Design method for dynamic energy efficiency optimization control in dynamic load changing process of coal-fired unit

Abstract

The invention discloses a design method for dynamic energy efficiency optimization control in a dynamic load changing process of a coal-fired unit, which belongs to the field of thermal control of a thermal power plant and comprises the following steps of: S1, establishing a unit nonlinear dynamic model based on a mass and energy balance law according to a typical coal-fired unit; s2, based on a multi-target genetic algorithm, performing parameter identification by taking minimization of an integral absolute error as a target to obtain parameters of the nonlinear dynamic model of the coal-fired unit, and based on stable operation point linearization, obtaining a linear state space model of the coal-fired unit; s3, designing an extended state Kalman observer, and estimating and compensating the state of the coal-fired unit model under the conditions of model uncertainty, external disturbance and noise; and S4, providing an optimization objective of comprehensively considering the tracking performance and the dynamic # imgabs0 # loss, and finishing the design of the control system in combination with an extended state Kalman observer. According to the invention, the anti-interference capability of the control system under unknown or sudden disturbance can be enhanced, and the coordination level control performance of the unit can be improved.

Description

Technical Field

[0001] The invention belongs to the field of thermal control of thermal power plants, and in particular relates to a design method for dynamic energy efficiency optimization control in a dynamic variable load process of a coal-fired unit. Background Art

[0002] The increasing installed capacity of new energy sources urgently requires coal-fired power generation units to have stronger flexible load response capabilities. The control system of coal-fired units plays an important role in production safety and economic operation. In the process of rapid load change, coal-fired units face the challenges of difficult coordination of adjustment actions, reduced dynamic efficiency and slow response speed. On the one hand, the traditional coal-fired unit control system focuses on stable and efficient operation in a specific load range when designed. However, it is difficult to effectively deal with various external interferences and load change rate uncertainties when facing wide-range and high-frequency load fluctuations, which makes the control performance deviate from expectations and restricts the rapidity and stability of its load change process. On the other hand, with the frequent adjustment of the output state of coal-fired units, the thermal system of the unit has been in a non-steady-state process of variable load, affecting the safe and economic operation of the unit and resulting in a decrease in the energy utilization efficiency of the system during dynamic operation. The adaptability and robustness of model predictive control can be effectively improved by online estimation and compensation of various uncertainties. At the same time, advanced dynamic energy efficiency optimization control technology can finely control and optimize the key parameters of coal-fired units in the process of variable load in real time, so as to improve the flexibility and efficiency of coal-fired units. Summary of the invention

[0003] In order to overcome the shortcomings of the above-mentioned technologies, the present invention aims to explore the coordination level control strategy of the unit in the dynamic load change process, targeting the various uncertainties and dynamic energy efficiency reduction problems faced by the control system of the coal-fired unit in the process of rapid load change, combined with the typical problems in the actual operation process of the power plant. The purpose of the present invention is to provide a dynamic energy efficiency optimization control method for uncertainty compensation in the dynamic load change process of a coal-fired unit, improve the load change response speed and stability of the unit, and thus improve the unit's operating flexibility and energy utilization efficiency.

[0004] In order to achieve the above object, the present invention adopts the following technical solution:

[0005] The design method of dynamic energy efficiency optimization control of coal-fired units in the dynamic variable load process includes the following steps:

[0006] Steps: S1. According to typical coal-fired units, a nonlinear model of the unit based on the mass and energy balance law is established;

[0007] S2. Based on the multi-objective genetic algorithm, the parameters of the nonlinear dynamic model of the coal-fired unit are obtained by parameter identification with the goal of minimizing the integral absolute error, and the linear state space model of the coal-fired unit is obtained based on the linearization of the stable operating point;

[0008] S3. Design an extended state Kalman observer to estimate and compensate for the model state of the coal-fired unit under model uncertainty, external disturbances and noise conditions;

[0009] S4. Comprehensive consideration of tracking performance and dynamics The optimization goal of the loss is combined with the extended state Kalman observer to complete the design of the uncertainty compensation dynamic energy efficiency optimization control system in the dynamic load variable process of the coal-fired unit.

