Thermal system combined cycle online operation optimization method and device based on decomposition algorithm

Abstract

The invention provides a thermodynamic system combined cycle online operation optimization method and device based on a decomposition algorithm. The method comprises the following steps: constructing a dynamic numerical model of a thermodynamic system; constructing a thermodynamic system operation optimization problem based on the thermodynamic system dynamic numerical model; decomposing a thermodynamic system operation optimization problem into a nonlinear programming sub-problem and a linear programming main problem; alternately and iteratively solving the nonlinear programming sub-problem and the linear programming main problem until convergence, and obtaining an optimal decision variable of the current optimization round; and operating the optimal decision variable and other operating parameters of the thermodynamic system under the optimal decision variable, and performing the next round of optimization in a rolling optimization mode. According to the method, the optimal decision variable of the operation optimization problem of the thermodynamic system is solved through a decomposition algorithm, the optimization calculation efficiency is improved while the model precision is ensured, the optimization calculation result is obtained in the time scale of the system operation working condition, online optimization under the time-varying boundary condition is achieved, and the optimization efficiency is improved. Therefore, the safety and the stability of the thermodynamic system are improved.

Description

Technical Field

[0001] This invention relates to the field of energy system optimization and control technology, and particularly to a method and apparatus for online operation optimization of combined cycle thermal systems based on a decomposition algorithm. Background Technology

[0002] With the large-scale grid integration of renewable energy, the volatility and uncertainty of the power system have increased significantly, placing higher demands on the grid's regulation capabilities. To address this, the power grid has successively introduced ancillary service market mechanisms such as peak shaving and frequency regulation, encouraging various power sources, including thermal power generation equipment, to participate in regulation, thereby improving the system's flexibility and stability. In frequency regulation services, the response timescale of thermal power equipment is typically on the order of seconds to minutes. Therefore, it is urgent to study the dynamic response characteristics of thermal power generation systems on this timescale and to optimize their online operation to meet the grid's rapid response requirements.

[0003] However, the dynamic operation optimization of thermal systems still faces many challenges. Existing optimization methods are mainly divided into three categories: model-driven optimization algorithms based on optimality conditions, model-driven algorithms based on heuristic search, and data-driven optimization algorithms.

[0004] The first type of method is based on modeling and solving the optimality conditions of Karush-Kuhn-Tucker (KKT). Although this method can theoretically obtain the optimal solution, the characteristics of thermodynamic systems, such as a large number of devices, complex structures, and drastic changes in the physical properties of the circulating working fluid with the operating conditions, result in a large number of variables to be optimized and complex constraints. The establishment and solution of the KKT equations are difficult, the computational efficiency is low, and it is difficult to meet the real-time optimization requirements of seconds to minutes.

[0005] The second type of method employs heuristic optimization algorithms, such as genetic algorithms and particle swarm optimization. While these methods can handle nonlinear and nonconvex problems to some extent, they often introduce a large number of hyperparameters. The influence of these parameters on the optimization results is unclear, and the global optimality of the solution cannot be guaranteed, resulting in poor stability and repeatability of the optimization results.

[0006] The third type of method is the data-driven optimization method, which has developed rapidly in recent years. It uses real-world operating data to train a surrogate model to approximate the input-output relationship of the thermodynamic system, and then optimizes the parameters. However, this type of method relies on high-quality training data that covers a wide range of operating conditions, but it is difficult to obtain real-world variable operating condition data. At the same time, the surrogate model lacks physical interpretability and extrapolation ability, and is prone to prediction bias under extreme operating conditions, which affects the reliability of optimization and makes it difficult to apply directly to online control systems.

[0007] Furthermore, to improve computational efficiency, many existing methods tend to simplify physical models, sacrificing model accuracy for increased computational speed. However, this simplification strategy can easily lead to optimization results deviating from the actual physical process when faced with complex boundary conditions or extreme operating states, posing a risk of violating safe operating boundaries and thus affecting the system's safety and stability.

[0008] In summary, current methods for dynamic online operation optimization of thermal systems cannot fully meet the requirements of power grid frequency regulation and peak shaving for rapid response and real-time optimization of thermal equipment in terms of model accuracy, solution efficiency, and safety assurance. Summary of the Invention

[0009] This invention provides a method and apparatus for online operation optimization of combined cycle thermodynamic systems based on decomposition algorithms. It overcomes the shortcomings of existing technologies, such as the complex structure of combined cycle thermodynamic systems, drastic changes in the physical properties of the circulating working fluid, and the high-dimensional and strong nonlinearity of the mathematical model, which prevent the dynamic online operation optimization of the system from simultaneously taking into account model accuracy, solution efficiency, and safety. While ensuring the accuracy of the physical model, it improves the optimization calculation efficiency and adapts to time-varying boundary conditions and complex operating environments, thereby enhancing the safety and stability of the system.

[0010] On one hand, this invention provides a method for optimizing the joint cycle operation of a thermal system based on a decomposition algorithm, comprising: constructing a dynamic numerical model of the thermal system; constructing a thermal system operation optimization problem based on the dynamic numerical model of the thermal system; decomposing the thermal system operation optimization problem into a nonlinear programming subproblem and a linear programming master problem; iteratively solving the nonlinear programming subproblem and the linear programming master problem alternately until convergence, obtaining the optimal decision variables for the current optimization round; running the optimal decision variables and other operating parameters of the thermal system under the optimal decision variables, and performing the next round of optimization using a rolling optimization approach.

[0011] Furthermore, the construction of the dynamic numerical model of the thermal system includes: constructing the dynamic numerical model of the thermal system based on the moving boundary method; wherein, the dynamic numerical model of the thermal system includes the moving boundary model of the evaporator, the moving boundary model of the condenser, the heat-work conversion process model of the expander, and the heat-work conversion process model of the working fluid pump.

[0012] Furthermore, the step of constructing a thermal system operation optimization problem based on the dynamic numerical model of the thermal system includes: taking the highest average thermal efficiency of the system within a set time interval as the optimization objective, the flow rate of the system working fluid pump as the decision variable, and the dynamic numerical model of the thermal system and the range of values ​​of the decision variable as constraints, to construct the thermal system operation optimization problem; wherein, the thermal system operation optimization problem is a nonlinear programming problem.

[0013] Furthermore, the step of decomposing the thermal system operation optimization problem into a nonlinear programming subproblem and a linear programming master problem includes: based on the generalized Benders decomposition algorithm, by selecting and fixing some complex variables, decomposing the thermal system operation optimization problem into the nonlinear programming subproblem and the linear programming master problem.

