A fuzzy modeling method for TS power of small integrated pressurized water reactor based on fuzzy active disturbance rejection control

Abstract

The present invention discloses a power T-S fuzzy modeling method of a small integrated pressurized water reactor based on fuzzy self-disturbance rejection control, comprising the following steps: step 1, constructing a core lumped parameter model of a small integrated pressurized water reactor; step 2, obtaining a second-order nonlinear model for the design of an self-disturbance rejection controller by performing model transformation on the core lumped parameter model in step 1; step 3, constructing a power T-S fuzzy model equivalent to step 2. The present invention is designed for LADRC, and does not take the simplest model structure or the highest model fitting accuracy as the modeling target. For a nonlinear model of a specific structure, by adopting appropriate premise variables and fuzzy set membership functions, the obtained T-S fuzzy model has an equivalent analytical form to the original nonlinear model, which not only improves the accuracy of the fuzzy model, but also reduces the difficulty of fuzzy model structure selection and parameter optimization.

Description

Technical Field

[0001] The invention belongs to the technical field of nuclear power and relates to a fuzzy modeling method of TS power of a small integrated pressurized water reactor based on fuzzy self-disturbance rejection control. Background Art

[0002] Pressurized water reactors are very important components in the field of nuclear power, and small integrated pressurized water reactors are currently a very popular design idea. However, since small integrated pressurized water reactors adopt an integrated design, this integrated design method reduces the construction cost, but it also brings some problems. First, the reactor core and steam generator of the small integrated pressurized water reactor do not have hot pipe sections and cold pipe sections like those of pressurized water reactors, and there is a strong interaction between the coolants of different equipment; second, relative to the core power, the volume of the pressure vessel is large, the coolant inventory is also large, and the inertia of the main coolant is large; third, compared with traditional steam generators, its water storage capacity is small, and the steam generator is very sensitive to disturbances. Therefore, when the power changes or there are other disturbances, the system under such a design responds slowly to the disturbance.

[0003] The controlled object of a small integrated pressurized water reactor is a nonlinear time-varying system. The model used in the controller design generally has large parameter uncertainties and unmodeled dynamics. There are multiple disturbances in the system, and multiple states are unmeasurable. In addition, due to the large differences in equipment type and equipment layout between small integrated pressurized water reactors and large pressurized water reactors, there are also large differences in regulation requirements and dynamic characteristics. It is generally believed that small integrated pressurized water reactors are sensitive to disturbances due to the large thermal inertia of the primary circuit and the small water storage capacity of the steam generator. In addition, the high mobility requirements of nuclear power equipment make the control of small integrated pressurized water reactors more challenging and have not yet been well solved.

[0004] Model is the basis of controller design. It is difficult to design a controller based on a high-order nonlinear model of a small integrated pressurized water reactor, and the designed controller is generally complex in structure and not convenient for engineering application. Therefore, a low-order model suitable for linear controller design is needed, which has high theoretical significance and great engineering application value.

[0005] Purpose of the Invention

[0006] The purpose of the present invention is to address the above-mentioned problems existing in the prior art and propose a method for constructing a TS fuzzy model, specifically a method for constructing a low-order model suitable for a linear controller. Summary of the invention

[0007] The present invention provides a fuzzy modeling method for TS power of a small integrated pressurized water reactor based on fuzzy auto-disturbance rejection control, comprising the following steps:

[0008] Step 1, constructing a core lumped parameter model of a small integrated pressurized water reactor;

[0009] Step 2: By performing model transformation on the core lumped parameter model in step 1, a second-order nonlinear model for designing an active disturbance rejection controller is obtained;

[0010] Step 3: Construct a power TS fuzzy model equivalent to step 2.