[0010] Furthermore, the typical coal-fired unit described in step S1 includes a boiler system and a steam turbine system. The boiler system includes blowing coal into the furnace through a pulverizer, burning coal powder to heat steam, and the steam passes through a superheater to become high-temperature and high-pressure product steam that is output to the steam turbine system.

[0011] Furthermore, based on the mass and energy balance law, the complete product steam generation process of a typical coal-fired unit is calculated as follows:

[0012]

[0013] Where: q f is the mass flow rate of pulverized coal; c0 is the coefficient to be identified; u B is the boiler coal supply instruction; t is a certain moment; τ is the time lag constant; D b is the steam generation flow; c5 is the coefficient to be identified; k4 is the gain coefficient; D b The steam generation flow rate; D i is the working medium flow rate at the drum inlet; c6 and c7 are coefficients to be identified; p b is the drum pressure; p T is the main steam pressure; k5 is the gain coefficient; D T is the main steam flow; k2 is the gain coefficient; p1 is the regulating stage pressure;

[0014] The key component to achieve dynamic variable load response is the regulating stage in the steam turbine system. The main steam flow is controlled by adjusting the opening of the regulating mechanism. The steam turbine system converts the thermal energy of steam into mechanical work output. The calculation formula for the regulation and power generation process is:

[0015]

[0016] N e =k3D T ;

[0017] Where: p1 is the regulating stage pressure; c1 is the coefficient to be identified; k1 is the gain coefficient; μ T is the main steam valve opening; p T Main steam pressure; N eis the output power; k3 is the gain coefficient; D T Main steam flow.

[0018] Furthermore, the five static gain coefficients k1 to k5 in step S1 are obtained by fitting the actual on-site steady-state data of the power plant, where k1 and k2 are 0.0083 and 74.7, and the expressions of k3, k4, and k5 are as follows:

[0019]

[0020] In the formula: DT is the main steam flow rate; qf is the coal powder mass flow rate; pb is the drum pressure.

[0021] Furthermore, a multi-objective genetic algorithm is used to perform system identification on the six parameters to be identified in step S1: c0-c7, and the time lag constant τ, and the optimization goal is the minimum integral absolute error:

[0022]

[0023] Where: N is the number of samples; Δt is the sampling time; y i (k) is the model output; y' i (k) is the measured value of the output.

[0024] The multi-objective genetic algorithm NSGA-Ⅱ is used for parameter identification. The relevant parameter population size of NSGA-Ⅱ is 100, the crossover probability is 80%, the mutation probability is 5%, and the Pareto score is 35%.

[0025] Furthermore, the system input u is defined as the coal feed quantity instruction u1 and the main steam valve opening u2, and the main process variables of the system are defined as the state variable x, including: coal powder mass flow x1, steam generation flow x2, drum pressure x3, main steam pressure x4, regulating stage pressure x5 and main steam flow x6. The system output y is defined as the load y1 and the main steam pressure y2. According to the formula of step S1 and the parameter identification result of S2, the complete energy conversion process of the coal-fired unit is described by the following nonlinear differential equations and system output as follows:

[0026]

[0027] The nonlinear system is linearized by Taylor expansion, and the nonlinear mechanism model is linearized at rated working conditions u1 = 112 t / h, u2 = 86.2% to obtain a continuous linear state space model:

[0028]

[0029] Where: x(t) is the state variable; u(t) is the input; y(t) is the output; A m , Bm , C m is the coefficient matrix.