[0014] Further, the iterative solution of the nonlinear programming subproblem and the linear programming master problem until convergence, to obtain the optimal decision variables for the current optimization round, includes: solving the nonlinear programming subproblem to obtain the values ​​of the variables to be solved; calculating the system average thermal efficiency and Benders cut within a set time interval based on the values ​​of the variables to be solved and fixed complex variables; updating the upper bound of the optimal solution for the thermal system operation optimization problem based on the system average thermal efficiency; solving the linear programming master problem with the Benders cut as a constraint and the introduced auxiliary variables as the optimization objective to obtain the updated complex variables and the values ​​of the auxiliary variables to be solved; updating the lower bound of the optimal solution for the thermal system operation optimization problem based on the values ​​of the auxiliary variables to be solved; wherein, the updated complex variables are used to replace the fixed complex variables in the next iteration round.

[0015] Furthermore, the iterative solution of the nonlinear programming subproblem and the linear programming master problem until convergence, to obtain the optimal decision variables for the current optimization round, includes: when the difference between the upper bound and the lower bound of the optimal solution of the thermal system operation optimization problem is less than a first threshold, and the difference between the average thermal efficiencies of adjacent systems calculated for a consecutive preset number of iterations is less than a second threshold, the thermal system operation optimization problem converges, and the optimal decision variables for the current optimization round are obtained.

[0016] Secondly, the present invention also provides an online operation optimization device for a combined cycle of thermal systems based on a decomposition algorithm, comprising: a thermal system dynamic numerical model construction module for constructing a dynamic numerical model of the thermal system; a thermal system operation optimization problem construction module for constructing an operation optimization problem of the thermal system based on the dynamic numerical model of the thermal system; a thermal system operation optimization problem decomposition module for decomposing the thermal system operation optimization problem into a nonlinear programming subproblem and a linear programming master problem; a thermal system operation optimization problem solving module for iteratively solving the nonlinear programming subproblem and the linear programming master problem until convergence, thereby obtaining the optimal decision variables for the current optimization round; and an optimal decision variable and related parameter operation module for operating the optimal decision variables and other operating parameters of the thermal system under the optimal decision variables, and performing the next round of optimization using a rolling optimization method.

[0017] Thirdly, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the online operation optimization method for combined cycle operation of a thermodynamic system based on the decomposition algorithm as described above.

[0018] Fourthly, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the online operation optimization method for combined cycle operation of a thermodynamic system based on a decomposition algorithm as described above.

[0019] Fifthly, the present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the online operation optimization method for combined cycle operation of a thermodynamic system based on the decomposition algorithm as described above.

[0020] This invention provides a method for online optimization of joint-cycle operation of a thermal system based on a decomposition algorithm. It constructs a dynamic numerical model of the thermal system and, based on this model, formulates an optimization problem for the system's operation. This problem is then decomposed into a nonlinear programming subproblem and a linear programming master problem. These subproblems are iteratively solved alternately until convergence, yielding the optimal decision variables for the current optimization round. The optimal decision variables and other operating parameters of the thermal system under these optimal decision variables are then used, and a rolling optimization approach is employed for the next optimization round. This method solves the optimal decision variables for the thermal system's operation optimization problem using a decomposition algorithm, improving computational efficiency while maintaining model accuracy. It obtains optimization results within the timescale of system operating conditions, achieving online optimization under time-varying boundary conditions, thereby improving the overall energy efficiency of the thermal system. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0022] Figure 1 This is a flowchart illustrating the online operation optimization method for combined cycle thermal systems based on decomposition algorithms provided in this embodiment of the invention.

[0023] Figure 2 This is a schematic diagram of a gas-steam-organic Rankine cycle combined power generation system provided in an embodiment of the present invention.

[0024] Figure 3This is a schematic diagram of the operation process of a typical ORC power generation system provided in an embodiment of the present invention.

[0025] Figure 4 This is a schematic diagram of the moving boundary model of the evaporator in the thermal system provided in the embodiment of the present invention.

[0026] Figure 5 This is a schematic diagram of the solution process for the thermal system operation optimization problem based on the Benders decomposition algorithm provided in an embodiment of the present invention.

[0027] Figure 6 This is a schematic diagram illustrating the basic principle of the rolling optimization method provided in this embodiment of the invention.

[0028] Figure 7 This is a schematic flowchart of the online operation optimization method for combined circulation of a thermal system based on the rolling optimization method provided in an embodiment of the present invention.

[0029] Figure 8 This is a schematic diagram of the structure of the online operation optimization device for combined cycle of thermodynamic system based on decomposition algorithm provided in an embodiment of the present invention.

[0030] Figure 9 This is a schematic diagram of the physical structure of the electronic device provided in an embodiment of the present invention. Detailed Implementation

[0031] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0032] To address the problem that existing technologies for combined cycle thermodynamic systems cannot simultaneously balance model accuracy, solution efficiency, and safety due to factors such as the complex structure of the combined cycle thermodynamic system, the drastic changes in the properties of the circulating working fluid, and the high-dimensional nonlinearity of the mathematical model, this invention proposes an online optimization method for combined cycle thermodynamic systems based on a decomposition algorithm.

[0033] Specifically, Figure 1 The diagram shows a flowchart of the online operation optimization method for combined cycle thermal systems based on decomposition algorithms provided in an embodiment of the present invention.

[0034] like Figure 1As shown, the method includes: S110, constructing a dynamic numerical model of the thermal system; S120, constructing a thermal system operation optimization problem based on the dynamic numerical model of the thermal system; S130, decomposing the thermal system operation optimization problem into a nonlinear programming subproblem and a linear programming master problem; S140, alternately iterating to solve the nonlinear programming subproblem and the linear programming master problem until convergence, obtaining the optimal decision variables; S150, running the optimal decision variables and other operating parameters of the thermal system under the optimal decision variables, and using a rolling optimization method for the next round of optimization.

[0035] The following will provide a detailed description of steps S110-S150 and related steps.

[0036] S110, construct a dynamic numerical model of the thermal system.

[0037] Dynamic numerical models of thermodynamic systems are used to simulate the behavior of thermodynamic systems over time, including mathematical descriptions of processes such as temperature distribution, fluid flow, and heat transfer. The foundations of dynamic numerical models of thermodynamic systems include physical laws such as the conservation of energy, mass, and momentum, as well as the heat conduction equation.

