Hydroelectric generating unit power regulation transition process anti-vibration control method and system

Abstract

The application discloses a kind of water turbine generator power regulation transition process vibration avoidance control method and system, method includes: constructing the control model of water power station water power generation system unit power regulation;Key parameters during simulation are introduced into the control model;Control parameters of control model are used as decision variable, constraint condition of decision variable is set and target function of vibration avoidance performance and power regulation performance is set;According to the target function and constraint condition set, the control parameters of the control model are optimized and solved, and an optimal solution set is obtained;The optimal solution set is evaluated and the final optimal solution is selected, and the final optimal solution is used as the optimal control parameter to control the power of the control model.The application can significantly reduce the number of unit vibration zone crossing and reduce the vibration zone operation time while meeting the standard requirements of power regulation performance, improve the dynamic quality of unit power regulation, and improve the service life of unit operation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of hydroelectric generating set control, in particular to a hydroelectric generating set power regulation transition process anti-vibration control method and system. BACKGROUND

[0002] When the water turbine runs away from the optimal working condition, the hydraulic characteristics will deteriorate obviously, causing hydraulic phenomena such as flow separation, vortex, cavitation, and increasing vibration of the water turbine. If the water turbine runs in the vibration zone for a long time, it will lead to fatigue damage of the water turbine structure, loosening of fasteners and other safety problems, and even cause a major safety accident of water flooding the plant. In order to ensure the safe operation of the hydroelectric generating set, the on-site vibration zone test will be carried out in each large-scale hydropower plant to divide the vibration zone of the water turbine. Under the stable operation condition of the hydroelectric generating set, the operation personnel and the monitoring system of the hydropower station will avoid running in the vibration zone of the water turbine. However, during the load regulation process of the generating set, the hydroelectric generating set may cross the vibration zone. Due to the poor parameter setting of the control system, the generating set may run in the vibration zone for a long time. In addition, due to the dead zone of the control system and hydraulic fluctuations, the hydroelectric generating set may run in the vibration zone during the load regulation process (as shown in FIG. 1), which may cause damage to the key components of the water turbine. Figure 1

[0003] The current hydroelectric generating set anti-vibration operation strategy mainly focuses on the anti-vibration operation under the stable operation of the generating set. Generally, the power grid will issue a total power instruction to the hydropower plant, and the plant will distribute the total load regulation instruction to each generating set according to the actual load and operation of each generating set. In order to avoid the generating set running in the vibration zone, it is necessary to consider avoiding running in the vibration zone. Currently, methods such as enumeration, dynamic programming, genetic algorithm, and particle swarm optimization algorithm are used to distribute the power instruction value to prevent the power instruction value issued to each generating set from falling into the vibration zone of the hydroelectric generating set.

[0004] The patent application with publication number CN 115453931 A discloses a method and device for optimizing active power control of a hydroelectric generating set monitoring system. The method includes: constructing a power control model of the hydroelectric generating set monitoring system, importing basic parameters and initial conditions for solving the model; setting related parameters of the multi-objective genetic algorithm; obtaining an individual represented by a decision variable set, inputting the model to obtain a hydroelectric generating set active power regulation thread, and constructing an objective function according to the optimization index obtained from the regulation thread; solving the control model by using an improved multi-objective genetic algorithm to obtain an optimal solution set; and obtaining a decision variable set with the optimal objective function value from the optimal solution set to control the active power of the hydroelectric generating set. However, the optimization target of this scheme is only the response speed and power counter-regulation of the generating set, and the requirement of anti-vibration operation of the generating set during power regulation is not considered, which may cause the hydroelectric generating set to run in the vibration zone for a long time during power regulation, resulting in excessive vibration of the hydroelectric generating set and affecting the safety of the generating set. ​

[0005] In summary, the current research results in the vibration avoidance operation of the unit mainly focus on the distribution of the load instruction of the hydroelectric unit, and the transition process in the power regulation of the hydroelectric power is not considered. Although the target value of the power regulation of the hydroelectric unit can be effectively avoided from entering the vibration area of the unit, the unit may frequently cross the vibration area, stay in the vibration area for too long, and the regulation quality cannot meet the requirements of the power grid during the regulation process. Therefore, the above comprehensive factors need to be considered to propose a vibration avoidance control method for the transition process of the power regulation of the hydroelectric unit. SUMMARY

[0006] The technical problem solved by the present application is to provide a vibration avoidance control method for the transition process of the power regulation of the hydroelectric unit, which can significantly reduce the number of times of crossing the vibration area of the unit, reduce the operation time in the vibration area, improve the dynamic quality of the power regulation of the unit, and prolong the service life of the unit while meeting the standard requirements of the power regulation performance.

[0007] In order to solve the above technical problems, the technical scheme adopted by the present application is:

[0008] A vibration avoidance control method for the transition process of the power regulation of the hydroelectric unit, comprising the following steps:

[0009] Constructing a control model of the power regulation of the unit of the water delivery power generation system of the hydroelectric station;

[0010] Importing key parameters during simulation of the control model;

[0011] Taking the control parameters of the control model as decision variables, setting the constraint conditions of the decision variables, and setting the target functions of the vibration avoidance performance and the power regulation performance;

[0012] Optimizing and solving the control parameters of the control model according to the set target functions and constraint conditions to obtain an optimal solution set;

[0013] Evaluating the optimal solution set and selecting a final optimal solution, and using the final optimal solution as the optimal control parameters to perform power control on the control model.

