Rectification control model generation method, rectification control method and rectification device

Abstract

The present application discloses a rectification control model generation method, a rectification control method, and a rectification device. The generation method includes: constructing a simulated rectification model and performing decoupling processing to generate a control loop between each device parameter and the simulated concentration information of the rectification product; constructing a to-be-tuned control model corresponding to the device parameter according to the control loop, and the to-be-tuned control model is used to adjust the device parameter to control the rectification process; constructing an equivalent model according to the to-be-tuned control model; obtaining a target performance index to perform parameter tuning on the equivalent model to obtain a parameter-tuned equivalent model; tuning the controller gain and observer gain of the to-be-tuned control model according to the parameter-tuned equivalent model to obtain a target control model. By combining the actual rectification process of the rectification device during the model establishment process to perform parameter tuning on the model parameters, a rectification control model with strong anti-interference ability, high precision, and high stability is generated, which is convenient to control the rectification process through the model to obtain the required product concentration.

Description

Technical Field

[0001] The present application relates to the field of distillation control, and particularly to a method for generating a distillation control model, a distillation control method, and a distillation device. Background Art

[0002] A distillation device is used to perform distillation operation on a mixture. Taking a distillation column as an example: The distillation column utilizes the different volatilities of each component in the mixture to transfer the light components (low-boiling substances) in the liquid phase to the gas phase, while transferring the heavy components (high-boiling substances) in the gas phase to the liquid phase, so as to achieve the separation of the mixture and obtain separated products.

[0003] Specifically, during the distillation process, the thermodynamic parameters and pressure parameters inside the distillation column will greatly affect the concentration of the separated products. Since mass transfer and heat transfer phenomena will occur during the distillation process of the distillation column, and there are certain external disturbances in the distillation process itself, such as errors in feed flow rate, feed temperature, feed components, etc., which will cause large disturbances in the internal state of the distillation column during the distillation process. Therefore, it is difficult to achieve high-precision and high-stability control of the distillation column using a traditional controller model, and it is necessary to tune multiple model parameters, resulting in complex parameter tuning. Summary of the Invention

[0004] The present application provides a method for generating a distillation control model, a distillation control method, and a distillation device, aiming to tune model parameters in combination with the actual distillation process of the distillation device during the model establishment process, generate a distillation control model with strong anti-external disturbance ability, so as to achieve high-precision and high-stability control, and facilitate controlling the distillation process through the model to obtain the required product concentration.

[0005] In a first aspect, the present application provides a method for generating a distillation control model, characterized in that the method includes:

[0006] Construct a simulated distillation model according to the distillation device, and the simulated distillation model is used to generate simulated concentration information of the distillation product according to the device parameters of the distillation device;

[0007] Decouple the simulated distillation model to generate a control loop between each device parameter and the simulated concentration information of the distillation product;

[0008] Construct a to-be-tuned control model corresponding to the device parameter according to the control loop, wherein the to-be-tuned control model is used to adjust the device parameter according to the controller gain, the observer gain, and the concentration target value of the distillation product to control the distillation process of the distillation device;

[0009] Construct an equivalent model including an equivalent filter and an equivalent controller according to the to-be-tuned control model;

[0010] Obtain the target performance indicators of the equivalent model, and perform parameter tuning on the equivalent filter and equivalent controller of the equivalent model according to the target performance indicators to obtain a parameter-tuned equivalent model;

[0011] Tune the controller gain and observer gain of the to-be-tuned control model according to the parameter-tuned equivalent model to obtain the target control model.

[0012] In a second aspect, the present application also provides a rectification control method, and the method includes:

[0013] When receiving a target rectification instruction, parse the target rectification instruction to obtain the concentration target value of the rectification product, and obtain the current device parameters of the rectification device;

[0014] Call the target control model according to the target rectification instruction, where the target control model is a target control model generated by using any rectification control model generation method provided in the embodiments of the present application;

[0015] Input the concentration target value and the device parameters of the rectification device into the target control model, so that the target control model generates the device parameter target value according to the concentration target value and the device parameters;

[0016] Adjust the corresponding device parameters in the rectification device according to the device parameter target value to control the rectification process of the rectification device.

[0017] In a third aspect, the present application also provides a rectification device, including:

[0018] A raw material supply mechanism for providing a mixture raw material;

[0019] A rectification reaction mechanism for providing a reaction site for the mixture raw material and performing rectification treatment on the mixture raw material in the reaction site to generate a rectification product, where the rectification reaction mechanism has device parameters to be adjusted;

[0020] A product delivery mechanism for delivering the generated rectification product to a target container;

[0021] A control mechanism, connected to the rectification reaction mechanism, and used to execute any rectification control method provided in the embodiments of the present application to adjust the device parameters of the rectification reaction mechanism to control the rectification process of the rectification device.

[0022] The present application provides a rectification control model generation method, a rectification control method, and a rectification device. The generation method includes: constructing a simulated rectification model and performing decoupling processing to generate a control loop between each device parameter and the simulated concentration information of the rectification product; constructing a to-be-tuned control model corresponding to the device parameter according to the control loop, where the to-be-tuned control model is used to adjust the device parameter to control the rectification process; constructing an equivalent model according to the to-be-tuned control model; obtaining a target performance index to perform parameter tuning on the equivalent model to obtain a parameter-tuned equivalent model; and tuning the controller gain and observer gain of the to-be-tuned control model according to the parameter-tuned equivalent model to obtain a target control model. By combining the actual rectification process of the rectification device during the model establishment process to perform parameter tuning, a rectification control model with strong anti-interference ability, high precision, and high stability is generated, which is convenient to control the rectification process through the model to obtain the required product concentration. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present application. For those skilled in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0024] Figure 1 It is a schematic diagram of the scenario of a rectification control model generation method provided by an embodiment of the present invention;

[0025] Figure 2 It is a schematic block diagram of the structure of a control component provided by an embodiment of the present application;