[0030] Furthermore, in step S3, the discrete state space equation considering various uncertainties and disturbances is described as:

[0031]

[0032] Where: x(k) is the state variable; u(k) is the input; y(k) is the output; w(k) is the sum of model uncertainty, external disturbance and noise, referred to as total disturbance; A m , B m , B e , C m is the coefficient matrix;

[0033] Considering various uncertainties and improving estimation accuracy, the total disturbance is expanded into a new state quantity, and the expanded state equation is obtained:

[0034]

[0035] Where: is the expansion state variable; AKF, BKF, CKF are the expansion system coefficient matrices;

[0036] According to the difference between the actual output and the estimated output, the extended state Kalman observation equation and various related calculation formulas are obtained to update the state estimate and covariance matrix. The formula is as follows:

[0037]

[0038] Where: is the estimated value of the expansion state; y(k) is the actual output; is the output estimate; K k is the Kalman gain; P k are the covariance matrices of the prior and posterior estimation errors respectively; Q k , R k are the covariance matrices of process noise and measurement noise, respectively.

[0039] Furthermore, the thermal-electric conversion process of coal-fired units will cause a decrease in energy quality, i.e. loss, defining the dynamics of the unit during load changes The loss is the input of coal Output of power generation The difference is expressed as follows:

[0040] V(k)=W u u(k)-W y y(k)

[0041] Where: V(k) represents the Loss; Wu, W y are coefficient matrices respectively; u(k) is input; y(k) is output,

[0042] Based on unit dynamics The loss calculation formula comprehensively considers the set value tracking effect and dynamics Loss, establish the optimization objective function of the coal-fired unit control system:

[0043]

[0044] Where: J is the optimization objective function; y(k) is the output at time k; r(k) is the output setting value at time k; Δu(k) is the input increment at time k; V(k) is the value at time k. Loss; S q , Sr, Sv are the error weight matrix, control weight matrix and Loss weight matrix;

[0045] Based on the real-time estimated values ​​of the extended state Kalman observer, a predictive control strategy for the coordinated economic model of the coal-fired unit is designed, including comprehensive consideration of the set value tracking effect and dynamic The objective function and related constraints of the loss optimization problem are:

[0046]

[0047] u(k)∈Ω1,k=1,…,N c

[0048] Δu(k)∈Ω2,k=1,…,N c ;

[0049] Where: J is the optimization objective function, Js and Je are the set value tracking target and dynamic loss target; Np is the prediction time domain; Nc is the control time domain; r(k) is the output setting value at time k; Δu(k) is the input increment at time k; V(k) represents the Loss; S q , S r , S v are the error weight matrix, control weight matrix and Loss weight matrix; x(k), u(k), w(k) are the state, input and total disturbance at time k respectively; S(Δ) is the value range of the input; is the estimated value of the expansion state at time k+1; is the output estimate at time k; f d 、h dare the functional relationships between the corresponding calculation state estimation and output estimation respectively; Ω1 and Ω2 represent the constraints of input quantity and input quantity increment.

[0050] A control system designed by the design method of the dynamic energy efficiency optimization system of the dynamic variable load process of the coal-fired unit;

[0051] The invention comprises a coal-fired unit, an expanded state observer and a dynamic energy efficiency model predictive controller. The coal-fired unit feeds back input power and main steam pressure to the expanded state observer. The expanded state observer outputs an estimated value and the expanded state estimated value to the dynamic energy efficiency model predictive controller. The dynamic energy efficiency model predictive controller outputs the control of coal feed amount and main steam valve opening to the coal-fired unit.

[0052] A computer-readable storage medium stores a computer program, which, when executed by a processor, is used to implement a design method for a dynamic energy efficiency optimization system for a dynamic load variation process of a coal-fired unit.

[0053] Compared with the prior art, the present invention has the following beneficial effects:

[0054] (1) Considering various uncertainties and disturbances in the rapid load change process of coal-fired units, an uncertainty compensation mechanism is introduced, which can effectively improve the estimation accuracy of unit state variables, enhance the anti-interference ability of the control system under unknown or sudden disturbances, and improve the coordination level control performance of the unit;

[0055] (2) In order to address the problem of decreased dynamic energy utilization efficiency during the load-changing process of coal-fired units, a coordinated model predictive control for dynamic energy efficiency optimization is implemented to improve the operating efficiency and economic benefits of the units and reduce energy waste. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 Schematic diagram of a typical coal-fired unit and control variables provided by the present invention.