[0014] Further, when constructing the control model of the power regulation of the unit of the water delivery power generation system of the hydroelectric station, it comprises:

[0015] Constructing a mathematical model of the monitoring system, and the control signal calculation formula of the PWM controller of the mathematical model of the monitoring system is as follows:

[0016]

[0017] In the formula, M is the output amplitude of the monitoring system, T is the pulse period, ΔP is the difference between the actual active power of the water turbine and the target value of the power regulation, sign is a symbol function, and Tk The pulse width of the PWM output, the calculation formula of sign is as follows:

[0018]

[0019] T k The calculation formula is as follows:

[0020]

[0021] In the formula, β is the adjustment parameter; T kmax is the maximum pulse width limit; T kmin is the minimum pulse width limit; and T kmin ≤T kmax ≤T.

[0022] The mathematical model of the speed regulator is constructed, and the transfer function is as follows:

[0023]

[0024] Wherein, u(s) is the Laplace transform of the input signal, y(s) is the Laplace transform of the output signal, T iy is the speed regulator opening given integrator time constant, T y is the governor system servomotor reaction time constant, and s is the Laplace operator;

[0025] The water turbine model is constructed, and the specific expression of the water turbine operation characteristic curve is as follows:

[0026] Q=f(α,H)

[0027] P=g(α,H)

[0028] In the formula, Q is the water turbine flow; P is the water turbine output; α is the guide vane opening; and H is the water turbine working water head;

[0029] The mathematical model of the water diversion system is constructed, and the formula is as follows:

[0030]

[0031] In the formula, h(s) is the Laplace transform of the water head of the water turbine, q(s) is the Laplace transform of the water turbine flow, T w is the water flow inertia time constant of the water conveying system, h f0 is the water head loss of the water conveying system, and H0 is the initial water turbine water head;

[0032] Combined with the monitoring system model, the speed regulator mathematical model, the water turbine mathematical model and the water diversion system model, the control model of the unit power regulation of the water and electricity station water conveying power generation system is obtained.

[0033] Further, the key parameters specifically include a water turbine operation comprehensive characteristic curve, a water flow inertia time constant T w , a speed regulating system servomotor reaction time constant T y , an initial operation water head H0, an initial output N t0 , an initial flow Q0, a water turbine water head, and a water turbine vibration zone under different water heads represented by a unit active power.

[0034] Further, the constraint conditions include a monitoring system and a speed regulator parameter magnitude constraint condition, and the expression is as follows:

[0035] M[T,T kmax ,T kmin ,β,T iy ]=0.1

[0036] Wherein, M[] represents a magnitude of a decision variable, T represents a pulse period, T kmax represents a maximum pulse width, T kmin represents a minimum pulse width, β represents an adjustment parameter, and T iy represents a speed regulator opening degree given integral time constant.

[0037] Further, the constraint conditions include a unit load distribution coefficient magnitude constraint condition, and the expression is as follows:

[0038] M[P1]=0.01

[0039] Wherein, M[] represents a magnitude of a decision variable, and P1 represents a load distribution coefficient.

[0040] Further, the constraint conditions include a monitoring system control parameter inequality constraint condition, and the expression is as follows:

[0041] T kmin ≤T kmax ≤T

[0042] Wherein, T represents a pulse period, T kmax represents a maximum pulse width, and T kmin represents a minimum pulse width.

[0043] Further, the vibration avoidance performance and power regulation performance objective functions include a vibration avoidance performance objective function, and the expression is as follows:

[0044] Minobj1(T,T kmax ,T kmin ,β,T iy ,P1)=ki+ti

[0045] Wherein, T represents a pulse period, T kmax represents a maximum pulse width, and T kminrepresents the minimum pulse width, β represents the adjustment parameter, T iy represents the speed governor opening given integral time constant, P1 represents the load distribution coefficient, T i represents the number of times of passing through the vibration zone corresponding to the regulating unit by the time domain curve of the regulating unit, t i represents the time of staying in the vibration zone corresponding to the regulating unit during the regulating process of the regulating unit.

[0046] Further, the vibration zone range corresponding to the regulating unit is expressed as follows:

[0047]

[0048] wherein, represents the vibration zone of the i-th regulating unit when the working water head is H0, P i1 min represents the lower limit of the first vibration zone of the i-th regulating unit, P i1 max represents the upper limit of the first vibration zone of the i-th regulating unit, P i2 min represents the lower limit of the second vibration zone of the i-th regulating unit, P i2 max represents the upper limit of the second vibration zone of the i-th regulating unit.

[0049] Further, the target function of the vibration avoidance performance and the power regulation performance includes the target function of the power regulation performance, and is expressed as follows:

[0050]

[0051] wherein, T represents the pulse period, T kmax represents the maximum pulse width, T kmin represents the minimum pulse width, β represents the adjustment parameter, T iy represents the speed governor opening given integral time constant, P1 represents the load distribution coefficient, T p represents the power regulation time, E p represents the power regulation accuracy, P s represents the power overshoot, T pmin , E pmin , P smin are minimum recommended values of the power regulation time, the power regulation accuracy and the power overshoot respectively.

[0052] Further, when the control parameters of the control model are optimized and solved according to the set target function and constraint condition, the control parameters of the control model are optimized and solved according to the set target function and constraint condition by using a multi-objective genetic algorithm, and the method comprises the following steps:

[0053] An initial decision variable satisfying the constraint condition is randomly generated, and the expression is as follows:

[0054] X i_0 =(T,T kmin ,T kmax ,T iy ,β,P1)

[0055] Wherein, X i_0 represents the i-th initial decision variable, i = 1, 2, …, N, N represents the total number of initial decision variables, T represents the pulse period, T kmax represents the maximum pulse width, T kmin represents the minimum pulse width, β represents the adjustment parameter, T iy represents the speed regulator opening given integral time constant, P1 represents the load distribution coefficient;

[0056] The decision variable is brought into the control model for simulation calculation to obtain the dynamic time domain waveform of the hydroelectric generating unit active power, and the number of vibration zone crossings k1, the time t1 of staying in the corresponding vibration region during unit regulation, the unit power regulation time T p , the power regulation accuracy E p , and the power overshoot P s are calculated according to the waveform curve;

[0057] The number of vibration zone crossings k1, the time t1 of staying in the corresponding vibration region during unit regulation, the unit power regulation time T p , the power regulation accuracy E p , and the power overshoot P s are brought into the objective function formula of the vibration avoidance performance and the power regulation performance to obtain the corresponding objective function values obj1 and obj2;

[0058] The priority of each decision variable is determined according to the non-dominated fast sorting and crowding degree evaluation of the two objective function values of the decision variable, and then the decision variables with high priority are screened to form a parent population according to the results of non-dominated fast sorting and crowding degree evaluation, and a new generation population is generated using the parent population;

[0059] If the maximum evolution generation has been reached, the new generation population is taken as the optimal solution set, otherwise, the new generation population is merged with the parent population, and for each individual in the merged population, the step of bringing the decision variable into the control model for simulation calculation is performed again until the maximum evolution generation is reached.

