A method for suppressing vibration during turbine startup based on MOPSO and stability test

Abstract

The present invention discloses a method for suppressing vibration during the startup process of a hydraulic turbine based on MOPSO and stability testing, comprising the following steps: S1, establishing a startup model for a Francis turbine; S2, obtaining an objective function and decision variables for optimizing the startup rule; S3, optimizing the startup rule using an optimization method to obtain a Pareto optimal frontier; S4, selecting multiple groups of solutions from the Pareto frontier for field testing at a power station to obtain the vibration status of the hydraulic machinery; and S5, establishing a comprehensive state evaluation system to determine the final optimal solution. The present invention establishes a simulation model of the hydropower unit's regulation system, optimizes the startup rule using a multi-objective particle swarm algorithm with unit speed overshoot and unit speed rise time as targets, obtains a Pareto optimal frontier, and selects the final optimal solution based on field testing and expert experience. The final optimal solution can reasonably avoid mechanical vibration during the unit startup process, reduce unit losses, and extend unit life.

Description

Technical Field

[0001] The present invention relates to the technical field of hydropower generation, and in particular to a method for suppressing vibration during a turbine startup process based on MOPSO and stability tests. Background Art

[0002] In the context of a diversified and integrated power grid, ensuring the stability of a high-penetration renewable energy power system is crucial. Hydropower units are highly efficient, flexible, and provide inertia to the grid, making them an indispensable component of maintaining grid stability. Current research on optimizing hydropower unit patterns, both domestically and internationally, focuses primarily on unit operation optimization and load distribution, but has focused less on the transition process during the unit startup phase. The startup phase can be defined as the process from receiving the startup command to grid connection. Currently, turbine speed governor startup patterns typically include open-loop, closed-loop, and open-loop combined with closed-loop patterns. Parameter selection in open-loop patterns is often unreliable and difficult to accurately select. Closed-loop patterns overestimate the speed tracking performance of the unit, resulting in a prolonged speed ramp-up time. The open-loop combined with closed-loop pattern effectively combines the advantages of both open-loop and closed-loop startup, resulting in a fast, stable, and effective control process. Hydropower stations often employ an "open-loop + closed-loop" startup schedule and use empirical tuning to adjust startup schedule parameters. These optimizations fail to account for hydraulic and mechanical vibrations. This results in excessive unit vibration and reduced stability when the guide vane opening fluctuates significantly during startup. Therefore, it is necessary to conduct research on optimizing unit startup schedules while suppressing unit vibration.

[0003] Common optimization methods mainly target startup modes or startup parameters. Taking into account the actual optimization cost of the power station, the parameters that can be directly adjusted in the program can be optimized, and the parameters obtained after model optimization can be put into field tests. Through the model + test method, actual costs can be saved and optimization efficiency can be improved.

[0004] The "Method and System for Optimizing Parameters of a Hydroturbine Governor in a Weakly Damped Low-Frequency Oscillation Mode," published in Chinese patent literature and with publication number CN110377970A, discloses a method and system for optimizing parameters of a hydroturbine governor in a weakly damped low-frequency oscillation mode. The method includes: determining the weakly damped mode and its oscillation frequency associated with the unit to be optimized; establishing an open-loop transfer function for the hydroturbine and its regulating system, and calculating the damping torque coefficient of the open-loop transfer function of the hydroturbine and its regulating system at the oscillation frequency of the weakly damped oscillation mode; calculating the squared error integral of the mechanical power output of the hydroturbine and its regulating system under a step signal based on the open-loop transfer function of the hydroturbine and its regulating system; calculating the damping ratio corresponding to the maximum eigenvalue of the primary frequency regulation closed-loop system; establishing an objective function based on the calculated damping torque coefficient, squared error integral, and damping ratio; and employing a particle swarm algorithm to determine the optimal solution to the objective function, which serves as the optimal parameters for the hydroturbine governor. This invention improves system frequency stability while avoiding the deterioration of the weakly damped low-frequency oscillation mode and can be widely applied in the field of power system optimization. However, the invention does not involve any research on the optimization of the unit startup phase. Summary of the Invention

