Process continuous operation method and system, electronic equipment, storage medium and program product

Abstract

The invention relates to the technical field of process safety operation, in particular to a process continuous operation method and system, electronic equipment, a storage medium and a program product, and the method comprises the steps: constructing a time domain model according to a dynamic mechanism of a process system; converting the time domain model into a complex frequency domain model through Laplace transformation; solving a complex frequency domain optimal continuous operation variable based on the complex frequency domain model, and converting the complex frequency domain optimal continuous operation variable into a time domain optimal continuous operation variable through inverse Laplace transformation; and performing continuous operation based on the time domain optimal continuous operation variable. According to the process continuous operation method, an analytical method based on a complex frequency domain is provided, continuous working condition switching of a complex chemical system is achieved, compared with a subsection control strategy, the continuous control strategy can describe the real situation as much as possible, feature details of operation variables are clearly presented, higher information density is achieved, and the method is suitable for large-scale popularization and application. And therefore, the safety and the stability of the production process can be improved.

Description

Technical Field

[0001] This invention relates to the field of process safety operation technology, and in particular to a process continuous operation method, system, electronic device, storage medium and program product. Background Technology

[0002] After a century of development and evolution, the oil refining and chemical industry has matured its production processes. Under the current dual context of energy security and energy innovation, the future development direction of the chemical industry will be raw material diversification, low-carbon technology, and high-end products. Chemical production facilities generally have long and complex processes, with production conditions including extreme conditions such as high temperature and high pressure, and low temperature and negative pressure. Abnormal operation will directly affect the entire production system, leading to production and safety accidents.

[0003] Therefore, there are high requirements for the reliability of the operation and regulation of production equipment, and research on its process safety and stable operation is of utmost importance.

[0004] In petrochemical enterprises, considering economic benefits, a single production unit may involve multiple process flows. Changes such as switching grades or adjusting loads can significantly alter the process flow, leading to frequent large-scale variations in operating conditions. Chemical production is typically a multi-faceted, continuous process; any change in any condition will cause fluctuations in other variables, causing the system to deviate from normal operating conditions and further impacting operational safety and product quality.

[0005] Therefore, it is necessary to adjust the operating variables of the production equipment in a timely manner to ensure that the system maintains safe and stable production throughout the entire operating cycle. How to safely and smoothly achieve a wide range of operational adjustments to the production equipment is a problem that enterprises urgently need to solve.

[0006] In practice, operators of petrochemical production units often sacrifice efficiency to ensure smooth operation during switching processes. This involves repeatedly adjusting controller settings to achieve large-scale changes in operating conditions. These large-scale changes are artificially divided into several smaller-scale changes, inserting intermediate transition points to reduce the amplitude of each change, thereby minimizing operational fluctuations and ensuring production safety.

[0007] Currently, operators typically select and apply intermediate transition points based on experience, meaning they need to adjust controller settings promptly based on on-site variable measurements and their own experience. However, this selection process is relatively arbitrary and requires extensive front-line production experience, placing high demands on operators' experience and skills, which hinders widespread application. Furthermore, discontinuous manipulated variables may overlook some detailed characteristics due to the influence of discrete precision. Especially when the ideal trajectory of the manipulated variable is non-standard, such as a non-monotonic trajectory, segmented manipulated variables struggle to reflect the variable's changing patterns, potentially negatively impacting the control performance of continuous industrial processes. Therefore, a novel, continuous, and scientific operating method is needed to adjust for large-scale changes in operating conditions, thereby providing technical support for operational issues in petrochemical enterprises based on process safety and stable operation. Summary of the Invention

[0008] This invention aims to at least partially solve one of the technical problems in the aforementioned technologies, and to this end, proposes a continuous process operation method, comprising:

[0009] Construct a time-domain model based on the dynamic mechanism of the process system;

[0010] The time-domain model is transformed into a complex frequency-domain model using the Laplace transform;

[0011] Based on the aforementioned complex frequency domain model, the optimal continuous operation variables in the complex frequency domain are solved, and then transformed into the optimal continuous operation variables in the time domain through the inverse Laplace transform.

[0012] Perform continuous operations based on the time-domain optimal continuous operation variables.