[0060] Further, when evaluating the optimal solution set and selecting the final optimal solution, the closeness degree of each optimal solution in the optimal solution set is calculated, and then the optimal solution with the largest closeness degree is selected, including the following steps:

[0061] The normalized initial matrix J is constructed, and the expression is as follows:

[0062]

[0063] Wherein, x ij represents the value of the i-th solution in the optimal solution set on the j-th objective function, i=1, 2, …, n, n is the number of solutions in the optimal solution set, j=1, 2, j is 1 represents the objective function of the vibration avoidance performance, j is 2 represents the objective function of the power regulation performance;

[0064] The maximum value of each column element in the matrix J is set as the positive ideal point J + , and the minimum value of each column element in J is set as the negative ideal point J - , and the Euclidean distance of each element with the positive ideal point J + and the negative ideal point J - is calculated respectively, and the expression is as follows:

[0065]

[0066] The proximity degree corresponding to each solution in the optimal solution set is calculated, and the expression is as follows:

[0067]

[0068] Wherein, T i represents the proximity degree corresponding to the i-th solution in the optimal solution set.

[0069] The application also provides a hydroelectric generating set power regulation transition process vibration avoidance control system, which comprises a microprocessor and a computer readable storage medium connected with each other, and the microprocessor is programmed or configured to execute any one of the hydroelectric generating set power regulation transition process vibration avoidance control methods.

[0070] Compared with the prior art, the application has the following advantages:

[0071] The application is based on the control model of the unit power regulation of the water power generation system of a hydropower station, constructs the vibration avoidance performance objective function and the power regulation performance objective function, optimizes the control parameters by using a multi-objective optimization algorithm to obtain an optimal solution set, then performs scheme evaluation and optimal parameter selection on the optimal solution set, and finally uses the obtained optimal parameters to regulate power, so that the power regulation performance can meet the requirements of the power regulation performance index, the number of times of passing through the vibration zone and the running time in the vibration zone of the unit are reduced, and the service life of the unit and the safety of the power station are improved.

[0072] The application divides the vibration area data of the hydroelectric generating set according to the actual situation, and uses the data as a target function of power regulation of the hydroelectric generating set, adopts the multi-target optimization strategy, and obtains the optimal parameters considering the power regulation performance and vibration avoidance demand, so as to shorten the running time of the hydroelectric generating set in the vibration area during the power regulation process and reduce the damage of vibration to the structural components of the hydroelectric generating set on the basis of improving the power regulation performance. BRIEF DESCRIPTION OF DRAWINGS

[0073] Figure 1 The figure is a schematic diagram of power regulation transition process of the hydroelectric generating set.

[0074] Figure 2 The figure is a flow chart of the method of the application.

[0075] Figure 3 The figure is a schematic diagram of the water delivery system of the hydroelectric station.

[0076] Figure 4 The figure is a power regulation principle in the power control mode of the monitoring system.

[0077] Figure 5 The figure is a schematic diagram of the mathematical model of the governor.

[0078] Figure 6 The figure is a model based on the operation characteristic curve of the water turbine.

[0079] Figure 7 The figure is a whole model.

[0080] Figure 8 The figure is a schematic diagram of the vibration area of the hydroelectric generating set.

[0081] Figure 9 The figure is a schematic diagram of the target function value of each solution in the optimal solution set.

[0082] Figure 10 The figure is a schematic diagram of the calculation result of the closeness degree. DETAILED DESCRIPTION

[0083] The application will be further described below in combination with the accompanying drawings and specific preferred embodiments, but the protection scope of the application is not limited by the embodiments.

[0084] Embodiment one

[0085] The embodiment proposes a power regulation transition process anti-vibration control method for a hydroelectric generating unit. Based on a power control model of a hydroelectric station monitoring system, an anti-vibration performance objective function and a power regulation performance objective function are constructed. Key control parameters are optimized by using an improved non-dominated sorting genetic algorithm II (NSGA-II). A TOPSIS (Technique for Order Preference by Similarity to Ideal Solution) decision theory is introduced to evaluate the optimal solution set and select parameters. The recommended parameters are used for power regulation, so that the power regulation performance can meet the requirements of the power regulation performance index, while reducing the number of times the unit passes through the vibration zone and the time of operation in the vibration zone, and improving the unit life and the safety of the power station. As shown in Figure 2 the following steps are included:

[0086] Step 1: Construct a control model for the power regulation of the hydroelectric station water delivery and power generation system.

