Reliability Redundancy Control Method for Speed ​​Control System of Large Hydropower Station

Abstract

The present invention discloses a reliability redundant control method for a speed regulating system of a large hydropower station, comprising: constructing a simulation environment, inputting signals, outputting command limiting, performance evaluation and optimizing control strategies. The method outputs speed, power and guide vane opening through a simulation environment, combines the desired control instructions, inputs them into the redundant control strategy of the speed regulating system, feeds them back to the simulation environment after processing by a limiting unit, and optimizes the control strategy to improve the performance of the speed regulating system of the hydropower station. The present invention can improve the stability and response speed of the speed regulating system, and enhance the reliability and efficiency of the operation of the hydropower station.

Description

Technical Field

[0001] The invention relates to the control field of a speed regulating system of a large hydropower station, and in particular to a reliability redundancy control method for a speed regulating system of a large hydropower station. Background Art

[0002] Existing speed control systems for large hydropower stations face challenges in reliability and stability in terms of control. These systems usually need to accurately control the speed and power output of the turbine generator sets under complex hydraulic conditions to ensure efficient and safe operation of the power station. Traditional speed control systems may lack sufficient redundancy and flexibility to cope with sudden equipment failures or changes in the external environment, which may lead to a decline in control performance or even system failure. In addition, due to the variability of the operating environment of hydropower stations, such as water flow changes, head fluctuations, etc., higher requirements are placed on the dynamic response and adaptability of the speed control system. In these systems, accurate control algorithms and strategies are crucial to maintaining the optimal operating state of the units, but existing technologies often find it difficult to fully consider and respond to these complex factors in real time.

[0003] In the process of implementing the embodiments of the present invention, the inventors found that there are at least the following problems or defects in the prior art: the existing speed regulation system control strategies often lack effective processing of uncertainty factors, resulting in difficulty in adapting to changing operating conditions in actual operation; at the same time, there is a lack of a systematic method to evaluate and optimize control performance in real time, making it difficult for the speed regulation system to achieve the best control effect and the highest reliability when facing complex and changeable operating conditions. Summary of the invention

[0004] The embodiment of the present invention aims to provide a reliability redundancy control method for a speed regulating system of a large hydropower station.

[0005] The embodiment of the present invention solves the technical problem by adopting the following technical solution:

[0006] A reliability redundancy control method for a large hydropower station speed regulation system is provided, comprising:

[0007] Construct a simulation environment for redundant control strategies of speed regulation systems in large hydropower stations;

[0008] According to the speed, power and guide vane opening output by the simulation environment, and the expected control instruction, an input signal of the redundant control strategy of the speed regulation system is generated, and the input signal is input into the redundant control strategy of the speed regulation system to obtain an output instruction;

[0009] Inputting the output instruction obtained from the redundant control strategy of the speed regulation system into the control limit unit, obtaining the output result of the control limit unit, and inputting the output result into the simulation environment;

[0010] Constructing a performance evaluation unit, wherein the performance evaluation unit feeds back the current moment evaluation result of the control performance of the simulated environment to the redundant control strategy of the speed regulation system, and optimizes the configuration of the redundant control strategy of the speed regulation system by optimizing the evaluation result of the control performance;

[0011] The redundant control strategy of the speed regulation system is configured and optimized to obtain a final redundant control strategy of the speed regulation system, and the speed regulation system of the hydropower station is controlled based on the redundant control strategy of the speed regulation system.

[0012] Furthermore, the simulation environment introduces random uncertainty influencing factors into the model coefficients of the hydro-generator set model;

[0013] Correspondingly, the model coefficients include control signals of speed, power and guide vane opening, and the control signals of speed, power and guide vane opening include corresponding control coefficients respectively;

[0014] The control coefficient is calculated by the following formula:

[0015] C=k1·H+k2·u+k3·α+k4·β

[0016] Among them, C is the control coefficient, k1, k2, k3, k4 are coefficients, H is the water head, u is the flow rate, α is the guide vane opening, and β is the guide vane angle.

[0017] Furthermore, the hydro-generator set model is a multi-parameter dynamic model; accordingly, the simulation environment for constructing the redundant control strategy of the speed regulation system includes:

[0018] The simulation environment is constructed according to the following expression:

[0019] P=ρgQH

[0020] Among them, P represents power, ρ is the density of water, g is the acceleration due to gravity, Q is the flow rate, and H is the water head.

[0021] Furthermore, the output result of the control limiting unit includes a limiting result obtained by limiting the action amplitude and the change rate of the output instruction;

[0022] The output instructions include a theoretical guide vane opening, a theoretical guide vane angle, and a theoretical guide vane deflection angle;

[0023] The limiting result is calculated according to the following formula:

[0024] For action amplitude limiting:

[0025] Δα max =max(Δα)

[0026] For rate of change limiting:

[0027]

[0028] Among them, Δα represents the change in the guide vane opening, Indicates the rate of change of guide vane opening.