[0038] For a combined cycle thermodynamic system, its dynamic numerical model can include the moving boundary model of the evaporator, the moving boundary model of the condenser, the heat-work conversion process model of the expander, and the heat-work conversion process model of the working fluid pump. Of course, the number of components involved in the dynamic numerical model of the thermodynamic system can be set according to the actual situation of the thermodynamic system, and no specific limitation is made here.

[0039] When constructing dynamic numerical models of thermodynamic systems, one can self-program based on physical equations in programming software such as MATLAB and Python, or one can use existing professional software such as ANSYS Fluent, COMSOL Multiphysics, and OpenFOAM.

[0040] The specific construction process and contents of the dynamic numerical model of the thermal system will be elaborated in detail in the following embodiments.

[0041] Based on the dynamic numerical model of the thermal system constructed in step S110, step S120 is further executed.

[0042] S120, Based on the dynamic numerical model of the thermal system, construct the thermal system operation optimization problem.

[0043] The thermal system operation optimization problem refers to finding an optimal sequence of decision variables to achieve a certain performance index, while satisfying the physical laws, operational limitations, and safety constraints of the thermal system. The thermal system operation optimization problem includes the optimization objective (objective function), dynamic model constraints, decision variables (control variables), and operational constraints.

[0044] Optimization objectives include minimizing fuel consumption, maximizing energy efficiency, and minimizing carbon emissions. Dynamic model constraints refer to the physical equations governing the evolution of the system state over time, derived from dynamic numerical models of thermodynamic systems. Decision variables are adjustable system operating parameters, such as valve opening, fan speed, and fuel supply rate. Operational constraints include upper and lower limits for controlling decision variables (e.g., maximum valve opening) and limits for controlling state variables (e.g., maximum operating temperature of equipment).

[0045] For a combined cycle thermal system, the optimization objective can be to maximize the system's average thermal efficiency over a certain time interval. The system's working fluid pump flow rate can be used as the decision variable, and the dynamic numerical model of the thermal system and the range of values ​​for the decision variables can be used as constraints to construct an optimization problem for the thermal system. The optimization problem for the thermal system constructed in this step is essentially a nonlinear programming problem.

[0046] The specific expression for the thermal system optimization operation problem will be elaborated in detail in the following embodiments.

[0047] After constructing the thermal system operation optimization problem based on the dynamic numerical model of the thermal system in step S120, steps S130-S140 are further executed sequentially.

[0048] S130, the thermal system operation optimization problem is decomposed into a nonlinear programming subproblem and a linear programming master problem.

[0049] S140, alternately iterate to solve the nonlinear programming subproblem and the linear programming master problem until convergence, and obtain the optimal decision variables for the current optimization round.

[0050] It is easy to understand that for the optimization problem of the operation of a constructed thermal system, the Generalized Benders Decomposition (GBD) algorithm can be used to solve the problem and obtain the optimal decision variables. Of course, other feasible decomposition algorithms can also be used to solve the problem, and no specific limitation is made here.

[0051] The generalized Benders decomposition algorithm is an iterative optimization method for solving mixed-integer nonlinear programming problems. Its basic idea is to decompose the original optimization problem into a main problem and subproblems, and then gradually approach the optimal solution by alternately solving these two problems: First, fix the integer variables or other selected complex variables, and solve the nonlinear programming subproblems with respect to continuous variables or other variables to be solved. Then, construct a Benders cut (i.e., feedback constraint) using the solution information from the subproblems and add it to the main problem. The main problem is a linear programming problem with complex variables as the main variables, used to update the fixed complex variables. Repeat the above steps until convergence.

[0052] In this embodiment, the original thermal system operation optimization problem is first decomposed into a nonlinear programming subproblem and a linear programming master problem. Specifically, the decision variables are selected as complex variables, and the given initial values ​​of the complex variables are substituted into the original thermal system operation optimization problem, which degenerates into a nonlinear programming subproblem. Correspondingly, the linear programming master problem takes the introduced auxiliary variables as the optimization objective and the set of Benders cuts obtained by solving the nonlinear programming subproblem as the constraints.

[0053] Then, the nonlinear programming subproblem and the linear programming master problem are solved iteratively to obtain the optimal decision variables for the current optimization round. Specifically, by solving the nonlinear programming subproblem, the average thermal efficiency of the system over a certain time interval can be obtained. Furthermore, a Benders cut can be constructed, and the upper bound of the optimal solution to the original thermal system operation optimization problem can be updated. Then, using the constructed Benders cut as a constraint and introducing auxiliary variables as the optimization objective, the linear programming master problem is solved to obtain the updated complex variables and update the lower bound of the optimal solution to the original thermal system operation optimization problem. The initial value of the upper bound of the optimal solution to the thermal system operation optimization problem can be set to positive infinity, and the initial value of the lower bound can be set to negative infinity; no specific limitations are imposed here.

[0054] Finally, the updated complex variables are substituted into the nonlinear programming subproblem, and the nonlinear programming subproblem and the linear programming main problem are solved iteratively until the convergence condition is met, thus obtaining the optimal decision variables for the current optimization round.

[0055] The definitions and solutions of the nonlinear programming subproblems and the linear programming master problem will be explained in detail in the following examples.

[0056] Based on the optimal decision variables obtained in step S140 for the current optimization round, step S150 is further executed.

[0057] S150, run the optimal decision variables and other operating parameters of the thermal system under the optimal decision variables, and perform the next round of optimization using a rolling optimization method.

[0058] Specifically, while obtaining the optimal decision variables for the current optimization round, we also learn about the other operating parameters of the thermal system under the optimal decision variables. In this case, by running the optimal decision variables and the other operating parameters of the thermal system under the optimal decision variables in the thermal system, we can realize the online operation optimization of the thermal system in the current optimization round, which is also the beginning of the next round of optimization. Then, we can use rolling optimization to realize the continuous online optimization of the thermal system.

[0059] In this embodiment, a dynamic numerical model of the thermal system is constructed, and based on this model, a thermal system operation optimization problem is established. This problem is then decomposed into a nonlinear programming subproblem and a linear programming master problem, which are iteratively solved alternately until convergence. This yields the optimal decision variables for the current optimization round, allowing the calculation of the optimal decision variables and other operating parameters of the thermal system under those variables. A rolling optimization approach is then used for the next optimization round. This method solves the thermal system operation optimization problem using a decomposition algorithm, improving computational efficiency while maintaining model accuracy. It obtains optimization results within the timescale of system operating conditions, achieving online optimization under time-varying boundary conditions, thereby improving the overall operational energy efficiency of the thermal system.