[0013] Furthermore, a time-domain model is constructed based on the dynamic mechanism of the process system, including:

[0014] The time-domain model is constructed based on the dynamic mechanism and constraints of the process system; wherein, the constraints include: the range of change of the process system's operating variables during the process system's transition from the current operating condition to the target operating condition.

[0015] Furthermore, the expression corresponding to the dynamic mechanism includes:

[0016]

[0017] c(t) = f c [x(t),u(t),d(t)]

[0018] Where t represents the running time; x represents a vector of time-domain state variables; denoted by , u represents the derivative of the time-domain state variables; denoted by u represents the vector of time-domain manipulated variables; denoted by d represents the vector of time-domain driving variables; denoted by c represents the vector of constraint variables that must remain unchanged during the switching process; f x (·) represents the equation representing the dynamic mechanism of the process object; f c (·) represents the equation for the constant constraint variable c(t).

[0019] Furthermore, the constraints include: endpoint constraints and process constraints; wherein,

[0020] The endpoint constraints, and the corresponding expressions, include:

[0021]

[0022] Where t0 represents the moment when all variables are at their steady-state values ​​before the change in operating conditions; t f This indicates the moment when each variable reaches its steady-state value after a change in operating conditions. This represents the steady-state value of the variable (·) at time t0; Indicates the variable (·) in t f steady-state value at time;

[0023] The process constraints, and the corresponding expressions, include:

[0024] c(t) = c * ,t∈[t0,t f ]

[0025] Among them, c * Let c(t) represent the expected value of c(t) during the process of changing operating conditions.

[0026] Furthermore, the expression corresponding to the complex frequency domain model includes:

[0027] ΔC(s)=[G u (s) G d (s)][ΔU(s) ΔD(s)] T

[0028] =G u (s)ΔU(s)+G d (s)ΔD(s)

[0029] ΔC(s)=ΔC * (s)=0

[0030] ΔD(s)=ΔD * (s)

[0031] Where Δ represents the Laplace transform of each variable under zero initial conditions; T represents the transpose of the matrix; G u (s) and G dΔU(s) represents the transfer function of the process system; ΔU(s) represents the complex frequency domain operation variable; ΔC(s) represents the complex frequency domain constraint variable; ΔC * (s) represents the expected value of the complex frequency domain constraint variable; ΔD(s) represents the complex frequency domain driving variable; ΔD * (s) represents the expected value of the driving variable in the complex frequency domain.

[0032] Furthermore, based on the aforementioned complex frequency domain model, the optimal continuous operational variables in the complex frequency domain are solved, and then transformed into optimal continuous operational variables in the time domain through inverse Laplace transform. The corresponding expressions include:

[0033]

[0034] Δu * (t)=L -1 [ΔU * (s)]

[0035]

[0036] Among them, L -1 Indicates the inverse Laplace transform; ΔU * (s) represents the expected value of the variable operated on in the complex frequency domain; u * (t) represents the expected value of the time-domain operated variable.

[0037] Furthermore, the constraint variables, operation variables, and driving variables also satisfy the following expression:

[0038]

[0039] Where, k u With k d Let G represent the transfer function respectively. u (s) and G d The steady-state gain of (s).

[0040] Furthermore, after transforming the time-domain optimal continuous operational variables through the inverse Laplace transform, the time-domain optimal continuous operational variables are simplified using a fitting method, including:

[0041] The time-domain optimal continuous operational variable time series data obtained by the solution are sampled;

[0042] By fitting the sampled data, an approximate control function is obtained;

[0043] The simplified time-domain optimal continuous operating variable is calculated based on the approximate control function.

[0044] This application also proposes a continuous process operating system for petrochemical enterprises, including:

[0045] The model building module is used to construct a time-domain model based on the dynamic mechanism of the process system.

[0046] The model conversion module is used to convert the time-domain model into a complex frequency-domain model using Laplace transform.

[0047] The variable calculation module is used to solve for the optimal continuous operation variable in the complex frequency domain based on the complex frequency domain model, and to transform it into the optimal continuous operation variable in the time domain through the inverse Laplace transform.

[0048] An operation control module is used to perform continuous operations based on the time-domain optimal continuous operation variables.