[0005] The present invention solves the problem in the prior art of not considering hydraulic and mechanical vibration factors during the turbine startup process, and proposes a vibration suppression method for the turbine startup process based on MOPSO and stability testing. The present invention establishes a simulation model of the regulation system of the hydropower unit, takes the unit speed overshoot and the unit speed rise time as targets, uses a multi-objective particle swarm algorithm to optimize the startup law, obtains the Pareto optimal frontier, and selects the final optimal solution based on field tests and expert experience. The final optimal solution can reasonably avoid mechanical vibration during the unit startup process, reduce unit losses, and extend the unit life.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a method for suppressing vibration during the startup process of a hydraulic turbine based on MOPSO and stability test, comprising the following steps:

[0007] S1, establish the startup model of the Francis unit;

[0008] S2, obtain the objective function and decision variables for startup rule optimization;

[0009] S3, optimize the startup rules using optimization methods to obtain the Pareto optimal frontier;

[0010] S4, select multiple groups of solutions in the Pareto front to conduct field tests at the power station to obtain the vibration conditions of the hydraulic machinery;

[0011] S5, establish a comprehensive status evaluation system to determine the final optimal solution. The method of the present invention is based on the hydraulic layout of a certain power station. First, a startup model of the unit is established. Since the startup condition belongs to a large fluctuation transition process, a characteristic line method is used to establish a water diversion system model. The rotation speed and water pressure are comprehensively considered. The absolute value integral of the relative error of the unit speed and the absolute value integral of the relative error of the volute water pressure are used as the optimization objective function. In combination with the characteristics of multi-stage startup, the parameters to be optimized are selected. The startup rule is optimized based on a multi-objective particle swarm algorithm to obtain the Pareto optimal frontier. Several groups of solutions are selected for field testing to obtain the mechanical vibration, swing and pressure pulsation of unit components such as the frame, guide bearings and tailwater pipes. A comprehensive state evaluation system is formed based on expert experience and information fusion to evaluate the unit stability corresponding to each group of optimized parameters. The group of parameters with the best evaluation results is selected as the final value of the parameters for optimizing the startup rule of the turbine under the corresponding water head. The present invention uses a multi-objective optimization algorithm to obtain a set of optimal solutions for the startup rule. In combination with the comprehensive state evaluation system, the influence of hydraulic-mechanical vibration is fully considered. This can not only improve the dynamic performance of the system, but also effectively suppress the hydraulic-mechanical vibration during the startup process and improve the smoothness of the unit transition.

[0012] Preferably, step S2 includes the following steps:

[0013] S21, select the absolute integral of the relative error of the speed of the unit and the absolute integral of the relative error of the volute water pressure as two objective functions. The specific objective functions are:

[0014]

[0015]

[0016] Where k is the total number of samples, n(i) represents the real-time speed value corresponding to the sampling process, n(∞) represents the stable value of the unit speed, Hvol(i) is the real-time value of the volute water pressure during the sampling process, and Hvol_average is the average value of the volute water pressure during the sampling process. The first objective J1 of the objective function is to minimize the absolute integral of the relative error of the unit speed, and the second objective J2 is to minimize the absolute integral of the relative error of the volute water pressure;

[0017] The constraints are:

[0018]

[0019] Where, t a is the adjustment time, T u is the upper limit of the adjustment time, θ l is the lower limit of the decision variable, θ u is the upper limit of the decision variable;

[0020] S22, obtain the decision variable, the decision variable θ is expressed as:

[0021] θ=(K p , K i , K d , Y1, Y2, Y3, Y4, t1, t2, t3)

[0022] Where K p , K i , K d These correspond to the three parameters in the proportional, integral, and differential phases of the unit's PID controller. Y1 and t1 are the maximum guide vane opening of the first stage and the time it takes for the guide vane to reach its maximum opening in the first stage, respectively. Y2 and t2 are the maximum guide vane opening of the second stage and the time it takes for the guide vane to reach its maximum opening in the second stage, respectively. t3 is the time it takes for the guide vane to close after opening. Y3 and t4 are the guide vane opening when the PID controller is activated and the time it takes for the guide vane to activate PID control, respectively. Y4 is the stable guide vane opening after the unit's smooth transition. In the present invention, the two objective functions are the absolute integral of the relative error in the unit's speed and the absolute integral of the relative error in the volute water pressure. These objective functions are optimized under the constraints of time and decision variables.

