Hydraulic turbine governor control method and device considering structural response

Abstract

The application belongs to the technical field of hydroelectric power generation, and specifically discloses a water turbine governor control method and device considering structural response. The method comprises the following steps: determining a multi-index control parameter planning model based on a plurality of performance indexes of a unit runner; the plurality of performance indexes comprise a pressure pulsation amplitude and a dynamic stress amplitude, and further comprise a steady-state rotating speed regulation time and / or a rotating speed fluctuation frequency; taking a governor control parameter of a water turbine as a variable, iteratively solving the multi-index control parameter planning model, and determining an optimal governor control parameter of the water turbine to control the operation of the water turbine governor. According to the application, the optimal control parameter of the unit governor can be accurately obtained, and the precision and regulation quality of the water turbine governor control are improved.

Description

Technical Field

[0001] This application belongs to the field of hydropower technology, specifically the field of fluid engine technology, and more specifically, to a turbine governor control method and device that takes into account structural response. Background Technology

[0002] Currently, existing research on the optimization of the control law of the speed regulation system during the start-up process of hydropower units usually uses a one-dimensional water hammer model to dynamically describe the hydraulic transition process, and uses the guide vane opening control law as the optimization variable. The optimal guide vane opening change law of the governor is obtained by solving the optimization method.

[0003] However, since the one-dimensional water hammer model can only reflect the one-dimensional dynamic characteristics of flow and pressure changes over time in the flow channel system, it cannot accurately capture the complex three-dimensional unsteady flow behavior inside the turbine. At the same time, it ignores the sudden load changes and abnormal disturbances that exist during the unit start-up process, resulting in low control accuracy of the above-mentioned existing governor control methods.

[0004] Therefore, how to better control the turbine governor has become a technical problem that the industry urgently needs to solve. Summary of the Invention

[0005] In view of the shortcomings of the existing technology, the purpose of this application is to better realize the control of the turbine governor and to solve the problem of low control accuracy of the existing governor control method.

[0006] The first aspect of this application relates to a turbine governor control method considering structural response, the turbine governor control method comprising: determining a multi-index control parameter planning model based on multiple performance indicators of the turbine runner; the multiple performance indicators include pressure pulsation amplitude and dynamic stress amplitude, and also include steady-state speed adjustment time and / or speed fluctuation frequency; Using the governor control parameters of the water turbine as variables, the multi-index control parameter planning model is iteratively solved to determine the optimal governor control parameters of the water turbine, so as to control the operation of the water turbine governor.

[0007] In some embodiments, the step of iteratively solving the multi-index control parameter programming model using the governor control parameters of the turbine as variables to determine the optimal governor control parameters of the turbine includes: Multiple particles are randomly generated; each particle includes a set of speed controller parameters. Based on the objective function of the multi-index control parameter planning model, the objective fitness function of the particle swarm optimization algorithm is determined; the objective function is determined based on the calculation formulas of the multiple performance indices. With the goal of minimizing the fitness value of the target fitness function, the optimal speed governor control parameters are obtained by iteratively solving the particle swarm optimization algorithm using the multiple particles.

[0008] In some implementations, the step of using the plurality of particles to iteratively solve a particle swarm optimization algorithm to obtain the optimal governor control parameters, with the objective of minimizing the fitness value of the target fitness function, includes: Step S101: Initialize each particle, the local best position of each particle, and the current global best particle; Step S102: Input the speed controller control parameters corresponding to each particle into the preset structure response-control system coupled model for simulation, obtain the corresponding model response output data, and determine the fitness value of each particle based on the target fitness function and the model response data corresponding to each particle. Step S103: With the goal of minimizing the fitness value, update the local best position and the global best particle for each particle in the current iteration. Step S104: Determine whether the current iteration count has reached the maximum iteration count or whether the fitness value of the current global best particle has converged; if not, update the position and velocity of each particle and jump to step S102; if yes, execute step S105. Step S105: Obtain the global optimal particle of the particle swarm optimization algorithm, and determine the optimal speed regulator control parameters based on the global optimal particle.

[0009] In some embodiments, the preset structural response-control system coupling model includes a one-dimensional transient flow speed regulation system model and a structural response prediction model; step S102 includes: The governor control parameters corresponding to each particle are input into the one-dimensional transient flow speed control system model to obtain the corresponding guide vane opening data and the corresponding unit speed data. The guide vane opening data and the corresponding unit speed data for each particle are input into the structural response prediction model to obtain the pressure pulsation amplitude data and the corresponding runner dynamic stress amplitude data for each particle output by the structural response prediction model; the structural response prediction model is trained based on the turbine governor control parameter samples and their corresponding pressure pulsation data labels and runner dynamic stress data labels. Based on the unit speed data, pressure pulsation amplitude data, and turbine dynamic stress amplitude data corresponding to each particle, the target fitness function is calculated to obtain the fitness value of each particle.