[0049] This application also proposes an electronic device, including a memory and a processor, wherein the memory stores a computer program or instructions, and when the computer program or instructions are executed by the processor, they are used to at least implement the above-described process continuous operation method.

[0050] This application also proposes a computer-readable storage medium storing a computer program or instructions, which, when executed by a processor, are at least used to implement the above-described continuous process operation method.

[0051] This application also proposes a computer program product stored in a computer-readable storage medium, which, when executed by a processor, is used to at least implement the above-described continuous process operation method.

[0052] Compared with the prior art, the beneficial effects of the present invention are:

[0053] (1) This invention proposes an analytical method based on the complex frequency domain, which realizes continuous operation switching of complex chemical systems. Compared with the segmented control strategy, the continuous control strategy can describe the real situation as much as possible, clearly present the characteristic details of the operating variables, and has higher information density, thereby improving the safety and stability of the production process.

[0054] (2) Taking into account practical application issues such as computational cost and operational complexity, this invention proposes to strictly satisfy the endpoint constraints of the switching process during the identification of the transfer function, thereby avoiding the addition of constraints in subsequent calculations. In addition, the obtained continuous control strategy is simplified through numerical analysis to improve its practicality.

[0055] (3) For the study of the working condition switching problem, an analysis approach based on the complex frequency domain is given, which gets rid of the existing time-domain dynamic optimization analysis approach based on time discretization and numerical approximation, and provides a technical route for solving the continuous control strategy of complex systems.

[0056] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention can be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings. The technical solutions of the invention will be further described below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0057] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0058] Figure 1 This is a schematic diagram of a continuous process operation method given in the embodiment;

[0059] Figure 2 This is a schematic diagram of the approximate function of the control trajectory of the manipulated variable, provided as an example.

[0060] Figure 3a The graph shown here illustrates the change in feed flow rate over time, obtained through process control based on the scheme described in this application, as an example.

[0061] Figure 3b The graph shown is a curve of ethylene yield versus time obtained by process control based on the scheme of this application, as given in the example.

[0062] Figure 3c The graph shown here illustrates the change in heat exchange of the reboiler at the bottom of the column over time, obtained by process control based on the scheme of this application, as an example.

[0063] Figure 3d The following is a graph showing the change of ethylene product composition over time obtained by process control based on the scheme of this application, as shown in the example.

[0064] Figure 4a The graph shows the change in ethylene yield over time obtained by process control based on a PID control strategy, as shown in the example.

[0065] Figure 4b The following is a graph showing the change of ethylene product composition over time obtained by process control based on a PID control strategy, as shown in the example.

[0066] Figure 4c The figure provided is a graph showing the change of ethylene yield over time obtained by process control based on a segmented control strategy, as shown in the example.

[0067] Figure 4d The example shows a graph illustrating the change of ethylene product composition over time, obtained through process control based on a segmented control strategy.

[0068] Figure 5 This is a schematic diagram of a continuous process operation method given in the embodiment;

[0069] Figure 6 A schematic diagram of an electronic device provided in an embodiment;

[0070] Figure 7 This is a schematic diagram of a readable storage medium provided for an embodiment. Detailed Implementation

[0071] The present invention will be described below with reference to the accompanying drawings. The preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0072] This invention provides a method for continuous process operation, comprising:

[0073] Construct a time-domain model based on the dynamic mechanism of the process system;

[0074] The time-domain model is transformed into a complex frequency-domain model using the Laplace transform;

[0075] Based on the aforementioned complex frequency domain model, the optimal continuous operation variables in the complex frequency domain are solved, and then transformed into the optimal continuous operation variables in the time domain through the inverse Laplace transform.

[0076] Perform continuous operations based on the time-domain optimal continuous operation variables.

[0077] According to some embodiments of this application, such as Figure 1 As shown, the method is as follows: First, a complex frequency domain mathematical model of the system, i.e., a transfer function model, is constructed based on the dynamic mechanism of the process system. Second, based on the time-domain analysis method, a complex frequency domain-based analysis method is proposed to map the optimization problem of the manipulated variables that cannot be solved in the t-plane to the s-plane, and the optimal continuous manipulated variables are solved in the complex frequency domain based on the transfer function. In addition, the endpoint constraints of the manipulated variables are satisfied during the identification of the transfer function, thereby avoiding the addition of constraints in subsequent calculations. Finally, numerical analysis is used to simplify the control strategy, and an approximate function composed of simple function combinations is used to replace the original complex function to improve its practicality.