[0023] As a preferred embodiment, the step S4 is to select multiple solutions in the Pareto front to conduct field tests in the power station to obtain the vibration, swing and pressure pulsation of the turbine components such as the upper and lower frames, guide bearings, and tailwater pipes. The Pareto front θ selected by the test is max，n It can be expressed as:

[0024] θ max，n =(K p，n , K i，n , K d，n , Y 1，n , Y 2，n , Y 3，n , Y 4，n , t 1，n , t 2，n , t 3，n )

[0025] n=1,2,...,N

[0026] N≤M

[0027] Where n corresponds to the optimal solution in the Pareto frontier selected, N is the number of optimal solutions selected in the experiment, and M is the number of optimal solutions in the Pareto frontier;

[0028] The stability state data set can be expressed as:

[0029] O n =(O n，1, O n，2 ,...,O n，q ,...,O n，Q )

[0030] Where, O n represents θ max，n The optimal solution corresponds to a set of hydraulic machinery vibrations, where Q represents the total number of measured points in the unit, q = 1, 2, ..., and Q corresponds to different measured points. In the present invention, several Pareto optimal front solutions are selected for field testing, ultimately forming a stability state dataset with a one-to-one correspondence between the Pareto front and the hydraulic machinery vibrations.

[0031] Preferably, step S5 specifically includes establishing a comprehensive state evaluation system based on the degradation index and expert experience to screen the Pareto front and determine the final optimal solution. When describing the evaluation system, the vibration, swing, and pressure pulsation values ​​of each measuring point are the values ​​of each index;

[0032] Substituting each indicator value into the degradation degree formula, it becomes a degradation degree value that describes the degree of quality of the indicator:

[0033]

[0034] Where, β n，q is the qth index value corresponding to the nth group of Pareto front solutions, is the normalized value, β0 is the good value of the indicator, β max is the limit value of the indicator;

[0035] The weight value M of each indicator is given based on expert experience:

[0036] M=(M1,M2,...,M q ,...,M Q )

[0037]

[0038] Where M q is the weight value corresponding to the qth indicator, and the sum of the weight values ​​of Q indicators is 1;

[0039] Calculate the comprehensive state evaluation value F of each group of Pareto solutions n，q :

[0040]

[0041] F=min(F 1，q , F 2，q ,...,F n，q )

[0042] The solution corresponding to F is the final solution obtained from the stability test. In the present invention, a comprehensive state evaluation system is formed based on expert experience and information fusion methods, an index evaluation model for vibration, swing, and pressure pulsation data sets is established, and the hierarchical analysis method is combined to evaluate the unit startup stability, and the optimal solution is selected. The corresponding variable parameters are the final values ​​of the parameters for optimizing the startup rules of the turbine under the corresponding water head. In addition, the test water head is different, and the weights of each indicator given based on expert experience are different. Considering the corresponding changes in the dominant indicators of hydraulic machinery vibration, the optimal solution of the parameters also changes accordingly, which helps to increase the flexibility and breadth of startup parameter optimization.

[0043] Preferably, the optimization method is a multi-objective particle swarm algorithm, which utilizes the concept of Pareto dominance to find solutions to multi-objective problems. In the present invention, based on the multi-objective particle swarm algorithm, the objective function for optimizing the startup pattern is optimized for the speed governor parameters and the guide vane opening pattern parameters to obtain a Pareto optimal frontier.

[0044] Preferably, the startup model of the Francis turbine includes a water diversion system model and boundary conditions, wherein the boundary conditions include the Francis turbine, generator, speed regulator, and reservoir. In the present invention, the Francis turbine utilizes a comprehensive characteristic curve to determine the turbine's torque and flow rate, and the generator equation is a first-order differential equation. The speed regulator comprises a parallel PID controller and an actuator, and the reservoir boundary conditions are constructed by combining known conditions and characteristic line equations.