[0026] Figure 3 It is a schematic flowchart of a rectification control model generation method provided by an embodiment of the present invention;

[0027] Figure 4 It is a structural block diagram of an equivalent model in a rectification control model generation method provided by an embodiment of the present invention;

[0028] Figure 5 It is a schematic flowchart of the parameter tuning of the equivalent model in a rectification control model generation method provided by an embodiment of the present invention;

[0029] Figure 6 It is a schematic flowchart of the parameter tuning of the equivalent controller in a rectification control model generation method provided by an embodiment of the present invention;

[0030] Figure 7 It is a schematic flowchart of the determination of the target point in the parameter tuning of the equivalent controller in a rectification control model generation method provided by an embodiment of the present invention;

[0031] Figure 8It is the open-loop frequency characteristic curve diagram of the target control model in a rectification control model generation method provided by an embodiment of the present invention;

[0032] Figure 9 It is a schematic flow chart of a rectification control method provided by an embodiment of the present invention;

[0033] Figure 10 It is a schematic diagram of the module structure of a rectification model provided by an embodiment of the present invention. Detailed implementation manners

[0034] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application. The flow charts shown in the accompanying drawings are only illustrative examples, and do not necessarily include all the contents and operations / steps, nor do they necessarily need to be executed in the described order. Some operations / steps can also be decomposed, combined, or partially merged.

[0035] It should be understood that the terms used in the specification are only for the purpose of describing specific embodiments and are not intended to limit the present application. Unless clearly specified otherwise in the context, the singular forms of "a", "an", and "the" are intended to include the plural forms. "And / or" refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0036] Next, some embodiments of the present application will be described in detail in conjunction with the accompanying drawings. Without conflict, the following embodiments and the features in the embodiments can be combined with each other.

[0037] It should be noted that the rectification device includes a rectification reaction mechanism, for example, a rectification column. A rectification column is a tower-type gas-liquid contact device for rectification. It utilizes the different volatilities of the components in the mixture to transfer the light components (low-boiling substances) in the liquid phase to the gas phase, while the heavy components (high-boiling substances) in the gas phase are transferred to the liquid phase, obtaining at least two rectification products to achieve the purpose of separation.

[0038] Please refer to Figure 1 , Figure 1 It is a schematic diagram of the scenario of a rectification control model generation method provided by an embodiment of the present invention.

[0039] Such as Figure 1As shown, the rectifying reaction mechanism is, for example, a rectifying column, which includes a main body, a condenser, a reboiler and a number of trays. Among them, the feed enters the column from a certain tray in the rectifying column, and this tray is called the feed tray. The feed tray divides the rectifying column into upper and lower sections. The part above the feed tray is called the rectifying section, and the part below the feed tray is called the stripping section.

[0040] An embodiment of the present application provides a method for generating a rectifying control model. This method for generating a rectifying control model can be applied to electronic devices such as tablet computers, laptop computers, wearable devices, etc. or control mechanisms. Among them, the control mechanism can be an independent control mechanism or a control mechanism cluster.

[0041] Taking the application to a control mechanism as an example, the purpose of implementing this method is to tune the model parameters in combination with the actual rectifying process of the rectifying device to generate a rectifying control model with strong anti-external disturbance ability. The rectifying device can call the generated rectifying control model to output adjusted device parameters, and control the rectifying process of the rectifying device according to the adjusted device parameters. Among them, the device parameters are, for example, pressure parameters and thermal parameters in the rectifying device.

[0042] Please refer to Figure 2 , Figure 2 which is a schematic flow chart of a method for generating a rectifying control model provided by an embodiment of the present invention.

[0043] As Figure 2 shown, the method for generating a rectifying control model specifically includes steps S1 to S6.

[0044] Step S1: Construct a simulated rectifying model according to the rectifying device. The simulated rectifying model is used to generate simulated concentration information of rectifying products based on the device parameters of the rectifying device.

[0045] It should be noted that the simulated rectifying model is used to simulate and imitate the rectifying process of the rectifying device. Specifically, the device parameters of the rectifying device are input into the simulated rectifying model, and the simulated rectifying model generates and outputs the simulated concentration information of the rectifying products based on the device parameters of the rectifying device.

[0046] Among them, the simulated concentration information of the rectifying products specifically includes the concentration values of the rectifying products obtained by simulating and imitating the rectifying process. For the same simulated rectifying model, when the device parameters change, the simulated concentration information of the rectifying products changes accordingly.

[0047] Step S2: Decouple the simulated rectifying model to generate control loops between each device parameter and the simulated concentration information of the rectifying products.

[0048] For example, the device parameters at least include pressure parameters and thermal parameters in the rectifying device, and the simulated concentration information at least includes the concentration values of various rectifying products.

[0049] It should be noted that during the rectification process, the thermal parameters and pressure parameters inside the rectification column will greatly affect the concentration of the rectification products. There is a coupled relationship among the thermal parameters, pressure parameters, and the concentration values of various generated rectification products. Adjusting a single device parameter will bring uncontrollable changes to the concentration values of various rectification products.

[0050] In contrast, the simulated rectification model constructed according to the rectification device is a multi-input and multi-output model. Among them, the input quantities of the simulated rectification model at least include the thermal parameters and pressure parameters in the rectification device, and the output quantities at least include the concentration values of various generated rectification products.

[0051] In order to simplify the model generation and parameter tuning process and avoid the mutual coupling and interference of multiple device parameters, the method first decouples the simulated rectification model to generate a control loop between each device parameter and the simulated concentration information of the rectification products.

[0052] Taking the rectification device including a rectification column as a specific example, during the rectification process of the rectification column, a top product of light components and a bottom product of heavy components will be produced at the top and bottom of the rectification column respectively. The device parameters of the rectification device at least include the pressure parameter and thermal parameter of the rectification column, and the simulated concentration information at least includes the first product concentration of the top product and the second product concentration of the bottom product. Further, the pressure parameter in the rectification column refers to the pressure difference between the rectifying section and the stripping section in the rectification column, and the thermal parameter is used to characterize the thermal condition in the rectification column.