[0057] Figure 2 This is a structural diagram of the uncertainty compensation dynamic energy efficiency optimization control provided by the present invention. DETAILED DESCRIPTION

[0058] A method for dynamic energy efficiency optimization control of a coal-fired unit during a dynamic load change process with uncertainty compensation comprises the following steps:

[0059] S1: Based on typical coal-fired units, a nonlinear dynamic model of the unit is established based on the mass and energy balance law;

[0060] Typical coal-fired units used in the present invention can be referenced Figure 2As shown, the coal-fired unit is divided into a boiler system and a steam turbine system. The boiler system includes blowing coal into the furnace through a coal mill and burning coal powder to heat steam. The steam passes through a superheater to become high-temperature and high-pressure product steam and is output to the steam turbine system. Based on the law of mass and energy balance, the relevant calculation formula for the complete product steam generation process is as follows:

[0061]

[0062] Where: q f is the mass flow rate of pulverized coal; c0 is the coefficient to be identified; u B is the boiler coal supply instruction; t is a certain moment; τ is the time lag constant; D b is the steam generation flow; c5 is the coefficient to be identified; k4 is the gain coefficient; D b The steam generation flow rate; D i is the working medium flow rate at the drum inlet; c6 and c7 are coefficients to be identified; p b is the drum pressure; p T is the main steam pressure; k5 is the gain coefficient; D T is the main steam flow; k2 is the gain coefficient; p1 is the regulating stage pressure.

[0063] The key component to achieve dynamic variable load response is the regulating stage in the steam turbine system, which controls the main steam flow rate by adjusting the opening of the regulating mechanism. The steam turbine system converts the thermal energy of steam into mechanical work output. The calculation formula for the regulation and power generation process is:

[0064]

[0065] N e =k3D T

[0066] Where: p1 is the regulating stage pressure; c1 is the coefficient to be identified; k1 is the gain coefficient; μ T is the main steam valve opening; p T Main steam pressure; N e is the output power; k3 is the gain coefficient; D T Main steam flow.

[0067] S2: Based on the multi-objective genetic algorithm (NSGA-Ⅱ), the parameters of the nonlinear dynamic model of the coal-fired unit are obtained by parameter identification with the goal of minimizing the integral absolute error (IEA); the linear state space model of the coal-fired unit is obtained based on the linearization of the stable operating point;

[0068] The five static gain coefficients k1 to k5 in step S1 are obtained by fitting the on-site steady-state data of a power plant. Among them, k1 and k2 are insensitive to changes in operating conditions and are constants of 0.0083 and 74.7 respectively. However, k3, k4, and k5 change significantly when the operating conditions change. By checking the physical meaning, it is found that the gain is closely related to the specific state variable. The expression is as follows:

[0069]

[0070] Where: D T is the main steam flow rate; q f is the mass flow rate of pulverized coal; p b is the steam drum pressure.

[0071] Based on the operation data of a period of time on site, a multi-objective genetic algorithm is used to identify the six parameters to be identified (c0~c7) and the hysteresis constant (τ) in step S1. The optimization goal is the minimum integral absolute error:

[0072]

[0073] Where: N is the number of samples; Δt is the sampling time; y i (k) is the model output; y' i (k) is the measured value of the output.

[0074] The multi-objective genetic algorithm (NSGA-Ⅱ) is used for parameter identification. The relevant parameter population size of NSGA-Ⅱ is 100, the crossover probability is 80%, the mutation probability is 5%, and the Pareto score is 35%. The optimized identification parameters are shown in Table 1.