[0045] The beneficial effects of the present invention are:

[0046] 1. The water diversion head of the present invention completes the optimization research on the startup rules of hydropower units for mechanical vibration suppression, effectively suppresses vibration, swing, and pressure pulsation, and improves the unit's grid connection speed and startup performance;

[0047] 2. This invention establishes a high-precision simulation model of the power plant unit speed regulation system based on the characteristic line method, providing a model foundation for the study of the unit speed governor control method and transition process. Through simulation, the number of real-machine tests during the transition process can be minimized. 3. This invention adopts a method that combines modeling and testing to ensure the rapid grid connection of hydropower units and the performance of the units during startup, thereby increasing economic benefits and helping to enhance the innovation capabilities of hydropower stations in the power generation industry.

[0048] 4. The optimization research results of the present invention can extend the life of the unit, ensure the economic benefits of the power station, and reduce the mechanical vibration during the startup process, which can reduce the wear of the unit to a certain extent, extend the life of the unit, and ensure the economic benefits of the power station to a certain extent. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is a schematic diagram of the process of the present invention;

[0050] Figure 2 This is a structural schematic diagram of a mixed flow power station according to the present invention;

[0051] Figure 3 Schematic diagram of the characteristic line method of the present invention;

[0052] Figure 4 This is a schematic diagram of the startup model of the unit of the present invention;

[0053] Figure 5 This is a schematic diagram of the startup rule of "multi-stage guide vane direct opening + PID" of the present invention;

[0054] Figure 6 This is a flowchart of startup parameter optimization based on the multi-objective particle swarm optimization algorithm of the present invention;

[0055] Figure 7 Schematic diagram of some measuring points for testing stability during the test of the present invention

[0056] Among them, 1. Upper guide swing measuring point 2, upper frame vibration measuring point 3, lower guide swing measuring point 4, lower frame vibration measuring point 5, tailwater pipe inlet measuring point 6, outlet measuring point 7, elbow pipe measuring point. DETAILED DESCRIPTION

[0057] Example:

[0058] This embodiment proposes a method for suppressing vibration during the turbine startup process based on MOPSO and stability test. Figure 1 , mainly includes the following steps:

[0059] Step 1: Establish a startup model for the Francis unit;

[0060] refer to Figure 2 In the simplified structural diagram of the Francis power station, the water diversion piping system is simplified to the upstream (before the runner) sections 1-6 and the downstream 7-9 sections. The unit startup model consists of two parts: the water diversion system model and the boundary conditions. Since the startup condition is a large fluctuation transition process, the characteristic line method is used to establish the water diversion system model. The characteristic line method diagram is shown below. Figure 3 As shown. The boundary conditions include: Francis unit, generator, governor, reservoir. The Francis unit uses the comprehensive characteristic curve to obtain the torque and flow of the unit. The generator equation is a first-order differential equation. The governor consists of a parallel PID controller and an actuator. The reservoir boundary conditions are composed of known conditions and characteristic line equations. The unit startup model is as follows Figure 4 As shown;

[0061] (1) Water diversion system model

[0062]

[0063]

[0064] Among them, C + and C - are the positive and negative characteristic line equations, respectively.

[0065] in

[0066]

[0067] Among them, Q p and H p are the flow rate and water head at point P at the current moment, Q i-1 and H i-1 are the flow rate and water head at the moment before point A, Q i+1 and H i+1 are the flow rate and water head at the moment before point B, A is the cross-sectional area of ​​the pipe, g is the acceleration of gravity, f is the loss coefficient along the way, Δt is the water hammer wave step length, a is the water hammer wave velocity, and D is the inner diameter of the pipe;

[0068] (2) Boundary conditions

[0069] 1. Reservoir boundary

[0070] When considering the local head loss at the inlet, the boundary equation of the upstream reservoir is:

[0071]