[0060] Figure 2 A schematic diagram of a gas-steam-organic Rankine cycle combined power generation system provided in an embodiment of the present invention is shown.

[0061] like Figure 2 As shown, the gas-steam-organic Rankine cycle combined power generation system includes a gas-fired power generation system, a steam power generation system, and an organic Rankine cycle (ORC) power generation system. In the gas-steam combined cycle, the exhaust gas temperature of the waste heat boiler can reach over 170°C, possessing high heat recovery value. The heat source for the ORC power generation system is the flue gas from the waste heat boiler. When the gas-steam combined cycle system participates in grid frequency regulation, its operating conditions often change drastically, leading to significant variations in the exhaust gas temperature and flow rate of the waste heat boiler. This results in pronounced dynamic characteristics of the ORC power generation system. The embodiments of this invention and subsequent embodiments will use the dynamic online operation optimization of the ORC power generation system as an example to explain the online operation optimization method for combined thermal system operation based on decomposition algorithms provided by this invention.

[0062] Based on the above embodiments, the following will further describe in detail the construction process of the dynamic numerical model of the thermal system.

[0063] Figure 3 A schematic diagram of the operation flow of a typical ORC power generation system provided in an embodiment of the present invention is shown. Figure 3 As shown, the ORC power generation system uses an organic circulating working fluid, and its components include an expander, a working fluid pump, a generator, an evaporator, and a condenser.

[0064] Accordingly, the dynamic numerical model of the ORC power generation system, namely the dynamic numerical model of the thermal system, is constructed, including: constructing the dynamic numerical model of the thermal system based on the moving boundary method; wherein, the dynamic numerical model of the thermal system includes the moving boundary model of the evaporator, the moving boundary model of the condenser, the heat-work conversion process model of the expander, and the heat-work conversion process model of the working fluid pump.

[0065] Specifically, based on the moving boundary method, the evaporator and condenser are divided into superheated, two-phase, and subcooled regions according to the different phase states of the working fluid. Mass conservation equations, energy conservation equations, and tube wall energy conservation equations are constructed for each of these three regions, with the length of each region used as a variable and continuously updated during the dynamic simulation. Simultaneously, based on the relationship between the flow rate of the working fluid through the working components and its rotational speed, volumetric displacement, working fluid density, and volumetric efficiency, flow equations are constructed for the expander and working fluid pump, respectively. Thermodynamic process equations for the working components are constructed based on the specific enthalpy and isentropic efficiency of the working fluid inlet and outlet. These equations together constitute the dynamic numerical model of the thermodynamic system.

[0066] Taking the moving boundary method applied to evaporators as an example, Figure 4 A schematic diagram of the moving boundary model of an evaporator in a thermal system provided by an embodiment of the present invention is shown. Figure 4 In the process, the working fluid (such as water or other organic working fluid) enters the evaporator from the left side, undergoes heating and evaporation processes inside, and changes from a subcooled state to a saturated liquid state, a two-phase state, a saturated gas state, and a superheated state, and finally is discharged from the right side in the form of steam. Figure 4 From left to right, these are the supercooled zones. Two-phase region and overheated areas .

[0067] Within the subcooled region, the working fluid is in a completely liquid state. Based on the lumped parameter method, the temperature of the subcooled region is expressed as: The density is enthalpy value The length of the supercooled zone is used This indicates that the heat source fluid transfers heat to the liquid working medium through the pipe wall, causing its temperature to gradually increase.

[0068] Within the two-phase region, the working fluid exists simultaneously in liquid and gaseous states, forming a state of two-phase coexistence. The saturated liquid temperature is expressed as... The density is Enthalpy value The mass flow rate is ; Saturated gas temperature is expressed as The density is Enthalpy value The mass flow rate is The temperature of the two-phase region is expressed as: The average density is Enthalpy value As heat is continuously input, the liquid working fluid gradually evaporates, producing steam. The length of the two-phase region is... express.

[0069] Within the superheated zone, the working fluid has completely transformed into a gaseous state, and the temperature is expressed as follows: The density is enthalpy value The length of the overheated zone is used as... This indicates that the steam continues to absorb heat, its temperature rises further, and it eventually leaves the evaporator as steam with a higher temperature and enthalpy.

[0070] In addition, Figure 4 middle, , These represent the mass flow rates of the working fluid entering and leaving the evaporator, respectively. , These represent the enthalpy values ​​of the working fluid entering and leaving the evaporator, respectively. , These represent the temperature and mass flow rate of the heat source fluid, respectively. , , These represent the temperatures of the pipe walls in the subcooled zone, two-phase zone, and superheated zone, respectively. , , These represent the temperatures of the heat source fluids in the subcooled, two-phase, and superheated regions, respectively. During the dynamic simulation of the heat exchanger, the lengths of these three regions change along the boundaries of the saturated liquid and saturated vapor.

[0071] Construct a moving boundary model of the evaporator and make the following assumptions: 1) Assume that the evaporator is a horizontal cylindrical pipe with a certain cross-sectional area, in which the working fluid and the heat source fluid are both one-dimensional flows; 2) Ignore the axial heat conduction between the working fluid and the pipe wall; 3) Ignore the changes in the gravitational potential energy and kinetic energy of the fluid; 4) Ignore the pressure drop of the fluid inside the pipe, and the pressure of the fluid inside the pipe is only related to time.

[0072] Based on the above assumptions, the mass conservation equation, energy conservation equation, and pipe wall energy conservation equation for the three regions of superheated zone, two-phase zone, and subcooled zone can be derived.

[0073] For the supercooled zone, the mass conservation moving boundary model, energy conservation moving boundary model and pipe wall energy conservation moving boundary model are respectively referred to in equations (1)-(3).

[0074] (1).

[0075] (2).

[0076] (3).