[0011] Preferably, step 1 further comprises:

[0012] The core lumped parameter model is constructed based on the Mann heat transfer model and the small integrated pressurized water reactor power model, which is expressed as shown in equations (1)-(7):

[0013]

[0014]

[0015]

[0016]

[0017]

[0018]

[0019]

[0020] Where P th is the relative power; ρ is the reactivity; Λ is the neutron generation time; C r is the relative precursor concentration; n, n0 are the neutron density and the initial equilibrium neutron density or the steady-state neutron density respectively; P, P0 are the reactor power and the reactor rated power respectively; λ is the time constant for the delayed neutron precursor decay; T f is the fuel temperature; f is the share of fission energy released in the fuel; θ1 and θ2 are the temperatures of coolant node 1 and coolant node 2 respectively; m c1 , m c2 are the mass of coolant node 1 and coolant node 2 respectively; c pc is the cooling specific heat capacity of a circuit; c pf A is the specific heat capacity of fuel; fc , A fc1 , A fc2 is the total heat transfer area of ​​fuel and coolant, the total heat transfer area of ​​fuel and coolant node 1 and the total heat transfer area of ​​fuel and coolant node 2; W c is the core coolant mass flow rate; U fc are the heat transfer coefficients of fuel and coolant; ρ, ρ rReactivity introduced for reactivity and control rods; G r is the control rod reactivity value; z r is the relative speed of the control rod.

[0021] Preferably, step 2 further comprises:

[0022] Sub-step S21: further express formula (1) as shown in formula (8):

[0023]

[0024] Taking the derivative of equation (1) with respect to time, we get equation (9):

[0025]

[0026] Substituting equations (2) and (8) into equation (9), we obtain the relationship shown in equation (10):

[0027]

[0028] It can be further expressed as shown in formulas (11)-(16):

[0029] P th =P th (0)+δP th (11)

[0030] c r =c r (0)+δc r (12)

[0031] T f =T f (0)+δT f (13)

[0032] θ1=θ1(0)+δθ1 (14),

[0033] θ2=θ2(0)+δθ2 (15),

[0034] ρ r =ρ r (0)+δρ r =δρ r (16)

[0035] Where P th (0), C r (0),ρ(0),ρ r (0), T f (0), θ1(0), θ2(0) are the initial values, and P th0 ,1,0,0,T f0 ,θ10 ,θ 20 ;

[0036] Sub-step S22, deriving equation (16), expressed as shown in equation (17):

[0037]

[0038] for and The expression of the two coolant control bodies has a small temperature difference, and the coolant specific heat capacity, density and heat transfer coefficient from fuel to coolant are not much different, which can be expressed as shown in equations (18)-(20):

[0039] μ c =μ c1 =μ c2 =m c1 c pc =m c2 c pc (18)

[0040] Ω=Ω1=Ω2=U fc A fc1 =U fc A fc2 (19)

[0041] M=W c c pc (20)

[0042] Substituting equations (11), (13), (14), (15), (19), and (20) into equations (2) and (5), respectively, we obtain the first-order expression of the coolant node temperature increment, as shown in equations (21)-(23):

[0043]

[0044]

[0045]

[0046] The coolant node temperatures δθ1 and δθ2 are simplified by using the core inlet and outlet temperatures, as shown in equations (24-(25):

[0047] δθ2≈δT outlet ≈δT hl (twenty four),

[0048] δT inlet ≈δT cl (25),

[0049] Sub-step S23: In steady state, subtract equation (4) from equation (5), and express it as shown in equation (26):

[0050] W c c pc (δT inlet -δθ1)-W c c pc (δθ1-δθ2)=0 (26),

[0051] When the temperature of the two coolant control bodies changes slowly during the transient process, they have the relationship shown in formula (27):

[0052]

[0053] From equations (22) and (27), we can get the relationship shown in equation (28):

[0054]

[0055] Substituting equation (28) into equation (23) yields the equation shown in equation (29):

[0056]

[0057] Sub-step S24: rewrite equation (3) as shown in equation (30):

[0058]

[0059] Among them, Ω f , μ f They are expressed as shown in equations (31)-(32):

[0060]