[0072] C - :Q S =C n +C a H S

[0073] Among them, S is the boundary point of the upstream reservoir, H S and Q S are the head and flow at point S, H u is the head of the upstream reservoir, k1 is the inlet loss coefficient;

[0074] When considering the local head loss at the outlet, the boundary equation of the downstream reservoir is:

[0075]

[0076] C + :Q X =C p -C a H X

[0077] Where X is the downstream reservoir boundary point, H X and Q Xare the water head and flow at point X, H d is the head of the downstream reservoir, k2 is the outlet loss coefficient;

[0078] 2. Turbine boundary

[0079] The turbine characteristics are generally given as a model comprehensive characteristic curve. In the transient process calculation, the model comprehensive characteristic curve needs to be converted into a curve of unit flow and unit torque, which is described as follows:

[0080] Q 11 =f(a, n 11 )

[0081] M 11 =g(α, n 11 )

[0082] Among them, Q 11 is the unit flow rate, M 11 is the unit torque of the turbine, a is the guide vane opening, n 11 is the unit speed of the turbine;

[0083] The turbine similarity equation is described as follows:

[0084]

[0085]

[0086] Among them, Q is the turbine discharge, M t is the turbine torque, D1 is the turbine runner diameter, H t The working water head of the turbine;

[0087] 3. Generator boundary

[0088]

[0089] Where n is the unit speed, J is the moment of inertia of the hydro-generator unit;

[0090] 4. Governor boundary model

[0091] The speed regulator consists of a parallel PID controller and an actuator.

[0092] The transfer function of the parallel PID controller is:

[0093]

[0094] Among them, K p , K i , K d are the proportional, integral and differential parameters of the controller respectively;

[0095] The transfer function of the actuator is:

[0096]

[0097] Among them, T y is the corresponding time constant of the relay.

[0098] Step 2: Obtain objective function and decision variables

[0099] (1) The two objective functions are the absolute integral of the relative error of the unit speed and the absolute integral of the relative error of the volute water pressure. The objective function is expressed as:

[0100]

[0101]

[0102] Among them, k represents the total number of samples, n(i) is the real-time speed value corresponding to the sampling process, n(∞) is the stable value of the unit speed, Hvol(i) represents the real-time value of the volute water pressure during the sampling process, and Hvol_average represents the average value of the volute water pressure during the sampling process. The objective function J1 represents the minimization of the absolute integral of the relative error of the unit speed, and J2 represents the minimization of the absolute integral of the relative error of the volute water pressure;

[0103] The constraints are

[0104]

[0105] θ l =(K p，min , K i，min , K d，min , Y 1，min , Y 2，min , Y 3，min , Y 4，min , t 1，min , t 2，min , t 3，min )

[0106] θ u =(K p，max , K i，max , K d，max , Y 1，max , Y 2，max , Y 3，max , Y 4，max , t 1，max , t 2，max , t 3，max )

[0107] Among them, t a Indicates the adjustment time, T uIndicates the upper limit of the adjustment time, θ l represents the lower limit of the decision variable, θ u represents the upper limit of the decision variable, K p , K i , K d They represent the proportional, integral, and differential parameters of the controller, respectively. The subscripts min and max are the lower and upper limits of each parameter, respectively.

[0108] (2) Decision variables, reference Figure 5 The startup rule is "multi-stage guide vane direct opening + PID control". The guide vane first opens to a certain opening at a certain speed, then opens to another opening at a smaller speed. After maintaining this opening for a period of time, the guide vane is closed at a certain speed until PID control is put into operation.

[0109] The decision variables are:

[0110] θ=(K p , K i , K d , Y1, Y2, Y3, Y4, t1, t2, t3)

[0111] Among them, Y1 and t1 are the maximum opening of the first guide vane and the time from the guide vane opening to the maximum opening of the first guide vane, respectively; Y2 and t2 are the maximum opening of the second guide vane and the time from the guide vane opening to the maximum opening of the second guide vane, respectively; t3 is the time from the guide vane opening to the start of closing; Y3 and t4 are the guide vane opening when the PID controller is put into operation and the time from the guide vane opening to the start of PID control, respectively; Y4 is the stable opening of the guide vane after the unit smoothly transitions, K p , K i , K d They are the proportional, integral and differential parameters of the controller respectively.

