A distributed intelligent predictive control method for flexible load variation of thermal power units

Abstract

The present invention relates to a distributed intelligent predictive control method for flexible load variation of a thermal power unit. This method addresses the nonlinear dynamic characteristics of the unit during load variation by constructing a nonlinear predictive control model using data-driven machine learning. To address the real-time nature and complexity of nonlinear predictive control, a distributed control strategy is employed to divide the unit into two parallel computational sub-control systems: the boiler and the turbine. Information exchange between the two sub-control systems is achieved using a Nash equilibrium game method, as well as global optimization decision-making for the entire thermal power unit. Compared to existing technologies, this method can reduce operating parameter fluctuations while ensuring the unit's tracking accuracy over a wide range of load variations, thereby achieving safe, economical, and stable operation of the thermal power unit.

Description

Technical Field

[0001] The present invention relates to the field of thermal automatic control, and in particular to a distributed intelligent predictive control method for a thermal power unit with flexible load variation. Background Art

[0002] The development trend of the energy and power industry is to build a clean, low-carbon, safe and efficient energy system and to construct a new power system with new energy as the main body. This puts higher requirements on the operating flexibility of thermal power units.

[0003] The boiler-turbine coordination system is one of the most critical systems in a thermal power plant. Its primary task is to regulate the unit's power output to meet the grid's load demands while maintaining various operating parameters (such as pressure and temperature) within reasonable ranges. On the one hand, the boiler-turbine coordination control system is characterized by multivariable, strong coupling, nonlinearity, and strict operational constraints. On the other hand, when a thermal power plant operates under a wide range of operating conditions, its operating parameters fluctuate significantly and can easily exceed their reasonable ranges, thus affecting the unit's stable operation. Summary of the Invention

[0004] The purpose of the present invention is to overcome the defects of the above-mentioned existing technologies and provide a distributed intelligent predictive control method for flexible variable load of thermal power units. This method can ensure the tracking accuracy of the unit's large-scale variable load while reducing the fluctuation of operating parameters, thereby realizing safe, economical and stable operation of the thermal power unit.

[0005] The purpose of the present invention can be achieved by the following technical solutions:

[0006] According to a first aspect of the present invention, a distributed intelligent predictive control method for flexible variable load of a thermal power unit is provided, the method comprising the following steps:

[0007] Step S1: Based on the control system mechanism of the thermal power unit, the system is decomposed into two parallel controlled boiler subsystems and steam turbine subsystems;

[0008] Step S2: A neural network is used to describe the nonlinear characteristics of the variable load operation of the thermal power unit. A distributed control strategy is adopted to construct nonlinear dynamic prediction models of the boiler subsystem and the turbine subsystem respectively, and a distributed predictive controller is used to perform predictive tracking control; wherein, the Nash equilibrium game method is used to control the information exchange between the two sub-control systems and the global optimization decision of the thermal power unit.

[0009] Preferably, the control parameters of the thermal power unit control system, boiler subsystem and steam turbine subsystem in step S1 are:

[0010] 1) Thermal power unit: The control variables are the valve openings for controlling the fuel flow, the valve openings for controlling the steam flow at the turbine inlet, and the valve openings for controlling the feedwater flow at the drum inlet; the controlled variables are the unit's generating load, boiler drum pressure, and drum water level deviation; the state variables are the unit's generating load, boiler drum pressure, and the density of the steam-water mixture in the boiler drum;

[0011] 2) Boiler subsystem: The controlled variables are the valve opening for controlling the fuel flow and the valve opening for controlling the feedwater flow at the drum inlet; the controlled variables are the boiler drum pressure and the drum water level deviation;

[0012] 3) Steam turbine subsystem: The controlled variable is the valve opening that controls the steam flow at the turbine inlet, and the controlled variable is the unit power generation load.

[0013] Preferably, in step S2, a neural network is used to describe the nonlinear characteristics of the variable load operation of the thermal power unit, specifically: a neural network is used to describe the nonlinear characteristics of the variable load operation of the thermal power unit, and the control variables and controlled variables at the previous moment are selected as the input of the neural network, and the output at the current moment is used as the output of the neural network;

[0014] Combined with the historical operating data of thermal power units, the error back propagation algorithm is used to train the neural network, and nonlinear dynamic prediction models of the boiler subsystem and the turbine subsystem are established respectively.