[0029] Further, the construction performance evaluation unit includes: the current moment evaluation is represented by the following linear form:

[0030]

[0031] Among them, e t is the current moment evaluation at time t, k1 and k2 are evaluation coefficients, s t is the system status, u t To control the amount, is the expected system state.

[0032] Furthermore, the reliability redundancy control method of the speed regulation system of a large hydropower station also includes:

[0033] The current evaluation score output by the performance evaluation unit can be changed by adjusting the evaluation coefficients k1 and k2 to ensure that the system state and control quantity meet the requirements;

[0034] When the preset number of control targets are completed consecutively, additional task completion rewards will be given to guide priority completion of tasks.

[0035] Furthermore, the configuration optimization of the redundant control strategy of the speed regulation system includes:

[0036] Determine evaluation parameters and control parameters;

[0037] Initialize the simulation environment, initialize random noise, and then loop through data sampling and parameter optimization;

[0038] Multiple independent parameter optimization processes are carried out to optimize the configuration of the redundant control strategy of the speed regulation system.

[0039] Furthermore, the performing of multiple independent parameter optimization processes to optimize the configuration of the redundant control strategy of the speed regulation system includes:

[0040] Each optimization step is as follows:

[0041] a) Extract the initialization input state s1;

[0042] b) If the current time t is less than or equal to the maximum optimization duration T, then execute step c), otherwise end this optimization and proceed to the next optimization;

[0043] c) Based on the initial input state s1, the control parameters are used to generate the action a at time t t =μ(s t |)+N t , N t is the random noise at time t used to increase the diversity of samples, denoted by N init is the noise size at the initial moment, and N t As the optimization process gradually becomes smaller, it satisfies: N t =0.995×N t-1 ;

[0044] The obtained action command is limited and applied to the simulation environment, and the current moment evaluation e is obtained according to the model characteristic integration. t and the state s at the next moment t+1 , and e t If the control target is met and the preset number of times is completed continuously, let e t =e goal =100;

[0045] e) The data pair (s t ,a t ,e t ,s t+1 ) is stored in the data resource pool. If the data resource pool is full, R If there are data pairs, execute step f) to start parameter optimization, otherwise jump back to step c);

[0046] f) Randomly sample N data pairs from the data resource pool;

[0047] g) Let θ calculate the evaluation parameter loss as follows:

[0048]

[0049] h) Use the gradient descent optimization algorithm to update the evaluation parameter θ. The optimization goal is to minimize the loss L. The learning rate of the gradient descent is λ, γ is the discount factor of the optimization algorithm, and N is the total number of data samples;

[0050] i) Calculate the gradient of the sampled data pair as follows:

[0051]

[0052] j) Update the control parameter θ using the gradient descent optimization algorithm based on the calculated objective function gradient;

[0053] k) Use soft update to update the target network parameters, that is:

[0054] θ target =τθ+(1-τ)θ target,old

[0055] Among them, τ is the soft update coefficient.

[0056] Furthermore, the flow-related component is obtained by the following formula:

[0057]

[0058] Among them, Q represents the flow rate, A is the flow area, g is the gravitational acceleration, and H is the water head.

[0059] Furthermore, the guide vane opening α and the guide vane angle β are calculated according to the following formulas:

[0060]

[0061] Among them, the value ranges of α and β are [-π,π] respectively, and arctan and arcsin are inverse trigonometric functions with a range defined between [-π,π].

[0062] The above-mentioned embodiments of the present invention have at least the following beneficial effects: the reliability redundancy control method of the speed control system of a large hydropower station proposed in the embodiments of the present invention can improve the control performance of the speed control system when facing the influence of random uncertainty. By constructing a simulation environment to introduce random uncertainty influencing factors, the method can simulate various uncertainties that may be encountered in actual operation, and can make the speed control system control strategy more adaptable to actual working conditions. In addition, through real-time feedback and optimized control performance evaluation of the performance evaluation unit, the method can dynamically adjust the control strategy to adapt to the changing operating conditions, ensuring that the hydropower station speed control system can maintain the optimal operating state under various conditions.

[0063] The method can also limit the action amplitude and change rate of the output command through the control limiter unit, further enhancing the stability and response speed of the system. Through this method, it can ensure that the adjustment of the guide vane opening and angle is within a safe and effective range, avoiding damage to the hydro-generator set due to drastic changes in the control command. At the same time, through multiple independent parameter optimization processes, the method can continuously iterate and improve the control strategy to achieve higher control accuracy and better system performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] One or more embodiments are exemplarily described by the figures in the corresponding drawings, and these exemplifications do not constitute limitations on the embodiments. Elements with the same reference numerals in the drawings represent similar elements, and unless otherwise stated, the figures in the drawings do not constitute proportional limitations.