[0078] Furthermore, a time-domain model is constructed based on the dynamic mechanism of the process system, including:

[0079] A time-domain model is constructed based on the dynamic mechanism and constraints of the process system; the constraints include the range of change of the process system's operating variables during the process of switching the process system from the current operating condition to the target operating condition.

[0080] Furthermore, the expressions corresponding to the dynamic mechanism include:

[0081]

[0082] c(t) = f c [x(t),u(t),d(t)]

[0083] Where t represents the running time; x represents a vector of time-domain state variables; denoted by , u represents the derivative of the time-domain state variables; denoted by u represents the vector of time-domain manipulated variables; denoted by d represents the vector of time-domain driving variables; denoted by c represents the vector of constraint variables that must remain unchanged during the switching process; f x (·) represents the equation representing the dynamic mechanism of the process object; f c (·) represents the equation for the constant constraint variable c(t).

[0084] Furthermore, the constraints include: endpoint constraints and process constraints; wherein,

[0085] Endpoint constraints, and their corresponding expressions, include:

[0086]

[0087] Where t0 represents the moment when all variables are at their steady-state values ​​before the change in operating conditions; t f This indicates the moment when each variable reaches its steady-state value after a change in operating conditions. This represents the steady-state value of the variable (·) at time t0; Indicates the variable (·) in t f steady-state value at time;

[0088] Process constraints, and their corresponding expressions, include:

[0089] c(t) = c * ,t∈[t0,t f ]

[0090] Among them, c * Let c(t) represent the expected value of c(t) during the process of changing operating conditions.

[0091] Furthermore, the complex frequency domain model, and its corresponding expressions, include:

[0092] ΔC(s)=[G u (s) G d (s)][ΔU(s) ΔD(s)] T

[0093] =G u (s)ΔU(s)+G d (s)ΔD(s)

[0094] ΔC(s)=ΔC * (s)=0

[0095] ΔD(s)=ΔD * (s)

[0096] Where Δ represents the Laplace transform of each variable under zero initial conditions; T represents the transpose of the matrix; G u (s) and G d ΔU(s) represents the transfer function of the process system; ΔU(s) represents the complex frequency domain manipulated variable; ΔC(s) represents the complex frequency domain constraint variable; ΔC * (s) represents the expected value of the complex frequency domain constraint variable; ΔD(s) represents the complex frequency domain driving variable; ΔD * (s) represents the expected value of the driving variable in the complex frequency domain.

[0097] According to some embodiments of this application, the process of constructing a complex frequency domain mathematical model (transfer function model) of the system based on the dynamic mechanism of the process system is as follows:

[0098] (1) Determine the range of change of the operating variables, i.e. the steady-state increment, based on the change from the current operating condition to the target operating condition through steady-state optimization;

[0099] (2) Establish a time-domain dynamic model of the production process and describe the optimization problem of the operational variables in the working condition switching process using mathematical language. First, establish a time-domain dynamic model of the production process. The mechanistic model of the process object can be expressed as:

[0100]

[0101] c(t) = f c [x(t),u(t),d(t)]

[0102] Where x is a vector of state variables; u is a vector of operational variables; d represents a vector of driving variables; and c represents a vector of constraint variables that must remain unchanged during the switching process. x (·) represents a system of differential-algebraic equations for a process object mechanism model; f c (·) represents the equation for the constant constraint variable c(t).