[0015] Preferably, the nonlinear dynamic prediction models of the boiler subsystem and the steam turbine subsystem are expressed as follows:

[0016]

[0017] Where Y 1,m is the prediction model output of the boiler subsystem, Y 1,m =[P m ,L m ] T , P m is the predicted value of boiler drum pressure, L m is the predicted value of drum water level deviation; Y 2,m is the prediction model output of the steam turbine subsystem, Y 2,m =E m , E m is the predicted value of the unit's power generation load; U1 is the control variable of the boiler subsystem, U1=[u1,u3] T , u1 is the valve opening that controls the fuel flow rate, u3 is the valve opening that controls the feedwater flow rate at the drum inlet; Y1 is the controlled variable of the boiler subsystem, Y1=[P,L] T, P is the boiler drum pressure, L is the drum water level deviation; U2 is the controlled variable of the turbine subsystem, U2=u2, u2 is the valve opening that controls the steam flow at the turbine inlet; Y2 is the controlled variable of the turbine subsystem, Y2=E, E is the unit power generation load.

[0018] Preferably, the use of a distributed predictive controller for predictive tracking control is specifically as follows: for the decomposed boiler subsystem and turbine subsystem, a distributed control strategy is adopted, and predictive controllers are designed respectively, so that the unit load, drum pressure and drum water level deviation of the thermal power unit track the set values.

[0019] Preferably, the objective function expression of the boiler subsystem optimization is:

[0020]

[0021] Where N p is the prediction time domain, N c To control the time domain, Q 1,i is the error weighting coefficient, R 1,j is the control weight matrix, U 1,max 、U 1,min are the upper and lower limits of the control variable U1, ΔU1 is the increment of the control variable U1, ΔU1=[Δu1,Δu3] T , ΔU 1,max , ΔU 1,min are the upper and lower limits of the control variable ΔU1; Y 1,p is the predicted output of the boiler subsystem, Y 1,p =[P p ,L p ] T , the expression is:

[0022] Y 1,p (k+i)=Y 1,m (k+i)+h1(Y1(k)-Y 1,m (k)) (3)

[0023] Where h1 is the correction coefficient, Y 1,m (k+i) is the multi-step prediction value of the boiler subsystem;

[0024] Y 1,r is the expected trajectory of the boiler subsystem, Y 1,r =[P r ,L r ] T , the expression is:

[0025]

[0026] Where α is the softening coefficient, R1 is the expected value of the boiler subsystem, R1 = [P set ,L set ] T , P set is the drum pressure setting value, L set It is the set value of the drum water level deviation.

[0027] Preferably, the objective function expression of the steam turbine subsystem optimization is:

[0028]

[0029] Where U 2,max 、U 2,min are the upper and lower limits of the control variable U2, ΔU2 is the increment of the control variable U2, ΔU2=Δu2, ΔU 2,max , ΔU 2,min are the upper and lower limits of the control variable ΔU2; Y 2,p is the predicted output of the steam turbine subsystem; Y 2,r is the expected trajectory of the boiler subsystem.

[0030] Preferably, in step S2, the information interaction between the two sub-control systems and the global optimization decision of the thermal power unit are performed using the Nash equilibrium game method, specifically:

[0031] Solve the optimization problem equations (2) and (5) and satisfy the iteration conditions to obtain the optimal solution To make global optimization decisions for thermal power units; the iterative condition expression is:

[0032]

[0033] Where l is the iteration coefficient; δ i is the iteration accuracy;

[0034] The calculation expressions of the control variables of the two sub-control systems are:

[0035] U(k)=U(k-1)+ΔU * (k) (7)

[0036] Where U(k)=[U1(k),U2(k)] T ,

[0037] According to a second aspect of the present invention, an electronic device is provided, comprising a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the program, any one of the methods described above is implemented.

[0038] According to a third aspect of the present invention, a computer-readable storage medium is provided, on which a computer program is stored, and when the program is executed by a processor, any one of the methods described above is implemented.