[0053] In some embodiments, decoupling the simulated rectification model to generate a control loop between each device parameter and the simulated concentration information includes:

[0054] Obtaining the transfer function matrix between the pressure parameter, thermal parameter, and simulated concentration information;

[0055] Decoupling the transfer function matrix to generate a first control loop between the pressure parameter and the simulated concentration information;

[0056] Decoupling the transfer function matrix to generate a second control loop between the thermal parameter and the simulated concentration information.

[0057] Specifically, first obtain the transfer function matrix between the pressure parameter, thermal parameter, and simulated concentration information according to the simulated rectification model, and then generate the first control loop and the second control loop based on the transfer function matrix.

[0058] In some embodiments, the rectification device is at least used to output a first rectification product and a second rectification product;

[0059] Obtaining the transfer function matrix between the pressure parameter, thermal parameter, and simulated concentration information includes:

[0060] Construct a first transfer function of the pressure parameter and the first product concentration of the first rectification product according to the rectification device, and construct a second transfer function of the pressure parameter and the second product concentration of the second rectification product according to the rectification device;

[0061] Construct a third transfer function of the thermal parameter and the first product concentration according to the rectification device, and construct a fourth transfer function of the thermal parameter and the second product concentration according to the rectification device;

[0062] Construct a transfer function matrix according to the first transfer function, the second transfer function, the third transfer function and the fourth transfer function.

[0063] Exemplarily, the simulation rectification model can be expressed as:

[0064]

[0065] Wherein, Y is the first product concentration of the overhead product of the rectification column, X is the second product concentration of the bottom product of the rectification column, P is the pressure parameter, Q is the thermal parameter, and p11, p12, p21 and p22 are respectively the first transfer function between the first product concentration and the pressure parameter, the second transfer function between the second product concentration and the pressure parameter, the third transfer function between the first product concentration and the thermal parameter, and the fourth transfer function between the second product concentration and the thermal parameter.

[0066] It should be noted that the transfer function matrix is specifically a transfer composed of the above multiple transfer functions. Then the transfer function matrix between the pressure parameter and the thermal parameter and the simulated concentration information can be expressed as:

[0067]

[0068] After generating the transfer function matrix, decouple the transfer function matrix to generate a first control loop between the pressure parameter and the simulated concentration information, and generate a second control loop between the thermal parameter and the simulated concentration information.

[0069] Furthermore, based on the simulated concentration parameters including the first product concentration of the overhead product and the second product concentration of the bottom product, the first control loop specifically includes:

[0070] A first sub-loop with the pressure parameter as the input and the first product concentration as the output;

[0071] A second sub-loop with the pressure parameter as the input and the second product concentration as the output;

[0072] A third sub-loop with the thermal parameter as the input and the first product concentration as the output;

[0073] and a fourth sub-loop with the input being a thermal parameter and the output being the concentration of the second product.

[0074] Based on the above corresponding sub-loops, adjusting the thermal parameter and / or the pressure parameter can control the concentration of the first product and the concentration of the second product produced by the distillation column.

[0075] Specifically, any one of the first sub-loop, the second sub-loop, the third sub-loop, and the fourth sub-loop can be expressed as:

[0076]

[0077] where i = 1, 2, 3, 4, and i is used to represent that this loop is the i-th sub-loop. is the input of the i-th sub-loop. is the output of the i-th sub-loop. is the total loop disturbance of the i-th sub-loop. is the input gain of the i-th sub-loop.

[0078] Taking i = 1 as an example, this formula is used to represent the first sub-loop. is the input of the first sub-loop, that is, the pressure parameter. is the output of the first sub-loop, that is, the concentration of the first product. is the total loop disturbance of the first sub-loop. is the input gain of the first sub-loop.

[0079] By decoupling the simulated distillation model to generate a control loop between each device parameter and the simulated concentration information of the distillation product, the simulated distillation model is simplified to a loop relationship in which each output corresponds one-to-one with each output, avoiding the mutual coupling interference of multiple device parameters during the subsequent process of tuning the parameters of each loop.

[0080] Step S3: Construct a to-be-tuned control model corresponding to the device parameters according to the control loop, where the to-be-tuned control model is used to adjust the device parameters according to the controller gain, the observer gain, and the concentration target value of the distillation product to control the distillation process of the distillation device.

[0081] It should be noted that the model parameters to be tuned in the to-be-tuned control model at least include the controller gain and the observer gain. The to-be-tuned control model is used to adjust the device parameters according to the controller gain, the observer gain, and the concentration target value of the distillation product to control the distillation process of the distillation device.

[0082] For example, after the control mechanism executing this method generates the control loop, constructing a to-be-tuned control model corresponding to the device parameters according to the control loop includes:

[0083] Construct a to-be-tuned control model between the pressure parameter and the first product concentration according to the first sub-loop in the first control loop; and / or, construct a to-be-tuned control model between the pressure parameter and the second product concentration according to the second sub-loop in the first control loop; and / or, construct a to-be-tuned control model between the thermal parameter and the first product concentration according to the third sub-loop in the second control loop; and / or, construct a to-be-tuned control model between the thermal parameter and the second product concentration according to the fourth sub-loop in the second control loop.

[0084] In some embodiments, constructing a to-be-tuned control model corresponding to the device parameters according to the control loop includes:

[0085] Construct a controller model according to the control loop, and the controller model is used to control the simulation rectification model to generate simulated concentration information according to the controller gain and the device parameters;

[0086] Construct an observer model according to the controller model, and the observer model is used to observe the simulated concentration information to generate observed concentration information according to the observer gain, and perform feedback compensation on the device parameters according to the concentration target value and the observed concentration information;

[0087] Construct a to-be-tuned control model based on the controller model and the observer model.