[0010] In some embodiments, before inputting the guide vane opening data and the corresponding unit rotation speed data for each particle into the structural response prediction model, the method further includes: Using a three-dimensional fluid-structure interaction model of a water turbine, the pressure pulsation amplitude data and dynamic stress amplitude data of the water turbine runner under different governor control parameter samples are determined, which are used as data labels for training a preset neural network model. Each speed governor control parameter sample and its corresponding data label are used as a set of training samples to obtain multiple sets of training samples. The preset neural network model is trained using multiple sets of training samples to obtain the structural response prediction model.

[0011] In some implementations, the governor control parameters include PID control parameters and guide vane opening segmentation setting parameters.

[0012] The second aspect of this application relates to a turbine governor control device that takes into account structural response, the turbine governor control device comprising: The first processing module is used to determine a multi-indicator control parameter planning model based on multiple performance indicators of the unit runner; the multiple performance indicators include pressure pulsation amplitude and dynamic stress amplitude, as well as steady-state speed adjustment time and / or speed fluctuation number; The second processing module is used to iteratively solve the multi-index control parameter planning model using the governor control parameters of the turbine as variables, to determine the optimal governor control parameters of the turbine, so as to control the operation of the turbine governor.

[0013] Thirdly, this application provides an electronic device, comprising: at least one memory for storing a program; and at least one processor for executing the program stored in the memory, wherein when the program stored in the memory is executed, the processor is configured to execute the method described in the first aspect or any possible implementation thereof.

[0014] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when run on a processor, causes the processor to perform the method described in the first aspect or any possible implementation thereof.

[0015] Fifthly, this application provides a computer program product that, when run on a processor, causes the processor to perform the method described in the first aspect or any possible implementation thereof.

[0016] It is understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant descriptions in the first aspect mentioned above, and will not be repeated here.

[0017] Overall, the technical solutions conceived in this application have the following beneficial effects compared with the prior art: This application provides a turbine governor control method and device considering structural response. By studying the impact of load abrupt changes and abnormal disturbances during unit start-up on the dynamic response characteristics of the unit structure, and combining the dynamic characteristics of the flow field and the dynamic response characteristics of the structure during the start-up process, a multi-index control parameter planning model is comprehensively constructed, including structural response data and flow field dynamic response data. The structural response data includes pressure pulsation amplitude and dynamic stress amplitude, while the flow field dynamic response data includes steady-state speed adjustment time and / or the number of speed fluctuations. This comprehensively characterizes the speed regulation response, stability, and structural safety of the turbine. Using the governor control parameters as decision variables, a multi-objective optimization strategy is employed to solve the model, which can accurately obtain the optimal control parameters of the unit governor, improve the accuracy and regulation quality of the turbine governor control, and effectively improve the safety and efficiency during unit start-up, speed regulation, and stable operation. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating the turbine governor control method considering structural response provided in an embodiment of this application. Figure 2 This is a schematic diagram of the structure of the turbine governor control device considering structural response provided in the embodiments of this application; Figure 3 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0020] In this application, the term "and / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent three cases: A existing alone, A and B existing simultaneously, and B existing alone. In this application, the symbol " / " indicates that the related objects are in an "or" relationship, for example, A / B means A or B.

[0021] In this application, the terms "first" and "second," etc., are used to distinguish different objects, not to describe a specific order of objects. For example, "first response message" and "second response message," etc., are used to distinguish different response messages, not to describe a specific order of response messages.

[0022] In the embodiments of this application, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design that is described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the terms "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0023] In the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more, for example, multiple processing units means two or more processing units, multiple elements means two or more elements, etc.

[0024] In existing technologies, optimization studies on the control laws of speed regulation systems during the start-up process of hydropower units typically employ a one-dimensional water hammer model. However, this model only reflects the one-dimensional dynamic characteristics of flow rate and pressure changes over time within the flow channel system, failing to accurately capture the complex three-dimensional unsteady flow behavior inside the turbine. During unit start-up and shutdown, load surges, and abnormal disturbances, the turbine impeller is subjected to hydraulic excitation, resulting in significant periodic dynamic stress and pressure pulsations. Research has revealed that existing optimization methods generally neglect the influence of governor control laws on structural responses such as impeller dynamic stress and structural vibration.

[0025] Specifically, during rapid start-up and shutdown of the unit and drastic load changes, the internal pressure distribution of the flow field adjusts rapidly. Complex unsteady turbulence is generated in the guide vane-impeller interaction zone, leading to significant dynamic stress and irregular vibration responses in the runner blades. On the one hand, relying solely on traditional fixed-parameter controllers (such as fixed PID parameters) cannot adjust the guide vane opening response in a timely manner according to real-time structural changes, easily causing frequency overshoot, overshoot, or even instability. On the other hand, neglecting the optimization of speed regulation patterns based on structural response over a long period will accelerate the development of impeller fatigue cracks and shorten the equipment's service life.