[0039] Compared with the prior art, the present invention has the following advantages:

[0040] The present invention adopts a data-driven neural network algorithm to establish a nonlinear predictive control model for the unit, which can better describe the nonlinear dynamic characteristics of the thermal power unit during the load change process, thereby meeting the tracking accuracy requirements of various operating parameters during the unit's load change operation. Taking into account the real-time and complexity of the nonlinear predictive control operation, a distributed control strategy is adopted to divide the thermal power unit into two parallel computing sub-control systems, namely the boiler and the turbine. Based on the Nash equilibrium game method, information interaction between the two sub-control systems and global optimization decision-making of the entire thermal power unit are realized. While ensuring the unit's large-scale load change tracking accuracy, the fluctuation of operating parameters is reduced, thereby achieving safe, economical and stable operation of the thermal power unit. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is the control system structure diagram;

[0042] Figure 2 The structural diagram of the prediction model for the boiler subsystem;

[0043] Figure 3 This is the structure diagram of the prediction model for the steam turbine subsystem;

[0044] Figure 4 This is the effect diagram of the prediction model;

[0045] Figure 5 is the error curve between the predicted value and the true value;

[0046] Figure 6 It is the output curve diagram of MPC and DMPC control;

[0047] Figure 7 Input curve diagram for MPC and DMPC control. DETAILED DESCRIPTION

[0048] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0049] Example

[0050] This embodiment provides a distributed intelligent predictive control method for flexible load variation of a thermal power unit, which includes the following steps:

[0051] Step S1: Figure 1 As shown in the figure, the Bell-Astrom oil-fired power plant model (hereinafter referred to as BA model) is selected as the controlled object of the distributed intelligent predictive control method. Its mechanism model is:

[0052]

[0053] Where, u1 is the valve opening for controlling the fuel flow, u2 is the valve opening for controlling the steam flow at the turbine inlet, u3 is the valve opening for controlling the feedwater flow at the drum inlet, E is the generating load of the unit, P is the boiler drum pressure, L is the drum water level deviation, and ρ f is the density of the steam-water mixture in the boiler drum, α s is a parameter that characterizes the steam quality, q e is the steam evaporation rate.

[0054] Control variables: valve opening to control fuel flow, valve opening to control steam flow at turbine inlet, valve opening to control feedwater flow at drum inlet;

[0055] Controlled variables: unit power generation load, boiler drum pressure, drum water level deviation;

[0056] State variables: unit power generation load, boiler drum pressure, and steam-water mixture density in the boiler drum.

[0057] Then, based on the system mechanism, the entire system is decomposed into a boiler subsystem and a steam turbine subsystem. The boiler subsystem's controlled variables are the valve openings for fuel flow and the valve openings for feedwater flow at the drum inlet, and the controlled variables are the boiler drum pressure and drum water level deviation. The turbine subsystem's controlled variables are the valve openings for steam flow at the turbine inlet, and the controlled variable is the unit's power generation load. Furthermore, the valve openings for steam flow at the turbine inlet and the boiler drum pressure serve as coupling variables between the boiler and turbine subsystems.

[0058] According to the BA model mechanism, the range of the model control variables u1, u2, and u3 (i.e., valve opening) is [0, 1], and the range of their change rates is:

[0059]

[0060] Step S2: A neural network is used to describe the nonlinear characteristics of the thermal power unit's large-scale variable load operation. Based on the mechanism analysis of the BA model, the control variables and controlled variables at the previous moment are selected as the input of the neural network, and the output at the current moment is selected as the output of the neural network. Combined with the unit's historical operating data, the error back propagation algorithm is used to train the neural network, and nonlinear dynamic prediction models for the boiler subsystem and the turbine subsystem are established respectively:

[0061]

[0062] Where Y 1,m is the prediction model output of the boiler subsystem, Y 1,m =[P m ,L m ] T , P m is the predicted value of boiler drum pressure, L m is the predicted value of drum water level deviation; Y 2,m is the prediction model output of the steam turbine subsystem, Y 2,m =E m , E m is the predicted value of the unit's power generation load; U1 is the control variable of the boiler subsystem, U1=[u1,u3] T ; Y1 is the controlled variable of the boiler subsystem, Y1=[P,L] T ; U2 is the controlled variable of the steam turbine subsystem, U2=u2; Y2 is the controlled variable of the steam turbine subsystem, Y2=E.