[0088] Please refer to Figure 3 , Figure 3 which is the model structure block diagram of the to-be-tuned control model provided by the embodiment of the present application.

[0089] As Figure 3 shown, the constructed to-be-tuned control model includes at least two parts: a controller model and an observer model. Among them, the controller model is used to control the simulation rectification model according to the controller gain and the device parameters to generate rectification products, while the observer model is used to observe the simulated concentration information of the simulation rectification model according to the observer gain to generate observed concentration information, and perform feedback compensation on the device parameters according to the concentration target value and the observed concentration information.

[0090] It should be noted that the simulated concentration information is the concentration value generated by the simulation rectification model under the control of the controller model, and the observed concentration information is the observed value generated by the observer model observing the above concentration value. Based on the to-be-tuned controller gain and observer gain, there is a deviation between the simulated concentration information and the observed concentration information.

[0091] Specifically, to construct the to-be-tuned control model, first construct a controller model according to the control loop, then construct an observer model according to the controller model, and then construct a to-be-tuned control model based on the controller model and the observer model.

[0092] Step S4: Construct an equivalent model including an equivalent filter and an equivalent controller according to the control model to be tuned.

[0093] After constructing the control model to be tuned corresponding to the device parameters according to the control loop, according to the model structure of the control model to be tuned, convert the control model to be tuned into an equivalent model equivalent thereto, where the equivalent model includes at least an equivalent filter and an equivalent controller.

[0094] In some embodiments, constructing an equivalent model including an equivalent filter and an equivalent controller according to the control model to be tuned includes:

[0095] Construct an equivalent controlled object according to the control model to be tuned, and the equivalent controlled object is used to generate simulated concentration information according to the input control instruction;

[0096] Construct an equivalent filter and an equivalent controller according to the equivalent controlled object, where the equivalent filter is used to filter the concentration target value to generate a concentration filtered value, and the equivalent controller is used to input a control instruction to the equivalent controlled object according to the concentration filtered value.

[0097] It should be noted that the construction process of the equivalent controlled object includes: according to the control model to be tuned, convert the simulated rectification model into an equivalent model, that is, the equivalent controlled object corresponds to the simulated rectification model. Among them, the equivalent controlled object is used to respond to the input control instruction and generate corresponding simulated concentration information under the control of the control instruction.

[0098] It should also be noted that according to the model structure of the control model to be tuned, specifically, the model structure of the control model to be tuned is converted into an equivalent model equivalent thereto, where the equivalent model includes at least an equivalent filter and an equivalent controller.

[0099] Please refer to Figure 4 , Figure 4 which is the model structure block diagram of the equivalent model provided by the embodiment of the present application.

[0100] The equivalent model includes an equivalent filter and an equivalent controller. The equivalent filter is used to filter the concentration target value to generate a concentration filtered value, and the equivalent controller is used to input a control instruction to the equivalent controlled object according to the concentration filtered value, so that the equivalent controlled object is used to generate simulated concentration information according to the input control instruction.

[0101] Further, when the equivalent filter filters the concentration target value, specifically, first obtain the simulated concentration information as feedback to perform filtering processing on the concentration target value corresponding to the simulated concentration information, so as to realize the loop control of the simulated concentration information.

[0102] Specifically, when constructing the equivalent model, first, an equivalent controlled object is constructed according to the control model to be tuned, and then an equivalent filter and an equivalent controller are constructed based on the controller model and the observer model in the equivalent controlled object. Among them, the equivalent filter has model parameters to be tuned, including the gain, zeros, and poles of the transfer function of the equivalent filter, and the equivalent controller has model parameters to be tuned, including the gain, zeros, and poles of the transfer function of the equivalent controller. It should be noted that during the process of constructing the equivalent filter, there is a preset first correspondence between the controller gain and the observer gain and the transfer function of the equivalent controller, and there is a preset second correspondence between the controller gain and the observer gain and the transfer function of the equivalent filter.

[0103] Specifically, the control instruction can be expressed as:

[0104]

[0105] where U is the control instruction, g is the transfer function of the equivalent controller, f is the transfer function of the equivalent filter, R is the concentration target value, and Y is the analog concentration information after Laplace transform.

[0106] Step S5: Obtain the target performance indicators of the equivalent model, and tune the parameters of the equivalent filter and the equivalent controller of the equivalent model according to the target performance indicators to obtain a parameter-tuned equivalent model.

[0107] As Figure 5 shown, in some embodiments, the target performance indicators include controller indicators and filter performance indicators. Correspondingly, in step S5, tuning the parameters of the equivalent controller and the equivalent filter of the equivalent model according to the target performance indicators to obtain a parameter-tuned equivalent model includes:

[0108] Step S51: Construct a closed-loop curve dynamically corresponding to the equivalent controller;

[0109] Step S52: Adjust the model parameters of the equivalent controller according to the closed-loop boundary indicators until the closed-loop curve corresponding to the equivalent controller is constrained by the controller indicators;

[0110] Step S53: Construct a filter performance function for characterizing the filter tracking performance;

[0111] Step S54: Adjust the model parameters of the equivalent filter until the output value of the filter performance function matches the filter performance indicators;

[0112] Step S55: Obtain a parameter-tuned equivalent model according to the parameter-tuned equivalent filter and equivalent controller.

[0113] Specifically, the control mechanism for implementing this method first tunes the parameters of the equivalent controller, which specifically includes: constructing a closed-loop curve dynamically corresponding to the equivalent controller according to the transfer function of the equivalent controller. It should be noted that when the gain, zeros, and poles of the equivalent controller change, the closed-loop curve changes accordingly; adjusting at least one of the gain, zeros, and poles of the equivalent controller according to the preset closed-loop boundary index, and monitoring in real time the closed-loop curve dynamically corresponding to the equivalent controller; when the closed-loop curve is constrained by the controller index, determining the model parameters of the equivalent controller as the target values, that is, the equivalent controller at this time is the equivalent controller with tuned parameters.

