line.

[0069] As a preferred example, the reference range for the outlet vapor temperature of the condenser when producing silicone oil with a preset target output, based on historical production processes, includes the following methods:

[0070] In the historical production process, after each batch of silicone oil is produced, the indicators of that batch of silicone oil (including viscosity and volatile matter) are tested and recorded as the historical indicators of that batch of silicone oil; and the average value of the outlet gas phase temperature of the condenser at all times during the production of that batch of silicone oil (only considering the time when the condenser is working, not the time when the condenser is not working) is obtained and recorded as the historical outlet temperature of that batch of silicone oil.

[0071] For all batches of silicone oil produced throughout all historical production processes, historical indicators and historical outlet temperatures were obtained for each batch. It should be noted that because the production process of silicone oil is a complex chemical process, the historical outlet temperatures for silicone oils with the same historical indicators (i.e., the same viscosity and volatile matter) may not be unique, but rather fall within a temperature range. From all historical indicators, several historical indicators most similar to the preset indicator were selected, and the temperature range formed by the maximum and minimum values ​​of the historical outlet temperatures corresponding to these indicators was recorded as the reference range. In some examples, for the several historical indicators most similar to the preset indicator, outliers were removed from the historical outlet temperatures corresponding to these indicators (e.g., using the LOF algorithm to remove outliers), and the temperature range formed by the maximum and minimum values ​​of these outliers was then recorded as the reference range.

[0072] As an example, the method for obtaining several historical indicators that are most similar to the preset indicator from all historical indicators includes:

[0073] Viscosity and volatile matter in all historical and preset indicators are normalized using the softmax formula. For any indicator object among all historical and preset indicators, the normalized (viscosity, volatile matter) is taken as an indicator vector. The Euclidean distance between any two indicator objects is denoted as k, and exp(-k) is taken as the similarity between any two indicator objects. exp() represents an exponential function with the natural constant as the base. The n historical indicators with the greatest similarity to the preset indicator are taken as the most similar historical indicators to the preset indicator. In this embodiment, n is taken as one-tenth of the total number of historical indicators (or the total number of historical productions).

[0074] As a preferred example, determining the target temperature within the reference range includes the following process:

[0075] Each temperature within the reference range corresponds to a historical operating efficiency. All temperatures and historical operating efficiencies within the reference range are clustered into several categories, and the mean of all temperatures within each category is denoted as the center temperature. An evaluation index is calculated for each category, which is positively correlated with both the temperature distribution density and the average historical operating efficiency within each category. The center temperature of the category with the highest evaluation index is denoted as the target temperature.

[0076] The larger the evaluation index of a category, the greater the temperature distribution density and historical average working efficiency. The more concentrated the temperature distribution within the category is and the greater the average working efficiency, the more likely the temperature of this category can be prioritized as the outlet gas phase temperature required by the condenser for stable and efficient production of silicone oil, which is a preset index extracted from historical production processes.

[0077] As an example, each temperature within the reference range corresponds to a historical operating efficiency, including the following methods:

[0078] For any historical production process, the average outlet vapor temperature of the condenser at all times (i.e., the historical outlet temperature of any batch of silicone oil mentioned above) is recorded as the average operating efficiency at all times during that historical production process. For all historical production processes, the historical operating efficiency corresponding to any temperature within the reference range is interpolated using a linear interpolation algorithm based on the historical operating efficiencies corresponding to all average outlet vapor temperatures. The linear interpolation algorithm is a well-known technique and will not be described in detail in this embodiment.

[0079] Within this reference range, any temperature corresponds to a historical operating efficiency.