[0063] During neural network training, the neural network model structures of the boiler subsystem and the turbine subsystem are as follows: Figure 2 and Figure 3 As shown, the number of neurons in the hidden layer is 14. The data set has 10002 groups, 70% of which are selected as training sets and 30% as test sets. The training effect is shown in Figure 4 As shown, the training error is Figure 5 As shown in the figure, it can be seen that the model performance is good and meets the control requirements.

[0064] Step S3: For the decomposed boiler and steam turbine subsystems, a distributed control strategy is used to design predictive controllers, respectively, so that the unit load, drum pressure, and drum water level deviation of the thermal power unit can quickly track the set values. To ensure that the system output can track the set value and avoid drastic changes in the control action increment during the control process, the objective functions for the optimization of the boiler and steam turbine subsystems are shown in Equations (4) and (7), respectively.

[0065]

[0066] Where N pis the prediction time domain, N c To control the time domain, Q 1,i is the error weighting coefficient, R 1,j is the control weight matrix, U 1,max 、U 1,min are the upper and lower limits of the control variable U1, ΔU1 is the increment of the control variable U1, ΔU1=[Δu1,Δu3] T , ΔU 1,max , ΔU 1,min are the upper and lower limits of the control variable ΔU1 respectively. 1,p is the predicted output of the boiler subsystem, Y 1,p =[P p ,L p ] T , calculated by formula (5); Y 1,r is the expected trajectory of the boiler subsystem, Y 1,r =[P r ,L r ] T , calculated by formula (6).

[0067] Y 1,p (k+i)=Y 1,m (k+i)+h1(Y1(k)-Y 1,m (k)) (5)

[0068] Where h1 is the correction coefficient, Y 1,m (k+i) is the multi-step prediction value of the boiler subsystem.

[0069]

[0070] Where α is the softening coefficient, R1 is the expected value of the boiler subsystem, R1 = [P set ,L set ] T , P set is the drum pressure setting value, L set It is the steam drum digital deviation setting value.

[0071]

[0072] Where U 2,max 、U 2,min are the upper and lower limits of the control variable U2, ΔU2 is the increment of the control variable U2, ΔU2=Δu2, ΔU 2,max , ΔU 2,min are the upper and lower limits of the control variable ΔU2 respectively. 2,p is the predicted output of the steam turbine subsystem, Y 2,p =E p , the calculation method is similar to formula (6); Y 2,ris the expected trajectory of the boiler subsystem, Y 2,r =E r , the calculation method is similar to formula (6).

[0073] Step S4: Since there is input coupling between the boiler subsystem and the steam turbine subsystem, the output of each subsystem is related to the input of other subsystems. Therefore, each subsystem will also consider the effects of other subsystems when solving its own optimization problem.

[0074] At time k, the boiler subsystem and the turbine subsystem initialize the input variables and exchange data information through the distributed controller information interaction layer. If the iteration coefficient l = 1, the initial value of the optimal solution is obtained. Solve optimization problems (4) and (7) and obtain the optimal solution Then, determine whether the two sub-controllers meet the iteration conditions, as shown in formula (8). If the iteration conditions are met, the optimal solution is obtained. Otherwise, let l = l + 1, Solve optimization problems (4) and (7) again until the iteration conditions are met.

[0075]

[0076] Where, δ i Iteration precision.

[0077] Step S5: Calculate the control variable of the system, as shown in formula (9). Apply the control variable to the thermal power unit so that the output of the unit can quickly track the set value.

[0078] U(k)=U(k-1)+ΔU * (k) (9)

[0079] Where U(k)=[U1(k),U2(k)] T ,

[0080] Next, a set of simulation comparison experiments are conducted to verify the control effect of the distributed intelligent predictive controller of the present invention, as follows:

[0081] The unit is tested with variable load, assuming the unit is running at a load of 85.06MW and a drum pressure of 118.8kg / cm 3 , the drum water level deviation is 0m, at k=5, the unit load is changed to 105.8MW and the drum pressure is 129.6kg / cm 3 , the drum water level deviation is 0m. The predictive controller (MPC) and distributed predictive controller (DMPC) are designed for the unit. The parameters of MPC are set as follows: sampling time T s=3s, prediction time domain N p =20, control time domain N c =5, error weighting coefficient q = 1, control weighting coefficient r = 0.01. The parameters of DMPC are set as follows: sampling time T s =3s, prediction time domain N p =20, control time domain N c =5, error weighting coefficient q i =0.5, control weighting coefficient r i =0.01, iteration accuracy δ i =0.01, (i=1,2).