[0026] It should be noted that the various defects in the technical solutions of the prior art are the result of the inventors’ careful practical research. Therefore, the discovery process of the above problems and the solutions proposed by the embodiments of this application in the following text should be the inventors’ contributions to this application in the process of realizing this application.

[0027] To address the aforementioned technical problems, this application provides a method and apparatus for controlling a turbine governor that considers structural response.

[0028] The embodiments of this application are described below with reference to the accompanying drawings.

[0029] Figure 1This is a flowchart illustrating the turbine governor control method considering structural response provided in this application embodiment, as shown below. Figure 1 As shown, it includes: Step S1: Based on multiple performance indicators of the unit runner, determine the multi-indicator control parameter planning model; the multiple performance indicators include pressure pulsation amplitude and dynamic stress amplitude, as well as steady-state speed regulation time and / or speed fluctuation number; Step S2: Using the governor control parameters of the turbine as variables, iteratively solve the multi-index control parameter programming model to determine the optimal governor control parameters of the turbine, so as to control the operation of the turbine governor.

[0030] Specifically, the multiple performance indicators described in the embodiments of this application include pressure pulsation amplitude and dynamic stress amplitude, as well as steady-state speed adjustment time and / or speed fluctuation number.

[0031] Among them, pressure pulsation amplitude refers to the amplitude of the periodic / random change of fluid pressure in the flow channel around the impeller with time or space, and is usually represented by the peak-to-peak value of pressure pulsation; dynamic stress amplitude refers to the stress fluctuation amplitude generated by the impeller under the combined action of fluid dynamic load, mechanical vibration and thermal stress.

[0032] Steady-state speed regulation time refers to the time it takes for the turbine unit speed to first enter and remain within the ±5% steady-state error band; the number of speed fluctuations refers to the number of zero-crossing points or peak-valley points that appear in the unit speed response curve n(t), which is specifically determined by analyzing the number of zero-crossing points or local extreme values ​​of the speed response curve.

[0033] The governor control parameters of the turbine described in this application embodiment may specifically include PID control parameters and / or guide vane opening segmentation setting parameters. The PID control parameters include proportional coefficients. Integral coefficient Differential coefficients The guide vane opening segment setting parameters include the first segment guide vane opening setting value. Second stage guide vane opening setting value ,here, and These are respectively used for nonlinear segmented guide vane adjustment strategies in speed governors. This indicates the initial control switching point, i.e., the guide vane adjustment speed in the first stage of the two-stage start-up strategy; This represents the threshold of the secondary control segment, used to define the fast opening rate of the second guide vane response in the two-stage start-up strategy, thereby enhancing the regulation performance under strong disturbances.

[0034] In the embodiments of this application, in step S1, based on multiple performance indicators of the unit runner, including pressure pulsation amplitude and dynamic stress amplitude, as well as steady-state speed adjustment time and / or speed fluctuation number, the objective function of the model is jointly determined according to their calculation formulas, and then a multi-index control parameter planning model can be constructed based on the objective function of the model.

[0035] Optionally, the governor control parameters include PID control parameters and guide vane opening segmentation setting parameters. Specifically, in the embodiments of this application, the governor control parameters... It can be represented as: = .

[0036] The method of this application embodiment, compared with the traditional control method that only uses the guide vane opening control law as the optimization variable to solve the optimal guide vane opening change law of the governor, or the speed regulation method that only fixes the PID parameters, constructs the governor control parameters by taking into account both the PID control parameters and the guide vane opening response parameters, and uses the governor control parameters as the optimization variable for optimization, can further improve the control accuracy of the turbine governor.

[0037] In the embodiments of this application, in step S2, various biomimetic optimization algorithms, such as particle swarm optimization and genetic algorithms, can be used to iteratively solve the multi-index control parameter planning model with the governor control parameters of the water turbine as variables. The optimal governor control parameters of the water turbine can be solved, and then the operation of the water turbine governor can be controlled according to the optimal governor control parameters.

[0038] The turbine governor control method considering structural response in this application study the impact of load abrupt changes and abnormal disturbances during unit start-up on the dynamic response characteristics of the unit structure. Combining the dynamic characteristics of the flow field and the dynamic response characteristics of the structure during turbine start-up, a multi-index control parameter planning model is comprehensively constructed, incorporating structural response data and flow field dynamic response data. The structural response data includes pressure pulsation amplitude and dynamic stress amplitude, while the flow field dynamic response data includes steady-state speed adjustment time and / or the number of speed fluctuations. This comprehensively characterizes the speed regulation response speed, stability, and structural safety of the turbine. Using governor control parameters as decision variables, a multi-objective optimization strategy is employed to solve the model, accurately obtaining the optimal control parameters of the turbine governor. This improves the accuracy and regulation quality of turbine governor control and effectively enhances the safety and efficiency during unit start-up, speed regulation, and stable operation.