[0061] μ f =m f c pf (32),

[0062] Substituting equations (11), (13), and (14) into equation (30), we obtain equation (33):

[0063]

[0064] Substituting (6), (17), (28), (29), and (32) into equation (9), we obtain equations (34) and (35):

[0065]

[0066]

[0067] Where f is the sum of the “unmodeled dynamics” and the internal disturbance, i.e., the fuel temperature change δT f Unmeasurable variables;

[0068] Formula (34) is simplified again as shown in formula (36):

[0069]

[0070] Among them, the relevant parameters are expressed as shown in formulas (37)-(47):

[0071]

[0072]

[0073]

[0074]

[0075]

[0076]

[0077]

[0078]

[0079]

[0080]

[0081] Preferably, step 3 further comprises the following sub-steps:

[0082] Sub-step S31: construct a TS fuzzy model in which the nonlinear terms of the two state quantities are equivalent, which is expressed as shown in formula (47):

[0083]

[0084] The fuzzy rules adopted by the TS fuzzy model are expressed as follows:

[0085]

[0086]

[0087] Among them, N and P are fuzzy sets, is the fuzzy representation of f, d i is a constant to be determined;

[0088] The fuzzy inference output of the TS fuzzy model is written as shown in formula (48):

[0089]

[0090] Among them, w i (δP th ) is the weight of the fuzzy rule, which is equal to the membership degree of the fuzzy set N or P, which is expressed as shown in formula (49):

[0091] w i (δP th )=μ i (δP th ), i = 1, 2 (49),

[0092] w i represents the degree of membership, and equations (48) and (49) are constrained by equation (50):

[0093] w1+w2=1 (50),

[0094] It has the relationship shown in formula (51):

[0095] w1d1+w2d2=δP th (51),

[0096] The values ​​of d1 and d2 satisfy that for all δP th ∈Ψ has w i (x i )∈[0,1], expressed as shown in formula (52):

[0097] Ψ≡{x(t)∈R n |||x(t)||≤ε} (52),

[0098] Combining (50) and (51) to obtain the solution shown in formula (53):

[0099]

[0100] Let d1 = -d2 = d, where d is expressed as shown in equation (54):

[0101] d=max(||δP th (t)||) (54),

[0102] Then, from equations (50) and (53), we can obtain the relationship shown in equations (55)-(56):

[0103]

[0104]

[0105] Sub-step S32: Select δT cl , δT hl,δP th and δz r As the premise variable, the fuzzy model corresponding to formula (36) is established, which is expressed as follows:

[0106]

[0107] The inference output is the following relationship:

[0108]

[0109] Where f represents the “total disturbance”, including unmodeled dynamics, parameter uncertainty, and external disturbances, and satisfies the relationship shown in equations (57)-(58):

[0110]

[0111]

[0112] Assume x is as shown in formula (59):

[0113] x=[x1 x2 x3 x4]=[δT cl δT hl δP th δz r ] (59),

[0114] Satisfies the relationship shown in equations (60)-(62):

[0115]

[0116]

[0117]

[0118] Among them, d j is the upper bound of each premise variable, expressed as shown in formula (63):

[0119] d j =max(|x j |)+ε j ,ε j >0 (63),

[0120] Among them, the relevant parameters are expressed as shown in equations (64)-(71):

[0121]

[0122]

[0123]

[0124]

[0125] b i,P-P =(-1) ceil(i / 2) B P-P d3 (68),

[0126] b i,z-P =(-1) i B z-P d4 (69),

[0127]

[0128] BRIEF DESCRIPTION OF THE DRAWINGS

[0129] Figure 1 It is a schematic diagram of the Mann heat transfer model of the small integrated pressurized water reactor core of the present invention. DETAILED DESCRIPTION

[0130] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0131] Those skilled in the art should understand that the step numbers used herein are only for convenience of description and are not intended to limit the order in which the steps are executed. The terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0132] As used in the present specification and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms unless the context clearly indicates otherwise. The terms "include" and "comprise" indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their collections. The term "and / or" refers to any combination of one or more of the associated listed items and all possible combinations, and includes these combinations.