[0112] Step 3: Based on the multi-objective particle swarm algorithm, the speed governor parameters and guide vane opening law parameters are optimized according to the objective function of the startup law optimization to obtain the Pareto optimal frontier. The process is as follows: Figure 6 As shown,

[0113] The solution process of the multi-objective particle swarm optimization algorithm is as follows:

[0114] S1: Set the parameters of the multi-objective particle swarm optimization algorithm (MOPSO), including the maximum number of iterations, population size, external archive set size, and the value range of the parameters to be optimized in the decision variables;

[0115] S2: Initialize the initial position value θ of the decision variable based on the upper and lower limits of the decision variable i (k), the current iteration number k = 1, i = 1, 2, ..., N, N is the particle swarm size;

[0116] S3: Calculate the absolute value integral of the relative error of the target function speed J1 i (k) and the absolute value integral of the relative error of the volute water pressure J2 i (k);

[0117] S4: Determine whether the particles meet the constraints. If so, jump to S5. Otherwise, return to S2 until all particles meet the constraints.

[0118] S5: Evaluate the fitness function values ​​of all particles and update the optimal position P of the individual particles best and the global optimal position G best ;

[0119] S6: Update the external archive set, insert the non-dominated solution into the external archive set, and remove the dominated solution from the external archive set;

[0120] S7: Use the formula to calculate the latest velocity v(k+1) and the latest position θ(k+1) of each particle,

[0121] v(k+1)=ωv(k)+c1r1(P best (k)-θ(k))+c2r2(G best (k)-θ(k))

[0122] θ(k+1)=θ(k)+v(k+1)

[0123] Where v(k) and v(k+1) are the current and latest velocities of the particle, ω is the inertia weight, c1 and c2 are acceleration constants, r1 and r2 are random numbers in the interval [0,1], and P best (k), G best (k) are the current optimal position of the particle and the global optimal position, respectively. θ(k) and θ(k+1) are the current position and the latest position of the particle, respectively.

[0124] S8: Determine whether the maximum number of iterations has been reached. If so, terminate the process and output the Pareto front.

[0125] Step 4: Randomly select several Pareto front solutions and conduct field experiments in the power station to obtain the vibration, swing and pressure pulsation data sets of the upper and lower frames, guide bearings, tailwater pipes and other unit components under different test conditions (decision variables). The locations of some measurement points are as follows: Figure 7 As shown, a stable state data set with one-to-one correspondence between Pareto front and hydraulic machinery vibration is formed;

[0126] The Pareto front selected is:

[0127] θ max，n =(K p，n, K i，n , K d，n , Y 1，n , Y 2，n , Y 3，n , Y 4，n , t 1，n , t 2，n , t 3，n )

[0128] n=1,2,...,N

[0129] N≤M

[0130] Among them, n corresponds to the optimal solution in the selected Pareto front, N represents the number of optimal solutions selected in the experiment, and M represents the number of optimal solutions in the Pareto front;

[0131] In addition, the stability state data set is specifically:

[0132] O n =(O n，1 , O n，2 ,...,O n，q ,...,O n，Q )

[0133] Among them, O n is θ max，n The hydraulic machinery vibration set corresponding to this optimal solution, Q represents the total number of measuring points of the unit to be tested, q = 1, 2, ..., Q, corresponding to different measuring points to be tested, a schematic diagram of some measuring points is shown in the figure below. Figure 7 , including upper guide swing measurement point 1, upper frame vibration measurement point 2, lower guide swing measurement point 3, lower frame vibration measurement point 4, draft tube inlet measurement point 5, outlet measurement point 6, and elbow measurement point 7. The stability state data set with a one-to-one correspondence between the Pareto front and the hydraulic machinery vibration is listed. Please refer to the table below for details.