[0065] Figure 1 A schematic flow chart of a reliability redundancy control method for a speed regulation system of a large hydropower station provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0066] In order to facilitate the understanding of the present invention, the present invention is described in more detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that when an element is described as "connecting" another element, it can be directly on another element, or there can be one or more centered elements therebetween. The orientation or positional relationship indicated by the terms "upper", "lower", "left", "right", "upper end", "lower end", "top" and "bottom" used in this specification is based on the orientation or positional relationship shown in the accompanying drawings, only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation of the present invention. In addition, the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0067] Unless otherwise defined, all technical and scientific terms used in this specification have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0068] Combine the following Figure 1 , the reliability redundancy control method 100 of the speed regulation system of a large hydropower station provided in the embodiment of the present application is described in detail through a specific embodiment.

[0069] Figure 1 The flowchart of the reliability redundancy control method of the speed regulation system of a large hydropower station provided by the present invention is as follows. The reliability redundancy control method of the speed regulation system of a large hydropower station provided by one embodiment of the present invention comprises:

[0070] Step 101, constructing a simulation environment for a redundant control strategy of a large hydropower station speed regulation system;

[0071] Step 102, generating an input signal of the redundant control strategy of the speed control system according to the speed, power and guide vane opening output by the simulation environment, and the desired control instruction, and inputting the input signal into the redundant control strategy of the speed control system to obtain an output instruction;

[0072] Step 103: input the output instruction obtained from the redundant control strategy of the speed regulation system to the control limit unit, obtain the output result of the control limit unit, and input the output result to the simulation environment;

[0073] Step 104, constructing a performance evaluation unit, wherein the performance evaluation unit feeds back the current moment evaluation result of the control performance of the simulated environment to the redundant control strategy of the speed regulation system, and optimizes the configuration of the redundant control strategy of the speed regulation system by optimizing the evaluation result of the control performance;

[0074] Step 105: optimizing the configuration of the redundant control strategy of the speed regulation system to obtain a final redundant control strategy of the speed regulation system, and controlling the speed regulation system of the hydropower station based on the redundant control strategy of the speed regulation system.

[0075] It should be noted that the method first involves building a simulation environment for the redundant control strategy of a large hydropower station speed control system. This simulation environment is virtual and is used to simulate the actual operation of the hydropower station speed control system, including key parameters such as speed, power and guide vane opening. The speed control system is a system in a hydropower station used to control the speed and direction of water flowing through the guide vanes to adjust the speed and output power of the generator set.

[0076] Specifically, the construction of the simulation environment involves introducing random uncertainty factors into the coefficients of the hydro-generator model, which include the control signals of speed, power and guide vane opening, as well as the corresponding control coefficients. These control coefficients are calculated using specific formulas and involve parameters such as head, flow, guide vane opening and guide vane angle. The setting of these parameters needs to be determined based on the actual hydropower station operation data and historical records to ensure that the simulation environment can accurately reflect the actual operation.

[0077] Preferably, the construction of the simulation environment can also include consideration of the multi-parameter dynamic characteristics of the hydro-generator model. This means that the model should not only consider static parameters, but also dynamic characteristics that change over time. For example, the state of the turbine at the next moment can be predicted by integrating the changes in the guide vane opening and guide vane angle. Such a dynamic model can more accurately simulate the actual operation of the turbine and provide a more accurate reference for the control of the speed control system.

[0078] Furthermore, the simulation environment can also increase the diversity of samples by introducing random noise. The initial size of the random noise and the rate at which it gradually decreases during the optimization process can be adjusted according to actual needs.

[0079] In some embodiments, the simulation environment introduces random uncertainty influencing factors into the model coefficients of the hydro-generator set model;

[0080] Correspondingly, the model coefficients include control signals of speed, power and guide vane opening, and the control signals of speed, power and guide vane opening include corresponding control coefficients respectively;

[0081] The control coefficient is calculated by the following formula:

[0082] C=k1·H+k2·u+k3·α+k4·β

[0083] Among them, C is the control coefficient, k1, k2, k3, k4 are coefficients, H is the water head, u is the flow rate, α is the guide vane opening, and β is the guide vane angle.

[0084] It should be noted that this implementation involves introducing random uncertainty influencing factors into the model coefficients of the hydro-generator set model. This means that when simulating the behavior of the hydro-generator set, we not only consider deterministic physical parameters, but also random factors to more realistically reflect the uncertainty in actual operation. Random uncertainty influencing factors refer to those unpredictable factors that may affect the performance of the hydro-generator set, such as environmental changes, equipment aging, etc.

[0085] Specifically, the model coefficients include control signals of speed, power and guide vane opening, and these control signals include corresponding control coefficients. The control coefficients are calculated by a specific formula, which involves parameters such as head, flow, guide vane opening and guide vane angle. The setting of these parameters needs to be determined based on the actual hydropower station operation data and historical records. For example, head refers to the energy of water when it falls from a certain height, flow refers to the volume of water passing through the guide vane per unit time, guide vane opening is the degree to which the guide vane is opened, and the guide vane angle is the angle of the guide vane relative to the water flow.

[0086] Preferably, the calculation formula of the control coefficient can be further refined. For example, the coefficients α1, α2, α3, and α4 in the formula can be optimized according to the design parameters and historical operation data of the turbine to improve the prediction accuracy of the model.