[0080] As an example, clustering all temperatures and historical operating efficiency within a reference range into several categories includes the following methods:

[0081] This example samples 500 temperatures at equal intervals within a reference range to represent all temperatures within that range. All temperatures within the reference range are then normalized, along with their corresponding historical operating efficiencies. This example uses the softmax formula for normalization. Each normalized temperature and its corresponding normalized historical operating efficiency are treated as a coordinate. For all coordinates within the reference range, K-Means clustering is used to create all possible clusters. In this example, the number of clusters is set to one-tenth of the total number of coordinates (rounded up). It should be noted that each cluster contains the temperature and historical operating efficiency before normalization; the normalized temperature and historical operating efficiency are not included in each cluster.

[0082] As an example, an evaluation metric is calculated for each category, which is positively correlated with the temperature distribution density within each category and the historical average work efficiency within each category, using the following methods:

[0083] Obtain the ratio of the number of coordinates in each category to the total number of coordinates across all categories, denoted as x. Use m×x / (q+1) as the evaluation metric for each category, where m represents the average historical work efficiency in each category, and q represents the standard deviation of all temperatures in each category. x / (q+1) represents the temperature distribution density; the more temperatures (coordinates) a category contains, the more concentrated the distribution (the smaller the standard deviation), and the higher the temperature distribution density of that category. q+1 is used to avoid a denominator of 0.

[0084] As a preferred example, the method for calculating work efficiency is as follows:

[0085] For any given moment in the production process, the evaporation rate of low-boiling-point substances in the de-lowering reactor at that moment is obtained. Similarly, the mass of low-boiling-point substances recovered by the condenser at each moment is recorded. The difference between the mass of low-boiling-point substances recovered by condensation at that moment and the mass of low-boiling-point substances recovered by condensation at the previous moment is recorded as the condensation recovery amount at that moment. The ratio of the condensation recovery amount to the evaporation rate is used as the working efficiency at that moment. The higher the working efficiency, the more energy-saving and efficient the recovery of low-boiling-point substances. For example, the de-lowering reactor does not need to use a large amount of heating power to evaporate low-boiling-point substances on a large scale to achieve the condensation recovery of more low-boiling-point substances. The lower the working efficiency, the lower the energy saving or recovery efficiency. For example, the de-lowering reactor evaporates a large amount of low-boiling-point substances with a large amount of heating power, but this large amount of low-boiling-point substances does not need to be condensed and recovered. In this case, the energy consumption in the de-lowering reactor is not used by the condenser for recovery, or the working process of the condenser cannot efficiently adapt to or follow the specific working conditions of the de-lowering reactor, resulting in low energy saving or recovery efficiency.

[0086] In special cases, if there is no previous time step, the work efficiency at that time step is not calculated.

[0087] As an example, the method for obtaining the evaporation rate of low-boiling-point substances is as follows:

[0088] The electric heating power at any given moment is collected in the de-caling vessel. A heat loss power is preset, which represents the power lost due to the insulation condition of the de-caling vessel when the electric heating power is used to heat the de-caling vessel. In this embodiment, the heat loss power is set to 10% of the electric heating power. In other embodiments, the power lost when heating the de-caling vessel with different electric heating powers can also be measured under laboratory conditions and used as the heat loss power.

[0089] Evaporation rate G = η × (PH - PL) / ΔH, where PH represents the electric heating power at that moment, PL represents the heat loss power, and η represents the thermal efficiency, which is a constant determined by the heat transfer properties of the de-boiling vessel; in this example, η = 0.87. ΔH represents the latent heat of vaporization of the low-boiling substance. As an example, the calculation method for this latent heat of vaporization is as follows: since the low-boiling substance mainly contains D4, the latent heat of vaporization of D4 under a specified environment is experimentally determined and used as the latent heat of vaporization of the low-boiling substance. The specified environment refers to the environment under the temperature and pressure conditions inside the de-boiling vessel at that moment.

[0090] Where η×(PH-PL) represents the effective power, in kilojoules per second; the latent heat of vaporization is in kilojoules per kilogram; and the evaporation rate represents the mass evaporated per second.