[0087] The hydroelectric station water delivery and power generation system, as shown in Figure 3 , includes a water delivery system, a water turbine, and control equipment. The active power control system of a hydroelectric station is a complex nonlinear system that integrates hydraulic, mechanical, and electrical processes. The control equipment includes a monitoring system and a water turbine governor. The active power response characteristics of the hydroelectric generating unit involved in the power regulation process are closely related to the characteristics of each subsystem, and each subsystem is coupled. To accurately simulate the power regulation transition process of the hydroelectric generating unit, the step S1 of the embodiment combines the results of the inventors' previous research (Fu L, Bao HY, Huang B. Simulation research on active power regulation of hydroelectric generating unit monitoring system [J]. Journal of Hydroelectric Engineering, 2020, 39(01): 62-71.) to establish a detailed nonlinear mathematical model of the hydroelectric generating unit power regulation that considers the nonlinear characteristics and mutual coupling of each link, including the following steps:

[0088] Step 1.1: Construct a mathematical model of the monitoring system:

[0089] The active power closed-loop regulation principle of the hydroelectric station monitoring system is shown in Figure 4 . After receiving the power regulation instruction from the power grid, the hydroelectric station monitoring system sends the active regulation instruction to the relevant unit local control unit (LCU). After receiving the power regulation instruction, the unit LCU adjusts the pulse width modulation (PWM) method. When the absolute value of the actual active power deviation from the target value is greater than the regulation dead zone, the LCU sends a power "increase / decrease" pulse signal to the governor in real time until the active power deviation enters the dead zone and stops the order. Finally, the governor adjusts the guide vane opening to change the output of the water turbine.

[0090] The control signal u(t) of the PWM controller is calculated as follows:

[0091]

[0092] where M is the output amplitude of the monitoring system, in MW; T is the pulse period, in s; ΔP is the difference between the actual hydraulic turbine active power and the power regulation target value, in MW; sign is the sign function; T k is the pulse width of the PWM output (k = 0, 1, 2,...). The calculation formula of sign is shown in equation (2):

[0093]

[0094] T k is calculated as follows:

[0095]

[0096] where β is the regulation parameter; T kmax is the maximum pulse width limit, in s; T kmin is the minimum pulse width limit, in s; and there are T kmin ≤ T kmax ≤ T, which can be manually set.

[0097] Step 1.2: Constructing the governor mathematical model:

[0098] The hydraulic turbine governor is composed of a regulator and an electro-hydraulic servo system, as shown in Figure 5 . Its transfer function is shown in equation (4).

[0099]

[0100] where u(s) is the Laplace transform of the input signal, y(s) is the Laplace transform of the output signal, T iy is the governor opening degree given integrator time constant, T y is the governor servo reaction time constant, and s is the Laplace operator.

[0101] In the power control opening degree mode, the main task of the governor is to act as an actuator, receive the power regulation instructions from the monitoring system, adjust the hydraulic turbine guide vane opening degree, and thus change the output to achieve the power regulation purpose.

[0102] Figure 5 The input signal u is the command output by the unit LCU, and the output signal y is the relative value of the guide vane opening degree. The regulator part K D is the differential gain; K P is the proportional gain; K IKp is integral gain; T lv Tb is differential link damping time constant, unit is second (s); b p Kd is steady-state droop coefficient; T iy Tf is governor opening given integrator time constant, unit is second (s); T y Tm is governing system servomotor response time constant, unit is second (s). The above parameters are generally set manually, wherein T y Tb is obtained by field measurement.

[0103] After the LCU outputs the increase or decrease pulse signal, the governor issues a guide vane opening step according to the accepted pulse command, and the step value is ±100%. For a step signal, a ramp signal is output, and the slope of the ramp signal is determined by T iy , and the calculated guide vane opening signal is output to the electro-hydraulic servo system.

[0104] The electro-hydraulic servo system controls the guide vane servomotor to adjust the guide vane opening after the input guide vane opening adjustment electrical signal is converted by the electro-hydraulic conversion link and amplified by the hydraulic amplifier.

[0105] Step 1.3: Constructing a hydraulic turbine model

[0106] In this embodiment, the hydraulic turbine operating characteristic curve is used for modeling, and the differential link in the water conveyance system is changed to an integral link to eliminate algebraic loops. The model block diagram based on the hydraulic turbine operating characteristic curve is shown in Figure 6 .

[0107] Since the hydraulic turbine is operated in the normal operating range in the active power regulation numerical simulation of the hydropower station, in order to prevent algebraic loops in the model, the hydraulic turbine operating characteristic curve is used to describe the nonlinear characteristics of the hydraulic turbine, and the head H and the power P are set as the longitudinal and transverse coordinates.

[0108] The hydraulic turbine operating characteristic curve reflects the relationship between the flow rate, output, head, and guide vane opening of the hydraulic turbine, and the specific expression is as follows:

[0109] Q = f (α, H) (5)

[0110] P = g (α, H) (6)

[0111] In the formula, Q is the flow rate of the hydraulic turbine, unit is m 3 / s; P is the output of the hydraulic turbine, unit is MW; α is the guide vane opening, unit is mm; H is the working head of the hydraulic turbine, unit is m.

[0112] Step 1.4: Constructing a water conveyance system mathematical model

[0113] The water conveyance system mathematical model is shown in formula (7).

[0114]

[0115] where h(s) is the Laplace transform of the water head of the water turbine, q(s) is the Laplace transform of the flow of the water turbine, T w is the water flow inertia time constant of the water delivery system, h f0 is the water head loss of the water delivery system, H0 is the initial water head of the water turbine, and s is the Laplace operator.

[0116] Step 1.5: Constructing the overall mathematical model:

[0117] Combining the monitoring system model, the governor mathematical model, the water turbine mathematical model, and the water delivery system model, the overall mathematical model of the power regulation of the hydroelectric generating set monitoring system is obtained, as shown in Figure 7 How to combine these models to obtain the overall mathematical model is known to those skilled in the art, and the embodiment does not involve improvement of the specific process, so the specific model combination process will not be described again.

[0118] Step 2: Importing the key parameters of the control model for the simulation of the power regulation of the water delivery and power generation system of the hydroelectric power station.

[0119] The characteristic parameters required when importing the simulation control model for simulation include the comprehensive characteristic curve of the water turbine operation, the water flow inertia time constant T w , the servomotor reaction time constant T y of the speed regulation system. At the same time, the initial working conditions of the model are set, including the initial working water head H0, the initial output N t0 , and the initial flow Q0.