[0133] Figure 1 Schematic diagram of the Mann heat transfer model of the small integrated pressurized water reactor core of the present invention, Figure 1 As shown, the model includes nuclear energy, fuel temperature T f , coolant node 1 temperature θ1, coolant node 2 temperature θ2, core inlet temperature T cl, core outlet temperature T hl These indicators.

[0134] The present invention provides a fuzzy modeling method for TS of a small integrated pressurized water reactor power based on fuzzy auto-disturbance rejection control, which specifically comprises the following steps:

[0135] Step 1, constructing a core lumped parameter model of a small integrated pressurized water reactor;

[0136] Step 2: By performing model transformation on the core lumped parameter model in step 1, a second-order nonlinear model for designing an active disturbance rejection controller is obtained;

[0137] Step 3: Construct a power TS fuzzy model equivalent to step 2.

[0138] Step 1 specifically includes:

[0139] The core lumped parameter model is constructed based on the Mann heat transfer model and the small integrated pressurized water reactor power model, which is expressed as shown in equations (1)-(7):

[0140]

[0141]

[0142]

[0143]

[0144]

[0145]

[0146]

[0147] Where P th is the relative power; ρ is the reactivity; Λ is the neutron generation time; C r is the relative precursor concentration; n, n0 are the neutron density and the initial equilibrium neutron density or the steady-state neutron density respectively; P, P0 are the reactor power and the reactor rated power respectively; λ is the time constant for the delayed neutron precursor decay; T f is the fuel temperature; f is the share of fission energy released in the fuel; θ1 and θ2 are the temperatures of coolant node 1 and coolant node 2 respectively; m c1 , m c2 are the mass of coolant node 1 and coolant node 2 respectively; c pc is the cooling specific heat capacity of a circuit; c pf A is the specific heat capacity of fuel; fc , A fc1 , Afc2 is the total heat transfer area of ​​fuel and coolant, the total heat transfer area of ​​fuel and coolant node 1 and the total heat transfer area of ​​fuel and coolant node 2; W c is the core coolant mass flow rate; U fc are the heat transfer coefficients of fuel and coolant; ρ, ρ r Reactivity introduced for reactivity and control rods; G r is the control rod reactivity value; z r is the relative speed of the control rod.

[0148] Step 2 specifically includes:

[0149] Sub-step S21: further express formula (1) as shown in formula (8):

[0150]

[0151] Taking the derivative of equation (1) with respect to time, we get equation (9):

[0152]

[0153] Substituting equations (2) and (8) into equation (9), we obtain the relationship shown in equation (10):

[0154]

[0155] It can be further expressed as shown in formulas (11)-(16):

[0156] P th =P th (0)+δP th (11)

[0157] c r =c r (0)+δc r (12)

[0158] T f =T f (0)+δT f (13)

[0159] θ1=θ1(0)+δθ1 (14),

[0160] θ2=θ2(0)+δθ2 (15),

[0161] ρ r =ρ r (0)+δρ r =δρ r (16)

[0162] Where P th(0), C r (0),ρ(0),ρ r (0), T f (0), θ1(0), θ2(0) are the initial values, and P th0 ,1,0,0,T f0 ,θ 10 ,θ 20 ;

[0163] Sub-step S22, deriving equation (16), expressed as shown in equation (17):

[0164]

[0165] for and The expression of the two coolant control bodies has a small temperature difference, and the coolant specific heat capacity, density and heat transfer coefficient from fuel to coolant are not much different, which can be expressed as shown in equations (18)-(20):

[0166] μ c =μ c1 =μ c2 =m c1 c pc =m c2 c pc (18)

[0167] Ω=Ω1=Ω2=U fc A fc1 =U fc A fc2 (19)

[0168] M=W c c pc (20)