[0134]

[0135]

[0136] Step 5: Create a comprehensive status evaluation system to obtain the final optimal solution

[0137] Specifically, a comprehensive condition evaluation system is created using patent experience and degradation indexes, and the optimal solution is determined by screening the Pareto front. When describing the evaluation system, the vibration, swing, and pressure pulsation values ​​of the measuring point are the values ​​of each index. Substituting the values ​​of each index into the degradation formula:

[0138]

[0139] Among them, β n，q represents the qth index value corresponding to the nth group of Pareto front solutions, represents the normalized value, β0 represents the good value of the indicator, β max Indicates the limit value of the indicator;

[0140] The weight of each indicator is given by expert experience:

[0141] M=(M1,M2,...,M q ,...,M Q )

[0142]

[0143] Among them, M q Indicates the weight value corresponding to the qth indicator, and the sum of the weight values ​​of Q indicators is 1;

[0144] Get the comprehensive status evaluation value of each group of Pareto solutions:

[0145]

[0146] F=min(F 1，q , F 2，q ,...,F n，q )

[0147] The solution of F is the final solution obtained from the stability test.

[0148] The test water heads are different, and the weights of various indicators given based on expert experience are different. Considering the corresponding changes in the dominant indicators of hydraulic machinery vibration, the optimal solution of the parameters will also change accordingly. The combination of subjective and objective methods of "model + test" increases the flexibility and breadth of startup parameter optimization.

[0149] The water diversion head of the present invention completes the optimization research on the startup rules of hydropower units for mechanical vibration suppression, effectively suppresses vibration, swing, and pressure pulsation, and improves the grid-connected speed of the unit and the performance of the startup process; the present invention establishes a high-precision simulation model of the power station unit speed regulation system based on the characteristic line method, providing a model basis for the unit speed regulator control method and transition process research, and can reduce the number of real machine tests in the transition process as much as possible through simulation; the present invention adopts a method that combines modeling and testing to ensure the rapid grid connection of the hydropower unit and the performance of the unit during the startup process, increase economic benefits, and help enhance the innovation ability of the hydropower station in the power generation industry; the optimization research results of the present invention can extend the life of the unit, ensure the economic benefits of the power station, and reducing mechanical vibration during the startup process can reduce unit wear to a certain extent, extend the life of the unit, and ensure the economic benefits of the power station to a certain extent.

[0150] The above embodiments are further elaborations and illustrations of the present invention for ease of understanding, and are not intended to limit the present invention in any way. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for suppressing vibration during the startup process of a hydraulic turbine based on MOPSO and stability test, characterized in that: The following steps are involved: S1, establish the startup model of the Francis unit; S2, obtain the objective function and decision variables for startup rule optimization; The two objective functions are the absolute value integral of the relative error of the unit speed and the absolute value integral of the relative error of the volute water pressure; the constraints are: the adjustment time t a Not greater than the upper limit of the adjustment time T u , the decision variable θ is not greater than the upper limit θ of the decision variable u And not less than the lower limit θ of the decision variable l ; S3, optimize the startup rules using optimization methods to obtain the Pareto optimal frontier; S4, select multiple groups of solutions in the Pareto front to conduct field tests at the power station to obtain the vibration conditions of the hydraulic machinery; S5, establish a comprehensive state evaluation system to determine the final optimal solution; based on the degradation index and expert experience, establish a comprehensive state evaluation system to screen the Pareto front and determine the final optimal solution. When describing the evaluation system, the vibration, swing, and pressure pulsation values ​​of each measuring point are the values ​​of each index.