[0120] For the vibration avoidance operation control of the hydroelectric generating set, the vibration zone data of the hydroelectric generating set also need to be imported, mainly including the water head of the water turbine [H1, H2, …H n ], and the vibration zone of the water turbine at different water heads represented by the active power of the generating set, for example represents the vibration zone of the first to fourth units of the water turbine at the first to n water heads.

[0121] Step 3: Setting the parameters of the multi-objective genetic algorithm, specifically taking the control parameters of the control model as the decision variables, setting the constraint conditions of the decision variables, and setting the objective functions of the vibration avoidance performance and the power regulation performance.

[0122] In this embodiment, the multi-objective genetic algorithm adopts the improved non-dominated gene sorting algorithm NSGA-II (Non-dominated Sorting Genetic Algorithms II), and when setting the parameters of the multi-objective genetic algorithm, the following is specifically included:

[0123] Step 3.1: Initializing the NSGA-II parameters:

[0124] The number of objective functions nobj is set to 2, the population quantity npop is set to 100, the maximum iteration number maxit is set to 60, the crossover ratio pc is set to 0.8, the mutation probability mu is set to 0.05, and the decision variable nvar is set to 6. In this embodiment, the decision variable nvar is set to 6 because the key control parameters to be solved include the pulse period T, the maximum pulse width T kmax , the minimum pulse width T kmin , the adjustment parameter β, the governor opening degree given integral time constant T iy , and the load distribution coefficient P1. After the key control parameters to be solved by the model are taken as the decision variables, the individual set constituted by the decision variables can be expressed as:

[0125] X = (T, T kmin , T kmax , T iy , β, P1)

[0126] Step 3.2: Setting the constraint condition, that is, setting the constraint condition of the decision variable:

[0127] Generally, the precision of the control parameter in the actual operation of the hydropower plant is set to one significant digit after the decimal point. The magnitude constraint condition of the control parameter of the monitoring system and the governor can be expressed as:

[0128] M[T, T kmax , T kmin , β, T iy ] = 0.1 (8)

[0129] The magnitude constraint adjustment of the unit load distribution coefficient can be expressed as

[0130] M[P1] = 0.01 (9)

[0131] In the formula (8) and the formula (9), M[] represents the magnitude of the decision variable.

[0132] The maximum pulse width T kmax , the minimum pulse width T kmin , and the pulse period T of the control parameter of the monitoring system have the following inequality constraint conditions:

[0133] T kmin ≤ T kmax ≤ T

[0134] Step 3.3: Setting the objective function, specifically setting the objective function of the anti-vibration performance and the power regulation performance, wherein:

[0135] For the vibration avoidance performance objective function of the present embodiment, there are multiple vibration zones for the hydroelectric plant unit, which are generally obtained through field tests after the unit is put into operation, and the vibration zones of the hydroelectric unit change with the working water head, as shown in the schematic diagram of the unit vibration zones. Figure 8 represents the vibration zone of the i th participating unit when the working water head is H 0, i1 min represents the lower limit of the first vibration zone of the i th participating unit, P i1 max represents the upper limit of the first vibration zone of the i th participating unit, P i2 min represents the lower limit of the second vibration zone of the i th participating unit, P i2 max represents the upper limit of the second vibration zone of the i th participating unit. When the unit participates in power regulation, the dynamic time-domain curve of the active power is tracked, and the number of times that the time-domain curve of the participating unit crosses the corresponding vibration zone of the unit is output as k i , the time that the unit stays in the corresponding vibration zone during regulation is output as t i , and the range of the corresponding vibration zone of the unit can be represented as:

[0136]

[0137] The above formula can represent the vibration zone range of the i th participating unit when the working water head is H 0.

[0138] The objective function of the unit vibration avoidance performance can be represented as:

[0139] Minobj1(T,T kmax ,T kmin ,β,T iy ,P1)=k i +t i (11)

[0140] All the indexes in formula (11) are the smaller the better.

[0141] For the power regulation performance objective function of the present embodiment, it is composed of three indexes: regulation time T p , regulation accuracy E p , and overshoot P s , and can be represented as:

[0142]

[0143] In the present embodiment, the power regulation time T p ​The total time length from the start of active power regulation action to the time when the active power of the hydroelectric generating unit enters the dead zone, and the active power of the hydroelectric generating unit entering the dead zone refers to the deviation of the active power of the hydroelectric generating unit from the target value being within ±1% of the rated power.

[0144] In this embodiment, the power regulation accuracy E is defined as p The ratio of the average value of the active power of the hydroelectric generating unit after entering the dead zone to the rated power of the unit.

[0145] In this embodiment, the power overshoot P is defined as s As follows:

[0146] Detecting whether the grid instruction is a load reduction command or a load increase command, if the regulation instruction is a load reduction command, obtaining the minimum value in the load reduction process, and setting the difference between the power regulation target value and the minimum value as the power overshoot; if the regulation instruction is a load increase command, obtaining the maximum value in the load increase process, and setting the difference between the maximum value and the power regulation target value as the power overshoot.

[0147] T pmin , E pmin , and P smin are respectively the minimum recommended values of the power regulation time, the power regulation accuracy, and the power overshoot specified in the standard.

[0148] Step 4: Using the improved non-dominated sorting genetic algorithm NSGA-II, the control parameters of the control model of the unit power regulation of the water power generation system of the hydropower station are optimized and solved according to the set objective function and constraint conditions, specifically including:

[0149] Step 4.1: generating initial decision variables:

[0150] Randomly generating N initial decision variables that satisfy the constraint conditions, and the expression is as follows:

[0151] X i_0 = (T, T kmin , T kmax , T iy , β, P1) (13)

[0152] Where X i_0 represents the i-th initial decision variable, i = 1, 2, …, N, T represents the pulse period, T kmax represents the maximum pulse width, T kmin represents the minimum pulse width, β represents the regulation parameter, T iy represents the integral time constant of the governor opening degree given, and P1 represents the load distribution coefficient.