[0091] The amount of low-boiling-point substance evaporated at that moment is obtained by multiplying the time from the previous moment to the present moment by the evaporation rate. It should be noted that when the evaporation amount is small, for example, less than or equal to 0.01 kg, the evaporation amount is set to 0.01 kg to avoid the denominator being 0 when calculating the efficiency above.

[0092] As an example, based on the variation of condenser efficiency with outlet gas temperature over several recent historical periods, and the difference in the variation trend of outlet gas temperature between several recent historical periods and several future periods, the target temperature is re-determined within a reference range based on the condenser efficiency during the historical production process. The methods include:

[0093] The ratio of the first rate of change of the outlet gas phase temperature over time at several future moments to the second rate of change of the outlet gas phase temperature over time at several recent historical moments is denoted as the trend difference. Specifically, when the first or second rate of change is in the range [-0.01, 0.01], if the first or second rate of change is less than 0, then the first or second rate of change is set to -0.01; if the first or second rate of change is greater than 0, then the first or second rate of change is set to 0.01. The purpose of this is to avoid the denominator of the above ratio being 0, and at the same time, to ensure that when the first or second rate of change approaches 0, the trend difference is equal to 1 (ensuring that the target temperature can still be updated when the first or second rate of change approaches 0).

[0094] When the change is greater than 0, there is no decline in working efficiency and the target temperature is not changed. At this time, the target temperature can be adapted to the working efficiency in historical production (that is, the target temperature is the outlet gas phase temperature required by the condenser when producing stable and efficient silicone oil as a preset index extracted from historical production process).

[0095] When the change is less than or equal to 0, it indicates that the work efficiency has declined during the historical control process, and the next step should be carried out:

[0096] A difference in the trend greater than 1 (including or equal to 1) indicates that, based on the historical increase (or decrease) in outlet gas phase temperature, the outlet gas phase temperature will continue to increase (or decrease) at a greater rate in the future. In this case, the downward trend in the condenser's efficiency may intensify further. Therefore, it is necessary to change the target temperature to suppress this decline. Specifically, the target temperature is updated within a reference range based on the condenser's historical production efficiency, ensuring that the change in target temperature is positively correlated with both the absolute value of the change and the difference in the trend. This process, by changing the target temperature, ensures that it not only adapts to historical production efficiency but also suppresses the decline in efficiency during the current production process.

[0097] When the difference in the trend is less than 1, it indicates that based on the increasing (or decreasing) outlet gas phase temperature in historical time periods, the increasing (or decreasing) trend of the outlet gas phase temperature in future time periods will slow down, or the outlet gas phase temperature will begin to decrease (or increase) in future time periods. In this case, the downward trend of working efficiency in the control process in future time periods will be suppressed. At this time, the target temperature is not changed, so that the target temperature can be adapted to the working efficiency in historical production.

[0098] As an example, methods for obtaining the variation of condenser operating efficiency with outlet gas phase temperature at several recent historical moments include:

[0099] The condenser outlet gas phase temperature at several recent historical moments (including the current moment) and the predicted outlet gas phase temperature at several future moments are normalized. In this example, the softmax formula is used for normalization. The operating efficiency at several recent historical moments (including the current moment) is also normalized using the softmax formula. The purpose of normalization is to remove dimensions and orders of magnitude.

[0100] In the most recent historical moments (including the current moment), the absolute value of the difference between the normalized outlet gas phase temperature at the last moment and the normalized outlet gas phase temperature at the first moment is denoted as the first temperature change y1. The difference between the normalized operating efficiency at the last moment and the normalized operating efficiency at the first moment is denoted as the first operating efficiency change y2. The ratio of y2 to y1 is denoted as the change, which describes the change in operating efficiency per unit change in outlet gas phase temperature. If the change is greater than 0, it indicates that the operating efficiency generally shows an increasing trend over a short period of time (i.e., over several historical moments). The larger the change, the more obvious the increasing trend in operating efficiency. If the change is less than 0, it indicates that the operating efficiency generally shows a decreasing trend over a short period of time (i.e., over several historical moments). The smaller the change, the more obvious the decreasing trend in operating efficiency.