[0087] Furthermore, more random uncertainty factors can be introduced, such as temperature changes, equipment maintenance status, etc., to enhance the robustness of the model. For the calculation of guide vane opening and guide vane angle, more accurate sensors and advanced measurement technologies can be used to ensure the accuracy of the control signal. These improvements can make the simulation environment closer to the actual working conditions, thereby improving the effectiveness and reliability of the speed control system control strategy.

[0088] In some embodiments, the hydro-generator set model is a multi-parameter dynamic model; accordingly, the simulation environment for constructing a redundant control strategy for a speed regulation system includes:

[0089] The simulation environment is constructed according to the following expression:

[0090] P=ρgQH

[0091] Among them, P represents power, ρ is the density of water, g is the acceleration due to gravity, Q is the flow rate, and H is the water head.

[0092] It should be noted that this embodiment describes the construction process of a multi-parameter dynamic model for the hydro-generator set model. A multi-parameter dynamic model refers to a mathematical model that can describe the interaction and influence between multiple parameters of a system that change over time. In a hydropower station speed control system, such a model can more accurately simulate and predict the dynamic response of a turbine, which is crucial for achieving precise control.

[0093] Specifically, building a simulation environment for the redundant control strategy of the speed control system involves building a simulation environment based on a specific expression. In this expression, power (P) is the main output parameter, which is related to the water density (ρ), gravity acceleration (g), flow rate (Q) and head (H). These parameters together determine the output power of the turbine. For example, the water density and gravity acceleration are constants, while the flow rate and head are variables that can be adjusted according to the operating conditions of the hydropower station. In practical applications, the settings of these parameters need to be determined based on the design parameters of the turbine and real-time hydrological data.

[0094] Preferably, the construction of the simulation environment can be further refined. For example, more environmental factors can be introduced, such as water temperature, air pressure, etc., which may also affect the efficiency and output of the turbine.

[0095] Furthermore, advanced numerical integration methods can be used to more accurately simulate the dynamic response of the turbine, such as the Runge-Kutta method. Machine learning techniques, such as neural networks, can also be used to optimize model parameters to better adapt them to actual operating data. These methods can improve the accuracy and robustness of the simulation environment, thereby providing a more reliable control strategy for the speed control system.

[0096] In some embodiments, the output result of the control limiting unit includes limiting the action amplitude and the change rate of the output instruction, and the limiting result obtained;

[0097] The output instructions include a theoretical guide vane opening, a theoretical guide vane angle, and a theoretical guide vane deflection angle;

[0098] The limiting result is calculated according to the following formula:

[0099] For action amplitude limiting:

[0100] Δα max =max(Δα)

[0101] For rate of change limiting:

[0102]

[0103] Among them, Δα represents the change in the guide vane opening, Indicates the rate of change of guide vane opening.

[0104] It should be noted that this implementation involves the output result of the control limiting unit, which includes the action amplitude limiting and change rate limiting of the output instruction, and the resulting limiting result. The control limiting unit is a key component in the speed regulation system. Its function is to limit the amplitude and change rate of the control instruction to ensure the safe and stable operation of the system. The action amplitude limiting refers to the limitation of the maximum and minimum values ​​of the control instruction, while the change rate limiting refers to the limitation of the change speed of the control instruction.

[0105] Specifically, the output instructions include theoretical guide vane opening, theoretical guide vane angle, and theoretical guide vane deflection angle. These instructions are calculated based on the operating status and control objectives of the turbine generator set. The calculation of the limiting results involves specific formulas, among which the guide vane opening change and the guide vane opening change rate are key parameters. The guide vane opening change refers to the difference between the actual change and the theoretical change of the guide vane opening, while the guide vane opening change rate refers to the speed at which this change occurs. The setting of these parameters needs to be determined based on the design parameters and operating conditions of the turbine to ensure that the control instructions are within a safe and effective range.

[0106] Preferably, the calculation of the limiting result can be further refined. For example, a more complex mathematical model can be used to more accurately describe the changes in the guide vane opening and angle, such as a nonlinear model or a dynamic system model.

[0107] Furthermore, an adaptive control strategy can be introduced to dynamically adjust the limit parameters according to the real-time feedback of the system to improve the response speed and stability of the system. It is also possible to consider introducing intelligent algorithms, such as fuzzy logic or neural networks, to handle complex limit logic. These methods can improve the intelligence level of the limit unit and make it better adapt to different operating conditions and environmental changes.

[0108] In some embodiments, the construction performance evaluation unit includes: the current moment evaluation is represented by the following linear form:

[0109]

[0110] Among them, e t is the current moment evaluation at time t, k1 and k2 are evaluation coefficients, s t is the system status, u t To control the amount, is the expected system state.

[0111] It should be noted that this implementation involves the process of constructing a performance evaluation unit, which evaluates the control performance at the current moment through a linear representation. The performance evaluation unit is a key component in the speed control system, which is responsible for evaluating the control performance of the system and providing feedback to optimize the control strategy. Linear representation refers to the use of linear equations to describe the relationship between the system state and the control quantity.