[0077] In equations (1)-(3), This indicates the cross-sectional area of ​​the heat exchange tube. This represents the average density of the working fluid in the subcooled region. This represents the density of a saturated liquid. Indicates the length of the supercooled region. Indicates time, Indicates pressure, Indicates the specific enthalpy of the working fluid. This indicates the specific enthalpy of the working fluid in the supercooled region. This represents the specific enthalpy of a saturated liquid. This represents the average specific enthalpy of the working fluid in the supercooled region. This indicates the inlet specific enthalpy of the working fluid in the supercooled region. This represents the inlet mass flow rate of the working fluid in the subcooled region. This represents the mass flow rate of a saturated liquid. This indicates the inner diameter of the heat exchange tube. This indicates the heat transfer coefficient within the subcooled zone tube. This indicates the temperature of the tube wall in the subcooled zone. Indicates the temperature of the supercooled zone. This represents the specific heat capacity at constant pressure. Indicates the density of the working fluid, subscript Indicates pipe wall, This indicates the heat transfer coefficient outside the subcooled zone tubes. Indicates the outer diameter of the heat exchange tube. This indicates the temperature of the heat source fluid in the subcooled zone.

[0078] The average density of the working fluid in the subcooled zone is calculated from the average value of the working fluid pressure and the inlet and outlet specific enthalpy, as shown in equations (4)-(5) below.

[0079] (4).

[0080] (5).

[0081] For the two-phase region, the mass conservation moving boundary model, energy conservation moving boundary model and pipe wall energy conservation moving boundary model are respectively referred to in equations (6)-(8).

[0082] (6).

[0083] (7).

[0084] (8).

[0085] In equations (6)-(8), Indicates the average porosity. Represents the density of a saturated gas. Indicates the length of the two-phase region. This represents the mass flow rate of the saturated gas. This represents the specific enthalpy of a saturated gas. This represents the heat transfer coefficient within the tube in the two-phase region. This indicates the temperature of the pipe wall in the two-phase region. This indicates the temperature of the two-phase region. This represents the heat transfer coefficient outside the tube in the two-phase region. This indicates the temperature of the heat source fluid in the two-phase region.

[0086] The average porosity is calculated from the density ratio μ of saturated gas and saturated liquid, as shown in equations (9)-(10).

[0087] (9).

[0088] (10).

[0089] For the superheated zone, its mass conservation moving boundary model, energy conservation moving boundary model and pipe wall energy conservation moving boundary model are respectively referred to in the following equations (11)-(13).

[0090] (11).

[0091] (12).

[0092] (13).

[0093] In equations (11)-(13), This represents the average density of the working fluid in the superheated zone. Indicates the length of the overheated zone. This represents the average specific enthalpy of the working fluid in the superheated zone. This indicates the outlet specific enthalpy of the working fluid in the superheated zone. This represents the outlet mass flow rate of the working fluid in the superheated zone. This indicates the heat transfer coefficient within the superheated zone of the pipe. This indicates the temperature of the pipe wall in the superheated zone. Indicates the temperature of the overheated zone. This indicates the heat transfer coefficient outside the tube in the superheated zone. This indicates the temperature of the heat source fluid in the superheated zone.

[0094] Total heat absorption of the evaporator It is calculated based on the enthalpy difference between the inlet and outlet of the circulating working fluid and the mass flow rate, as shown in the following formula (14).

[0095] (14).

[0096] In equation (14), Indicates the mass flow rate of the heat source fluid. The subscript indicates the isobaric specific heat capacity of the heat source fluid. No. Each region Indicates the first The inlet temperature of the heat source fluid in each region. Indicates the first The outlet temperature of the heat source fluid in each region. Indicates the first Mass flow rate of the working fluid in each region Indicates the first Enthalpy ratio of working fluid exports in each region Indicates the first The enthalpy ratio of imported working fluids in each region.

[0097] Mass flow rate of working fluid in expander According to the machine's speed Volumetric displacement Imported working fluid density and volumetric efficiency The calculation is shown in the following formula (15).

[0098] (15).

[0099] In equation (15), the subscript This refers to an expander. Characteristic parameters such as volumetric displacement and efficiency can be obtained from the machine design manual and by fitting operational data.

[0100] Expander output power Based on the working fluid mass flow rate and the enthalpy difference between the working fluid inlet and outlet. The calculation is shown in the following formula (16).

[0101] (16).

[0102] in, Indicates the enthalpy value of the working fluid import. This represents the enthalpy value at the outlet of the working fluid. The enthalpy difference at the outlet of the working fluid is calculated from the thermodynamic process equation, as shown in equation (17).

[0103] (17).

[0104] In equation (17), This represents the isentropic efficiency of the expander. This represents the specific enthalpy value obtained from an isentropic process in an expander.

[0105] For working fluid pumps, the formula for calculating their mass flow rate can be found in the following formula (18).

[0106] (18).

[0107] In equation (18), Indicates the pump speed of the working fluid. This indicates the volumetric displacement of the working fluid pump. This indicates the density of the working fluid at the pump inlet. Indicates the volumetric efficiency of the working fluid pump, subscript This indicates a working fluid pump.

[0108] The power consumption of the working fluid pump can be calculated based on the enthalpy difference between the inlet and outlet of the working fluid, as shown in the following formula (19).

[0109] (19).

[0110] in, This indicates the enthalpy value at the outlet of the working fluid pump. This represents the enthalpy value at the inlet of the working fluid pump. The specific enthalpy of the working fluid at the outlet of the working fluid pump can be calculated using the following formula (20).

[0111] (20).

[0112] In equation (20), This represents the specific enthalpy value obtained from an isentropic process using a pump. This indicates the isentropic efficiency of the working fluid pump.

[0113] In the above equations (1)-(20), the density, heat transfer coefficient, specific enthalpy, etc. of the working fluid are obtained by querying the physical property library, such as CoolProp, Refprop, etc., based on the temperature and pressure values.

[0114] Based on the above formulas (1)-(20), the dynamic numerical model of the ORC power generation system, i.e., the dynamic numerical model of the thermal system, can be constructed.

[0115] It should be noted that for an ORC power generation system in a real thermodynamic combined cycle, the heat source temperature, heat source flow rate, cooling water temperature and flow rate can be monitored in real time at every moment, and therefore are assumed to be known in the subsequent solution.

[0116] In this embodiment, a dynamic numerical model of the thermal system is constructed, and based on this model, a thermal system operation optimization problem is established. This problem is then decomposed into a nonlinear programming subproblem and a linear programming master problem, which are iteratively solved alternately until convergence. This yields the optimal decision variables for the current optimization round, allowing the calculation of the optimal decision variables and other operating parameters of the thermal system under those variables. A rolling optimization approach is then used for the next optimization round. This method solves the thermal system operation optimization problem using a decomposition algorithm, improving computational efficiency while maintaining model accuracy. It obtains optimization results within the timescale of system operating conditions, achieving online optimization under time-varying boundary conditions, thereby improving the overall operational energy efficiency of the thermal system.