[0153] Step 4.2: calculating the objective function:

[0154] The decision variable is brought into the control model built in step 1 to perform simulation calculation, to obtain the dynamic time domain waveform of the active power of the hydroelectric generating unit, and the number of times k1 of passing through the vibration zone, the time t1 of staying in the corresponding vibration zone during the unit regulation process, and the unit power regulation time T are calculated according to the waveform curve p , the power regulation accuracy E p , and the power overshoot P s . How to perform simulation calculation on the control model to obtain the dynamic time domain waveform of the active power of the hydroelectric generating unit, and how to calculate these parameters according to the waveform curve of the dynamic time domain waveform of the active power of the hydroelectric generating unit are all known to those skilled in the art, and the improvement of the specific calculation process is not involved in the embodiment, so the specific calculation process will not be described here.

[0155] The number of times k1 of passing through the vibration zone of each decision variable and the time t1 of staying in the corresponding vibration zone during the unit regulation process are brought into formula (11) to obtain the value obj1 of the vibration avoidance performance objective function corresponding to each decision variable, and the unit power regulation time T p , the power regulation accuracy E p , and the power overshoot P s are brought into formula (12) to obtain the value obj2 of the power regulation performance objective function corresponding to each decision variable.

[0156] Step 4.3: Non-dominated fast sorting and crowding distance evaluation

[0157] Each decision variable is taken as an individual, and non-dominated fast sorting and crowding distance evaluation are performed according to the two objective function values of the individual to determine the priority of the individual.

[0158] In the embodiment, non-dominance is defined as: for two different individuals X i and X j in the population, the two objective function values obtained by bringing the two individuals into the model both satisfy Minobj k (X i )≤Minobj k (X j ), then X i is said to dominate X j . The embodiment generates multiple “levels” or “fronts” (Pareto fronts) of individuals in the population through non-dominated fast sorting. The first front contains all non-dominated solutions, the second front contains solutions dominated by the first front, and so on, thereby helping to identify better solutions in the population, and the less dominated the solution is by other solutions, the higher the sorting level of the solution.

[0159] In this embodiment, the crowdedness is defined as follows: first, the population is sorted according to the ascending order of the size of each objective function value, and then the crowdedness is evaluated, that is, the boundary solution of each objective function (i.e., the solution with the maximum value and the minimum value) is assigned a value of infinite distance, and all other intermediate solutions are assigned an absolute difference value equal to the normalized target function value of the two adjacent solutions. In this embodiment, through the crowdedness evaluation, the "crowdedness" between individuals in the same level of the front is further evaluated to measure the density of individuals in the target space.

[0160] In this embodiment, after determining the priority of each individual, N high-priority individuals need to be selected, specifically, individuals in the higher level of the non-dominated quick sort are selected, and individuals with low crowdedness are preferentially retained in the same front to enhance the diversity of the population. Thus, the first generation population G = (X1, X2, X3, …, XN) containing N individuals is obtained. n ), and X i = (T, T kmax , T kmin , T iy , β, P1), i = 1, 2, …, N, the individuals in the population are divided into priority through non-dominated quick sorting of the objective function value and crowdedness evaluation, thereby helping to select individuals with high priority in the evolutionary algorithm to generate the next generation population.

[0161] Step 4.4: generating a new generation population:

[0162] The first generation population is used as the parent population, and genetic operations are performed to generate a new generation population, and the new generation population is used as the child population of the parent population. How to perform genetic operations on the parent population to generate a new generation population is known to those skilled in the art, and the embodiment does not involve improvements in the specific implementation process of genetic operations, and the specific process is not described here.

[0163] After generating the new generation population, it is detected whether the current evolution generation number reaches the maximum evolution generation number. If yes, the new generation population is output as the optimal solution set, otherwise, the parent population and the child population are merged, and steps 4.2 to 4.4 are performed on the merged population to recalculate the objective function value, and the next generation population is obtained through non-dominated quick sorting and crowdedness evaluation. The next generation population is used as the parent population, and a new generation population is generated through genetic operations, and the new generation population is used as the child population of the parent population, until the current evolution generation number reaches the maximum evolution generation number, and the optimal population is output as the optimal solution set. As shown in Figure 9 The two objective function values of each solution in the optimal solution set show a trend that the power regulation performance objective function value increases as the vibration avoidance performance objective function value decreases in the two-dimensional coordinate system with the vibration avoidance performance objective function as the vertical axis and the power regulation performance objective function as the horizontal axis.

[0164] Step 5: evaluating the optimal solution set and selecting a final optimal solution, and using the final optimal solution as an optimal control parameter to perform power control on the control model.

[0165] In this embodiment, in order to obtain an optimal solution in the optimal solution set shown in Figure 9 , a TOPSIS (Technique for Order Preference by Similarity to Ideal Solution) decision theory is used to obtain the optimal solution, and specifically includes the following steps:

[0166] Step 5.1: homogenization processing is performed on the optimal solution set:

[0167] A normalized initial matrix J is constructed, and the formula is:

[0168]

[0169] In the formula, x ij represents the value of the i-th solution in the optimal solution set on the j-th objective function, i = 1, 2, …, n, n is the number of solutions in the optimal solution set, j = 1, 2, j = 1 represents the objective function of the vibration avoidance performance, and j = 2 represents the objective function of the power adjustment performance.