[0112] Specifically, the evaluation process of the performance evaluation unit includes using a specific linear equation to calculate the evaluation value at the current moment. In this equation, the system state (X) and the control quantity (U) are key variables, and the desired system state (X*) is the control target. The evaluation coefficients (α1 and α2) are parameters used to adjust the evaluation results, and they can be set according to the specific requirements and historical data of the system. The setting of these parameters needs to take into account the dynamic characteristics and control objectives of the system to ensure that the evaluation results can accurately reflect the actual performance of the system.

[0113] Preferably, the evaluation process of the performance evaluation unit can be further refined. For example, more system state variables and control quantity variables can be introduced to more comprehensively describe the operating state of the system. In addition, the evaluation coefficient can be dynamically adjusted using an adaptive algorithm to adapt to changes in system performance and the influence of the external environment.

[0114] Furthermore, it is also possible to consider using machine learning methods, such as regression analysis or reinforcement learning, to optimize the setting of the evaluation coefficients. These methods can improve the accuracy and adaptability of the evaluation unit, so that it can better serve the optimization of the control strategy.

[0115] In some embodiments, the large hydropower station speed regulation system reliability redundancy control method further includes:

[0116] The current evaluation score output by the performance evaluation unit can be changed by adjusting the evaluation coefficients k1 and k2 to ensure that the system state and control quantity meet the requirements;

[0117] When the preset number of control targets are completed consecutively, additional task completion rewards will be given to guide priority completion of tasks.

[0118] It should be noted that this implementation involves a mechanism for adjusting the evaluation coefficient and providing a reward for completing the task. This mechanism is intended to change the current evaluation score output by the performance evaluation unit by adjusting the evaluation coefficients α1 and α2 to ensure that both the system state and the control amount meet the requirements. The reward for completing the task is an incentive measure used to guide the system to prioritize completing the task.

[0119] Specifically, adjusting the evaluation coefficients α1 and α2 means that the values ​​of these two coefficients can be dynamically adjusted according to the actual operation of the system and the performance target. In this way, the performance evaluation unit can be more flexible and can adapt to different operating conditions and control requirements. The task completion reward is a positive incentive. When the system completes the control target for a preset number of times in a row, an additional reward will be given to encourage the system to continue to maintain good performance.

[0120] More specifically, the specific settings of these parameters need to be determined based on the system's historical performance data and expected goals to ensure that the evaluation results can accurately reflect the system's actual performance and motivate the system to achieve higher performance levels.

[0121] Preferably, the mechanism for adjusting the evaluation coefficient and providing rewards for completing tasks can be further refined. For example, an adaptive adjustment algorithm can be designed to automatically adjust the evaluation coefficient according to the real-time feedback of the system to achieve more precise performance control.

[0122] Furthermore, the reward mechanism for completing tasks can be designed to be multi-layered, not just simple rewards, but also include rewards of different levels to accommodate tasks of different difficulty and different performance. Intelligent algorithms, such as reinforcement learning, can also be introduced to optimize the reward mechanism to make it more intelligent and efficient. These refinements and alternatives can make the system's performance evaluation and incentive mechanisms more flexible and effective, thereby improving the control performance and reliability of the entire speed control system.

[0123] In some embodiments, the configuration optimization of the redundant control strategy of the speed regulation system includes:

[0124] Determine evaluation parameters and control parameters;

[0125] Initialize the simulation environment, initialize random noise, and then loop through data sampling and parameter optimization;

[0126] Multiple independent parameter optimization processes are carried out to optimize the configuration of the redundant control strategy of the speed regulation system.

[0127] It should be noted that this implementation involves the process of configuring and optimizing the redundant control strategy of the speed control system. Configuration optimization refers to improving the performance and reliability of the speed control system by determining evaluation parameters and control parameters and performing multiple independent parameter optimization processes. Evaluation parameters and control parameters are key variables in the speed control system for evaluating and adjusting system behavior.

[0128] Specifically, determining the evaluation parameters and control parameters includes selecting appropriate parameter values ​​to optimize system performance. The evaluation parameters may include indicators such as the system's response time, stability, and efficiency, while the control parameters may involve adjustment variables such as guide vane opening and guide vane angle.

[0129] More specifically, the initialization of the simulation environment and random noise is to simulate the uncertainty in actual operation, where random noise can increase sample diversity and help the system better adapt to different operating conditions. The setting of these parameters should be based on the system's historical data and design requirements to ensure the effectiveness of the optimization process.

[0130] Preferably, the configuration optimization process can be further refined. For example, genetic algorithms, particle swarm optimization or other evolutionary algorithms can be used to perform parameter optimization, which can find the optimal solution in a complex parameter space.

[0131] Furthermore, an adaptive learning rate and discount factor can be set to dynamically adjust the speed and amplitude of parameter updates in the gradient descent optimization algorithm, thereby improving the optimization efficiency. A multi-objective optimization strategy can also be introduced to consider multiple evaluation parameters at the same time to achieve comprehensive optimization of system performance. These refinements and alternatives can make the configuration optimization process of the speed control system more accurate and efficient.