[0117] Based on the above embodiments, the following will further describe in detail the construction and solution process of the thermal system operation optimization problem, still taking the ORC power generation system as an example.

[0118] Based on the dynamic numerical model of the thermal system, a thermal system operation optimization problem is constructed, including: taking the highest average thermal efficiency of the system within a set time interval as the optimization objective, the system working fluid pump flow rate as the decision variable, and the dynamic numerical model of the thermal system and the range of values ​​of the decision variable as constraints, and constructing a thermal system operation optimization problem; wherein, the thermal system operation optimization problem is a nonlinear programming problem.

[0119] It is easy to understand that, since fluctuations in the boundary conditions of the ORC power generation system will affect the system operating parameters and the overall thermal efficiency of the system, in order to ensure that the system can adjust its operating parameters in a timely manner when the boundary conditions change and maintain high energy efficiency, this embodiment selects a period of time ( Average thermal efficiency of the internal ORC power generation system With the highest optimization objective as the system's working fluid pump flow rate within the specified time period as the decision variable, an ORC (Organizational Controlled Reaction) power generation system operation optimization problem is constructed, which is essentially a thermal system operation optimization problem. Solving this thermal system operation optimization problem yields the following results. The optimal decision variable within a given time period is the optimal system working fluid pump flow rate.

[0120] The definition equation for the average thermal efficiency of the system can be found in equation (21), and the definition of the thermal system operation optimization problem can be found in equation (22).

[0121] (twenty one).

[0122] (twenty two).

[0123] In equation (22), This indicates the lower limit of the working fluid pump flow rate. This indicates the upper limit of the working fluid pump flow rate. The set of constraint equations represents the problem of optimizing the operation of a thermal system.

[0124] After conceiving the thermal system operation optimization problem, this embodiment uses the generalized Benders decomposition algorithm to solve the thermal system operation optimization problem. Figure 5 A schematic diagram of the solution process for the thermal system operation optimization problem based on the Benders decomposition algorithm provided in an embodiment of the present invention is shown.

[0125] like Figure 5 As shown, firstly, given the time length considered in the optimization calculation... Set the computation time step in the solver of the dynamic numerical model of the thermal system, and select the decision variable as the complex variable, denoted as . y The remaining variables to be determined, excluding the complex variables, are denoted as... In the ORC power generation system operation optimization problem, the working fluid pump flow rate is selected as a complex variable, i.e. .

[0126] By fixing the complex variables, the optimization problem of the thermal system operation is decomposed into: 1) solving the nonlinear programming problem under the fixed complex variables; 2) updating the fixed complex variables and solving the problem. When the complex variables... When the time limit is reached, the thermal system operation optimization problem will be decomposed into a nonlinear programming subproblem and a linear programming master problem. The nonlinear programming subproblem takes the form of: The problem of optimizing the operation of a thermal system under fixed conditions (22).

[0127] Initializing complex variables , Then, the nonlinear programming subproblems and the linear programming main problem obtained by decomposition are solved alternately and iteratively until convergence, and the optimal decision variables for the current optimization round can be obtained.

[0128] In each iterative solution process, the nonlinear programming subproblem is solved first, including: solving the nonlinear programming subproblem to obtain the values ​​of the variables to be solved; calculating the system average thermal efficiency and Benders cut within a set time interval based on the values ​​of the variables to be solved and the fixed complex variables; and updating the upper bound of the optimal solution of the thermal system operation optimization problem based on the system average thermal efficiency.

[0129] Specifically, complex variables Once fixed, the nonlinear programming subproblem is actually an upper bound problem of the original thermodynamic system operation optimization problem, and the solution to the nonlinear programming subproblem is a feasible solution to the original thermodynamic system operation optimization problem. In the nth iteration, let ,when When the variables are fixed, the difficulty of solving the nonlinear programming subproblems is greatly reduced. Differential equation solvers, such as ode15s on the MATLAB platform, can be used to obtain the values ​​of the remaining variables. Then and Substitute directly into (Equation 21) to calculate the value of the optimization objective, that is... The system's average thermal efficiency over the time period Next, the upper bound of the optimal solution of the thermal system operation optimization problem is updated according to the following equation (23), which is defined by equation (23), and the Benders cut is calculated at the current point according to the following equation (24).

[0130] (twenty three).

[0131] (twenty four).

[0132] in, These are introduced auxiliary variables. It is the Lagrangian function of a nonlinear programming subproblem. It is a Lagrange multiplier vector. express about The gradient is shown in equations (25)-(26).

[0133] (25).

[0134] (26).

[0135] Generally, in systems where the properties of the working fluid change drastically, the gradient value can be solved using numerical methods, as shown in equation (27).

[0136] (27).

[0137] In each iterative solution process, after solving the nonlinear programming subproblem, the main linear programming problem is constructed and solved, including: using Benders cut as a constraint and the introduced auxiliary variable as the optimization objective, constructing and solving the main linear programming problem to obtain the updated complex variable and the value of the auxiliary variable to be solved; updating the lower bound of the optimal solution of the thermal system operation optimization problem according to the value of the auxiliary variable to be solved; wherein, updating the complex variable is used to replace the fixed complex variable in the next iteration round.

[0138] Specifically, based on the framework of the GBD algorithm, the main problem of linear programming can be expressed as the following equation (28), and its optimization objective is the introduced auxiliary variable. The constraints are the set of Benders cuts obtained by solving each step of the nonlinear programming subproblem. The main linear programming problem is a relaxation problem of the original thermal system operation optimization problem. Therefore, the optimal solution of the main linear programming problem is always less than the optimal solution of the original thermal system operation optimization problem, i.e., the optimization objective is... It is the lower bound LB of the optimal solution of the original thermal system operation optimization problem. The main linear programming problem can be solved using solvers in existing computing software such as linprog, and the lower bound LB of the optimal solution can be updated by the following equation (29).

[0139] (28).

[0140] (29).

[0141] In equation (29), To optimize the objective.

[0142] The above describes a single iterative solution process. By repeatedly iterating between the nonlinear programming subproblem and the linear programming main problem, the optimal solution to the original thermodynamic system operation optimization problem converges, thus yielding the solution. The optimal decision variable for the thermal system within a given time period is also the optimal working fluid pump flow rate for the ORC power generation system.