[0114] After that, the control mechanism of this method tunes the parameters of the equivalent filter, which specifically includes: constructing a filter performance function for characterizing the tracking performance of the filter, adjusting the model parameters of the equivalent filter, and monitoring in real time the output value of the filter performance function. When the output value of the filter performance function matches the filter performance index, determining the model parameters of the equivalent filter as the target values, that is, the equivalent filter at this time is the equivalent filter with tuned parameters.

[0115] After completing the parameter tuning of the equivalent filter and the equivalent controller, the equivalent filter and the equivalent controller with tuned parameters are integrated to obtain a fixed-parameter equivalent model. Specifically, the actual model structure of the fixed-parameter equivalent model can refer to the equivalent model as shown in Figure 5 shown.

[0116] Please refer to Figure 6 , Figure 6 which is a schematic flow chart of the parameter tuning of the equivalent controller in a rectification control model generation method provided by an embodiment of the present invention.

[0117] As shown in Figure 6 , in some embodiments, the controller index at least includes a closed-loop boundary index. Adjusting the model parameters of the equivalent controller according to the closed-loop boundary index in step S52 includes:

[0118] Step S521: Construct a constraint boundary according to the closed-loop boundary index to determine the constraint region corresponding to the closed-loop curve according to the constraint boundary;

[0119] Step S522: Construct a target performance function for characterizing the performance of the equivalent controller;

[0120] Step S523: Select a target point in the constraint region to make the output of the target performance function match the target performance index;

[0121] Step S524: Use the model parameters corresponding to the target point as the model parameters after the parameter tuning of the equivalent controller.

[0122] Specifically, the closed-loop boundary indicators include the closed-loop resonance peak, the disturbance signal frequency characteristics and the tracking boundary. Constructing the constraint boundary according to the closed-loop boundary indicators includes: first generating a robust stability boundary according to the closed-loop resonance peak, generating a disturbance boundary according to the disturbance signal frequency characteristics, and then generating a constraint boundary based on the integration of the robust stability boundary, the disturbance boundary and the constraint boundary, so as to determine the constraint area according to the constraint boundary.

[0123] For example, the robust stability boundary can be expressed as:

[0124]

[0125] in, is the closed-loop amplitude-frequency characteristic curve, and are the frequency characteristics of the equivalent controller and the equivalent controlled object, is the closed-loop resonance peak, is the angular frequency variable.

[0126] For example, the perturbation bound can be expressed as:

[0127]

[0128] in, For the preset frequency characteristics, is the frequency characteristic of the disturbance signal, is the disturbance boundary frequency characteristic model, which can be a constant or a polynomial form. is the angular frequency variable, is the selected frequency domain threshold.

[0129] For example, the transfer function of the first sub-loop, the second sub-loop, the third sub-loop, or the fourth sub-loop may be expressed as:

[0130]

[0131] The tracking boundary can be expressed as:

[0132]

[0133] in, is the closed-loop amplitude-frequency characteristic curve, is the transfer function constant, and are the transfer functions of the equivalent controller and the equivalent controlled object, respectively. and They are the upper bound and the lower bound respectively, and their common constraints form the tracking constraint boundary. is the angular frequency variable, is the selected frequency domain threshold, and s is the Laplace operator.

[0134] After that, the robust stability boundary, the perturbation boundary and the constraint boundary are integrated to generate the constraint boundary, so as to determine the constraint region according to the constraint boundary. For example, the three expressions of the robust stability boundary, the perturbation boundary and the constraint boundary can be integrated and transformed into Constraints, thus we can get the image corresponding to the closed loop curve that meets the constraints The area is the constraint area.

[0135] After determining the constraint region, a target performance function is constructed to characterize the performance of the equivalent controller. A target point is selected in the constraint region so that the output of the target performance function matches the target performance index. The model parameters corresponding to the target point are used as the model parameters after the equivalent controller parameters are tuned.

[0136] In some embodiments, the target performance function includes a first performance function for characterizing the open-loop gain of the equivalent controller, a second performance function for characterizing the distance between the closed-loop curve and the constraint boundary, and a third performance function for characterizing the monotonicity of the closed-loop curve.

[0137] like Figure 7 As shown, step S523 of selecting a target point in the constraint area includes:

[0138] Step S5231: determining a number of candidate points in the constraint area;

[0139] Step S5232: inputting the model parameters corresponding to the candidate points into the first performance function, the second performance function, and the third performance function respectively to generate function output fusion information;

[0140] Step S5233: Determine the target point among the candidate points based on the fusion information output by the function.

[0141] Specifically, first, several candidate points are determined in the constraint area, and then the model parameters corresponding to the candidate points are respectively input into the first performance function, the second performance function and the third performance function to generate function output fusion information. It should be noted that when the model parameters corresponding to different candidate points in the constraint area are respectively input into the first performance function, the second performance function and the third performance function, the function output fusion information obtained is different.

[0142] Specifically, the function output fusion information includes the output value of the first performance function, the output value of the second performance function, and the output value of the third performance function.

[0143] For example, the output value of the first performance function can be expressed as:

[0144]

[0145] in, is the output value of the first performance function, used to characterize the open-loop gain of the equivalent controller. is the open-loop gain of the equivalent controller. It should be noted that the open-loop gain of the designed equivalent controller should be as small as possible, that is, should be as small as possible to improve the system stability margin and reduce the system design cost.

[0146] Exemplarily, the output value of the second performance function can be expressed as:

[0147]

[0148] where is the output value of the second performance function, used to characterize the distance between the closed-loop curve and the constraint boundary, is the angular frequency variable, is the selected frequency-domain threshold, represents the phase angle of the constraint boundary range at and represents the phase angle of the open-loop frequency characteristic curve of the equivalent model at It should be noted that the open-loop frequency characteristic curve should be as close as possible to the constraint boundary at the selected frequency, that is, should be as small as possible, so as to improve the system stability margin and reduce the system design cost.