[0170] Step 5.2: calculating the Euclidean distance:

[0171] The maximum value of each column element in the matrix J is set as the positive ideal point J + , and the minimum value of each column element in J is set as the negative ideal point J - . The Euclidean distance of each element from the positive ideal point J + and the negative ideal point J - is calculated, and the expression is as follows:

[0172]

[0173] Step 5.3: calculating the closeness degree:

[0174] According to the Euclidean distance from the positive ideal point J + and the negative ideal point J - , the closeness degree of each solution in the optimal solution set is calculated, and the expression is as follows:

[0175]

[0176] In the above formula, T i represents the closeness degree corresponding to the i-th solution in the optimal solution set. The greater the value of the closeness degree, the better the control scheme of the solution in the optimal solution set. As shown in Figure 10 , the figure is a graph according to Figure 9The closeness of each solution calculated by the optimal solution set calculation is pasted in by Figure 10 It can be seen that the closeness of the 20th solution is the largest, indicating that the control scheme of the 20th solution is the optimal scheme, so the pulse period T, the maximum pulse width T kmax , the minimum pulse width T kmin , the adjustment parameter β, and the governor opening degree given integral time constant T iy , the load distribution coefficient P1 are pasted in by the control model constructed in step S1 or the actual control system corresponding to the control model to perform power control, so that the best power regulation performance and vibration avoidance performance can be achieved.

[0177] Embodiment two

[0178] The embodiment provides a hydroelectric generating set power regulation transition process vibration avoidance control system, which comprises a microprocessor and a computer readable storage medium connected with each other, and the microprocessor is programmed or configured to execute the hydroelectric generating set power regulation transition process vibration avoidance control method in the embodiment one.

[0179] To sum up, the embodiment provides a hydroelectric generating set power regulation transition process vibration avoidance control method and system, which has the following advantages:

[0180] 1. The power regulation performance target function and the vibration avoidance performance target function are set at the same time, so that the power regulation performance and the vibration avoidance demand are considered.

[0181] 2. The NSGA-II is used for optimizing the control parameters, and the TOPSIS decision theory is used for evaluating and selecting the optimal solution set, so that the optimal solution of the control parameters considering the power regulation performance and the vibration avoidance performance is obtained. The embodiment provides a new technical means for the active power control of the hydroelectric generating set, and improves the power regulation quality of the hydroelectric generating set monitoring system and the operation safety of the hydroelectric generating set.

[0182] The above only describes the preferred embodiments of the present application, and the protection scope of the present application is not limited to the above-mentioned embodiments. Any technical scheme falling within the idea of the present application belongs to the protection scope of the present application. It should be noted that, for ordinary skilled persons in the technical field, some improvements and decorations without departing from the principle of the present application are also considered as the protection scope of the present application.

Claims

1. A method of vibration avoidance control for power regulating transient processes of a hydroelectric power unit, characterized by The method comprises the following steps: A control model of power regulation of a water power station water delivery power generation system is constructed; Key parameters in simulation of the control model are introduced; Control parameters of the control model are taken as decision variables, constraint conditions of the decision variables are set, and an objective function of vibration avoidance performance and power regulation performance is set; Optimization solving of the control parameters of the control model is performed according to the set objective function and constraint conditions, and an optimal solution set is obtained; The optimal solution set is evaluated, and a final optimal solution is selected, and the final optimal solution is used as an optimal control parameter to perform power control on the control model.

2. The hydroelectric generating unit power regulation transient process anti-hunting control method of claim 1, wherein, When the control model of power regulation of the water power station water delivery power generation system is constructed, the following steps are included: A mathematical model of a monitoring system is constructed, and a control signal calculation formula of a PWM controller of the mathematical model of the monitoring system is as follows: wherein M is the output amplitude of the monitoring system; T is the pulse period; Δ P is the difference between the actual water turbine active power and the power regulation target value; sign is the sign function; T k is the pulse width of the PWM output, k = 0, 1, 2, …, and the calculation formula of sign is as follows: T k The calculation formula is as follows: wherein β is a tuning parameter; T kmax is a maximum pulse width limit; T kmin is a minimum pulse width limit; and T kmin ≤ T kmax ≤ T ; A mathematical model of a speed regulator is constructed, and a transfer function of the mathematical model of the speed regulator is as follows: wherein is the Laplace transform of the input signal, is the Laplace transform of the output signal, T iy is the governor opening set integrator time constant, T y is the governor system servomotor response time constant, s is the Laplace operator; A water turbine model is constructed, and a specific expression of a water turbine operation characteristic curve is as follows: In the formula: Q Q is the flow of the hydraulic turbine; P P is the output of the hydraulic turbine; α θ is the guide vane opening; H H is the working water head of the hydraulic turbine; A mathematical model of a water delivery system is constructed, and a formula of the mathematical model of the water delivery system is as follows: wherein is the Laplace transform of the water turbine head, is the Laplace transform of the water turbine flow rate, is the water flow inertial time constant of the water delivery system, is the head loss of the water delivery system, is the initial water turbine head; The control model of power regulation of the water power station water delivery power generation system is obtained in combination of the mathematical model of the monitoring system, the mathematical model of the speed regulator, the water turbine mathematical model, and the mathematical model of the water delivery system.

3. The hydroelectric generating unit power regulation transition process anti-hunt control method of claim 1, wherein, The key parameters specifically include a water turbine operation comprehensive characteristic curve, a water flow inertia time constant T w , a governing system servomotor reaction time constant T y , an initial working water head H 0, an initial output N t0 , an initial flow Q 0, a water turbine water head, and a vibration zone of the water turbine under different water heads represented by a unit active power.