[0103] Constraints are divided into endpoint constraints and process constraints. The former represents the conditions that each variable must meet at the beginning and end of the switching process, while the latter represents the requirements that each variable must meet throughout the entire process. Endpoint constraints can be expressed as:

[0104]

[0105]

[0106] Among them, the superscripts t0 and t f Let represent the steady-state values ​​of each variable before and after the change in operating conditions. The process constraints can be expressed as:

[0107] c(t) = c * ,t∈[t0,t f ]

[0108] In the formula, c * Let c(t) represent the expected value of c(t) during the switching process, which is a constant. Then, the optimization problem of the operating variables during the switching process can be expressed as:

[0109] c(t) = c *

[0110]

[0111] c(t) = f c [x(t),u(t),d(t)]

[0112] d(t)=d * (t)

[0113]

[0114] t∈[t0,t f ]

[0115] (13) Based on the discussion of system mathematical models in automatic control theory, a transfer function model based on the complex frequency domain is constructed, and an optimization problem of the operating variables in the operating condition switching process in the complex frequency domain is established. This problem can be expressed as:

[0116] ΔC(s)=[G u (s) G d (s)][ΔU(s) ΔD(s)] T

[0117] =G u (s)ΔU(s)+G d (s)ΔD(s)

[0118] ΔC(s)=ΔC * (s)=0

[0119] ΔD(s)=ΔD * (s)

[0120] Where Δ represents the Laplace transform of each variable under zero initial conditions, that is, the increment of each variable. G u (s) and G d ΔU(s) represents the system's transfer function; ΔU(s) represents the operation variable; ΔD(s) represents the driving variable.

[0121] Furthermore, the optimal continuous operational variables in the complex frequency domain are solved based on the complex frequency domain model, and then transformed into optimal continuous operational variables in the time domain through inverse Laplace transform. The corresponding expressions include:

[0122]

[0123] Δu * (t)=L -1 [ΔU * (s)]

[0124]

[0125] Among them, L -1 Indicates the inverse Laplace transform; ΔU * (s) represents the expected value of the variable operated on in the complex frequency domain; u * (t) represents the expected value of the time-domain operated variable.

[0126] According to some embodiments of this application, the analysis method based on the complex frequency domain maps the optimization problem of the operands that cannot be solved in the t-plane to the s-plane, and the process of solving for the optimal continuous operands based on the transfer function in the complex frequency domain is as follows:

[0127] (1) For the above-mentioned optimization problem of the operational variables, the time-domain mathematical model in the t-plane is transformed into a transfer function model in the s-plane through the Laplace transform. The transfer function is actually a function of complex variables and is a rational proper fractional function. In other words, the differential equation in the time domain is transformed into an algebraic equation in the complex frequency domain through the Laplace transform.

[0128] (2) Calculate the system of algebraic equations to obtain the manipulated variables in the complex frequency domain. The manipulated variables in the complex frequency domain are the ideal control trajectory increment ΔU. * (s) can be obtained by solving the algebraic equations of the operation variable optimization problem of the operating condition switching process in the complex frequency domain.

[0129] (3) Perform an inverse Laplace transform on the complex frequency domain result to obtain the corresponding result in the time domain. ΔU * (s) is converted into the ideal control trajectory increment Δu in the time domain after Laplace inverse transform. * (t), and thus obtain the ideal control trajectory u in the time domain. * (t), as shown below:

[0130]

[0131] Δu * (t)=L -1 [ΔU * (s)]

[0132]

[0133] Furthermore, the constraint variables, operating variables, and driving variables also satisfy the following expression:

[0134]

[0135] Where, k u With k d Let G represent the transfer function respectively. u (s) and G d The steady-state gain of (s).

[0136] According to some embodiments of this application, the following must be satisfied during the identification process of the transfer function:

[0137]

[0138] In this way, endpoint constraints are implicitly included in the system's transfer function, thus avoiding the need to add constraints in subsequent calculations. Regardless of how the dynamic characteristics of the operands change during the switching process, they will eventually stabilize at their final values.

[0139] Furthermore, after transforming the time-domain optimal continuous operational variables through the inverse Laplace transform, the time-domain optimal continuous operational variables are simplified using a fitting method, including:

[0140] The time-domain optimal continuous operational variable time series data obtained by the solution are sampled;

[0141] By fitting the sampled data, an approximate control function is obtained;

[0142] The simplified time-domain optimal continuous operating variable is calculated based on the approximate control function.