[0149] Exemplarily, the output value of the third performance function can be expressed as:

[0150]

[0151] where is a judgment function. When , , and vice versa, when , .

[0152] where is the output value of the third performance function, used to characterize the monotonicity of the open-loop frequency characteristic of the equivalent control model, is the next selected frequency-domain threshold larger than , represents the phase angle of the constraint boundary range at and represents the phase angle of the open-loop frequency characteristic curve of the equivalent model at It should be noted that the open-loop frequency characteristic of the designed equivalent control model should be kept as monotonically decreasing as possible, that is, should be as small as possible, so as to determine the good dynamic characteristics of the control system.

[0153] ​​It should be noted that the function output fusion information includes the outputs of the first performance function, the second performance function, and the third performance function. This method generates the output values of the corresponding functions according to the model parameters corresponding to each candidate point and the above expressions to generate the function output fusion information, and then determines the target point among the candidate points according to the function output fusion information.

[0154] Specifically, step S5233 determines the target point among the candidate points according to the function output fusion information as follows: comparing the function output fusion information corresponding to multiple candidate points, and selecting the candidate point that minimizes the outputs of the first performance function, the second performance function, and the third performance function as the target point, so that the equivalent controller open-loop gain, the distance between the closed-loop curve and the constraint boundary, and the [unclear term] corresponding to the target point all reach the minimum values.

[0155] Exemplarily, to determine the target point among the candidate points, a selection function as shown below can be designed:

[0156]

[0157]

[0158]

[0159] Among them, is the model parameter corresponding to the candidate point, , and are the output values of the first performance function, the second performance function, and the third performance function respectively, , , are the gain, zero point, and pole point of the equivalent controller respectively, is the constraint region expression obtained by integrating and deforming the three expressions of the above robust stability boundary, disturbance boundary, and constraint boundary. It is easy to know that after the model parameter corresponding to the candidate point in the above function is determined, the gain , zero point and pole point of the equivalent controller are determined accordingly.

[0160] Based on the above selection function, the candidate point with the minimum output values of the first performance function, the second performance function, and the third performance function is obtained as the target point.

[0161] It should be understood that in the case where the control model to be tuned corresponds to multiple control loops, such as including the first sub-loop, the second sub-loop, the third sub-loop, or the fourth sub-loop, the model parameter corresponding to the candidate point can be expressed as:

[0162]

[0163] It should also be understood that in the process of comparing the function output fusion information corresponding to multiple candidate points, there may not exist model parameters that can minimize the outputs of the first performance function, the second performance function, and the third performance function, or the model parameters that minimize the outputs of the first performance function, the second performance function, and the third performance function may not be the same.

[0164] Based on this, the method can also be to solve the selection function using a multi-objective evolutionary algorithm, determine whether the model parameters meet the preset design requirements after obtaining the Pareto optimal solution. If not, increase the system degrees of freedom and obtain the Pareto optimal solution again until the model parameters corresponding to the Pareto optimal solution meet the preset design requirements.

[0165] In some embodiments, in step S53 of constructing a filter performance function for characterizing the filter tracking performance, the filter performance function can be constructed in the following form;

[0166]

[0167]

[0168] Among them, is the filter performance function, and are the tracking upper bound and the tracking lower bound respectively, , and are the frequency characteristics of the equivalent filter, the equivalent controller, and the equivalent controlled object respectively, is the angular frequency variable.

[0169] After constructing the filter performance function, adjust the model parameters of the equivalent filter until the output value of the filter performance function matches the filter performance index. Specifically, the simplex method or the interior point method can be used to solve the model parameters of the equivalent filter to determine the optimal gain, zeros, and poles of the equivalent filter.

[0170] Please refer to Figure 8 , Figure 8 which is the open-loop frequency characteristic curve diagram of the target control model in a rectification control model generation method provided by an embodiment of the present invention.

[0171] As Figure 8As shown, the open-loop frequency characteristic curve specifically uses a Nichols plot. The open-loop frequency curves corresponding to multiple model parameters are plotted on the Nichols plot. Among them, the Nichols plot uses the phase as the abscissa and the frequency amplitude as the ordinate, and the unit of the phase is degree, and the unit of the frequency amplitude is decibel. Among them, curve L is the open-loop frequency characteristic curve of the target control model.

[0172] Step S6: Tune the controller gain and observer gain of the control model to be tuned according to the parameter-determined equivalent model to obtain the target control model.

[0173] Based on the fact that during the process of constructing the equivalent filter, there is a preset first correspondence between the controller gain and observer gain and the transfer function of the equivalent controller, and there is a preset second correspondence between the controller gain and observer gain and the transfer function of the equivalent filter. After tuning the transfer functions of the equivalent controller and the equivalent filter, adjust the controller gain and observer gain of the control model to be tuned according to the first correspondence, the second correspondence, and the tuned equivalent controller and equivalent filter to obtain the target control model.

[0174] Exemplarily, the model structure of the target control model can specifically refer to the structure of the control model to be tuned as shown in Figure 2 shown.

[0175] It should be understood that since mass transfer and heat transfer phenomena will occur during the rectification process of the rectification column, and the rectification process itself has certain external disturbances, such as errors in feed flow rate, feed temperature, feed composition, etc., which will cause large disturbances in the internal state of the rectification column during the rectification process. Based on this, in this method, the model parameters are tuned by combining the actual rectification process of the rectification device during the model establishment process to generate a rectification control model with strong anti-disturbance ability, high precision, and high stability, which is convenient to control the rectification process through the generated target control model to obtain the required product concentration.