[0039] Based on the above embodiments, as an optional embodiment, step S2 involves iteratively solving a multi-index control parameter programming model using the governor control parameters of the turbine as variables to determine the optimal governor control parameters of the turbine, including: Multiple particles are randomly generated; each particle includes a set of governor control parameters. The objective fitness function of the particle swarm optimization algorithm is determined based on the objective function of the multi-index control parameter planning model; the objective function is determined based on the calculation formula of multiple performance indices. With the goal of minimizing the fitness value of the objective fitness function, the optimal governor control parameters are obtained by iteratively solving the problem using a particle swarm optimization algorithm with multiple particles.

[0040] Specifically, in the embodiments of this application, each particle can represent a combination of governor control parameters. .

[0041] Furthermore, the optimization objectives include: The rapidity index—steady-state speed settling time—is calculated using the following formula:

[0042] In the formula, For the change in rotational speed, For steady-state speed, The allowable error band is generally ±5% steady-state error band.

[0043] Stability index – the number of speed fluctuations in a one-dimensional transient flow speed control system model, calculated using the following formula: ; In the formula, Reference speed; For discrete sampling times; This is an indicator function; it takes the value 1 if the value inside the parentheses is true, and 0 otherwise.

[0044] It should be noted that the one-dimensional transient flow speed control system model is a commonly used one-dimensional transient flow mathematical model in existing hydro turbine speed control systems. Specifically, it includes: a governor, a servo hydraulic mechanism, a water intake pipe, and a unit load module, represented by the guide vane opening curve. As input, it can be used to output the unit's speed response curve. .

[0045] Structural safety indicators—peak-to-peak value of pressure pulsation and amplitude of dynamic stress in the impeller under a three-dimensional coupled model. The calculation formulas for the peak-to-peak value of pressure pulsation and amplitude of dynamic stress in the impeller are as follows: ; In the formula, ; The dynamic stress amplitude of the runner can be taken as the maximum dynamic stress amplitude of the runner, and its calculation formula is as follows: ; ; In the formula, p Indicates the index of the measuring point on the wheel. Indicates each measuring point p The dynamic stress amplitude, The maximum stress in a stress cycle, The minimum stress in a stress cycle, This represents the maximum value among all measuring points.

[0046] The above indicators constitute the evaluation function for multi-objective optimization. It constitutes the objective function of the multi-index control parameter planning model and can be used as the objective fitness function of the subsequent particle swarm optimization algorithm.

[0047] Furthermore, in the embodiments of this application, with the goal of minimizing the fitness value of the objective fitness function, multiple particles are used to perform iterative solutions using a particle swarm optimization algorithm. Through iterative updates using a multi-objective particle swarm optimization algorithm, a set of governor parameter solutions that approximates the Pareto optimal front is obtained. The final non-dominated solution is a set of optimal governor control parameters that take into account both regulation performance and structural safety, providing parameter reference and decision support for practical applications.

[0048] The method in this application proposes a multi-objective particle swarm optimization algorithm based on governor control parameters, on the basis of constructing a multi-index control parameter planning model. The particle swarm optimization algorithm is iteratively solved with the objective of minimizing the fitness value of the multi-index function, which can quickly and accurately solve for the optimal governor control parameters.

[0049] Based on the above embodiments, as an optional embodiment, with the objective of minimizing the fitness value of the target fitness function, the optimal governor control parameters are obtained by iteratively solving a particle swarm optimization algorithm using multiple particles, including: Step S101: Initialize each particle, the local best position of each particle, and the current global best particle; Step S102: Input the speed controller control parameters corresponding to each particle into the preset structure response-control system coupled model for simulation, obtain the corresponding model response output data, and determine the fitness value of each particle based on the target fitness function and the model response data corresponding to each particle. Step S103: With the goal of minimizing the fitness value, update the local best position and the global best particle for each particle in the current iteration. Step S104: Determine whether the current iteration count has reached the maximum iteration count or whether the fitness value of the current global best particle has converged; if not, update the position and velocity of each particle and jump to step S102; if yes, execute step S105. Step S105: Obtain the globally optimal particle of the particle swarm optimization algorithm, and determine the optimal speed regulator control parameters based on the globally optimal particle.

[0050] Specifically, in the embodiments of this application, M particles are randomly generated, and each particle represents a set of decision variables. In the Multi-Objective Particle Swarm Optimization (MOPSO) framework, each particle represents a set of governor control parameters to be optimized. An initial optimization population is formed by randomly generating M initial particles in the five-dimensional parameter space. In step S101, each particle, its local optimum position, and the current global optimum particle are initialized.

[0051] In step S102, the speed controller control parameters corresponding to each particle are... The input is fed into the preset structure response-control system coupled model for simulation, and the corresponding model response output data can be obtained. Based on the aforementioned target fitness function and the model response data corresponding to each particle, the relevant fitness value can be calculated, and the fitness value of each particle can be determined.