[0075] Table 1. Parameter identification results

[0076] parameter <![CDATA[c0]]> <![CDATA[c1]]> <![CDATA[c2]]> <![CDATA[c5]]> <![CDATA[c6]]> <![CDATA[c7]]> τ Numeric 22 5 5 380 4057 5101 43

[0077] Through the above mechanism modeling and optimization solution, the dynamic model and related parameters of the coal-fired unit based on energy and mass balance are obtained. In order to design the control strategy, the system input u is further defined as the coal feed quantity instruction (u1) and the main steam valve opening (u2). The main process variables of the system are defined as the state variable x, including: coal powder mass flow rate (x1), steam generation flow rate (x2), drum pressure (x3), main steam pressure (x4), regulating stage pressure (x5) and main steam flow rate (x6). The system output y is defined as load (y1) and main steam pressure (y2). According to the formula of step S1 and the parameter identification results of S2, the complete energy conversion process of the coal-fired unit is described by the following nonlinear differential equations and system output as follows:

[0078]

[0079] The nonlinear system is linearized by Taylor expansion. The nonlinear mechanism model is linearized at rated working condition u1 = 112 t / h, u2 = 86.2% to obtain a continuous linear state space model:

[0080]

[0081] Where: x(t) is the state variable; u(t) is the input; y(t) is the output; A m , B m , C m is the coefficient matrix.

[0082] S3: Design an extended state Kalman observer to achieve model state estimation and compensation of coal-fired units under model uncertainty, external disturbances and noise conditions;

[0083] The actual coal-fired unit has complex dynamic characteristics, and it is difficult to establish an accurate state space model. In addition, it faces various external uncertainties and disturbances during operation, which reduces the accuracy of state parameter estimation. The discrete state space equation considering various uncertainties and disturbances is described as:

[0084]

[0085] Where: x(k) is the state variable; u(k) is the input; y(k) is the output; w(k) is the sum of model uncertainty, external disturbance and noise, referred to as total disturbance; A m , B m , B e , C m is the coefficient matrix.

[0086] Further considering various uncertainties and improving estimation accuracy, the total disturbance is expanded into a new state quantity, and the expanded state equation is obtained:

[0087]

[0088] Where: is the expansion state variable; A KF , B KF , C KF is the coefficient matrix of the expansion system.

[0089] According to the difference between the actual output and the estimated output, the extended state Kalman observation equation and various related calculation formulas are obtained to update the state estimate and covariance matrix. The formula is as follows:

[0090]

[0091] Where: is the estimated value of the expansion state; y(k) is the actual output; is the output estimate; K k is the Kalman gain; P k are the covariance matrices of the prior and posterior estimation errors respectively; Q k , R k are the covariance matrices of process noise and measurement noise, respectively.

[0092] S4: Propose a comprehensive consideration of tracking performance and dynamics Based on the optimization goal of loss, a predictive control strategy for the coordinated economic model of coal-fired units is designed; combined with the expanded state Kalman observer, the design of the uncertainty compensation dynamic energy efficiency optimization control system in the dynamic variable load process of the coal-fired unit is completed.

[0093] The thermal-electric conversion process of coal-fired units will cause a decrease in energy quality, that is, Loss. Define the dynamics of the unit during load change. The loss is the input of coal Output of power generation The difference is expressed as follows:

[0094] V(k)=W u u(k)-W y y(k)

[0095] Where: V(k) represents the Loss; W u , W y are coefficient matrices respectively; u(k) is input; y(k) is output.

[0096] Based on unit dynamics The loss calculation formula comprehensively considers the set value tracking effect and dynamics Loss, establish the optimization objective function of the coal-fired unit control system:

[0097]

[0098] Where: J is the optimization objective function; y(k) is the output at time k; r(k) is the output setting value at time k; Δu(k) is the input increment at time k; V(k) is the value at time k. Loss; S q , S r , S v are the error weight matrix, control weight matrix and Loss weight matrix.