[0176] As Figure 9 shown, an embodiment of the present application also provides a rectification control method, and the method includes:

[0177] Step S7: When receiving the target rectification instruction, parse the target rectification instruction to obtain the concentration target value of the rectification product, and obtain the current device parameters of the rectification device;

[0178] Step S8: Invoke the target control model according to the target rectification instruction, where the target control model is the target control model generated by using any rectification control model generation method provided in the embodiment of the present application;

[0179] Step S9: Input the concentration target value and the device parameters of the rectification device into the target control model, so that the target control model generates the target value of the device parameters according to the concentration target value and the device parameters;

[0180] Step S10: Adjust the corresponding device parameters in the rectification device according to the target value of the device parameters to control the rectification process of the rectification device.

[0181] Taking the rectification control method applied to the control mechanism as an example for illustration, the control mechanism is connected to the rectification reaction mechanism to control the rectification process of the rectification reaction mechanism. Among them, the rectification reaction mechanism includes a rectification device, and the rectification device is, for example, a rectification column.

[0182] Specifically, when the control mechanism receives the input target rectification instruction, it parses the target rectification instruction to obtain the concentration target value of the rectification product, and obtains the current device parameters of the rectification device. Among them, the device parameters of the rectification device are, for example, the pressure parameters and thermal parameters of the rectification column. Further, the pressure parameter of the rectification column is used to characterize the pressure difference between the rectification section and the stripping section in the rectification column, and the thermal parameter of the rectification column is used to characterize the thermal condition in the rectification column.

[0183] Then the control mechanism calls the target control model according to the target rectification instruction. It should be noted that the target control model is a target control model generated by using any of the rectification control model generation methods provided in the embodiments of the present application. Then, the concentration target value and the device parameters of the rectification device are input into the target control model, so that the target control model generates the target value of the device parameters according to the concentration target value and the device parameters.

[0184] After generating the target value of the device parameters, adjust the corresponding device parameters in the rectification device according to the target value of the device parameters to control the rectification process of the rectification device. Specifically, the control mechanism adjusts the corresponding device parameters in the rectification device to the target value of the device parameters.

[0185] Thus, by calling the generated target control model and controlling the rectification process of the rectification device according to the concentration target value and the target control model to obtain the required product concentration, the anti-interference ability, accuracy and stability of the rectification process control are improved.

[0186] As Figure 10 shown, an embodiment of the present application further provides a rectification device 100, including:

[0187] A raw material supply mechanism 110 for providing a mixture raw material;

[0188] A rectification reaction mechanism 120, which is used to provide a reaction site for the mixture raw material and carry out rectification treatment on the mixture raw material in the reaction site to generate a rectification product. Among them, the rectification reaction mechanism has device parameters to be adjusted;

[0189] A product delivery mechanism 130 for delivering the generated rectification product to a target container;

[0190] A control mechanism 140 is connected to the rectification reaction mechanism 120 and is configured to execute any one of the rectification control methods provided in the embodiments of the present application to adjust the device parameters of the rectification reaction mechanism 120 to control the rectification process of the rectification device 100.

[0191] Exemplarily, the rectification reaction mechanism 120 includes a rectification device, such as a rectification column, and the reaction site is arranged inside the rectification column. Among them, the specific structure of the rectification column can be referred to Figure 1 . The raw material supply mechanism 110 supplies a mixture raw material to the reaction site inside the rectification column, and the control mechanism 140 analyzes the target rectification instruction to obtain the concentration target value of the rectification product, and obtains the current device parameters of the rectification device. Among them, the device parameters of the rectification device are, for example, the pressure parameter and the thermal parameter of the rectification column. Then, the control mechanism 140 calls the target control model according to the target rectification instruction. It should be noted that the target control model is a target control model generated by using any one of the rectification control model generation methods provided in the embodiments of the present application. Then, the concentration target value and the device parameters of the rectification device are input into the target control model, so that the target control model generates a device parameter target value according to the concentration target value and the device parameters, and then adjusts the corresponding device parameters in the rectification device according to the device parameter target value to control the rectification process of the rectification device.

[0192] In summary, the present application provides a rectification control model generation method, a rectification control method, and a rectification device. The generation method includes: constructing a simulation rectification model and performing decoupling processing to generate a control loop between each device parameter and the simulated concentration information of the rectification product; constructing a to-be-tuned control model corresponding to the device parameter according to the control loop, and the to-be-tuned control model is used to adjust the device parameter to control the rectification process; constructing an equivalent model according to the to-be-tuned control model; obtaining a target performance index to perform parameter tuning on the equivalent model to obtain a parameter-tuned equivalent model; and tuning the controller gain and the observer gain of the to-be-tuned control model according to the parameter-tuned equivalent model to obtain a target control model. By tuning the model parameters in combination with the actual rectification process of the rectification device during the model establishment process, a rectification control model with strong anti-interference ability, high precision, and high stability is generated, which is convenient to control the rectification process through the generated target control model to obtain the required product concentration.

[0193] It should be understood that the term "and / or" refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations. As used herein, "including", "comprising", or any other variation thereof is intended to cover non-exclusive inclusion, including not only those elements but also elements inherent to such a process, method, article, or system.

[0194] The serial numbers of the above embodiments do not represent the superiority or inferiority of the embodiments. Any person skilled in the art within the technical scope disclosed in this application can easily think of various equivalent modifications or substitutions, which should all be covered within the protection scope of this application. Therefore, the protection scope of this application shall be subject to the protection scope of the claims.

Claims

1. A method for generating a rectification control model, characterized in that, The method includes: Constructing a simulated rectification model according to a rectification device, where the simulated rectification model is used to generate simulated concentration information of rectification products based on device parameters of the rectification device; Performing decoupling processing on the simulated rectification model to generate control loops between each device parameter and the simulated concentration information; Constructing a to-be-tuned control model corresponding to the device parameters according to the control loops, where the to-be-tuned control model is used to adjust the device parameters according to a controller gain, an observer gain, and a concentration target value of the rectification product to control the rectification process of the rectification device; Constructing an equivalent model including an equivalent filter and an equivalent controller according to the to-be-tuned control model; Obtaining a target performance index of the equivalent model, and performing parameter tuning on the equivalent filter and the equivalent controller of the equivalent model according to the target performance index to obtain a parameter-tuned equivalent model; Tuning the controller gain and the observer gain of the to-be-tuned control model according to the parameter-tuned equivalent model to obtain a target control model.