[0082] The simulation effect is as follows Figure 6 and Figure 7 As shown in the figure, it can be seen that DMPC can ensure the tracking accuracy of the unit over a wide range of load changes while reducing the fluctuation of operating parameters, thereby achieving safe, economical and stable operation of the thermal power unit.

[0083] The electronic device of the present invention includes a central processing unit (CPU), which can perform various appropriate actions and processes according to computer program instructions stored in a read-only memory (ROM) or loaded from a storage unit into a random access memory (RAM). In the RAM, various programs and data required for device operation can also be stored. The CPU, ROM, and RAM are connected to each other via a bus. An input / output (I / O) interface is also connected to the bus.

[0084] Many components in a device are connected to the I / O interface, including: input units, such as a keyboard and mouse; output units, such as various types of displays and speakers; storage units, such as magnetic disks and optical disks; and communication units, such as network cards, modems, and wireless communication transceivers. The communication unit allows the device to exchange information / data with other devices via computer networks such as the Internet and / or various telecommunication networks.

[0085] The processing unit performs the various methods and processes described above, such as methods S1 to S2. For example, in some embodiments, methods S1 to S2 can be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit. In some embodiments, part or all of the computer program can be loaded and / or installed on the device via a ROM and / or a communication unit. When the computer program is loaded into the RAM and executed by the CPU, one or more steps of methods S1 to S2 described above can be performed. Alternatively, in other embodiments, the CPU can be configured to execute methods S1 to S2 by any other appropriate means (for example, by means of firmware).

[0086] The functions described above herein may be performed, at least in part, by one or more hardware logic components. For example, and without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), and the like.

[0087] The program code for implementing the method of the present invention can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device so that when the program code is executed by the processor or controller, the functions / operations specified in the flow chart and / or block diagram are implemented. The program code can be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0088] In the context of the present invention, a machine-readable medium can be a tangible medium that can contain or store a program for use with an instruction execution system, device or equipment or used in combination with an instruction execution system, device or equipment. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared semiconductor system, device or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium can include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0089] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and such modifications or substitutions are intended to be within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be subject to the scope of protection of the claims.

Claims

1. A distributed intelligent predictive control method for flexible load variation of thermal power units, characterized in that: The method comprises the following steps: Step S1: Based on the control system mechanism of the thermal power unit, the system is decomposed into two parallel controlled boiler subsystems and steam turbine subsystems; Step S2: Using a neural network to describe the nonlinear characteristics of the variable load operation of the thermal power unit, adopting a distributed control strategy, constructing nonlinear dynamic prediction models for the boiler subsystem and the steam turbine subsystem respectively, and using a distributed predictive controller to perform predictive tracking control; wherein, a Nash equilibrium game method is used for information exchange between the two sub-control systems and for global optimization decision-making of the thermal power unit; The objective function expression of the boiler subsystem optimization is: Where N p is the prediction time domain, N c To control the time domain, Q 1,i is the error weighting coefficient, R 1,j is the control weight matrix, U 1,max 、U 1,min are the upper and lower limits of the control variable U1, ΔU1 is the increment of the control variable U1, ΔU1=[Δu1,Δu3] T , ΔU 1,max , ΔU 1,min are the upper and lower limits of the control variable ΔU1; Y 1,p is the predicted output of the boiler subsystem, Y 1,p =[P p ,L p ] T , the expression is: Y 1,p (k+i)=Y 1,m (k+i)+h1(Y1(k)-Y 1,m (k)) (3) Where h1 is the correction coefficient, Y 1,m (k+i) is the multi-step prediction value of the boiler subsystem; Y 1,r is the expected trajectory of the boiler subsystem, Y 1,r =[P r ,L r ] T , the expression is: Where α is the softening coefficient, R1 is the expected value of the boiler subsystem, R1 = [P set ,L set ] T , P set is the drum pressure setting value, L set It is the set value of the drum water level deviation.