[0052] Based on the above embodiments, as an optional embodiment, the preset structural response-control system coupling model includes a one-dimensional transient flow speed regulation system model and a structural response prediction model; step S102 includes: The governor control parameters corresponding to each particle are input into the one-dimensional transient flow speed control system model to obtain the corresponding guide vane opening data and the corresponding unit speed data. The guide vane opening data and the corresponding unit speed data for each particle are input into the structural response prediction model to obtain the pressure pulsation amplitude data and the corresponding runner dynamic stress amplitude data for each particle output by the structural response prediction model. The structural response prediction model is trained based on the turbine governor control parameter samples and their corresponding pressure pulsation data labels and runner dynamic stress data labels. Based on the unit speed data, pressure pulsation amplitude data, and turbine dynamic stress amplitude data corresponding to each particle, the target fitness function is calculated to obtain the fitness value of each particle.

[0053] Specifically, in the embodiments of this application, the preset structural response-control system coupling model specifically includes a one-dimensional transient flow speed regulation system model and a structural response prediction model.

[0054] First, a one-dimensional transient flow speed control system model is invoked, and the governor control parameters corresponding to each particle are input into the model to obtain the guide vane opening response curve y(t) under the corresponding governor control parameters. Then, the unit speed response curve n(t) is further solved, yielding the corresponding guide vane opening data and the corresponding unit speed data. Subsequently, the time-domain data of y(t) and n(t) can be used as input to invoke a pre-trained structural response prediction model, such as a BP neural network model, to predict the corresponding runner pressure pulsation amplitude data and dynamic stress amplitude data.

[0055] For each particle (i.e., a set of governor control parameters), a complete simulation evaluation needs to be performed using the structure-response-control system coupled model. The specific calculation process is as follows: Step 1: Input the parameters corresponding to the current particle into the one-dimensional transient flow speed control system model to simulate and generate the unit operation response, including the guide vane opening response curve data y(t) and the unit speed response curve data n(t).

[0056] Step two: Extract dynamic response information from the unit speed response curve data n(t) and calculate speed characteristic indicators, including: Steady-state speed adjustment time The time required for the rotational speed to enter the ±5% steady-state error band and remain stable; Number of speed fluctuations : Estimated by the number of zero-crossing or peak-valley points appearing in the speed response curve data n(t).

[0057] Specifically, the unit speed response curve data n(t) can be used and substituted into the calculation formula for steady-state speed settling time and the number of speed fluctuations in the aforementioned target fitness function to calculate the steady-state speed settling time. and the number of rotational speed fluctuations .

[0058] Step 3: Invoke the structural response prediction model. Input the guide vane opening data and the corresponding unit speed data corresponding to the current particle into the structural response prediction model, and output the corresponding pressure pulsation amplitude data and the corresponding turbine runner dynamic stress amplitude data. Using the pressure pulsation amplitude data and the corresponding turbine runner dynamic stress amplitude data corresponding to this particle, combined with the calculation formulas for the peak-to-peak value of pressure pulsation and the turbine runner dynamic stress amplitude in the aforementioned target fitness function, the peak-to-peak value of pressure pulsation can be calculated. With the dynamic stress amplitude of the impeller .

[0059] Step four, finally, can construct the fitness evaluation vector F= , which serves as the fitness index for the current particle.

[0060] The method in this application embodiment embeds a trained structural response prediction model into a one-dimensional transient flow speed control system model of a water turbine, forming a coupled structural response-control system model. When the system inputs governor control parameters, it can simultaneously calculate the speed response and structural response, which facilitates subsequent multi-objective evaluation and can greatly improve the efficiency of governor control.

[0061] Furthermore, in step S103, with the goal of minimizing the fitness value, the local best position and the global best particle of each particle in the current iteration are updated.

[0062] Specifically, in this embodiment, non-dominated sorting is used: the evaluation vectors of all particles are input into the multi-objective optimization decision module, and non-dominated sorting is performed according to the Pareto dominance relationship. Based on the sorting results, the individual optimal position of each particle (i.e., its historical best fitness value), the global non-dominated solution set (Pareto optimal set) in the overall population, and the external archive set (external elite library) are updated to provide a reference for guiding the particle's motion direction in the next step.

[0063] The specific implementation method for updating particles is as follows: Following the velocity and position update rules set in the MOPSO algorithm, and under the combined effect of inertia weighting, individual optimal guidance, and global optimal guidance, the next generation of particle swarms is generated. For each dimension (i.e., the governor control parameters), updates are performed using the following formula: ; ; in, xi This refers to the current position of the particle (i.e., a set of control parameters). vi For velocity vector, pi and gi These represent the individual historical best and the globally non-dominated guiding positions, respectively. w , c 1, r 1, c 2, r 2 represents the velocity and position update coefficients for the corresponding parameters.

[0064] Further, in step S104, it is determined whether the current iteration number has reached the maximum iteration number or whether the fitness value of the current global best particle has converged (or the Pareto front solution set in the external elite archive has stabilized after several generations); if not, the position and velocity of each particle are updated, and the process jumps to step S102 above; if yes, the iteration terminates, and step S105 is executed below.