2. The method for suppressing vibration during the startup process of a hydraulic turbine based on MOPSO and stability test according to claim 1 is characterized in that: The step S2 comprises the following steps: S21, select the absolute integral of the relative error of the speed of the unit and the absolute integral of the relative error of the volute water pressure as two objective functions. The specific objective functions are: Where k is the total number of samples, n(i) represents the real-time speed value corresponding to the sampling process, n(∞) represents the stable value of the unit speed, Hvol(i) is the real-time value of the volute water pressure during the sampling process, and Hvol_average is the average value of the volute water pressure during the sampling process. The first objective J1 of the objective function is to minimize the absolute integral of the relative error of the unit speed, and the second objective J2 is to minimize the absolute integral of the relative error of the volute water pressure; S22, obtain the decision variable, the decision variable θ is expressed as: <h2 style=";text-align:left;direction:ltr">θ=(K<h2 style=";text-align:left;direction:ltr"> p <h2 style=";text-align:left;direction:ltr"> K<h2 style=";text-align:left;direction:ltr"> i <h2 style=";text-align:left;direction:ltr"> K<h2 style=";text-align:left;direction:ltr"> d <h2 style=";text-align:left;direction:ltr"> (Y1, Y2, Y3, Y4, t1, t2, t3) Where K p , K i , K d They correspond to three parameters in the proportional link, integral link and differential link of the unit PID controller respectively. Y1 and t1 are the maximum value of the first guide vane opening and the time from the guide vane opening to the maximum value of the first guide vane opening, respectively. Y2 and t2 are the maximum value of the second guide vane opening and the time from the guide vane opening to the maximum value of the second guide vane opening, respectively. t3 is the time from the guide vane opening to the start of closing. Y3 and t4 are the guide vane opening when the PID controller is put into operation and the time from the guide vane opening to the start of PID control, respectively. Y4 is the stable opening of the guide vane after the unit has smoothly transitioned.

3. The method for suppressing vibration during the startup process of a hydraulic turbine based on MOPSO and stability test according to claim 1 is characterized in that: The step S4 is specifically to select multiple solutions in the Pareto front to conduct field tests in the power station to obtain the vibration, swing and pressure pulsation of the upper and lower frames, guide bearings and draft tube components of the turbine unit. The Pareto front θ selected by the test is max，n It can be expressed as: θ max，n =(K p，n ,K i，n ,K d，n ,Y 1，n ,Y 2，n ,Y 3，n ,Y 4，n ,t 1，n ,t 2，n ,t 3，n ) n=1,2,...,N N≤M Where n corresponds to the optimal solution in the Pareto frontier selected, N is the number of optimal solutions selected in the experiment, and M is the number of optimal solutions in the Pareto frontier; The stability state data set can be expressed as: The n =(O n，1 ,The n，2 ,...,O n，q ,...,O n，Q ) Where, O n represents θ max，n The hydraulic machinery vibration set corresponding to this optimal solution is Q, which is the total number of measured points of the unit to be tested, q = 1, 2, ..., and Q corresponds to different measured points to be tested.

4. The method for suppressing vibration during the startup process of a hydraulic turbine based on MOPSO and stability test according to claim 1 is characterized in that: The step S5 is specifically as follows: Substituting each indicator value into the degradation degree formula, it becomes a degradation degree value that describes the degree of quality of the indicator: Where, β n，q is the qth index value corresponding to the nth group of Pareto front solutions, is the normalized value, β0 is the good value of the indicator, β max is the limit value of the indicator; The weight value M of each indicator is given based on expert experience: M=(M1,M2,...,M q ,...,M Q ) Where M q is the weight value corresponding to the qth indicator, and the sum of the weight values ​​of Q indicators is 1; Calculate the comprehensive state evaluation value F of each group of Pareto solutions n，q : F=min(F 1，q ,F 2，q ,...,F n，q ) The solution corresponding to F is the final solution obtained from the stability test.

5. The method for suppressing vibration during the startup process of a hydraulic turbine based on MOPSO and stability test according to claim 1, characterized in that: The optimization method is a multi-objective particle swarm optimization algorithm, which uses the concept of Pareto dominance to find solutions to multi-objective problems.

6. The method for suppressing vibration during the startup process of a hydraulic turbine based on MOPSO and stability test according to claim 1, characterized in that: The startup model of the Francis turbine unit includes a water diversion system model and boundary conditions.

7. The method for suppressing vibration during the startup process of a hydraulic turbine based on MOPSO and stability test according to claim 6, characterized in that: The boundary conditions include a Francis turbine, a generator, a speed regulator and a reservoir.

Images