[0169] Substituting equations (11), (13), (14), (15), (19), and (20) into equations (2) and (5), respectively, we obtain the first-order expression of the coolant node temperature increment, as shown in equations (21)-(23):

[0170]

[0171]

[0172]

[0173] The coolant node temperatures δθ1 and δθ2 are simplified by using the core inlet and outlet temperatures, as shown in equations (24-(25):

[0174] δθ2≈δT outlet ≈δThl (twenty four),

[0175] δT inlet ≈δT cl (25),

[0176] Sub-step S23: In steady state, subtract equation (4) from equation (5), and express it as shown in equation (26):

[0177] W c c pc (δT inlet -δθ1)-W c c pc (δθ1-δθ2)=0 (26),

[0178] When the temperature of the two coolant control bodies changes slowly during the transient process, they have the relationship shown in formula (27):

[0179]

[0180] From equations (22) and (27), we can get the relationship shown in equation (28):

[0181]

[0182] Substituting equation (28) into equation (23) yields the equation shown in equation (29):

[0183]

[0184] Sub-step S24: rewrite equation (3) as shown in equation (30):

[0185]

[0186] Among them, Ω f , μ f They are expressed as shown in equations (31)-(32):

[0187]

[0188] μ f =m f c pf (32),

[0189] Substituting equations (11), (13), and (14) into equation (30), we obtain equation (33):

[0190]

[0191] Substituting (6), (17), (28), (29), and (32) into equation (9), we obtain equations (34) and (35):

[0192]

[0193]

[0194] Where f is the sum of the “unmodeled dynamics” and the internal disturbance, i.e., the fuel temperature change δT f Unmeasurable variables;

[0195] Formula (34) is simplified again as shown in formula (36):

[0196]

[0197] Among them, the relevant parameters are expressed as shown in formulas (37)-(47):

[0198]

[0199]

[0200]

[0201]

[0202]

[0203]

[0204]

[0205]

[0206]

[0207]

[0208] Step 3 specifically includes the following sub-steps:

[0209] Sub-step S31: construct a TS fuzzy model in which the nonlinear terms of the two state quantities are equivalent, which is expressed as shown in formula (47):

[0210]

[0211] The fuzzy rules adopted by the TS fuzzy model are expressed as follows:

[0212]

[0213]

[0214] Where N and P are fuzzy sets, is the fuzzy representation of f, d i is a constant to be determined;

[0215] The fuzzy inference output is written as shown in formula (48):

[0216]

[0217] Among them, w i (δP t ) is the weight of the fuzzy rule. Since there is only one premise variable, it is equal to the membership degree of the fuzzy set N or P, which can be expressed as shown in formula (49):

[0218] w i (δP th )=μ i (δP th ),i=1,2 (49),

[0219] For the convenience of description, the following uses w i represents the degree of membership, and equations (48) and (49) are constrained by equation (50):

[0220] w1+w2=1 (50),

[0221] It has the relationship shown in formula (51):

[0222] w1d1+w2d2=δP th (51),

[0223] The values ​​of d1 and d2 must satisfy for all δP t ∈Ψ has w i (x i )∈[0,1], expressed as shown in formula (52):

[0224] Ψ≡{x(t)∈R n |||x(t)||≤ε} (52),

[0225] Combining (50) and (51) to obtain the solution shown in formula (53):

[0226]

[0227] Take d1=-d2=d, where d is expressed as shown in formula (54):

[0228] d=max(||δP th (t)||) (54),

[0229] Then, from equations (50) and (53), we can obtain the relationship shown in equations (55)-(56):

[0230]

[0231]

[0232] It can be seen that if the premise variables, fuzzy subsets, membership functions and linear fuzzy consequents are used, the inference output of the constructed TS fuzzy model can be made equivalent to the given nonlinear terms. In this way, the nonlinear terms in the nonlinear second-order model of the small reactor core can be represented by the TS fuzzy model with linear fuzzy consequents (i.e. local submodels).