[0143] For this type of non-convex dynamic optimization problem, the convergence condition is: 1) the upper and lower bounds of the original optimization problem are sufficiently close; 2) the optimization objective values ​​obtained from three consecutive iterations are sufficiently close. More specifically, in the optimization problem of a thermal system operation, the difference between the upper and lower bounds of the optimal solution is less than a first threshold. Furthermore, the difference between the average thermal efficiencies of adjacent systems calculated in a preset number of iterations (e.g., 3 times) is less than the second threshold. Under these conditions, the thermal system operation optimization problem converges, and the optimal decision variables for the current optimization round are obtained.

[0144] The convergence condition can be expressed as equation (30).

[0145] (30).

[0146] Among them, threshold and Can be set to .

[0147] In this embodiment, a dynamic numerical model of the thermal system is constructed, and based on this model, a thermal system operation optimization problem is established. This problem is then decomposed into a nonlinear programming subproblem and a linear programming master problem, which are iteratively solved alternately until convergence. This yields the optimal decision variables for the current optimization round, allowing the calculation of the optimal decision variables and other operating parameters of the thermal system under those variables. A rolling optimization approach is then used for the next optimization round. This method solves the thermal system operation optimization problem using a decomposition algorithm, improving computational efficiency while maintaining model accuracy. It obtains optimization results within the timescale of system operating conditions, achieving online optimization under time-varying boundary conditions, thereby improving the overall operational energy efficiency of the thermal system.

[0148] Based on the above embodiments, the following will further describe in detail the rolling optimization method adopted for online operation optimization of power systems.

[0149] It is easy to understand that for a real-world dynamic optimization problem of a thermal system, the dynamic optimization applicable to online guidance is not a single offline optimization, but a "rolling optimization" in which the sampling time also advances as the system continues to run.

[0150] Figure 6 A schematic diagram illustrating the basic principle of the rolling optimization method provided in an embodiment of the present invention is shown. Figure 6 As shown, in the first x In round-robin optimization, rolling optimization at a certain moment The optimization objective is to determine the performance index from this moment to a finite time interval (i.e., the "prediction time domain"). At each sampling moment, the values ​​of the decision variables are recalculated based on the prediction error of the current state, with a certain rolling step size. The process is continuously scrolled forward for online optimization. In this embodiment, a finite time interval is selected as the time length for optimization consideration. Rolling step size The relationship between the prediction time domain and the sampling time satisfies the following equation (31).

[0151] (31).

[0152] In the At any time, the system with The optimization results obtained in each round are used to adjust the system's operating variables, and this process continues. The optimization objective is to maximize the system's average thermal efficiency over a given time period until the optimization ends. Generally, in rolling optimization, the time consumed by each round of optimization calculations must be less than the prediction time domain and the rolling step size. The aforementioned GBD-based dynamic optimization algorithm has high computational efficiency, which can meet the computational efficiency requirements of rolling optimization.

[0153] Figure 7 A schematic flowchart of the online operation optimization method for combined cycle thermal systems based on rolling optimization, provided in an embodiment of the present invention, is shown. Figure 7 As shown, firstly, the length of the time interval considered for single-round rolling optimization is set. Rolling step size and optimize total duration At the beginning of each optimization round, the decision variable values ​​obtained in the previous round are substituted to initialize the system parameters; using the GBD algorithm, following the process described in the above embodiment, the single-round optimization problem is decomposed, and the calculations are performed. The optimal decision variable that yields the highest average thermal efficiency within a given time period is used to run the optimal decision variable and other operating parameters of the system under the optimal decision variable, and then the next round of optimization is performed until the set total optimization time is completed, thus achieving online optimization of the thermal system.

[0154] Corresponding to the online operation optimization method for combined thermal cycle systems based on decomposition algorithms described in the above embodiments, the present invention also provides an online operation optimization device for combined thermal cycle systems based on decomposition algorithms.

[0155] Specifically, Figure 8 A schematic diagram of the structure of the online operation optimization device for combined cycle of thermodynamic system based on decomposition algorithm provided in an embodiment of the present invention is shown.

[0156] like Figure 8As shown, the device includes: a thermal system dynamic numerical model construction module 810, used to construct a thermal system dynamic numerical model; a thermal system operation optimization problem construction module 820, used to construct a thermal system operation optimization problem based on the thermal system dynamic numerical model; a thermal system operation optimization problem decomposition module 830, used to decompose the thermal system operation optimization problem into a nonlinear programming subproblem and a linear programming master problem; a thermal system operation optimization problem solving module 840, used to alternately iteratively solve the nonlinear programming subproblem and the linear programming master problem until convergence, obtaining the optimal decision variable for the current optimization round; and an optimal decision variable and related parameter operation module 850, used to operate the optimal decision variable and other operating parameters of the thermal system under the optimal decision variable, and to perform the next round of optimization using a rolling optimization method.

[0157] In this embodiment, a dynamic numerical model of the thermal system is constructed by a dynamic numerical model construction module 810. Based on this model, a thermal system operation optimization problem construction module 820 constructs an operation optimization problem. A thermal system operation optimization problem decomposition module 830 then decomposes the problem into a nonlinear programming subproblem and a linear programming master problem. A thermal system operation optimization problem solving module 840 iteratively solves the nonlinear programming subproblem and the linear programming master problem until convergence, obtaining the optimal decision variables for the current optimization round. The optimal decision variables and related parameters operation module 850 then operates on the optimal decision variables and other operating parameters of the thermal system under those variables, and performs the next round of optimization using a rolling optimization approach. This device solves the optimal decision variables for the thermal system operation optimization problem through a decomposition algorithm, improving both model accuracy and computational efficiency. It obtains optimization results within the timescale of system operating conditions, achieving online optimization under time-varying boundary conditions, thereby improving the overall operational energy efficiency of the thermal system.

[0158] It should be noted that the online operation optimization device for combined thermal system operation based on decomposition algorithm provided in this embodiment of the invention can be referred to in correspondence with the online operation optimization method for combined thermal system operation based on decomposition algorithm described in the above embodiments, and will not be repeated here.

[0159] Figure 9 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 9As shown, the electronic device may include: a processor 910, a communication interface 920, a memory 930, and a communication bus 940, wherein the processor 910, the communication interface 920, and the memory 930 communicate with each other through the communication bus 940. The processor 910 can call logical instructions in the memory 930 to execute a thermal system joint loop online operation optimization method based on a decomposition algorithm. This method includes: constructing a dynamic numerical model of the thermal system; constructing a thermal system operation optimization problem based on the dynamic numerical model of the thermal system; decomposing the thermal system operation optimization problem into a nonlinear programming subproblem and a linear programming master problem; iteratively solving the nonlinear programming subproblem and the linear programming master problem alternately until convergence, obtaining the optimal decision variables for the current optimization round; running the optimal decision variables and other operating parameters of the thermal system under the optimal decision variables, and performing the next round of optimization using a rolling optimization method.