[0143] According to some embodiments of this application, for complex systems such as chemical processes, although the proposed method can obtain the ideal control trajectory u of the manipulated variable... * The function expression for (t) is provided, but this expression is usually not a combination of simple functions, which brings great inconvenience to practical use. Therefore, in this embodiment, numerical analysis is used to simplify the control strategy, and an approximate function composed of simple functions is used to replace the original complex function for ease of application. That is, first in continuous u * Take a set of measurement points u on (t) j and the corresponding time point t j Subsequently, an approximate control function composed of simple functions was derived through numerical analysis. In this embodiment, since the approximate function obtained by the fitting method has a finite number of parameters, the fitting method is used to determine or measure the possible form of the function through the mechanism model, and all measurement points are approximated as a whole, thereby improving the practicality of the control strategy.

[0144] This invention also selects a process with large fluctuations in the feed flow rate of an in-service ethylene distillation column as an example to verify the effectiveness of the above technical solution:

[0145] In petrochemical enterprises, both grade switching and load changes are planned alterations to process conditions, meaning the system's steady-state operating point before and after the change is known. Therefore, steady-state optimization can be used to first obtain the current and target steady-state operating conditions. This allows the initial values, target values, and ranges of variation of the manipulated variables to be determined.

[0146] Ethylene, as the main product of the ethylene distillation column, will continue to participate in subsequent production processes. This means that the quality of the ethylene product should remain stable under all circumstances to ensure the smooth operation of subsequent processes. Therefore, this paper selects the quality of the ethylene product as a constant constraint variable in the ethylene distillation column operating condition switching problem, which can be expressed as:

[0147] ΔC(s)=G u (s)ΔU(s)+G d (s)ΔD(s)=0

[0148] U(s) = [U1(s) U2(s)] T D(s)=D(s)

[0149] U1(s)≡F D (s),U2(s)≡Q w (s),D(s)≡F F (s)

[0150] The model includes two operational variables: ethylene product output F. D (s) and heat exchange with the reboiler at the bottom of the column Q W (s), denoted by U1(s) and U2(s) respectively; and a driving variable, namely the feed flow rate F. F (s), denoted by D(s).

[0151] By transforming the time-domain mathematical model in the t-plane into the transfer function model in the s-plane using the Laplace transform, the following complex frequency-domain algebraic equations can be obtained:

[0152] 0 = [G u1 (s) G u2 [(s)][ΔU1(s) ΔU2(s)] T +G d (s)ΔD(s)

[0153] =G u1 (s)ΔU1(s)+G u2 (s)ΔU2(s)+G d (s)ΔD(s)

[0154] ΔU1(s)≡ΔF D (s),ΔU2(s)≡ΔQ W (s),ΔD(s)≡ΔF F (s)

[0155] Where, ΔF D (s) represents the increment of product output from the sideline; ΔQ w (s) is the increase in heat exchange in the reboiler at the bottom of the column; ΔF F (s) represents the increment of the feed flow rate; G u1 (s), G u2 (s) and G d (s) represent the transfer functions of ethylene product quality to side stream product output, the transfer function of heat exchange in the reboiler at the bottom of the tower, and the transfer function of feed flow rate, respectively.

[0156] Then, the system of algebraic equations is calculated to obtain the manipulated variable in the complex frequency domain, that is, the ideal control trajectory increment ΔU in the complex frequency domain. * (s).

[0157] Then, an inverse Laplace transform is performed on the complex frequency domain result to finally obtain the corresponding result in the time domain. ΔU * (s) is converted into the ideal control trajectory increment Δu in the time domain after Laplace inverse transform. * (t), and thus obtain the ideal control trajectory u in the time domain. * (t).

[0158] In this embodiment, the endpoint constraints of the operands are satisfied during the identification of the transfer function, thereby avoiding the addition of constraints in subsequent calculations. That is, the following formula must be satisfied during the identification of the transfer function:

[0159]

[0160] In the formula, k u and k d They represent the transfer function G respectively. u (s) and G d The steady-state gain of (s). In this way, the endpoint constraints are implicitly included in the system's transfer function, thus avoiding the need to add constraints in subsequent calculations. Regardless of how the dynamic characteristics of the operands change during the switching process, they will eventually stabilize at the final value.

[0161] Finally, numerical analysis is used to simplify the control strategy, replacing the original complex function with an approximate function composed of simpler functions to improve its practicality. For the large-range fluctuation of the ethylene distillation column feed flow rate in this example, a simplified approximate function of the control trajectory of the operating variables is obtained using a fitting method, such as... Figure 2 As shown.