[0101] It should be noted that when calculating the ratio of y2 to y1, a constant with a value of 1 is added to the denominator to avoid the denominator being equal to 0, and to ensure that the change is within the interval [-1, 1].

[0102] As an optional example, the target temperature is updated within a reference range based on the historical condenser operating efficiency during the production process, such that the change in target temperature is positively correlated with both the absolute value and the difference in the trend of the change. This includes the following steps:

[0103] The update change is determined based on the absolute value and trend difference of the change, and the update change represents the magnitude of the increase or decrease in the target temperature when updating the target temperature. The update change is positively correlated with both the absolute value and trend difference of the change, indicating that the larger the absolute value of the change or the larger the trend difference, the more significantly the target temperature needs to be increased or decreased, so as to significantly suppress the downward trend of work efficiency.

[0104] Provided that the change amount is less than or equal to 0 and the difference in the change trend is greater than or equal to 1, when both the first change rate and the second change rate are greater than or equal to 0, it indicates that during the historical control process, the working efficiency declined as the outlet gas phase temperature increased in historical moments. Further increases in the outlet gas phase temperature in future moments may exacerbate the downward trend of working efficiency. At this time, it is necessary to reduce the target temperature to suppress the downward trend of working efficiency. The target temperature minus the update change amount is then used as the initial update result.

[0105] When both the first and second rates of change are less than 0, it indicates that during the historical control process, the working efficiency declined as the outlet gas phase temperature decreased in historical time periods. Further decreases in the outlet gas phase temperature in future time periods may exacerbate the downward trend of working efficiency. In this case, it is necessary to increase the target temperature to suppress the downward trend of working efficiency. At this time, the target temperature plus the aforementioned update change is used as the initial update result.

[0106] For all temperatures and historical operating efficiencies within the aforementioned reference range, and for all clusters into which they are clustered, when it is necessary to reduce the target temperature, the category with a center temperature less than or equal to the target temperature is designated as the reference category; when it is necessary to increase the target temperature, the category with a center temperature greater than or equal to the target temperature is designated as the reference category.

[0107] Obtain the absolute value of the difference between the center temperature of each reference category and the initial update result, and denote it as the selection index for each reference category. Among all reference categories, obtain the reference category with the smallest selection index, and use the center temperature of this reference category as the updated target temperature (i.e., the redefined target temperature).

[0108] The smaller the selected index, the closer the center temperature of the reference category is to the temperature value after increasing or decreasing the target temperature.

[0109] As a preferred example, the target temperature is updated within a reference range based on the historical condenser operating efficiency during the production process, such that the change in target temperature is positively correlated with both the absolute value and the difference in the trend of the change. This includes the following steps:

[0110] For the reference categories obtained from the above optional examples, each reference category corresponds to an evaluation index. The larger the evaluation index, the greater the temperature distribution density and historical average working efficiency. The temperature of the reference category can be given priority as the outlet gas phase temperature of the condenser when producing the preset index of silicone oil, which is extracted from the historical production process.

[0111] Meanwhile, each of the above reference categories corresponds to a selection metric. In this embodiment, the evaluation metrics of all reference categories are normalized using the softmax formula, and the selection metric of each reference category is reduced by a specific percentage, which is equal to the normalized evaluation metric of each reference category. Thus, each reference category corresponds to a reduced selection metric. Among all reference categories, the reference category with the smallest reduced selection metric is obtained, and the center temperature of this reference category is used as the updated target temperature (i.e., the redefined target temperature).