[0099] Based on the real-time estimated values ​​of the extended state Kalman observer, a predictive control strategy for the coordinated economic model of the coal-fired unit is designed, including comprehensive consideration of the set value tracking effect and dynamic The objective function and related constraints of the loss optimization problem are:

[0100]

[0101] u(k)∈Ω1,k=1,…,N c

[0102] Δu(k)∈Ω2,k=1,…,N c

[0103] Where: J is the optimization objective function, where J s , J e Set the value tracking target and dynamic Damage target; N p is the prediction time domain; N c is the control time domain; r(k) is the output setting value at time k; Δu(k) is the input increment at time k; V(k) is the value at time k. Loss; S q , S r , S v are the error weight matrix, control weight matrix and Loss weight matrix; x(k), u(k), w(k) are the state, input and total disturbance at time k respectively; S(Δ) is the value range of the input; is the estimated value of the expansion state at time k+1; is the output estimate at time k; f d 、h d are the functional relationships between the corresponding calculation state estimation and output estimation respectively; Ω1 and Ω2 represent the constraints of input quantity and input quantity increment.

[0104] In summary, a dynamic energy efficiency optimization control method with uncertainty compensation in the dynamic load change process of coal-fired units takes into account various uncertainties and disturbances in the rapid load change process of coal-fired units, introduces an uncertainty compensation mechanism, effectively improves the estimation accuracy of the unit state variables, and enhances the anti-interference ability of the control system under unknown or sudden disturbances. Aiming at the problem of decreased dynamic energy utilization efficiency in the load change process of coal-fired units, a coordinated level model predictive control is used to realize dynamic energy efficiency optimization by combining accurate estimation values, which comprehensively improves the energy utilization efficiency and energy-saving benefits of the units.

[0105] Reference Figure 2As shown, the present invention also provides a control system, which is designed by the design method of the dynamic energy efficiency optimization system of the dynamic variable load process of the coal-fired unit mentioned in the claim; it includes a coal-fired unit, an expanded state observer and a dynamic energy efficiency model prediction controller, the coal-fired unit feeds back input power and main steam pressure to the expanded state observer, the expanded state observer outputs an estimated value and the expanded state estimated value to the dynamic energy efficiency model prediction controller, and the dynamic energy efficiency model prediction controller outputs a control of the coal feed amount and the main steam valve opening to the coal-fired unit.

[0106] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, is used to implement a design method for a dynamic energy efficiency optimization system for a dynamic variable load process of a coal-fired unit.

[0107] The design method of the dynamic energy efficiency optimization system of the dynamic variable load process of the coal-fired unit can be stored in a computer-readable storage medium if it is implemented in the form of a software functional unit and sold or used as an independent product.

[0108] Based on this understanding, the present invention implements all or part of the processes in the design method of the dynamic energy efficiency optimization system of the dynamic variable load process of the above-mentioned coal-fired unit, and can also be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the vehicle security access level control method based on the public key certificate can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or preset intermediate form, etc.

[0109] Computer-readable storage media may include: any entity or device that can carry computer program code, recording media, USB flash drives, mobile hard disks, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.

[0110] It should be noted that the content contained in computer-readable storage media can be appropriately increased or decreased according to the requirements of legislation and patent practices in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practices, computer-readable storage media do not include electrical carrier signals and telecommunication signals.

[0111] The above embodiment is only one of the implementation methods that can realize the technical solution of the present invention. The scope of protection claimed by the present invention is not limited only to this embodiment, but also includes changes, replacements and other implementation methods that can be easily thought of by any technician familiar with the technical field within the technical scope disclosed by the present invention.

Claims

1. A design method for a dynamic energy efficiency optimization system for a coal-fired unit during a dynamic load change process, characterized in that: The steps include: S1. According to typical coal-fired units, a nonlinear dynamic model of the unit based on the mass and energy balance law is established; S2. Based on the multi-objective genetic algorithm, the parameters of the nonlinear dynamic model of the coal-fired unit are obtained by parameter identification with the goal of minimizing the integral absolute error, and the linear state space model of the coal-fired unit is obtained based on the linearization of the stable operating point; S3. Design an extended state Kalman observer to estimate and compensate for the model state of the coal-fired unit under model uncertainty, external disturbances and noise conditions; S4. Comprehensive consideration of tracking performance and dynamics The optimization goal of the loss is combined with the extended state Kalman observer to complete the design of the uncertainty compensation dynamic energy efficiency optimization control system in the dynamic load variable process of the coal-fired unit.