[0160] Furthermore, the logical instructions in the aforementioned memory 930 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0161] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the online operation optimization method for a thermal system based on a decomposition algorithm provided by the above methods. The method includes: constructing a dynamic numerical model of the thermal system; constructing a thermal system operation optimization problem based on the dynamic numerical model of the thermal system; decomposing the thermal system operation optimization problem into a nonlinear programming subproblem and a linear programming master problem; iteratively solving the nonlinear programming subproblem and the linear programming master problem alternately until convergence, obtaining the optimal decision variables for the current optimization round; running the optimal decision variables and other operating parameters of the thermal system under the optimal decision variables, and performing the next round of optimization using a rolling optimization method.

[0162] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program implements the online operation optimization method for a thermal system based on a decomposition algorithm provided by the above methods. The method includes: constructing a dynamic numerical model of the thermal system; constructing a thermal system operation optimization problem based on the dynamic numerical model of the thermal system; decomposing the thermal system operation optimization problem into a nonlinear programming subproblem and a linear programming master problem; iteratively solving the nonlinear programming subproblem and the linear programming master problem alternately until convergence, obtaining the optimal decision variables for the current optimization round; running the optimal decision variables and other operating parameters of the thermal system under the optimal decision variables, and performing the next round of optimization using a rolling optimization method.

[0163] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0164] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0165] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for online optimization of combined cycle operation of a thermodynamic system based on a decomposition algorithm, characterized in that, include: Constructing a dynamic numerical model of the thermal system; Based on the dynamic numerical model of the thermal system, a thermal system operation optimization problem is constructed. The optimization problem of the thermal system operation is decomposed into a nonlinear programming subproblem and a linear programming master problem; The nonlinear programming subproblem and the linear programming master problem are solved alternately and iteratively until convergence, thus obtaining the optimal decision variables for the current optimization round; The optimal decision variables and other operating parameters of the thermal system under the optimal decision variables are run, and the next round of optimization is carried out using a rolling optimization method.

2. The method for online operation optimization of combined cycle thermodynamic systems based on decomposition algorithm according to claim 1, characterized in that, The construction of the dynamic numerical model of the thermal system includes: A dynamic numerical model of the thermodynamic system is constructed based on the moving boundary method. The dynamic numerical model of the thermal system includes the moving boundary model of the evaporator, the moving boundary model of the condenser, the heat-work conversion process model of the expander, and the heat-work conversion process model of the working fluid pump.

3. The method for online operation optimization of combined cycle thermodynamic systems based on decomposition algorithm according to claim 1, characterized in that, The process of constructing an optimization problem for the operation of the thermal system based on the dynamic numerical model of the thermal system includes: The optimization objective is to maximize the average thermal efficiency of the system within a set time interval. The system working fluid pump flow rate is the decision variable. The dynamic numerical model of the thermal system and the range of values ​​of the decision variable are the constraints. The thermal system operation optimization problem is constructed. The thermal system operation optimization problem is a nonlinear programming problem.

4. The method for online operation optimization of combined cycle thermodynamic systems based on decomposition algorithm according to claim 1, characterized in that, The decomposition of the thermal system operation optimization problem into nonlinear programming subproblems and linear programming master problems includes: Based on the generalized Benders decomposition algorithm, by selecting and fixing some complex variables, the thermal system operation optimization problem is decomposed into the nonlinear programming subproblem and the linear programming main problem.

5. The method for online operation optimization of combined cycle thermodynamic systems based on decomposition algorithm according to claim 4, characterized in that, The alternating iterative solution of the nonlinear programming subproblem and the linear programming master problem until convergence yields the optimal decision variables for the current optimization round, including: Solve the nonlinear programming subproblem to obtain the values ​​of the variables to be solved; calculate the system average thermal efficiency and Benders cut within a set time interval based on the values ​​of the variables to be solved and the fixed complex variables; update the upper bound of the optimal solution of the thermal system operation optimization problem based on the system average thermal efficiency. Using the Benders cut as a constraint and the introduced auxiliary variable as the optimization objective, the linear programming master problem is solved to obtain the updated complex variable and the value of the auxiliary variable to be determined; based on the value of the auxiliary variable to be determined, the lower bound of the optimal solution of the thermal system operation optimization problem is updated; wherein, the updated complex variable is used to replace the fixed complex variable in the next iteration round.

6. The method for online operation optimization of combined cycle thermodynamic systems based on decomposition algorithm according to claim 5, characterized in that, The alternating iterative solution of the nonlinear programming subproblem and the linear programming master problem until convergence yields the optimal decision variables for the current optimization round, including: When the difference between the upper bound and the lower bound of the optimal solution to the thermal system operation optimization problem is less than a first threshold, and the difference between the average thermal efficiencies of adjacent systems calculated for a consecutive preset number of iterations is less than a second threshold, the thermal system operation optimization problem converges, and the optimal decision variables for the current optimization round are obtained.

7. A thermodynamic system combined cycle online operation optimization device based on decomposition algorithm, characterized in that, include: The module for constructing dynamic numerical models of thermal systems is used to build dynamic numerical models of thermal systems. A module for constructing a thermal system operation optimization problem is used to construct a thermal system operation optimization problem based on the dynamic numerical model of the thermal system. A thermal system operation optimization problem decomposition module is used to decompose the thermal system operation optimization problem into nonlinear programming sub-problems and linear programming master problems; The module for solving the thermal system operation optimization problem is used to alternately and iteratively solve the nonlinear programming subproblem and the linear programming main problem until convergence, and obtain the optimal decision variables for the current optimization round; The optimal decision variable and related parameter operation module is used to run the optimal decision variable and other operating parameters of the thermal system under the optimal decision variable, and to perform the next round of optimization in a rolling optimization manner.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the online operation optimization method for combined cycle operation of a thermodynamic system based on a decomposition algorithm as described in any one of claims 1 to 6.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the online operation optimization method for combined cycle operation of a thermodynamic system based on a decomposition algorithm as described in any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the online operation optimization method for combined cycle operation of a thermodynamic system based on a decomposition algorithm as described in any one of claims 1 to 6.

Images