[0112] In this preferred example, the selected indicators are updated (i.e., reduced) based on the evaluation metrics. The larger the evaluation metric, the greater the reduction. This ensures that the center temperature of the reference category with higher temperature distribution density and historical average working efficiency can be preferentially used as the updated target temperature, avoiding the use of the center temperature of the reference category with lower temperature distribution density and historical average working efficiency as the updated target temperature. Compared to the optional example, this preferred example not only suppresses the downward trend in working efficiency when updating the target temperature, but also further references historical production processes with higher temperature distribution density and historical average working efficiency, ensuring the reliability of the target temperature update.

[0113] As an example, determining the updated change amount based on the absolute value of the change and the difference in the trend of change includes the following methods:

[0114] Normalizing the difference in trend includes dividing the difference in trend by a constant, such as 10, and using the result as the normalization result, where if the normalization result is greater than 1, the normalization result is set to 1.

[0115] The product of the absolute value of the change and the normalized difference in the trend of change is recorded as the update percentage, denoted as h1. The difference between the maximum and minimum center temperatures of all reference categories is recorded as the first difference h2, and h1 × h2 is used as the update change. This process ensures that the update change is no greater than the difference between the maximum and minimum center temperatures of all reference categories (i.e., h2), so that the updated target temperature does not exceed the reference range. Specifically, when there is only one reference category, the target temperature before and after the update is the same.

[0116] As an example, the first rate of change and the second rate of transformation are obtained as follows:

[0117] Plot the predicted export gas phase temperature over time for several recent future moments, and fit the curve to a straight line using the least squares method. The slope of this straight line is denoted as the first rate of change.

[0118] Plot the curves of the outlet gas phase temperature versus time at several recent historical moments, and fit the curves into a straight line using the least squares method. The slope of this straight line is denoted as the second rate of change.

[0119] 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 principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An adaptive control method for condenser operation in an organosilicon oil production line, characterized in that, The method includes the following steps: The reference range of the condenser outlet vapor temperature is determined based on the historical production process to produce silicone oil with preset specifications. The target temperature is then determined based on the condenser's operating efficiency within the reference range during the historical production process. The operating efficiency is determined by the evaporation rate of the de-evaporation reactor and the condensation recovery rate of the condenser. During the production of silicone oil with preset specifications, the MPC algorithm is used to control the condenser's operating parameters in real time, causing the outlet vapor temperature predicted by the MPC algorithm for several future moments to change towards the target temperature. Simultaneously, based on the change in condenser operating efficiency with outlet vapor temperature over several recent historical moments, and the difference in the changing trends of outlet vapor temperature between several recent historical moments and several future moments, the target temperature is re-determined within the reference range based on the condenser's operating efficiency during the historical production process, and the MPC algorithm is continued to be used to control the condenser's operating parameters in real time.

2. The adaptive control method for condenser operation in an organosilicon oil production line according to claim 1, characterized in that, The specific steps involved in determining the target temperature within a reference range based on the condenser's operating efficiency during historical production processes are as follows: Each temperature within the reference range is assigned a historical operating efficiency. All temperatures and historical operating efficiencies within the reference range are clustered into several categories. The mean of all temperatures within each category is recorded as the center temperature. An evaluation index is calculated for each category. The evaluation index is positively correlated with the temperature distribution density and the average historical operating efficiency within each category. The center temperature of the category with the largest evaluation index is recorded as the target temperature.

3. The adaptive control method for condenser operation in an organosilicon oil production line according to claim 2, characterized in that, The specific steps for re-determining the target temperature within a reference range based on the condenser's operating efficiency over several recent historical periods, and the difference in the changing trends of the outlet gas phase temperature between these periods and future periods, are as follows: The ratio of the first rate of change of the outlet gas phase temperature over time at several future moments to the second rate of change of the outlet gas phase temperature over time at several recent historical moments is denoted as the trend difference. When the change is greater than 0, the target temperature remains unchanged; When the change is less than or equal to 0, if the difference in the trend of change is greater than or equal to 1, the target temperature is updated within the reference range based on the working efficiency of the condenser in the historical production process, so that the change in the target temperature is positively correlated with the absolute value of the change and the difference in the trend of change, respectively. If the difference in the trend is less than 1, the target temperature will not be changed.