[0233] Sub-step S32: Select δT cl , δT l ,δP t and δz r As the premise variable, the fuzzy model corresponding to formula (36) is established, which is expressed as follows:

[0234]

[0235] The inference output is the following relationship:

[0236]

[0237] Where f represents the “total disturbance”, including unmodeled dynamics, parameter uncertainty, and external disturbances, and satisfies the relationship shown in equations (57)-(58):

[0238]

[0239]

[0240] Assume x is as shown in formula (59):

[0241] x=[x1 x2 x3 x4]=[δT cl δT hl δP th δz r ] (59),

[0242] Similarly, it can be expressed as shown in equations (60)-(62):

[0243]

[0244]

[0245]

[0246] where d jis the upper bound of each premise variable, expressed as shown in formula (63):

[0247] d j =max(|x j |)+ε j ,ε j >0 (63),

[0248] Among them, the relevant parameters are expressed as shown in equations (64)-(71):

[0249]

[0250]

[0251]

[0252]

[0253] b i,P-P =(-1) ceil(i / 2) B P-P d3 (68), b i,z-P =(-1) i B z-P d4 (69),

[0254]

[0255]

Claims

1. A fuzzy modeling method for TS power of a small integrated pressurized water reactor based on fuzzy active disturbance rejection control, characterized in that: The following steps are involved: Step 1, constructing a core lumped parameter model of a small integrated pressurized water reactor; Step 2: By performing model transformation on the core lumped parameter model in step 1, a second-order nonlinear model for designing an active disturbance rejection controller is obtained; Step 3: construct a power TS fuzzy model equivalent to step 2; Step 1 further comprises: The core lumped parameter model is constructed based on the Mann heat transfer model and the small integrated pressurized water reactor power model, which is expressed as shown in equations (1)-(7): Where P th is the relative power; ρ is the reactivity; Λ is the neutron generation time; C r is the relative precursor concentration; n, n0 are the neutron density and the initial equilibrium neutron density or the steady-state neutron density respectively; P, P0 are the reactor power and the reactor rated power respectively; λ is the time constant for the delayed neutron precursor decay; T f is the fuel temperature; f is the share of fission energy released in the fuel; θ1 and θ2 are the temperatures of coolant node 1 and coolant node 2 respectively; m c1 , m c2 are the mass of coolant node 1 and coolant node 2 respectively; c pc is the cooling specific heat capacity of a circuit; c pf A is the specific heat capacity of fuel; fc , A fc1 , A fc2 is the total heat transfer area of ​​fuel and coolant, the total heat transfer area of ​​fuel and coolant node 1 and the total heat transfer area of ​​fuel and coolant node 2; W c is the core coolant mass flow rate; U fc are the heat transfer coefficients of fuel and coolant; ρ, ρ r Reactivity introduced for reactivity and control rods; G r is the control rod reactivity value; z r is the relative speed of the control rod; Step 2 further includes: Sub-step S21: further express formula (1) as shown in formula (8): Taking the derivative of equation (1) with respect to time, we get equation (9): Substituting equations (2) and (8) into equation (9), we obtain the relationship shown in equation (10): It can be further expressed as shown in formulas (11)-(16): P th =P th (0)+δP th (11), c r =c r (0)+δc r (12), T f =T f (0)+δT f (13), θ1=θ1(0)+δθ1 (14), θ2=θ2(0)+δθ2 (15), r r =ρ r (0)+dr r =dr r (16), Where P th (0), C r (0),ρ(0),ρ r (0), T f (0), θ1(0), θ2(0) are the initial values, and P th0 ,1,0,0,T f0 ,θ 10 ,θ 20 ; Sub-step S22, deriving equation (16), expressed as shown in equation (17): for and The expression of the two coolant control bodies has a small temperature difference, and the coolant specific heat capacity, density and heat transfer coefficient from fuel to coolant are not much different, which can be