[0132] In some embodiments, the performing of multiple independent parameter optimization processes to optimize the configuration of the redundant control strategy of the speed regulation system includes:

[0133] Each optimization step is as follows:

[0134] a) Extract the initialization input state s1;

[0135] b) If the current time t is less than or equal to the maximum optimization duration T, then execute step c), otherwise end this optimization and proceed to the next optimization;

[0136] c) Based on the initial input state s1, the control parameters are used to generate the action a at time t t =μ(s t |)+N t , N t is the random noise at time t used to increase the diversity of samples, denoted by N init is the noise size at the initial moment, and N t As the optimization process gradually becomes smaller, it satisfies: N t =0.995×N t-1 ;

[0137] The obtained action command is limited and applied to the simulation environment, and the current moment evaluation e is obtained according to the model characteristic integration. t and the state s at the next moment t+1 , and e t If the control target is met and the preset number of times is completed continuously, let e t =e goal =100;

[0138] e) The data pair (s t ,a t ,e t ,s t+1 ) is stored in the data resource pool. If the data resource pool is full, R If there are data pairs, execute step f) to start parameter optimization, otherwise jump back to step c);

[0139] f) Randomly sample N data pairs from the data resource pool;

[0140] g) Let θ calculate the evaluation parameter loss as follows:

[0141]

[0142] h) Use the gradient descent optimization algorithm to update the evaluation parameter θ. The optimization goal is to minimize the loss L. The learning rate of the gradient descent is λ, γ is the discount factor of the optimization algorithm, and N is the total number of data samples;

[0143] i) Calculate the gradient of the sampled data pair as follows:

[0144]

[0145] j) Update the control parameter θ using the gradient descent optimization algorithm based on the calculated objective function gradient;

[0146] k) Use soft update to update the target network parameters, that is:

[0147] θ target =τθ+(1-τ)θ target,old

[0148] Among them, τ is the soft update coefficient.

[0149] It should be noted that this implementation method describes in detail the specific steps of carrying out multiple independent parameter optimization processes to optimize the configuration of the redundant control strategy of the speed regulation system. This process includes extracting the initialization input state, judging the optimization duration, generating actions, integral evaluation, data storage and parameter optimization. The initialization input state refers to the initial conditions of the system set at the beginning of the optimization process, and the maximum optimization duration refers to the maximum time allowed for the optimization process to run.

[0150] Specifically, each optimization step includes the following steps: first extract the initial input state, which involves setting the initial conditions of the system, such as the guide vane opening and water head; then determine whether the current moment is less than or equal to the maximum optimization time. If so, continue to the next step, otherwise end this optimization and start the next one; then use the control parameters to generate the action at the current moment according to the initial input state, and add random noise to increase sample diversity; then get the current moment evaluation and the state at the next moment according to the model characteristic integral; if the control target is completed for a preset number of times in a row, set a specific reward value; store the data pairs in the data resource pool, if the data resource pool is full of a certain number of data pairs, start parameter optimization, otherwise return to step c to continue data collection; randomly sample a certain number of data pairs from the data resource pool; calculate the evaluation parameter loss, and use the gradient descent optimization algorithm to update the evaluation parameters; calculate the gradient of the sampled data pairs, and use the gradient descent optimization algorithm to update the control parameters; finally, use the soft update method to update the target network parameters. The specific settings of these parameters need to be determined according to the actual operation of the system and the optimization goals.

[0151] Preferably, the optimization process can be further refined or alternatives can be provided. For example, different random noise generation strategies, such as Gaussian noise or uniform noise, can be used to simulate different uncertainty conditions; the learning rate and discount factor in the gradient descent optimization algorithm can be adjusted to adapt to different optimization speed and stability requirements; different parameter optimization algorithms, such as Adam or RMSprop, can be used to improve optimization efficiency; regularization techniques, such as L1 or L2 regularization, can also be introduced to prevent model overfitting. These refinements and alternatives can make the parameter optimization process of the speed control system more flexible and effective.

[0152] In some embodiments, the flow-related component is obtained by the following formula:

[0153]

[0154] Among them, Q represents the flow rate, A is the flow area, g is the gravitational acceleration, and H is the water head.

[0155] It should be noted that this embodiment describes a method for calculating flow-related components. Flow-related components refer to parameters directly related to flow in the speed control system of a hydropower station, which are crucial for controlling the operation of a hydro-generator set. Flow is the volume flow of water passing through the guide blades, which directly affects the output power of the turbine.

[0156] Specifically, the calculation of flow-related components involves the formula Q = A g H, where Q represents the flow, A is the flow area, g is the gravitational acceleration, and H is the head. The flow area refers to the cross-sectional area of ​​the water when it passes through the guide vane, the gravitational acceleration is a constant near the surface of the earth, and the head refers to the energy of water when it falls from a certain height. The specific settings of these parameters need to be determined according to the design parameters of the turbine and the actual operating conditions to ensure the accuracy of the calculation results.