[0065] Step S105: The final output is a set of non-dominated controller parameter solutions. Each set of parameters corresponds to an optimal strategy configuration that balances speed regulation performance and structural response. That is, the global optimal particle of the particle swarm optimization algorithm is obtained, and the optimal speed controller control parameters are determined based on the global optimal particle, providing diversified adjustment strategy support for subsequent engineering selection and implementation.

[0066] The method in this application embodiment, by fully combining the dynamic characteristics of the flow field and the dynamic response characteristics of the structure during the start-up process of the hydropower unit, and by using the PSO algorithm to continuously iteratively optimize the control parameters of the turbine governor, can further greatly improve the accuracy and efficiency of governor control parameter optimization.

[0067] Based on the above embodiments, as an optional embodiment, before inputting the guide vane opening data and the corresponding unit rotation speed data for each particle into the structural response prediction model, the method further includes: Using a three-dimensional fluid-structure interaction model of a water turbine, the pressure pulsation amplitude data and dynamic stress amplitude data of the water turbine runner under different governor control parameter samples are determined, which are used as data labels for training a preset neural network model. Each speed governor control parameter sample and its corresponding data label are used as a set of training samples to obtain multiple sets of training samples. A pre-defined neural network model is trained using multiple sets of training samples to obtain a structural response prediction model.

[0068] Specifically, in the embodiments of this application, in order to quickly predict the impact of governor control parameters on structural response, a three-dimensional fluid-structure interaction model of the turbine is used. After completing a large number of turbine speed regulation simulations under different start-up conditions, simulation data of corresponding turbine pressure pulsation amplitude and dynamic stress amplitude are collected. A preset neural network model is then constructed, using governor control parameters as samples and structural response indicators as outputs. This preset neural network model can specifically adopt a backpropagation (BP) neural network model.

[0069] The BP neural network model is described as follows: The formula for calculating the input layer to the hidden layer is as follows: ; in, This is the weight matrix from the input layer to the hidden layer. For bias vectors, is the activation function for the hidden layer.

[0070] Formula for calculating the hidden layer to output layer: ; in, This is the weight matrix from the input layer to the hidden layer. For bias vectors, This is the activation function for the output layer.

[0071] Specifically, firstly, a three-dimensional fluid-structure interaction (FSI) model of the volute, guide vanes, impeller, and tailrace pipe is constructed based on the CAD drawings of the entire flow path of the turbine unit. Transient flow field numerical simulation is then performed using CFD software to obtain the instantaneous pressure distribution and velocity field under various typical operating conditions. Next, the flow field load data output from the CFD simulation is mapped onto the surface of the structural finite element model to construct the fluid-structure interaction boundary. Then, the dynamic stress response of the impeller structure under different loads is solved in finite element analysis software to obtain the time-history response data of the runner's dynamic stress amplitude. Finally, the above process is repeated under multiple combinations of guide vane opening, flow rate, and speed to extract the pressure pulsation amplitude and runner dynamic stress amplitude data corresponding to different governor control parameter samples. The flow rate and speed are determined based on the unit's PID control parameters. Thus, the pressure pulsation amplitude and dynamic stress amplitude data of the turbine runner under different governor control parameter samples can be obtained and used as data labels for training a pre-set neural network model.

[0072] Furthermore, each governor control parameter sample (guide vane opening and each PID control parameter) is paired with the corresponding structural response data labels (pressure pulsation amplitude and dynamic stress amplitude) obtained from the aforementioned simulation, thereby constructing multiple sets of training samples.

[0073] Furthermore, by calling the one-dimensional transient flow speed control system model, and using the corresponding governor control parameter samples from each training sample as input to the one-dimensional transient flow speed control system model, simulation generates unit operation response data samples, including guide vane opening data samples and unit speed data samples. Using the guide vane opening data samples, unit speed data samples, and corresponding structural response data labels corresponding to each training sample, the preset neural network model is iteratively trained until the maximum number of iterations is reached or the model convergence condition is met, thus obtaining the trained structural response prediction model.

[0074] The method in this application uses each governor control parameter sample and its corresponding structural response data label as a set of training samples. Multiple sets of training samples are used to train a preset neural network model to form a nonlinear mapping relationship between control parameters and structural response. This allows the trained structural response prediction model to be efficiently called during the optimization iteration process, avoiding the need to run complex three-dimensional CFD simulations for each evaluation, which is beneficial to further improve the efficiency of turbine governor control.

[0075] The turbine governor control device considering structural response provided in this application is described below. The turbine governor control device considering structural response described below and the turbine governor control method considering structural response described above can be referred to in correspondence.