[0162] Since the approximate function obtained by the fitting method has a finite number of parameters, the fitting method is used to determine or measure the possible form of the function through a mechanistic model, and all measurement points are approximated as a whole, thereby improving the practicality of the control strategy. The obtained control strategy is applied to the ethylene distillation column in this example, and the simulation results are as follows: Figure 3a -d is shown; it is compared and analyzed with conventional PID control strategies and piecewise control strategies, such as Figure 4a As shown in -d. In the ethylene production industry, the maximum permissible deviation in ethylene product quality is typically ±0.01%. This paper also uses this value to measure the feasibility of control strategies. Figure 3d The results show that the continuous control strategy obtained by the analytical method based on the complex frequency domain can meet the requirements of ethylene product quality during the switching of operating conditions. Therefore, the proposed method is an effective solution method for continuous control strategies.

[0163] Figure 4a and Figure 4b This is the result of process switching using conventional control strategies. Figure 4c and Figure 4d This represents the process switching result obtained using a segmented control strategy. Figure 4b The data clearly shows that the ethylene product quality exceeded the permissible range. In other words, conventional process control strategies are not suitable for processes with large-scale operating condition changes. This also underscores the necessity of researching process switching control strategies. Figure 4d This indicates that the segmented control strategy obtained through dynamic optimization can meet the quality requirements of ethylene products; therefore, dynamic optimization is a feasible solution method. However, compared to... Figure 3b and Figure 3c , Figure 4c The trajectory of the manipulated variable in the initial stage does not exhibit a reverse process, proving that the piecewise control strategy based on numerical optimization loses important information during calculation. This is reflected in the operating results as follows: Figure 4d As shown, with Figure 3d In contrast, ethylene product quality fluctuates more drastically and frequently, resulting in poor switching performance. This underscores the necessity and superiority of researching continuous control strategies. Compared to existing segmented control strategies, continuous control strategies can fully represent the characteristic details of the operating variables, thereby reducing operational fluctuations and ensuring smooth switching operations.

[0164] Based on the same technological concept, such as Figure 5 As shown, this application also proposes a continuous process operating system for petrochemical enterprises, comprising:

[0165] The model building module is used to construct a time-domain model based on the dynamic mechanism of the process system.

[0166] The model conversion module is used to convert the time-domain model into a complex frequency-domain model using Laplace transform.

[0167] The variable calculation module is used to solve for the optimal continuous operation variables in the complex frequency domain based on the complex frequency domain model, and then transform them into the optimal continuous operation variables in the time domain through the inverse Laplace transform.

[0168] The operation control module is used to perform continuous operations based on the optimal continuous operation variables in the time domain.

[0169] like Figure 6 As shown, this application also proposes an electronic device, including a memory and a processor, wherein the memory stores a computer program or instructions, and when the computer program or instructions are executed by the processor, they are used to implement at least the above-described process continuous operation method.

[0170] like Figure 7 As shown, this application also proposes a computer-readable storage medium storing a computer program or instructions, which, when executed by a processor, are used to implement at least the above-described continuous process operation method.

[0171] This application also proposes a computer program product, which is stored in a computer-readable storage medium and, when executed by a processor, is used to implement at least the above-described continuous process operation method.

[0172] It is obvious that those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for continuous operation of a process, characterized in that, include: Construct a time-domain model based on the dynamic mechanism of the process system; The time-domain model is transformed into a complex frequency-domain model using the Laplace transform; Based on the aforementioned complex frequency domain model, the optimal continuous operation variables in the complex frequency domain are solved, and then transformed into the optimal continuous operation variables in the time domain through the inverse Laplace transform. Perform continuous operations based on the time-domain optimal continuous operation variables.

2. The continuous operation method of the process as described in claim 1, characterized in that, A time-domain model is constructed based on the dynamic mechanism of the process system, including: The time-domain model is constructed based on the dynamic mechanism and constraints of the process system; wherein, the constraints include: the range of change of the process system's operating variables during the process system's transition from the current operating condition to the target operating condition.