[0157] Preferably, the calculation of the flow-related components can be further refined or provided with alternatives. For example, more accurate measurement techniques can be used to determine the flow area, such as using laser scanning or ultrasonic measurement techniques. In addition, the dynamic characteristics of the water flow can be taken into account, such as uneven velocity distribution or turbulence, which may affect the calculation of the flow.

[0158] Furthermore, real-time hydrological data, such as rainfall and upstream reservoir water levels, can be introduced to help more accurately predict flow changes. These refinements and alternatives can make the calculation of flow-related components more precise, thereby improving the control accuracy and response speed of the speed regulation system.

[0159] In some embodiments, the guide vane opening α and the guide vane angle β are calculated according to the following formulas:

[0160]

[0161] Among them, the value ranges of α and β are [-π,π] respectively, and arctan and arcsin are inverse trigonometric functions with a range defined between [-π,π].

[0162] It should be noted that this embodiment involves a method for calculating the guide vane opening and guide vane angle. The guide vane is a component used to control the direction and flow of water flow in a turbine, and the guide vane opening and guide vane angle are key parameters for regulating water flow, which directly affect the working efficiency and output power of the turbine.

[0163] Specifically, the calculation of the guide vane opening and guide vane angle involves the use of inverse trigonometric functions. The guide vane opening is the degree to which the guide vane is opened, while the guide vane angle is the angle of the guide vane relative to the water flow. The calculation formulas for these parameters are and The value ranges of θ and φ are [-90°, 90°], arctan and arcsin are inverse trigonometric functions defined between [-1, 1]. The specific settings of these parameters need to be determined according to the design parameters and actual operating conditions of the turbine to ensure the accuracy of the calculation results and the optimal operating state of the turbine.

[0164] Preferably, the calculation of the guide vane opening and guide vane angle can be further refined or alternative solutions can be provided. For example, a higher precision sensor can be used to monitor the actual opening and angle of the guide vane in real time to improve the real-time and accuracy of the calculation. In addition, machine learning algorithms can be introduced to predict the optimal opening and angle of the guide vane. These algorithms can optimize the adjustment strategy of the guide vane based on historical operating data and current hydrological conditions.

[0165] Furthermore, the dynamic characteristics of the guide vanes can also be considered, such as the inertia and response time of the guide vanes, which may affect the real-time and accuracy of the guide vane adjustment. These refinements and alternatives can make the calculation of the guide vane opening and angle more accurate, thereby improving the operating efficiency and response speed of the turbine.

[0166] The above-mentioned various embodiments of the present invention have the following beneficial effects: the reliability redundancy control method of the speed control system of a large hydropower station described in the present invention can improve the reliability and control accuracy of the speed control system of the hydropower station. By constructing a simulation environment and a performance evaluation unit, the method can introduce random uncertainty influencing factors in the simulation environment, simulate various uncertainties in actual operation, and make the speed control system control strategy more adaptable to actual working conditions. At the same time, by controlling the limiter unit to limit the action amplitude and change rate of the output instruction, it can ensure that the guide vane opening and angle are adjusted within a safe and effective range, avoid damage to the hydro-turbine generator set, and enhance the stability and response speed of the system.

[0167] In addition, this method can continuously iterate and improve the control strategy through multiple independent parameter optimization processes to achieve higher control accuracy and better system performance. By adjusting the evaluation coefficient and providing rewards for completing tasks, the system can be guided to prioritize completing tasks and further improve control efficiency. This method can also update the target network parameters through soft updates, reduce system performance fluctuations caused by parameter updates, and ensure that the hydropower station speed control system can maintain the optimal operating state under various conditions.

[0168] Furthermore, the storage medium of the embodiment of the present application stores program instructions that can implement all the above methods, wherein the program instructions can be stored in the above storage medium in the form of a software product, including several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) or a processor to execute all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, or terminal devices such as a computer, a server, a mobile phone, and a tablet.

[0169] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them. Under the concept of the present invention, the technical features in the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other changes in different aspects of the present invention as above, which are not provided in detail for the sake of simplicity. Although the present invention is described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A reliability redundancy control method for a large hydropower station speed control system, characterized in that: include: Construct a simulation environment for redundant control strategies of speed regulation systems in large hydropower stations; According to the speed, power and guide vane opening output by the simulation environment, and the expected control instruction, an input signal of the redundant control strategy of the speed regulation system is generated, and the input signal is input into the redundant control strategy of the speed regulation system to obtain an output instruction; Inputting the output instruction obtained from the redundant control strategy of the speed regulation system into the control limit unit, obtaining the output result of the control limit unit, and inputting the output result into the simulation environment; Constructing a performance evaluation unit, wherein the performance evaluation unit feeds back the current moment evaluation result of the control performance of the simulated environment to the redundant control strategy of the speed regulation system, and optimizes the configuration of the redundant control strategy of the speed regulation system by optimizing the evaluation result of the control performance; The redundant control strategy of the speed regulation system is configured and optimized to obtain a final redundant control strategy of the speed regulation system, and the speed regulation system of the hydropower station is controlled based on the redundant control strategy of the speed regulation system.