[0076] Figure 2 This is a schematic diagram of the structure of the turbine governor control device considering structural response provided in the embodiments of this application, as shown below. Figure 2 As shown, it includes: The first processing module 10 is used to determine a multi-indicator control parameter planning model based on multiple performance indicators of the unit runner; the multiple performance indicators include pressure pulsation amplitude and dynamic stress amplitude, and also include at least one of steady-state speed regulation time and speed fluctuation number; The second processing module 20 is used to iteratively solve the multi-index control parameter planning model with the governor control parameters of the turbine as variables, and determine the optimal governor control parameters of the turbine in order to control the operation of the turbine governor.

[0077] It is understood that the detailed functional implementation of each of the above units / modules can be found in the description in the aforementioned method embodiments, and will not be repeated here.

[0078] It should be understood that the above-described device is used to execute the methods in the above embodiments. The implementation principle and technical effect of the corresponding program modules in the device are similar to those described in the above methods. The working process of the device can be referred to the corresponding process in the above methods, and will not be repeated here.

[0079] The turbine governor control device considering structural response in this application study the impact of load abrupt changes and abnormal disturbances during unit start-up on the dynamic response characteristics of the unit structure. Combining the dynamic characteristics of the flow field and the dynamic response characteristics of the structure during turbine start-up, a multi-index control parameter planning model is comprehensively constructed, including structural response data and flow field dynamic response data. The structural response data includes pressure pulsation amplitude and dynamic stress amplitude, while the flow field dynamic response data includes steady-state speed adjustment time and / or the number of speed fluctuations. This comprehensively characterizes the speed regulation response speed, stability, and structural safety of the turbine. Using governor control parameters as decision variables, a multi-objective optimization strategy is employed to solve the model, accurately obtaining the optimal control parameters of the turbine governor. This improves the accuracy and regulation quality of turbine governor control and effectively enhances the safety and efficiency during unit start-up, speed regulation, and stable operation.

[0080] Based on the methods in the above embodiments, this application provides an electronic device, such as... Figure 3As shown, the electronic device may include a processor 310, a communications interface 320, a memory 330, and a communication bus 340, wherein the processor 310, the communications interface 320, and the memory 330 communicate with each other via the communication bus 340. The processor 310 can call logical instructions in the memory 330 to execute the methods in the above embodiments.

[0081] Furthermore, the logical instructions in the aforementioned memory 330 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application.

[0082] Based on the methods in the above embodiments, this application provides a computer-readable storage medium storing a computer program that, when run on a processor, causes the processor to execute the methods in the above embodiments.

[0083] Based on the methods in the above embodiments, this application provides a computer program product that, when run on a processor, causes the processor to execute the methods in the above embodiments.

[0084] It is understood that the processor in the embodiments of this application can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. A general-purpose processor can be a microprocessor or any conventional processor.

[0085] The method steps in this application embodiment can be implemented in hardware or by a processor executing software instructions. The software instructions can consist of corresponding software modules, which can be stored in random access memory (RAM), flash memory, read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), registers, hard disks, portable hard disks, CD-ROMs, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can reside in an ASIC.

[0086] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially as a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted through the computer-readable storage medium. The computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state disk (SSD)).

[0087] It is understood that the various numerical designations used in the embodiments of this application are merely for the convenience of description and are not intended to limit the scope of the embodiments of this application.

[0088] It should be understood that expressions such as “comprising” and “may include” used in this application indicate the existence of the disclosed functions, operations, or constituent elements, and do not limit one or more additional functions, operations, and constituent elements. In this application, terms such as “comprising” and / or “having” are to be interpreted as indicating a particular characteristic, number, operation, constituent element, component, or combination thereof, but not to exclude the existence or possibility of adding one or more other characteristics, numbers, operations, constituent elements, components, or combinations thereof.

[0089] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A turbine governor control method considering structural response, characterized in that, include: Based on multiple performance indicators of the unit runner, a multi-indicator control parameter planning model is determined; the multiple performance indicators include pressure pulsation amplitude and dynamic stress amplitude, as well as steady-state speed regulation time and / or speed fluctuation frequency. Using the governor control parameters of the water turbine as variables, the multi-index control parameter planning model is iteratively solved to determine the optimal governor control parameters of the water turbine, so as to control the operation of the water turbine governor. The step of iteratively solving the multi-index control parameter planning model using the governor control parameters of the turbine as variables to determine the optimal governor control parameters of the turbine includes: Multiple particles are randomly generated; each particle includes a set of speed controller parameters. Based on the objective function of the multi-index control parameter planning model, the objective fitness function of the particle swarm optimization algorithm is determined; the objective function is determined based on the calculation formulas of the multiple performance indices. With the goal of minimizing the fitness value of the target fitness function, the optimal speed governor control parameters are obtained by iteratively solving the particle swarm optimization algorithm using the multiple particles. The step of minimizing the fitness value of the target fitness function, and using the multiple particles to iteratively solve a particle swarm optimization algorithm to obtain the optimal governor control parameters, includes: Step S101: Initialize each particle, the local best position of each particle, and the current global best particle; Step S102: Input the speed controller control parameters corresponding to each particle into the preset structure response-control system coupled model for simulation, obtain the corresponding model response output data, and determine the fitness value of each particle based on the target fitness function and the model response data corresponding to each particle. Step S103: With the goal of minimizing the fitness value, update the local best position of each particle in the current iteration and the global best particle; Step S104: Determine whether the current iteration count has reached the maximum iteration count or whether the fitness value of the current global best particle has converged; if not, update the position and velocity of each particle and jump to step S102; if yes, execute step S105. Step S105: Obtain the global optimal particle of the particle swarm optimization algorithm, and determine the optimal speed regulator control parameters based on the global optimal particle.