3. The continuous operation method of the process as described in claim 2, characterized in that, The expression corresponding to the dynamic mechanism includes: c(t)=f c [x(t),u(t),d(t)] Where t represents the running time; x represents a vector of time-domain state variables; denoted by , u represents the derivative of the time-domain state variables; denoted by u represents the vector of time-domain manipulated variables; denoted by d represents the vector of time-domain driving variables; denoted by c represents the vector of constraint variables that must remain unchanged during the switching process; f x (·) represents the equation representing the dynamic mechanism of the process object; f c (·) represents the equation for the constant constraint variable c(t).

4. The continuous operation method of the process as described in claim 3, characterized in that, The constraints include: endpoint constraints and process constraints; wherein, The endpoint constraints, and the corresponding expressions, include: Where t0 represents the moment when all variables are at their steady-state values ​​before the change in operating conditions; t f This indicates the moment when each variable reaches its steady-state value after a change in operating conditions. This represents the steady-state value of the variable (·) at time t0; Indicates the variable (·) in t f steady-state value at time; The process constraints, and the corresponding expressions, include: c(t)=c * ,t∈[t0,t f ] Among them, c * Let c(t) represent the expected value of c(t) during the process of changing operating conditions.

5. The continuous operation method of the process as described in claim 4, characterized in that, The complex frequency domain model, and the corresponding expressions, include: ΔC(s)=[G u (s) G d (s)][ΔU(s) ΔD(s)] T =G u (s)ΔU(s)+G d (s)ΔD(s) ΔC(s)=ΔC * (s)=0 ΔD(s)=ΔD * (s) Where Δ represents the Laplace transform of each variable under zero initial conditions; T represents the transpose of the matrix; G u (s) and G d ΔU(s) represents the transfer function of the process system; ΔU(s) represents the complex frequency domain operation variable; ΔC(s) represents the complex frequency domain constraint variable; ΔC * (s) represents the expected value of the complex frequency domain constraint variable; ΔD(s) represents the complex frequency domain driving variable; ΔD * (s) represents the expected value of the driving variable in the complex frequency domain.

6. The continuous operation method of the process as described in claim 5, characterized in that, Based on the aforementioned complex frequency domain model, the optimal continuous operational variables in the complex frequency domain are solved, and then transformed into optimal continuous operational variables in the time domain through inverse Laplace transform. The corresponding expressions include: Δu * (t)=L -1 [ΔU * (s)] Among them, L -1 Indicates the inverse Laplace transform; ΔU * (s) represents the expected value of the variable operated on in the complex frequency domain; u * (t) represents the expected value of the time-domain operated variable.

7. The continuous operation method of the process as described in claim 6, characterized in that, The constraint variables, manipulation variables, and driving variables also satisfy the following expression: Where, k u With k d Let G represent the transfer function respectively. u (s) and G d The steady-state gain of (s).

8. The continuous operation method of the process according to any one of claims 1-7, characterized in that, After transforming the time-domain optimal continuous operational variables through the inverse Laplace transform, the time-domain optimal continuous operational variables are simplified using a fitting method, including: The time-domain optimal continuous operational variable time series data obtained by the solution are sampled; By fitting the sampled data, an approximate control function is obtained; The simplified time-domain optimal continuous operating variable is calculated based on the approximate control function.

9. A continuous process operating system for petrochemical enterprises, characterized in that, include: The model building module is used to construct a time-domain model based on the dynamic mechanism of the process system. The model conversion module is used to convert the time-domain model into a complex frequency-domain model using Laplace transform. The variable calculation module is used to solve for the optimal continuous operation variable in the complex frequency domain based on the complex frequency domain model, and to transform it into the optimal continuous operation variable in the time domain through the inverse Laplace transform. An operation control module is used to perform continuous operations based on the time-domain optimal continuous operation variables.

10. An electronic device comprising a memory and a processor, characterized in that, The memory stores a computer program or instructions, which, when executed by the processor, are used to implement at least the method described in any one of claims 1-8.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program or instructions, which, when executed by a processor, are at least used to implement the method described in any one of claims 1-8.

12. A computer program product, said computer program product being stored in a computer-readable storage medium, characterized in that, When the computer program product is executed by a processor, it is used to implement at least the method described in any one of claims 1-8.

Images