4. The adaptive control method for condenser operation in an organosilicon oil production line according to claim 3, characterized in that, The specific steps involved in updating the target temperature within a reference range based on the condenser's historical production efficiency are as follows: When both the first rate of change and the second rate of change are greater than or equal to 0, among all the categories, the category with a center temperature less than or equal to the target temperature is recorded as the reference category, and the target temperature minus the update change amount is used as the initial update result; when both the first rate of change and the second rate of change are less than 0, the category with a center temperature greater than or equal to the target temperature is recorded as the reference category, and the target temperature plus the update change amount is used as the initial update result. The difference between the center temperature of each reference category and the initial update result is the selection metric for each reference category; The selection metrics for each reference category are reduced, with the reduction amount being positively correlated with the evaluation metrics for each reference category; Among all reference categories, the center temperature of the reference category with the smallest reduced selection index is used as the updated target temperature.

5. The adaptive control method for condenser operation in an organosilicon oil production line according to claim 4, characterized in that, The specific steps for updating the change amount are as follows: The update percentage is determined, and the update percentage is positively correlated with the absolute value of the change and the difference in the trend of change, respectively; the product of the first difference between the maximum and minimum center temperatures of all reference categories and the update percentage is recorded as the update change.

6. The adaptive control method for condenser operation in an organosilicon oil production line according to claim 1, characterized in that, The specific steps involved in determining the reference range of the condenser outlet vapor phase temperature when producing silicone oil with preset indicators based on historical production processes are as follows: In the historical production process, after each batch of silicone oil is produced, the indicators of that batch of silicone oil are tested, including viscosity and volatile matter, and recorded as the historical indicators of that batch of silicone oil; the average value of the outlet vapor phase temperature of the condenser at all times during the production of that batch of silicone oil is obtained and recorded as the historical outlet temperature of that batch of silicone oil. For all batches of silicone oil in all historical production processes, the historical indicators and historical outlet temperatures are used to obtain the most similar historical indicators to the preset indicators. The temperature range formed by the maximum and minimum values ​​of the historical outlet temperatures corresponding to the most similar historical indicators is recorded as the reference range.

7. The adaptive control method for condenser operation in an organosilicon oil production line according to claim 2, characterized in that, Each temperature within the reference range is assigned a historical operating efficiency, and the specific steps involved are as follows: For any historical production process, the average outlet vapor temperature of the condenser at all times is recorded as the average operating efficiency at all times of the historical production process. For all historical production processes, the historical operating efficiency corresponding to any temperature within the reference range is interpolated using a linear interpolation algorithm based on the historical operating efficiency corresponding to all average outlet vapor temperatures.

8. The adaptive control method for condenser operation in an organosilicon oil production line according to claim 3, 4, or 7, characterized in that, The specific steps for obtaining the work efficiency are as follows: For any given moment in the production process, obtain the evaporation rate and the condensation recovery rate in the de-evaporation vessel at that moment; the ratio of the condensation recovery rate to the evaporation rate is taken as the working efficiency at that moment.

9. The adaptive control method for condenser operation in an organosilicon oil production line according to claim 3, characterized in that, The specific steps for obtaining the change in condenser efficiency with outlet gas phase temperature at the most recent historical moments are as follows: The first temperature change is determined based on the differences in outlet gas phase temperature over several recent historical periods; the first efficiency change is determined based on the differences in operating efficiency over several recent historical periods, and the ratio of the first efficiency change to the first temperature change is taken as the change in condenser operating efficiency with outlet gas phase temperature.

10. The adaptive control method for condenser operation in an organosilicon oil production line according to claim 1, characterized in that, The operating parameters include the electric regulating valve on the cooling medium pipeline of the condenser and the operating frequency of the compressor of the variable frequency chiller in the condenser.

Images