2. The turbine governor control method according to claim 1, characterized in that, The preset structural response-control system coupling model includes a one-dimensional transient flow speed regulation system model and a structural response prediction model; step S102 includes: The governor control parameters corresponding to each particle are input into the one-dimensional transient flow speed control system model to obtain the corresponding guide vane opening data and the corresponding unit speed data. The guide vane opening data and the corresponding unit speed data for each particle are input into the structural response prediction model to obtain the pressure pulsation amplitude data and the corresponding runner dynamic stress amplitude data for each particle output by the structural response prediction model; the structural response prediction model is trained based on the turbine governor control parameter samples and their corresponding pressure pulsation data labels and runner dynamic stress data labels. Based on the unit speed data, pressure pulsation amplitude data, and turbine dynamic stress amplitude data corresponding to each particle, the target fitness function is calculated to obtain the fitness value of each particle.

3. The turbine governor control method according to claim 2, characterized in that, Before inputting the guide vane opening data and the corresponding unit rotation speed data for each particle into the structural response prediction model, the method further includes: Using a three-dimensional fluid-structure interaction model of a water turbine, the pressure pulsation amplitude data and dynamic stress amplitude data of the water turbine runner under different governor control parameter samples are determined, which are used as data labels for training a preset neural network model. Each speed governor control parameter sample and its corresponding data label are used as a set of training samples to obtain multiple sets of training samples. The preset neural network model is trained using multiple sets of training samples to obtain the structural response prediction model.

4. The turbine governor control method according to any one of claims 1-3, characterized in that, The speed governor control parameters include PID control parameters and guide vane opening segment setting parameters.

5. A turbine governor control device considering structural response, characterized in that, include: The first processing module is used to determine a multi-indicator control parameter planning model based on multiple performance indicators of the unit runner; the multiple performance indicators include pressure pulsation amplitude and dynamic stress amplitude, as well as steady-state speed adjustment time and / or speed fluctuation number; The second processing module is used to iteratively solve the multi-index control parameter planning model using the governor control parameters of the turbine as variables, to determine the optimal governor control parameters of the turbine, so as to control the operation of the turbine governor. The step of iteratively solving the multi-index control parameter planning model using the governor control parameters of the turbine as variables to determine the optimal governor control parameters of the turbine includes: Multiple particles are randomly generated; each particle includes a set of speed controller parameters. Based on the objective function of the multi-index control parameter planning model, the objective fitness function of the particle swarm optimization algorithm is determined; the objective function is determined based on the calculation formulas of the multiple performance indices. With the goal of minimizing the fitness value of the target fitness function, the optimal speed governor control parameters are obtained by iteratively solving the particle swarm optimization algorithm using the multiple particles. The step of minimizing the fitness value of the target fitness function, and using the multiple particles to iteratively solve a particle swarm optimization algorithm to obtain the optimal governor control parameters, includes: Step S101: Initialize each particle, the local best position of each particle, and the current global best particle; Step S102: Input the speed controller control parameters corresponding to each particle into the preset structure response-control system coupled model for simulation, obtain the corresponding model response output data, and determine the fitness value of each particle based on the target fitness function and the model response data corresponding to each particle. Step S103: With the goal of minimizing the fitness value, update the local best position of each particle in the current iteration and the global best particle; Step S104: Determine whether the current iteration count has reached the maximum iteration count or whether the fitness value of the current global best particle has converged; if not, update the position and velocity of each particle and jump to step S102; if yes, execute step S105. Step S105: Obtain the global optimal particle of the particle swarm optimization algorithm, and determine the optimal speed regulator control parameters based on the global optimal particle.

6. An electronic device, characterized in that, Includes memory and one or more processors; The memory is coupled to the one or more processors, and the memory is used to store computer program code, the computer program code including computer instructions; The one or more processors invoke the computer instructions to cause the electronic device to perform the method as described in any one of claims 1-4.

7. A computer-readable storage medium comprising instructions, characterized in that: When the instructions are executed on an electronic device, the electronic device causes the electronic device to perform the method as described in any one of claims 1-4.

8. A computer program product, comprising a computer program or instructions, characterized in that: When the computer program or instructions are run on an electronic device, the electronic device causes the electronic device to perform the method as described in any one of claims 1-4.

Images