4. The hydroelectric generating unit power regulation transition process anti-hunt control method of claim 1, wherein, The constraint conditions include a constraint condition of magnitudes of parameters of the monitoring system and the speed regulator, and an expression of the constraint condition of the magnitudes of the parameters is as follows: A constraint condition of magnitudes of load distribution coefficients of units, and an expression of the constraint condition of the magnitudes of the load distribution coefficients is as follows: where M[ ] represents the magnitude of the decision variable, T represents the pulse period, T kmax represents the maximum pulse width, T kmin represents the minimum pulse width, β represents the adjustment parameter, T iy represents the speed regulator opening given integral time constant; Wherein, M[] represents a magnitude of a decision variable, and P1 represents a load distribution coefficient. The constraint conditions include an inequality constraint condition of control parameters of the monitoring system, and an expression of the inequality constraint condition of the control parameters of the monitoring system is as follows: The objective function of the vibration avoidance performance and the power regulation performance includes a vibration avoidance performance objective function, and an expression of the vibration avoidance performance objective function is as follows: where T represents a pulse period, T kmax represents a maximum pulse width, T kmin represents a minimum pulse width.

5. The hydroelectric generating unit power regulation transition process anti-hunt control method of claim 1, wherein, An expression of a vibration zone range corresponding to a reference unit is as follows: where T represents the pulse period, T kmax represents the maximum pulse width, T kmin represents the minimum pulse width, β represents the adjustment parameter, T iy represents the speed regulator opening degree given integral time constant, P1 represents the load distribution coefficient, represents the number of times the time-domain curve of the reference regulating unit crosses the corresponding vibration zone of the unit, represents the time spent in the corresponding vibration zone during the adjustment process of the reference regulating unit.

6. The hydroelectric generating unit power regulation transition process anti-hunt control method of claim 5, wherein, The objective function of the vibration avoidance performance and the power regulation performance includes a power regulation performance objective function, and an expression of the power regulation performance objective function is as follows: wherein, Hwi represents the working head of the i-th participating unit, H Hwi represents the working head of the i-th participating unit, Hwi represents the lower bound of the first vibration zone of the i-th participating unit, Hwi represents the upper bound of the first vibration zone of the i-th participating unit, Hwi represents the lower bound of the second vibration zone of the i-th participating unit, Hwi represents the upper bound of the second vibration zone of the i-th participating unit.

7. The hydroelectric generating unit power regulation transition process anti-hunt control method of claim 1, wherein, When the optimization solving of the control parameters of the control model is performed according to the set objective function and constraint conditions, the optimization solving of the control parameters of the control model is performed according to the set objective function and constraint conditions by using a multi-objective genetic algorithm, and the following steps are included: where T represents a pulse period, T kmax represents a maximum pulse width, T kmin represents a minimum pulse width, β represents an adjustment parameter, T iy represents a speed governor opening degree given integral time constant, P1 represents a load distribution coefficient, T p represents a power adjustment time, E p represents a power adjustment accuracy, P s represents a power overshoot amount, T pmin , E pmin , P smin are minimum recommended values of the power adjustment time, the power adjustment accuracy, and the power overshoot amount, respectively.

8. The hydroelectric generating unit power regulation transition process anti-hunt control method of claim 1, wherein, An initial decision variable satisfying the constraint conditions is randomly generated, and an expression of the initial decision variable is as follows: Non-dominated fast sorting and congestion evaluation are performed according to two objective function values of the decision variables to determine a priority of each decision variable, then a high-priority decision variable is screened according to a result of the non-dominated fast sorting and the congestion evaluation to form a parent population, and a new generation population is generated by using the parent population; X i_0 =( T , T kmin , T kmax , T iy , β ， P 1) wherein X i_0 denotes the i-th initial decision variable, i = 1, 2,..., N , N denotes the total number of initial decision variables, T denotes the pulse period, T kmax denotes the maximum pulse width, T kmin denotes the minimum pulse width, β denotes an adjustment parameter, T iy denotes the governor opening degree given integral time constant, P1 denotes a load distribution coefficient; The decision variable is brought into the control model for simulation calculation to obtain a dynamic time domain waveform of the active power of the hydroelectric generating unit, and the number of times of vibration zone crossing is calculated according to the waveform curve k 1, time of staying in the corresponding vibration region in the unit regulating process t 1, unit power regulating time T p , power regulating accuracy E p , power overshoot P s ; the number of times the vibration region is crossed k 1, the time spent in the corresponding vibration region during unit regulation t 1, unit power regulation time T p , power regulation accuracy E p , power overshoot P s , the objective function formula of the vibration avoidance performance and the power regulation performance, to obtain the corresponding objective function values obj1 and obj2; If a maximum evolution number has been reached, the new generation population is taken as the optimal solution set, otherwise, the new generation population and the parent population are combined, and the step of bringing the decision variables into the control model to perform simulation calculation is performed again for each individual of the combined population until the maximum evolution number is reached. When the optimal solution set is evaluated and a final optimal solution is selected, each optimal solution in the optimal solution set is calculated for closeness, and then an optimal solution with the maximum closeness is selected, and the following steps are included:

9. The hydroelectric generating unit power regulation transition process anti-hunt control method of claim 1, wherein, A normalized initial matrix J is constructed, and an expression of the normalized initial matrix J is as follows: Closeness corresponding to each solution in the optimal solution set is calculated, and an expression of the closeness is as follows: wherein denotes the value of the i th solution in the set of optimal solutions on the j th objective function, , n is the number of solutions in the set of optimal solutions, , j is 1 for the objective function of the vibration avoidance performance, j is 2 for the objective function of the power regulation performance; Set the maximum value of each column element in matrix J as the positive ideal point J + , and set the minimum value of each column element in J as the negative ideal point J - , respectively calculate the Euclidean distance of each element with the positive ideal point J + and the negative ideal point J - , and the expression is as follows: ​ wherein, indicates the closeness of the i-th solution in the optimal solution set. i set.

10. A hydroelectric generating unit power regulation transient vibration avoidance control system characterized by, The computer readable storage medium comprises a microprocessor and a computer readable storage medium connected to each other, and the microprocessor is programmed or configured to execute the method for avoiding vibration during power adjustment transition process of a hydroelectric generating set according to any one of claims 1-9.

Images