2. The rectification control model generation method as shown in claim 1, characterized in that, The constructing the to-be-tuned control model corresponding to the device parameters according to the control loops includes: Constructing a controller model according to the control loops, where the controller model is used to control the simulated rectification model to generate the simulated concentration information according to a controller gain and the device parameters; Constructing an observer model according to the controller model, where the observer model is used to observe the simulated concentration information according to the observer gain to generate observed concentration information, and perform feedback compensation on the device parameters according to the concentration target value and the observed concentration information; Constructing the to-be-tuned control model based on the controller model and the observer model.

3. The method for generating a rectification control model according to claim 1, wherein The device parameters at least include a pressure parameter and a thermal parameter in the rectification device, and the performing decoupling processing on the simulated rectification model to generate control loops between each device parameter and the simulated concentration information includes: Obtaining a transfer function matrix between the pressure parameter, the thermal parameter and the simulated concentration information; Performing decoupling processing on the transfer function matrix to generate a first control loop between the pressure parameter and the simulated concentration information; Performing decoupling processing on the transfer function matrix to generate a second control loop between the thermal parameter and the simulated concentration information.

4. The method for generating a rectification control model according to claim 3, wherein The rectification device is at least used to output a first rectification product and a second rectification product; The obtaining the transfer function matrix between the pressure parameter, the thermal parameter and the simulated concentration information includes: Constructing a first transfer function between the pressure parameter and a first product concentration of the first rectification product according to the rectification device, and constructing a second transfer function between the pressure parameter and a second product concentration of the second rectification product according to the rectification device; Constructing a third transfer function between the thermal parameter and the first product concentration according to the rectification device, and constructing a fourth transfer function between the thermal parameter and the second product concentration according to the rectification device; Constructing the transfer function matrix according to the first transfer function, the second transfer function, the third transfer function and the fourth transfer function.

5. The method for generating a rectification control model according to any one of claims 1-4, characterized in that Constructing an equivalent model including an equivalent filter and an equivalent controller according to the to-be-tuned control model, including: Constructing an equivalent controlled object according to the to-be-tuned control model, where the equivalent controlled object is used to generate the simulated concentration information according to an input control instruction; Constructing the equivalent filter and the equivalent controller according to the equivalent controlled object, where the equivalent filter is used to perform filtering processing on the concentration target value to generate a concentration filtered value, and the equivalent controller is used to input the control instruction to the equivalent controlled object according to the concentration filtered value.

6. The method for generating a rectification control model according to any one of claims 1-4, characterized in that The target performance index includes a controller index and a filter performance index, and the controller index at least includes a closed-loop boundary index; Performing parameter tuning on the equivalent controller and the equivalent filter of the equivalent model according to the target performance index to obtain a parameter-tuned equivalent model, including: Constructing a closed-loop curve dynamically corresponding to the equivalent controller; Adjusting the model parameters of the equivalent controller according to the closed-loop boundary index until the closed-loop curve corresponding to the equivalent controller is constrained by the controller index; Constructing a filter performance function for characterizing the tracking performance of the filter; Adjusting the model parameters of the equivalent filter until the output value of the filter performance function matches the filter performance index; Obtaining the parameter-tuned equivalent model according to the equivalent filter and the equivalent controller after parameter tuning.

7. The rectification control model generation method according to claim 6, characterized in that The adjusting the model parameters of the equivalent controller according to the closed-loop boundary index includes: Constructing a constraint boundary according to the closed-loop boundary index to determine a constraint region corresponding to the closed-loop curve according to the constraint boundary; Constructing a target performance function for characterizing the performance of the equivalent controller; Selecting a target point in the constraint region so that the output of the target performance function matches the target performance index; Taking the model parameters corresponding to the target point as the model parameters after parameter tuning of the equivalent controller.

8. The rectification control model generation method according to claim 7, wherein The target performance function includes a first performance function for characterizing the open-loop gain of the equivalent controller, a second performance function for characterizing the distance between the closed-loop curve and the constraint boundary, and a third performance function for characterizing the monotonicity of the closed-loop curve; The selecting a target point in the constraint region includes: Determining a number of candidate points in the constraint region; Respectively inputting the model parameters corresponding to the candidate points into the first performance function, the second performance function and the third performance function to generate function output fusion information; Determining the target point from the candidate points according to the function output fusion information.

9. A rectification control method, characterized in that, The method includes: When receiving a target rectification instruction, parsing the target rectification instruction to obtain a concentration target value of the rectification product, and acquiring the current device parameters of the rectification device; Invoking a target control model according to the target rectification instruction, where the target control model is a target control model generated by using the rectification control model generation method according to any one of claims 1-8. Input the concentration target value and the device parameters of the rectification device into the target control model, so that the target control model generates a target value of the device parameters according to the concentration target value and the device parameters; Adjust the corresponding device parameters in the rectification device according to the target value of the device parameters to control the rectification process of the rectification device.

10. A rectification device, characterized in that, Comprising: A raw material supply mechanism for providing a mixture raw material; A rectification reaction mechanism for providing a reaction site for the mixture raw material and performing rectification treatment on the mixture raw material in the reaction site to generate a rectification product, wherein the rectification reaction mechanism has device parameters to be adjusted; A product conveying mechanism for conveying the generated rectification product to a target container; A control mechanism, connected to the rectification reaction mechanism and used to execute the rectification control method as described in claim 9 to adjust the device parameters of the rectification reaction mechanism to control the rectification process of the rectification device.

Images