2. The reliability redundancy control method for a large hydropower station speed regulation system according to claim 1 is characterized in that: The simulation environment introduces random uncertainty influencing factors into the model coefficients of the hydro-generator set model; The model coefficients include control signals of speed, power and guide vane opening, and the control signals of speed, power and guide vane opening include corresponding control coefficients respectively, which are calculated as follows: C=k1·H+k2·u+k3·α+k4·β Among them, C is the control coefficient, k1, k2, k3, k4 are coefficients, H is the water head, u is the flow rate, α is the guide vane opening, and β is the guide vane angle.

3. The reliability redundancy control method for a large hydropower station speed regulation system according to claim 2 is characterized in that: The hydro-generator set model is a multi-parameter dynamic model; the simulation environment for constructing a redundant control strategy for a speed regulation system includes: The simulation environment is constructed according to the following formula: P=ρgQH Among them, P is power, ρ is the density of water, g is the acceleration of gravity, Q is the flow rate, and H is the water head.

4. The reliability redundancy control method for a large hydropower station speed regulation system according to claim 1 is characterized in that: The output result of the control limiting unit includes limiting the action amplitude and the change rate of the output instruction to obtain a limiting result; The output instructions include a theoretical guide vane opening, a theoretical guide vane angle, and a theoretical guide vane deflection angle; The limiting result is calculated according to the following formula: For action amplitude limiting: Da max =max(D) For rate of change limiting: Among them, Δα represents the change in the guide vane opening, Indicates the rate of change of guide vane opening.

5. The reliability redundancy control method for a large hydropower station speed regulation system according to claim 1 is characterized in that: The construction performance evaluation unit is shown in the following formula: Among them, e t is the current moment evaluation at time t, k1 and k2 are evaluation coefficients, s t is the system status, u t To control the amount, is the expected system state.

6. The reliability redundancy control method for a large hydropower station speed regulation system according to claim 5 is characterized in that: The reliability redundancy control method of the speed regulation system of a large hydropower station also includes: By adjusting the evaluation coefficients k1 and k2, the current evaluation score output by the performance evaluation unit is changed to ensure that the system state and control quantity meet the requirements; When the preset number of control targets are completed consecutively, additional task completion rewards will be given to guide priority completion of tasks.

7. The reliability redundancy control method for a large hydropower station speed regulation system according to claim 1 is characterized in that: The configuration optimization of the redundant control strategy of the speed regulation system includes: Determine evaluation parameters and control parameters; Initialize the simulation environment, initialize random noise, and then loop through data sampling and parameter optimization; Multiple independent parameter optimization processes are carried out to optimize the configuration of the redundant control strategy of the speed regulation system.

8. The reliability redundancy control method for a large hydropower station speed regulation system according to claim 7 is characterized in that: The carrying out of multiple independent parameter optimization processes to optimize the configuration of the redundant control strategy of the speed regulation system includes: Each optimization step is as follows: a) Extract the initialization input state s1; b) If the current time t is less than or equal to the maximum optimization duration T, then execute step c), otherwise end this optimization and proceed to the next optimization; c) Based on the initial input state s1, the control parameters are used to generate the action a at time t t =μ(s t |)+N t , N t is the random noise at time t used to increase the diversity of samples, denoted by N init is the noise size at the initial moment, and N t As the optimization process gradually becomes smaller, it satisfies: N t =0.995×N t-1 ; For the action instruction a t The obtained action command is limited and applied to the simulation loop. The current moment evaluation e is obtained according to the model characteristic integral. t and the state s at the next moment t+1 , and e t If the control target is met and the preset number of times is completed continuously, let e t =e goal =100; e) The data pair (s t ,a t ,e t ,s t+1 ) is stored in the data resource pool. If the data resource pool is full, R If there are data pairs, execute step f) to start parameter optimization, otherwise jump back to step c); f) Randomly sample N data pairs from the data resource pool; g) Let θ calculate the evaluation parameter loss as follows: h) Use the gradient descent optimization algorithm to update the evaluation parameter θ. The optimization goal is to minimize the loss L. The learning rate of the gradient descent is λ, γ is the discount factor of the optimization algorithm, and N is the total number of data samples; i) Calculate the gradient of the sampled data pair as follows: j) Update the control parameter θ using the gradient descent optimization algorithm based on the calculated objective function gradient; k) Use soft update to update the target network parameters, that is: i target =τθ+(1-τ)θ target,old Among them, τ is the soft update coefficient.

9. The reliability redundancy control method for a large hydropower station speed regulation system according to claim 2 is characterized in that: The relevant components of the flow are obtained by the following formula: Among them, Q represents the flow rate, A is the flow area, g is the gravitational acceleration, and H is the water head.

10. The reliability redundancy control method for a large hydropower station speed regulation system according to claim 9, characterized in that: The guide vane opening α and guide vane angle β are calculated according to the following formulas: Among them, the value ranges of α and β are [-π,π] respectively, and arctan and arcsin are inverse trigonometric functions with a range defined between [-π,π].

Images