2. A distributed intelligent predictive control method for flexible load variation of thermal power units according to claim 1, characterized in that: In step S1, the control parameters of the thermal power unit control system, boiler subsystem and steam turbine subsystem are: 1) Thermal power unit: The control variables are the valve openings for controlling the fuel flow, the valve openings for controlling the steam flow at the turbine inlet, and the valve openings for controlling the feedwater flow at the drum inlet; the controlled variables are the unit's generating load, boiler drum pressure, and drum water level deviation; the state variables are the unit's generating load, boiler drum pressure, and the density of the steam-water mixture in the boiler drum; 2) Boiler subsystem: The controlled variables are the valve opening for controlling the fuel flow and the valve opening for controlling the feedwater flow at the drum inlet; The controlled variables are the boiler drum pressure and drum water level deviation; 3) Steam turbine subsystem: The controlled variable is the valve opening that controls the steam flow at the turbine inlet, and the controlled variable is the unit power generation load.

3. A distributed intelligent predictive control method for flexible load variation of thermal power units according to claim 2, characterized in that: In step S2, a neural network is used to describe the nonlinear characteristics of the variable load operation of the thermal power unit, specifically: a neural network is used to describe the nonlinear characteristics of the variable load operation of the thermal power unit, and the control variable and the controlled variable at the previous moment are selected as the input of the neural network, and the output at the current moment is used as the output of the neural network; Combined with the historical operating data of thermal power units, the error back propagation algorithm is used to train the neural network, and nonlinear dynamic prediction models of the boiler subsystem and the turbine subsystem are established respectively.

4. A distributed intelligent predictive control method for flexible load variation of thermal power units according to claim 3, characterized in that: The nonlinear dynamic prediction models of the boiler subsystem and the steam turbine subsystem are expressed as follows: Where Y 1,m is the prediction model output of the boiler subsystem, Y 1,m =[P m ,L m ] T , P m is the predicted value of boiler drum pressure, L m is the predicted value of drum water level deviation; Y 2,m is the prediction model output of the steam turbine subsystem, Y 2,m =E m , E m is the predicted value of the unit's power generation load; U1 is the control variable of the boiler subsystem, U1=[u1,u3] T , u1 is the valve opening that controls the fuel flow rate, u3 is the valve opening that controls the feedwater flow rate at the drum inlet; Y1 is the controlled variable of the boiler subsystem, Y1=[P,L] T , P is the boiler drum pressure, L is the drum water level deviation; U2 is the controlled variable of the turbine subsystem, U2=u2, u2 is the valve opening that controls the steam flow at the turbine inlet; Y2 is the controlled variable of the turbine subsystem, Y2=E, E is the unit power generation load.

5. A distributed intelligent predictive control method for flexible load variation of thermal power units according to claim 4, characterized in that: The use of a distributed predictive controller for predictive tracking control is specifically as follows: for the decomposed boiler subsystem and turbine subsystem, a distributed control strategy is adopted, and predictive controllers are designed respectively to achieve tracking control of the unit load, drum pressure and drum water level deviation of the thermal power unit to the set value.

6. A distributed intelligent predictive control method for flexible load variation of thermal power units according to claim 1, characterized in that: The objective function expression of the steam turbine subsystem optimization is: Where U 2,max 、U 2,min are the upper and lower limits of the control variable U2, ΔU2 is the increment of the control variable U2, ΔU2=Δu2, ΔU 2,max , ΔU 2,min are the upper and lower limits of the control variable ΔU2; Y 2,p is the predicted output of the steam turbine subsystem; Y 2,r is the expected trajectory of the boiler subsystem.

7. A distributed intelligent predictive control method for flexible load variation of thermal power units according to claim 6, characterized in that: In step S2, the Nash equilibrium game method is used to implement information interaction between the two sub-control systems and the global optimization decision of the thermal power unit, specifically: Solve the optimization problem equations (2) and (5) and satisfy the iteration conditions to obtain the optimal solution To make global optimization decisions for thermal power units; the iterative condition expression is: Where l is the iteration coefficient; δ i is the iteration accuracy; The calculation expressions of the control variables of the two sub-control systems are: U(k)=U(k-1)+ΔU * (k) (7) Where U(k)=[U1(k),U2(k)] T , 8. An electronic device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor executes the program, the method according to any one of claims 1 to 7 is implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images