2. The design method of the dynamic energy efficiency optimization system of the dynamic variable load process of the coal-fired unit according to claim 1 is characterized in that: The typical coal-fired unit described in step S1 includes a boiler system and a steam turbine system. The boiler system includes blowing coal into the furnace through a coal mill, burning coal powder to heat steam, and the steam passes through a superheater to become high-temperature and high-pressure product steam that is output to the steam turbine system.

3. The design method of the dynamic energy efficiency optimization system of the dynamic variable load process of the coal-fired unit according to claim 2 is characterized in that: Based on the mass and energy balance law, the complete product steam generation process of a typical coal-fired unit is calculated as follows: Where: q f is the mass flow rate of pulverized coal; c0 is the coefficient to be identified; u B is the boiler coal supply instruction; t is a certain moment; τ is the time lag constant; D b is the steam generation flow; c5 is the coefficient to be identified; k4 is the gain coefficient; D b The steam generation flow rate; D i is the working medium flow rate at the drum inlet; c6 and c7 are coefficients to be identified; p b is the drum pressure; p T is the main steam pressure; k5 is the gain coefficient; D T is the main steam flow; k2 is the gain coefficient; p1 is the regulating stage pressure; The key component to achieve dynamic variable load response is the regulating stage in the steam turbine system. The main steam flow is controlled by adjusting the opening of the regulating mechanism. The steam turbine system converts the thermal energy of steam into mechanical work output. The calculation formula for the regulation and power generation process is: N e =k3D T ; Where: p1 is the regulating stage pressure; c1 is the coefficient to be identified; k1 is the gain coefficient; μ T is the main steam valve opening; p T Main steam pressure; N e is the output power; k3 is the gain coefficient; D T Main steam flow.

4. The design method of the dynamic energy efficiency optimization system of the dynamic variable load process of the coal-fired unit according to claim 3 is characterized in that: The five static gain coefficients k1 to k5 in step S1 are obtained by fitting the actual on-site steady-state data of the power plant, where k1 and k2 are 0.0083 and 74.7, and the expressions of k3, k4, and k5 are as follows: In the formula: DT is the main steam flow rate; qf is the coal powder mass flow rate; pb is the drum pressure.

5. The design method of the dynamic energy efficiency optimization system of the dynamic variable load process of the coal-fired unit according to claim 4 is characterized in that: A multi-objective genetic algorithm is used to perform system identification on the six parameters to be identified in step S1: c0-c7 and the time lag constant τ, and the optimization goal is the minimum integral absolute error: Where: N is the number of samples; Δt is the sampling time; y i (k) is the model output; y' i (k) is the measured value of the output. The multi-objective genetic algorithm NSGA-Ⅱ is used for parameter identification. The relevant parameter population size of NSGA-Ⅱ is 100, the crossover probability is 80%, the mutation probability is 5%, and the Pareto score is 35%.

6. The design method of the dynamic energy efficiency optimization system of the dynamic variable load process of the coal-fired unit according to claim 5 is characterized in that: Define the system input u as the coal feed quantity instruction u1 and the main steam valve opening u2, define the main process variables of the system as the state variable x, including: coal powder mass flow x1, steam generation flow x2, drum pressure x3, main steam pressure x4, regulating stage pressure x5 and main steam flow x6, define the system output y as load y1 and main steam pressure y2, according to the formula of step S1 and the parameter identification result of S2, the complete energy conversion process of the coal-fired unit is described by the following nonlinear differential equations and system output as follows: The nonlinear system is linearized by Taylor expansion, and the nonlinear mechanism model is linearized at rated working conditions u1 = 112 t / h, u2 = 86.2% to obtain a continuous linear state space model: Where: x(t) is the state variable; u(t) is the input; y(t) is the output; A m , B m , C m is the coefficient matrix.