expressed as shown in equations (18)-(20): m c =μ c1 =μ c2 =m c1 c pc =m c2 c pc (18), Ω=Ω1=Ω2=U fc A fc1 =U fc A fc2 (19), M=W c cp c (20), Substituting equations (11), (13), (14), (15), (19), and (20) into equations (2) and (5), respectively, we obtain the first-order expression of the coolant node temperature increment, as shown in equations (21)-(23): The coolant node temperatures δθ1 and δθ2 are simplified by using the core inlet and outlet temperatures, as shown in equations (24)-(25): δθ2≈δT outlet ≈δT hl (24), δT inlet ≈δt cl (25), Sub-step S23: In steady state, subtract equation (4) from equation (5), and express it as shown in equation (26): W c c pc (δT inlet -δθ1)-W c c pc (δθ1-δθ2)=0 (26), When the temperature of the two coolant control bodies changes slowly during the transient process, they have the relationship shown in formula (27): From equations (22) and (27), we can get the relationship shown in equation (28): Substituting equation (28) into equation (23) yields the equation shown in equation (29): Sub-step S24: rewrite equation (3) as shown in equation (30): Among them, Ω f , μ f They are expressed as shown in equations (31)-(32): m f =m f c pf (32), Substituting equations (11), (13), and (14) into equation (30), we obtain equation (33): Substituting (6), (17), (28), (29), and (32) into equation (9), we obtain equations (34) and (35): Where f is the sum of "unmodeled dynamics" and internal disturbances, i.e., the fuel temperature change δT f Unmeasurable variables; Formula (34) is simplified again as shown in formula (36): Among them, the relevant parameters are expressed as shown in formulas (37)-(47): Step 3 further includes the following sub-steps: Sub-step S31: construct a TS fuzzy model in which the nonlinear terms of the two state quantities are equivalent, which is expressed as shown in formula (47): The fuzzy rules adopted by the TS fuzzy model are expressed as follows: IFδP th is N,THEN IFδP th is P,THEN Among them, N and P are fuzzy sets, is the fuzzy representation of f, d i is a constant to be determined; The fuzzy inference output of the TS fuzzy model is written as shown in formula (48): Among them, w i (δP th ) is the weight of the fuzzy rule, which is equal to the membership degree of the fuzzy set N or P, which is expressed as shown in formula (49): w i (δP th )=μ i (δP th ), i=1.2 (49), w i represents the degree of membership, and equations (48) and (49) are constrained by equation (50): w1+w2=1 (50), It has the relationship shown in formula (51): w1d1+w2d2=δP th (51), The values ​​of d1 and d2 satisfy that for all δP th ∈Ψ has w i (x i )∈[0,1], expressed as shown in formula (52): Ψ≡{x(t)∈R n |||x(t)||≤ε} (52), Combining (50) and (51) to obtain the solution shown in formula (53): Let d1 = -d2 = d, where d is expressed as shown in equation (54): d=max(||δP th (t)||) (54), Then, from equations (50) and (53), we can obtain the relationship shown in equations (55)-(56): Sub-step S32: Select δT cl , δT hl ,δP th and δz r As the premise variable, the fuzzy model corresponding to formula (36) is established, which is expressed as follows: IF δT cl is F 1i andδT hl is F 2i andδP th is F 3i andδz r i sF 4i ,THEN, i＝1,2,3,4,…,16， The inference output is the following relationship: Where f represents the "total disturbance", including unmodeled dynamics, parameter uncertainty and external disturbances, and satisfies the relationship shown in equations (57)-(58): Assume x is as shown in formula (59): x=[x1 x2 x3 x4]=[δT cl δT hl δP th δz r ] (59), Satisfies the relationship shown in equations (60)-(62): Among them, d j is the upper bound of each premise variable, expressed as shown in formula (63): d j =max(|x j |)+e j ,he j >0 (63), Among them, the relevant parameters are expressed as shown in equations (64)-(71): b i,P-P =(-1) ceil(i / 2) B P-P d3 (68),b i,z-P =(-1) i B z-P d4 (69),

Images