7. The design method of the dynamic energy efficiency optimization system of the dynamic variable load process of the coal-fired unit according to claim 6 is characterized in that: In step S3, the discrete state space equation considering various uncertainties and disturbances is described as: Where: x(k) is the state variable; u(k) is the input; y(k) is the output; w(k) is the sum of model uncertainty, external disturbance and noise, referred to as total disturbance; A m , B m , B e , C m is the coefficient matrix; Considering various uncertainties and improving estimation accuracy, the total disturbance is expanded into a new state quantity, and the expanded state equation is obtained: Where: is the expansion state variable; AKF, BKF, CKF are the expansion system coefficient matrices; According to the difference between the actual output and the estimated output, the extended state Kalman observation equation and various related calculation formulas are obtained to update the state estimate and covariance matrix. The formula is as follows: Where: is the estimated value of the expansion state; y(k) is the actual output; is the output estimate; K k is the Kalman gain; P k are the covariance matrices of the prior and posterior estimation errors respectively; Q k , R k are the covariance matrices of process noise and measurement noise, respectively.

8. The design method of the dynamic energy efficiency optimization system of the dynamic variable load process of the coal-fired unit according to claim 1 is characterized in that: The thermal-electric conversion process of coal-fired units will cause a decrease in energy quality, that is, loss, defining the dynamics of the unit during load variation The loss is the input of coal Output of power generation The difference is expressed as follows: V(k)=W u u(k)-W y y(k) Where: V(k) represents the Loss; Wu, W y are coefficient matrices respectively; u(k) is input; y(k) is output, Based on unit dynamics The loss calculation formula comprehensively considers the set value tracking effect and dynamics Loss, establish the optimization objective function of the coal-fired unit control system: Where: J is the optimization objective function; y(k) is the output at time k; r(k) is the output setting value at time k; Δu(k) is the input increment at time k; V(k) is the value at time k. Loss; S q , Sr, Sv are the error weight matrix, control weight matrix and Loss weight matrix; Based on the real-time estimated values ​​of the extended state Kalman observer, a predictive control strategy for the coordinated economic model of the coal-fired unit is designed, including comprehensive consideration of the set value tracking effect and dynamic The objective function and related constraints of the loss optimization problem are: Where: J is the optimization objective function, Js and Je are the set value tracking target and dynamic loss target; Np is the prediction time domain; Nc is the control time domain; r(k) is the output setting value at time k; Δu(k) is the input increment at time k; V(k) represents the Loss; S q , Sr, Sv are the error weight matrix, control weight matrix and Loss weight matrix; x(k), u(k), w(k) are the state, input and total disturbance at time k respectively; S(Δ) is the value range of the input; is the estimated value of the expansion state at time k+1; is the output estimate at time k; f d 、h d are the functional relationships between the corresponding calculation state estimation and output estimation respectively; Ω1 and Ω2 represent the constraints of input quantity and input quantity increment.

9. A control system, characterized in that: Designed by the design method of the dynamic energy efficiency optimization system of the dynamic variable load process of the coal-fired unit according to any one of claims 1 to 8; The invention comprises a coal-fired unit, an expanded state observer and a dynamic energy efficiency model predictive controller. The coal-fired unit feeds back input power and main steam pressure to the expanded state observer. The expanded state observer outputs an estimated value and the expanded state estimated value to the dynamic energy efficiency model predictive controller. The dynamic energy efficiency model predictive controller outputs the control of coal feed amount and main steam valve opening to the coal-fired unit.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, it is used to implement the design method of the dynamic energy efficiency optimization system of the dynamic load variation process of the coal-fired unit as described in any one of claims 1-8.

Images